Predicting properties of underground formation

FIELD: physics.

SUBSTANCE: method comprises steps for: obtaining seismic data for an area of interest; obtaining an initial seismic cube using said seismic data, wherein the initial seismic cube is a three-dimensional representation of the seismic data; generating a plurality of shifted seismic cubes within the area of interest using said seismic data and a shifting parameter, wherein each of the plurality of shifted seismic cubes is shifted from the initial seismic cube; and wherein the shifting parameter defines a direction and a range that the initial seismic cube should be shifted; generating a neural network using the initial seismic cube, the plurality of shifted seismic cubes, and well log data; and applying the neural network to said seismic data to obtain a model for the area of interest, the model being configured for use in adjusting an operation of the wellsite.

EFFECT: high accuracy.

20 cl, 19 dwg

 

CROSS-REFERENCE TO RELATED APPLICATION

This application claims the priority filing date of the provisional application U.S. No. 60/986249 "System and method for predicting properties of subsurface formations, registered on 7 November 2007

The technical FIELD

The present invention relates to forecasting properties of underground formations.

PRIOR art

Work, such as surveying, drilling, testing on wireline, completions, production, planning and production analysis, usually performed to identify and select valuable downhole fluid. Surveys are usually performed with the use of such data collection methodology, as seismic scanning for mapping underground formations. These formations are often examined to determine the presence of underground reserves, such as reserves of valuable fluid or minerals, or to determine whether formation characteristics suitable for storing fluid.

During the operations of drilling and production usually collect data for analysis and/or monitoring operations. Such data may include, for example, information relating to subterranean formations, equipment, and statistical and/or other data.

Data relating to a subterranean formation, collected from various sources. The same is these formations can be static or dynamic. Static data include, for example, to the structure formation and geological stratigraphy, forming the geological structure of the subsurface formation. Dynamic data includes, for example, for fluid media passing through the geological structure of the subsurface formations over time. Such a static and/or dynamic data can be collected for further information on the formation and stocks contained in them.

The INVENTION

In one example implementation, the forecasting properties of subsurface formations forecasting system properties of subsurface formations drilling site includes a data processing module, configured to receive seismic data of the region of interest. The system further includes the modelling unit, configured to obtain the initial seismic cube using seismic data and obtaining a number of shifted seismic cubes using seismic data, where each of the number of shifted seismic cubes is shifted from the initial seismic cube. The system further includes a module configured to create a neural network using the initial seismic cube, the number of shifted seismic cubes and data logging chart wells is. A training module is additionally configured to apply neural networks to seismic data to obtain a model of the region of interest, where the model can be used to adjust the operation of the drilling platform.

Other aspects and advantages of the forecasting properties of subsurface formations should become clear from the following description of the accompanying claims.

BRIEF DESCRIPTION of DRAWINGS

So the above characteristics and advantages of the forecasting properties of subsurface formations can be understood in detail, the following description of typical embodiments and not limiting the scope of protection, forecasting properties of subsurface formations, briefly described above with reference to embodiments of shown in the accompanying drawings, in which:

figa-1D depict examples of oil from underground formations, including collectors, and various operations in the field;

figa depicts the execution of geophysical research station seismic studies on the truck;

figv depicts drilling operations drilling tool suspended from a rig and spokeman in the underground formation;

figs depicts the operation on wireline running tool for logging the cable, suspended from the drilling rig and lowered into the wellbore;

fig.1D depicts the operation of extraction performed by the operating tool, deployed operational setup and completed well bore to move the fluid from the reservoir into structures on the surface;

figa-2D depict a graphical representation of data collected by instruments figa-1D, respectively;

figa depicts the track seismograms underground formations on figa;

figv represents the test result of the sample of core figv;

figs depicts a wireline logs of the borehole of an underground formation figs;

fig.2D curve depicts the fall of the production level of the fluid passing through the underground formation fig.1D;

figure 3 depicts a schematic view, partially in section, of the oil with many data collection instruments that are installed on different locations of the operations in the field for data collection underground formations;

figa-4C depict three-dimensional (3D) static models based on data collected by the data collection instruments 3;

figure 5 depicts a graph of the probability distribution of the static models figure 4;

6 depicts an example schematic diagram of the system of forecasting properties of subsurface formations for the operation of the drilling site;

Fig.7-8 image is tons of examples of block diagrams of the sequence of operations of the methods for predicting properties of subsurface formations for the operation of the drilling site;

Fig.9 depicts an example representation of groups of shifted seismic cubes;

figure 10 depicts an example representation of seismic data;

11 depicts an example of a view model.

A DETAILED DESCRIPTION of the PREFERRED EMBODIMENT VARIANTS of the INVENTION

Specific embodiments of the forecasting properties of subsurface formations are described in detail below with reference to the accompanying drawings.

In the following detailed description of embodiments of the forecasting properties of subsurface formations set forth numerous specific details to provide a more comprehensive understanding of the forecasting properties of subsurface formations. In other instances, well-known signs are not described in detail to avoid shading concepts of forecasting properties of subsurface formations.

On figa-1D schematically illustrates simplified to represent the views of the oilfield (100) with an underground formation (102)containing a reservoir (104), and various operations performed in the field, which includes at least one drilling pad. On figa shown geophysical research, performed by a station (106a) of seismic surveys on the truck, measure the properties of underground formations. Studies are seismic geophysical research production of acoustic vibrations is rd. On figa such acoustic oscillations (112)generated by the source (110), recorded from a number of horizons (114) in the strata of rock formations (116). Acoustic oscillations (112) are sensors, such as geophones (118)located on the surface. Geophones (118) generate electrical output signals, referred to as received data (120) on figa.

In response to received acoustic oscillations (112), representing different parameters (such as amplitude and/or frequency) acoustic oscillations (112), the geophones (118) generate electrical output signals containing data on subsurface formations. Received data (120) are passed as input data to the computer (122a) station (106a) of seismic surveys on the truck. In response to the input data computer (122a) creates an output record (124) seismic data. Seismic data can be saved, transferred, or further processed, as necessary, for example, by compressing the data. For example, you can apply a digital filtering of the signal to the output seismic data (i.e. paths gathers) for noise reduction and the creation and application of correction parameters for summarizing single tracks of seismic signals having a common geographic location, to improve the signal-to-noise ratio. the optional, you can use special algorithms to display seismic reflections in three-dimensional space. In this example, the result data may be a seismic cube, representing the underground geological layers as accurately as possible, and which can be used by geologists to interpret the main geological structures. In addition, the seismic cube can be used in combination with the data from the logs of wells for analysis of geological structures.

On FIGU shows the drilling operation of the drill tool (106b), suspended from the drilling rig (128) and spokeman in subterranean formations (102) to form the trunk (136) wells. Capacity (130) of the drilling fluid used for supply of drilling mud in the drill tools on-line (132) filing to implement the circulation of drilling mud through the drill pipe, up the wellbore (136) of the borehole and back to the surface. Drilling mud is usually filtered and return in the capacity of the drilling fluid. The circulation system can be used to store, control and filtration of the drilling fluid. Drilling tools down into the underground formation to achieve the collector (104). Each well can be aimed at one or more collectors. Drilling tools are made with the ability and the intent of the downhole properties using one or more probes logging while drilling. Probes logging while drilling can also be adapted for selection of the sample (133) of the core, as shown, or to remove, in order to ensure the selection of the core sample using another tool.

Surface unit (134) control is used for communication with drill tool and operations off-site. Ground control unit is able to communicate with drill tools for sending commands to the drilling tools and for receiving data from them. Ground control unit can be provided with computer equipment for receiving, storing, processing and analyzing data from the field. Ground control unit collects data generated during the drilling operations, and generates output data (135), which you can save or send. Computer equipment, such as equipment ground control unit, can be installed on various platforms in the field and/or in remote locations.

Sensors (S), such as measuring devices, can be installed everywhere in the field to collect data related to the various operations described above. As shown, the sensor (S) installed in one or more locations on the drilling tools and/or on a rig to measure drilling parameters, such as axial load on bit, torque is a bit, pressure, temperature, cost, composition, speed of rotation of the rotor, and/or other parameters of the operations. The sensors can also be installed in one or more places in the circulation system.

Information collected by the sensors may collect ground control unit and/or other means of collecting data for analysis or processing. The information gathered by the sensors can be used individually or in combination with other data. Data can be collected in one or more databases and/or transfer of venue or beyond. All data or selected data fragments can be selectively used for analysis and/or forecasting operations on current and/or other well holes. Data can be statistical data, data in real time or their combinations. Data in real time can be used in real time or save for later use. Data can also be combined with statistical data or other input data for additional analysis. Data can be placed in separate databases or combine them into one database.

The collected data can be used to perform analysis, such as modeling operation. For example, seismic output can be used is to perform geological, geophysical engineering or design of the reservoir development. Data of the reservoir, wellbore, obtained on the surface, and/or data field treatment of production wells can be used to perform the simulation of the reservoir, wellbore, geological, geophysical or other imitations. Output operation in the field can be transmitted by the sensors directly or after some pre-processing or simulation. These outputs can be used as input for further analysis.

Data can be collected and saved in the surface unit (134) management. One or more ground control units can be placed in the field or to be remotely associated with it. Ground control unit may be a single unit or as an integrated network of blocks used to perform the necessary control functions data transfer everywhere on the field. Ground control unit may be a system with manual or automatic control. Ground control unit, the user can control and/or configure it.

Ground control unit can be equipped with a transceiver (137), which provides communication between the block surface and different parts of the field or other places about the eraci. Ground control unit may also be equipped or functionally connected to one or more controllers to actuate mechanisms in the field. Ground control unit can send signals to the control commands in the field in response to the received data. Ground control unit may receive commands via a transceiver or may himself execute commands on the controller. To analyze the data (locally or remote), decision and/or to actuate the controller can be equipped with a processor. Thus, in the field it is possible to perform selective adjustment based on the collected data. This technique can be used to optimize areas of operations, such as the regulation of drilling, the axial load on the bit, the feeding speed of drilling pumps or other parameters. This adjustment can be done automatically, on the basis of the computer program or manually by the operator. In some cases, the drilling plans can be adjusted for optimum conditions of operation or to prevent problems.

On figs shows the operation on wireline running tool (106c) on wireline suspended from the drilling rig (128) in the wellbore (136) well figv. Tool (106c) on wireline adapted on what I deployment in the wellbore (136) wells to perform logging, perform testing on the face and/or sampling. Tool on wireline can be used in other ways and as a device performing seismic geophysical research. Tool on wireline pigs may have a source (144) blast, radiation, electrical energy, or acoustic waves, transmitting electrical signals in subterranean formations (102) and fluids in them and/or receiving signals from them.

Tool on wireline may be functionally connected, for example, geophones (118) and a computer (122a) station (106a) of seismic surveys on the truck figa. Tool on wireline may also transfer the data in the surface unit (134) management. Ground control unit collects data generated during the operation on wireline, and produces output data (135), which may be stored or transported. Tool (106c) on wireline can be installed at various depths in the wellbore (136) wells to provide studies or other information relating to a subterranean formation.

Sensors (S), such as measuring devices, can be installed in the field to collect data relating to various operations as described above. As shown, the sensor (S) to ascertain what the tool on wireline for measuring downhole parameters, related to, for example, porosity, permeability, composition of the fluid and/or other parameters of the operation.

On fig.1D shows the operation of extraction performed by the operating tool (106d), deployed on the operating unit or wellhead (129) completed wellbore (136) well pigs to move the fluid from the collector to the facilities (142) on the surface. Fluid passes from the reservoir (104) through the perforation tunnels in the casing (not shown) in the operating tool (106d) in the wellbore (136) wells and facilities (142) on the surface via the network (146) consolidated pipelines.

Sensors (S), such as measuring devices, can be installed in the field to collect data related to the various operations in the field, as described above. As shown, the sensor (S) can be installed on the operating tool (106d) or associated equipment, such as wellhead fittings, network prefabricated piping, equipment and structures on the surface and/or operational equipment and facilities, for measuring parameters of a fluid medium, such as the composition of the fluid, flow rate, pressure, temperature, and/or other operational parameters.

Although shown only the simplified configuration of the drilling site, it should be clear that the field can cover the substance of the plot of land, marine and/or aquatic places, with the placement of one or more drilling sites. The operation may also include the operation of injection wells (not shown) for additional production. One or more prefabricated structures can be functionally connected to one or more drill sites for selective collection of the downhole fluid from the drilling site (sites).

Although figw-1D shows the tools used to measure properties of the oil, it should be clear that the tools can be used in relation to operations in the oil fields, such as operations at the mines, aquifers, reservoirs, or other underground structures. Also, although shows some of the data collection instruments, it should be clear that it is possible to use various measuring instruments capable of measuring properties such as the total mileage seismic waves, density, specific resistance, the rate of production etc., underground formation and/or geological structures. Various sensors (S) and/or monitoring tools for collecting and/or monitoring data can be placed at various locations in the wellbore. You can also create other data retrieval from off-site.

Configuration field (figa-1D) cu presents dim description example field, where applicable, the forecasting properties of underground formations. Part or all of the fields can be on land, water and/or sea. Also, although shown one field with measurements in one place, the forecasting properties of underground formations can be used with any combination of one or more fields, one or more structures field treatment of production wells and one or more drilling sites.

On figa-2D show examples of graphical representation of data collected by instruments figa-1D, respectively. On figa shows an example of tracks (202) seismograms underground formations figa performed station (106a) of seismic surveys on the truck. Track seismograms can be used to obtain data such as data of bilateral mileage waves over a period of time. On FIGU shows an example of the sample (133) core formation, selected drill tools (106b). The core can be used to obtain data, such as graph density, porosity, permeability and other physical properties of the core sample along its length. Tests of density and viscosity can be performed on fluids in the core and at varying pressures and temperatures. On figs shown wireline logs (204) wells in subterranean formations figs executed by the tool (s) on the Abele. Logging chart executed by the tool on the cable, usually gives a measurement of the resistivity of a formation or other measurements at different depths. On fig.2D shown curve (206) decline in the production fluid flowing from an underground formation fig.1D obtained by ground equipment (142). Curve falling production levels typically provides a tempo Q production as a function of time t.

The corresponding charts figa-2C show examples of static measurements, which can describe the physical characteristics of the formation and collectors contained in it, or to give information about them. Measurement data can be analyzed to better define the properties of formations (formations) and determine the accuracy of the measurements and/or to check for errors. Diagrams of each of the respective measurements can be combined and scaled for comparison and verification of properties.

On fig.2D shows an example of a dynamic measurement of the properties of the fluid in the wellbore. When fluid passes through the borehole, measured properties of the fluid, such as flow rates, pressure, composition, etc. As described below, the static and dynamic measurements can be analyzed and used to create models underground formation to determine its characteristics. Similar measurements can also use the SQL to measure changes in the characteristics of the formation in time.

Figure 3 shows a schematic view, partially in section, of the oilfield (300) with tools (302a), (302b), (302c) and (302d) data collection installed at various locations of operations in the field for data collection subterranean formation (304). Tools 302a-302d data collection may be similar instrument 106a-106d data collection figa-1D, respectively, or the other, not shown. As shown, the tools 302a-302d data collection form chart 308a-308d data or measurements, respectively. The above chart data compiled across the field to demonstrate data generated by various operations.

Chart 308a-308c data are examples of diagrams of static data that can generate tools 302a-302d data collection, respectively. Chart (308a) static data is the display time of bilateral mileage seismic waves and may be similar to the track (202) seismograms shown in figa. Chart (308b) static data is formed on the core data measured on the model of core formation (304), the same sample (133) core figv. Chart (308c) static data is logging track, similar logging chart (204) well figs. Curve decline or chart (308d) is a chart dynamic data flow rate of the fluid over time is similar to the chart (206) fig.2D. You can also collect other data, such as statistical data, user input, economic information, other data dimensions and other parameters of interest.

The subterranean formation (304) has a variety of geological structures 306a-306d. As shown, the formation has a layer (306a) Sandstone layer (306b) limestone layer (306c) shale layer (306d) sand. Line (307) fault passes through formations (306a), (306b). In one embodiment, the implementation tools collect statistical data made with the possibility of measuring and detecting characteristics of the formations.

Although the specific subterranean formation with specific geological structures, it should be clear that the field can contain a variety of geological structures and/or formation, in some cases with extraordinary complexity. In some places, usually below the path of flow, fluid can occupy the pore space of the formation. Each measuring device can be used to measure properties of the formation and/or its geological characteristics. Although each instrument of data collection are shown in a specific place for operations in the formation, it should be clear that the measurement of one or more types can be performed at one or more locations of operations on one or more fields or on the natives areas for comparison and/or analysis.

Data collected from various sources, such as data collection tools shown in figure 3, can then be processed and/or evaluated. Typically, seismic data displayed on the chart (308a) static data from the instrument (302a) of data collection that uses a geophysicist to determine characteristics and characteristics of the underground formation. Data core, shown in a static diagram (308b), and/or data logging chart (308c) wells typically uses a geologist to determine various characteristics of the subterranean formation (304). Data extraction from a chart (308d) typically uses a development engineer on the field to determine the characteristics of the flow rate of the fluid manifold. The data analyzed by a geologist, geophysicist and engineer for the development of the field can be analyzed using the techniques of modeling. Examples of the modelling methods described in US 5992519, WO 2004/049216, WO 1999/064896, US 6313837, US 2003/0216897, US 7248259, US 2005/0149307 and US 2006/0197759. System to perform such modeling techniques are described, for example, in issued US 7248259, the content of which is fully incorporated herein by reference.

On figa-4C shows a three-dimensional graphical representation of the geological environment, considered as a static model. The static model can be created based on one or more of the models created, the example according to the data collected using the tool 302a-d data collection. On the shown figures of the static model 402a-c created tools 302a-data collection figure 3, respectively. Data static model can represent a two-dimensional view of the subsurface formation based on the data collected on this site.

Static models can have different accuracy, based on the type of measurement, data quality, place of operations and other factors. Although the static model figa-4C obtained using specific data collection instruments in one area of the oil field, one or more similar or different data collection instruments can be used to perform measurements at one or more locations around the field to create different models. You can choose different methods of analysis and modeling depending on the required data type and/or location of operations.

Each of the static models 402a-c are shown as three-dimensional representation of the oil with one or more collectors and structures surrounding formation. Data volumetric representations are forecast geological structure of the subsurface formation in exactly the prescribed place on the basis of the available measurements. In one possible implementation representation are possible what canariae, created using the same input data (statistical and/or in real time), but with a different interpretation, interpolation and modeling techniques. As shown, the static models contain geological layers in the subsurface formation. In particular, fault (307) 3 passes through each of the models. Each static model also has points A, B and C bindings placed in prescribed positions on each of the static models. Data static model and set the anchor point of the static models can be analyzed. For example, comparison of various static models may show differences in the structure of the fault (307) and the adjacent layer (306a). Each of the anchor points can facilitate the comparison of various static models. You can perform adjustments to models based on the analysis of various static models figa-C, and adjusted the layer formation can be created, as described additionally below.

Figure 5 is a diagram of the probability distribution of numerous static models, such as model 402A-C figure 4. The chart shows the percentage probability of a particular variable model for each of the static models as a function of the variable (V) collector, such as volume settings, the rate of production or other parameters. Piramanayagam be appreciate any static or dynamic component (components), such as volume settings, the total thickness of the breed, the effective thickness of the formation, the rate of production, cumulative production, etc. In one possible implementation variables in the implementation of the simulation is retained within a reasonable opportunity forecast real manifold (headers) or that observed in similar reservoirs. This chart is a histogram that shows the implementation in numerous models that can be generated from the data. Variables the results you can create by changing the parameters in numerous models. The chart can then generate through review and assess the likelihood of the models and their application to the chart.

The histogram shows that the static model (402a) gives a ten percent probability of coincidence with the actual parameter collector. The histogram also indicates that the static model (402b) has a fifty percent probability of coincidence, and the static model (402c) is a ninety percent probability. This diagram assumes that the static model (402c) is a more conservative model estimates variable (V), but has a higher likelihood of correctness, and the static model (402a) gives less confidence and should be considered more optimistic assessment. Static m is Delhi and their likelihood can be used for example, in defining plans and mining model ground processing facilities. View 402a-402c static models can be selected on the basis of need and risk and/or economic assumptions.

Model figa-4C adjusted based on dynamic data, obtained by extraction of the chart (308d) figure 3. Dynamic data collected by the tool (302d) data collection applied in each of the static models figa-4C. As shown, the dynamic data indicate that the fault (307) and the layer (306a), the predicted static models may need to be adjusted. Layer (306a) adjusted in each model, as shown by the dotted lines. A modified layer shows the positions 306a', 306a and 306a"' for static models figa-4C, respectively.

Dynamic data may indicate that some static models give the best representation of the field. The static model to match the statistics of the rate of production can be considered a good indication that the model can give accurate forecasts of future production. In such cases, you can choose the preferred static model. In this case, although the static model on figs may have the highest overall probability of accuracy based only on the static model, as shown in figure 5, the dynamic analysis the standard model assumes that what model figv has the best match. As shown in figa-4C, the comparison layer (306a) with layers 306a', 306a and 306a"' indicates that the fault (307) with the corresponding throughput of fluid through the rift most closely coincides with the prediction given by the static model (402b).

In this example, the selected static model (402b) modified based on dynamic data. The resulting correction model (402b) adjusted to better fit with the data extraction. As shown, the position of the geological structure (306a) is shifted in the position of the layer 306a" the difference, shows the dynamic data. As a result, the static model can be adapted to best meet both static and dynamic models.

In determining the best, in General, models of the geological environment can be considered static and/or dynamic data. In this case, when considering both static and dynamic data, static model (402b) FIGU selected model of the geological environment with the highest probability of accuracy on the basis of both static probabilities and dynamic input data. To obtain the best overall model may be necessary consideration of static and dynamic data from multiple sources, locations and/or data types.

Evaluation of different the static and dynamic data figure 3 includes the consideration of static data such as seismic data (308a), considered by the geophysicist, geological data (308b, 308c), considered by the geologist, and the data (308d) production, considered by the engineer on the development of the field. Everyone usually looks at data relating to a particular function, and creates a model based on this particular feature. However, as shown in figa-4C, each of the individual models may influence the decision on selecting the best overall model of the geological environment. Moreover, information from other models or sources may also affect the adjustment of model and/or the selection of the best overall model of the geological environment. Model of the geological environment, created as described for figa-5, is the basic model of the geological environment, a specific analysis of the different models you have created.

Another source of information that can affect the model (model)is an economic information. For all transactions shown on figa-1D, there are numerous commercial considerations. For example, the equipment shown in each of these figures has a different value and/or risks associated with it. At least some of the data collected in the field, refer to commercial factors, such as price risk. Commercial data may include, for example, the cost of production, time is I drilling, payment of storage, the price of oil/gas, weather factors, political stability, tax burden, availability of equipment, geological environment, the accuracy and sensitivity of the measuring instruments, data representation, and other factors that affect the cost of operations, or potential costs relating thereto. You can make a decision and to develop strategic business plans taking into account the possible reduction of costs and risks. For example, the project for the development of the oil can be based on data from commercial considerations. This project is the development of the oil may, for example, to determine the location of drilling rigs, as well as depth, number of wells, duration of operation, the rate of production, types of equipment and other factors that should influence the costs and risks associated with the operation.

Figure 6 shows a schematic view of a system (600) predict the properties of underground formations for the operation of the drilling site. Shows a system (600) includes a ground unit (602) control, functionally coupled with the drilling system (604) drilling site, the server (606), functionally connected to the ground block (602) management, and tools (608) modeling, functionally connected to the servers (606). As shown, the channels (610) connection established between the drilling system (604) super rig is, ground block (602) management servers (606) and the tool (608) modeling. Different communication channels can be created to perform data transfer in the system. For example, channels (610) communication can provide continuous, intermittent, in one direction, in two directions and/or selective communication system (600). Channels (610) connection can be of any type, such as wired, wireless, etc.

A drilling system (604) drilling site and the ground unit (602) control can be identical with the drilling system and ground control unit figw-1C. Ground unit (602) control can be equipped with a component (612) registration controller (614), block (616) display information processor (618) and a transceiver (620). Component (612) registration collects and/or stores the data field, which includes the drilling floor. The data may represent data measured by the sensors (S) drilling site, shown in figa-1D. These data may also be data received from other sources. Data can also be stored on a machine-readable medium such as a compact disk, digital video disk, optical media, a volatile memory device, the nonvolatile storage device, or any other media, made with the possibility of storing the data is.

Controller (614) able to enter in the command field. Controller (614) can be equipped with actuators that can perform operations such as guidance, promotion or other action on the rig floor. Commands can be created based on the logic processor (618) or by commands received from other sources. The processor (618) can be created with attributes for data processing and analysis. The processor (618) may be created with additional functionality operations.

Block (616) display information can be equipped on the rig floor and/or in a remote location for viewing the oilfield data (not shown). Oilfield data reproduced by the block (616) display information may be raw data, processed data and/or output data generated from different data. In one embodiment, the implementation unit (616) information display is made with the possibility of creating certain kinds of data, so the screens, if necessary, can be performed on request. The user can define the required method steps during drilling based on the consideration shown oilfield data. The operation can be selectively adjusted in response to the actions of block (616) display information. Blo is (616) display information can be two-dimensional display for the consideration of the oilfield data or creating field events. For example, a two-dimensional display may correspond to the output from a printer, plotter, monitor, or other device configured to transfer a two-dimensional output data. Block (616) display information may also be three-dimensional display to review the various aspects of operations. At least some aspects of the operations considered in real-time in three-dimensional display. For example, a three-dimensional display may correspond to the output from a printer, plotter, monitor, or other device configured to transfer the three-dimensional output data.

Transceiver (620) configured to create a data access other sources and/or data on them. Transceiver (620) is also configured to enable communication with other components, such as servers (606), the drilling system (604) drilling site, the ground unit (602) control and/or tool (608) simulation.

Servers (606) can be used to transfer data from one or more drilling sites on the tool (608) modeling. As shown, the server (606) includes platform servers (622), the remote server (624) and the server (626) third-party access. Platform servers (622) can be installed on the rig floor and/or other blizlezhashchikh to distribute data from a ground block (602) control. The remote server (624) installed at the site away from the field and provides data from remote sources. Server (626) third party access may be a pad or remote, and it works a third party, such as a customer.

In one implementation, the servers (606) able to read and write data drilling (for example, the logs), the events of the drilling trajectory and/or other oilfield data (for example, seismic data, statistics, economic data, or other data that can be used during analysis). Type of server is not intended to limit the forecasting properties of subsurface formations. For example, the system may be adapted for operation with any type of server that you can use.

Servers (606) communicate with the instrument (608) simulation, as specified channels (610) connection. As stated numerous arrows servers (606) may have separate channels (610) communication tool (608) modeling. One or more servers (606) you can combine or merge to create a combined channel (610) connection.

Servers (606) collect a wide range of data. Data can be collected from different channels, which give some data types, such as wireline logs of wells. Data from the servers (606) is upravlyaut tool (608) modeling for processing. Servers (606) can also be used for storing and/or forwarding data.

Tool (608) modeling functionally connected to the ground block (602) control to receive data from it. In some cases, the tool (608) modeling and/or the server (s) (606) can be installed on the rig floor. Tool (608) modeling and/or the server (s) (606) can also be installed on various platforms. Tool (608) modeling can be functionally connected to the ground control unit via the server (s) (606). Tool (608) modeling may also be included in the composition of the ground block (602) control or placed beside it.

Tool (608) modeling includes the interface (630), block (632) data processing unit (648) modeling archive (634) and data block (636) data visualization. Interface (630) communicates with other components such as servers (606). Interface (630) may also provide communication with other oil or no oil sources. Interface (630) takes the data and converts the data for processing. Data from the servers (606) usually go on the set of channels that can be selected by the interface (630).

As shown in Fig.6, the interface (630) selects the data channel server (s) (606) and receives the data. Interface (630) also converts the channels d is the R for data from the drilling site. Interface (630) may also receive data from data files (for example, file, extensible markup language (XML)file, a database or data files in a different format). The data can then be sent to the modules (642) of the processing tool (608) modeling. The data can be immediately entered in the tool (608) simulation for modeling sessions in real-time. Interface (630) generates requests for data (e.g., geophysical exploration, well logging charts, and risks), displays the user interface and manages the event status of the connection diagrams. Interface (630) also specifies the data in the information object for processing. Interface (630) can accept a request from a ground block (602) control to retrieve data from servers (606), a unit of wells and/or data files.

Block (632) data includes modules (640) formatting modules (642) processing, auxiliary modules (646) and modules (650) training. These modules are designed to manipulate oilfield data for analysis in real-time.

Modules (640) format is used to reconcile the data with the required format for processing. The input data may need formatting, translation, conversion and other manipulations to use. Modules (640) formatting the implementation of the trading with the possibility of formatting and use of data from different sources, data process and display in real time.

Auxiliary modules (646) provide support functions for the drilling system. Auxiliary modules (646) include component logging (not shown) and the component controlling device user interface. Component logging creates a common challenge for all data logging. Component logging provides the installation according to the application destination. Component logging can also be created with other signs, such as system debugging, distribution and prevention, among others. Debug system sends a debug message to those who use the system. Distribution system sends the information to a subsystem, users and others. Information may interrupt or not to interrupt the operation and can be extended to a variety of platforms and/or different users throughout the system. The warning system can be used to send error messages and warnings on a variety of platforms and/or different users throughout the system. In some cases, warning messages may interrupt the process of the operation and display alarms.

Component controlling device user interface creates an interface for the elements of the display devices. Component control device interface user the La forms screens for user input, such as menu items, context menus, toolbars, and window settings. Component control device can also be used to control events related to the data entry screens of the user.

Module (642) processing can be used to analyze the data and generate output data. As described above, the data may include static data, dynamic data, statistical data, in real-time or data of other types. Additionally, the data may relate to various aspects of operations, such as structure formation, geological stratigraphy, sampling, core logging wells, density, resistivity, composition of the fluid, flow rate, conditions downhole, surface, condition of equipment, or other aspects of operations.

Modules (642) processing can be used for analysis of these data to create models of the geological environment. For example, seismic data or the borehole trajectory can be determined from analysis of these data. Seismic data can be obtained from different seismic sources for the region of interest in the geological environment. Examples of seismic sources include, but are not limited to, the station seismic studies on the truck such as shown in figa, seismic charge or the seismic survey vessel. Additionally, the trajectory may include data logging charts wells collected through borehole logging probes figa-1D. Examples of data wireline logs of wells include a chart full of acoustic impedance, density, porosity, resistivity, etc. at different depths of the borehole trajectory.

Archive (634) data can store data for block modeling. Data can be saved in a format suitable for use in real-time (i.e. to update information at a speed approximately equal to the speed of receiving information). Data, in General, refer to the archive data from the processing component. Data can be kept in the file system (for example, as a file, an extensible markup language (XML)or in the database. System (600) may determine what conservation is the most appropriate to use for a specific part of the data, and stores the data in a way that enables the automatic data flow passage through the rest of the system in a seamless and integrated mode. System (600) may also carry information flow in the mode of manual and automated (such as data flow modeling, geological and GE is physical) based on the held data.

Block (636) data visualization performs the calculation of the visualization algorithm to create one or more maps for data visualization. Display can be presented to the user in block (616) display information. Block (636) imaging data may include two-dimensional "canvas" (display), three-dimensional "canvas", "canvas" section of the borehole or other canvases, if necessary.

Block (636) data visualization can selectively generate a display composed of any combination of one or more "canvases". "Canvas" can be synchronized or not synchronized with each other during the show. Block (636) data visualization can be equipped with mechanisms to actuate different "canvas" or other functions in the system. Additionally, the block (636) data visualization can be configured to generate mappings reflecting events in the field, created in real time drilling data collected in real time during drilling, events on the field, created according to the statistics of neighboring boreholes collected over time, the current trajectory of the borehole during drilling, the model of the geological environment, created by static data subsurface geological characteristics, and/or any the e combinations thereof. In addition, the block (636) data visualization can be configured for selective adjustment of the representations on the basis of drilling data in real time, such as these promote the drilling tool drilling system (604) in the underground formation.

Block (648) modeling the function of the simulation to generate the output field. Block (648) modeling can be a common simulation tool capable of modeling functions, such as creating, analyzing models of the geological environment and manipulation. Models of the geological environment typically include data exploration and production, such as shown in figa-2D. Block (648) modeling can be used to perform relative comparisons of the objects of the geological environment. Block (648) modeling can also be used to update models of the geological environment on the basis of the relative comparisons of the objects of the geological environment. Alternatively, the block (648) modeling can be used to update models of the geological environment on the basis of user input.

Block (648) modeling may also receive seismic data. Seismic data may be a data figa-1B for the area of interest. In this case, the seismic cube may include himself seismic data for the subsystem field, of interest. Block (648) modeling can also get shifted seismic cubes by shifting the seismic cube in the area of interest. For example, the seismic cube can be moved in three-dimensional coordinate system along the axis x, y and/or z to get out of the seismic cube. Alternatively, the block (648) modeling can perform modeling functions in an alternative coordinate system (for example, a two-dimensional coordinate system, an orthogonal coordinate system, and so on). Block (648) modeling can also be used to obtain graphs of seismic well logging seismic cubes.

Chart of seismic well logging may represent seismic data on the borehole trajectory. Specifically, the chart of seismic well logging may include seismic cubes resulting from the borehole trajectory. For example, if the trajectory of the vertical hole chart of seismic well logging can match the track seismograms, including all seismic cubes along the vertical axis. May be present, at least one chart of seismic hole logs for each well and for each seismic cube. In the case of out of the seismic cube numerous charts seism the economic well logging (number of cubes) can exist for each well. Seismic cubes can be moved at equal intervals in the vertical and horizontal axis according to the shift parameters. Alternatively, seismic cubes can be joined at unequal intervals on the vertical axis. In another example, seismic cubes can be moved along different trajectories (e.g., horizontal axis, diagonal axis, and so on).

Modules (650) training can educate and train the neural network. Specifically, the modules (650) learning can form the input levels and output levels for use in neural networks. The input level can correspond to the input data subject to processing in a neural network. The output level can correspond to the target output to be created for training the neural network. Each level (that is, input levels, output levels, and so on) may include any number of neurons, where each neuron corresponds to the sample data relating to the level.

The input and output layers can be created on the data taken from modules (642) processing. For example, the input level can be created by seismic logging charts wells (as indicated above), derived from the original and shifted seismic cubes. Additionally, the output level can be created according to the wireline logs of wells (e.g., acoustic impedance, density is STI, resistivity and so on).

Modules (650) training can also train the neural network. In this case, the neural network can be trained similarly to the training described in U.S. patent No. 5444619 entitled "SYSTEM AND METHOD OF PREDICTING RESERVOIR PROPERTIES" ("SYSTEM AND METHOD for PREDICTING RESERVOIR PROPERTIES"). Specifically, the modules (650) training can create the weight matrix for neural networks. For example, the weight matrix can describe the relative importance of neurons in the input layer. Additionally, the weight matrix can be created using different ways of learning (e.g., back-propagation learning errors, genetic evolution or some other ways of learning).

If necessary, a training module (650) may perform statistical analysis of the input level of the neural network to identify potential deviations (for example, excessively close fit, not enough close-fitting), which can result in the output level of the trained neural network. Specifically, a training module (650) may determine that the correct proportion have training data to create the input level to the weights of the weight matrix to minimize the possibility of rejection. Alternatively, if the number of training data is fixed, a training module (650) may use other techniques (e.g. the R, the choice of the model, creating a fast fluctuations, early stopping, the weight decomposition, baiesovskoe training etc) to minimize the possibility of rejection. For example, training module (650) may use the weight decomposition to reduce the size of larger scales (i.e. higher relative importance in the neural network. With decreasing size of the larger scales generalization of a neural network can be improved by reducing the variation of the output level.

After training, the neural network can be used to create the model. For example, modules (642) processing can apply the trained neural network to seismic data to create a full model of acoustic impedance. In this example, the full model of acoustic impedance may correspond to a three-dimensional view of data full of acoustic impedance for the region of interest associated with the seismic data.

Although the specific components shown and/or described for use in blocks and/or modules of the tool (608) modeling, it should be clear that a variety of components with various functions may be used to create formatting, processing, assisted and coordination functions necessary to ensure processing in real-time instrument (08) modeling. Components can be combined functionality and can be implemented in software, hardware, and software-hardware or their combinations.

Additionally, components (e.g. modules (642) and processing unit (636) data visualization tool (608) modeling can be placed on the server (622) on the site of the operations or on distributed platforms where the composition can include a remote server (624) and/or server (626) third party. The server (622) on-site operations can be placed in ground unit (602) control.

7 shows a block diagram of the operational sequence of the method for predicting properties of subsurface formations for the operation of the drilling site. The method can be performed using, for example, the system 6. The method can include obtaining seismic data for the region of interest (block 702), the receipt of the initial seismic cube using seismic data (block 704), many of shifted seismic cubes using seismic data, where each set of shifted seismic cubes is shifted from the initial seismic cube (block 706), the creation of a neural network using the initial seismic cube, the set of shifted seismic cubes and data kartanegara wells (block 708), application of neural networks to seismic data to obtain a model of the region of interest (block 710), and the adjustment operation on the basis of the model (block 712).

Seismic data can be obtained (block 702) from a variety of sources. As described figa-1B and 6, the seismic data associated with the region of interest can be generated by the sensor (S) at the drill site or obtained from other sources. Seismic data can be sent directly to the tool 608 modeling (6) or forward on the modeling tool, at least one of the servers 606 (6). Seismic data is then, in General, adopts the interface of the modeling tool. Seismic data can be obtained from different seismic sources module 642 processing (6). For example, the module 642 processing (6) can receive seismic data from stations of seismic surveys on the truck, shown in figa. In another example, the module 642 processing (6) can receive seismic data from the seismic charges. Seismic data can then be reproduced as output. Specifically, the output can create block 636 visualization (6) data modeling tool and presented to the user in block 616 display (6) ground unit (602) control.

About the art, of interest may correspond to the geological environment. Additionally, the region of interest may include any number of underground formations, as described above for figa-1D.

7 initial seismic cube can be obtained (block 704) based on seismic data. The initial seismic cube can meet certain area parameters of interest. The user can work with block 616 display (6) to define the initial seismic cube. Specifically, the user can define the number of parameters region of interest, using block 616 display (6) to obtain the initial seismic cube. Alternatively, the block 648 modeling (6) may determine the initial seismic cube based on seismic data.

Then, the set of shifted seismic cubes can be obtained on the basis of seismic data (block 706). Block 648 modeling (6) can create a lot of shifted seismic cubes by shifting the initial seismic cube in the area of interest. Additionally, the user can define the shift parameters (i.e. direction, limits, successive approximation, and so on), describing how the initial seismic cube should be shifted. For example, the user can define the memory limits of the shift in the vertical direction and/or extent of shift in the horizontal direction. In this case, the module 642 processing (6) can get a lot of shifted seismic cubes using user defined parameters.

The neural network can then be created (block 708) in a variety of ways. For example, a neural network can be created by using the initial seismic cube, the set of shifted seismic cubes and data wireline logs of wells. Additionally, the neural network can be trained using different training algorithms (e.g. genetic evolution, back-propagation learning errors, iterative inversion or some other learning algorithms). The trained neural network can describe the nonlinear relationship between seismic data and geological properties (for example, full acoustic impedance, porosity, density, etc) of the region of interest.

Then, the neural network can be applied to seismic data to obtain a model of the region of interest (block 710). More specifically, the neural network can be applied to seismic data to evaluate different geological properties (for example, full of acoustic impedance, porosity, density, and so on)associated with the region of interest. In one example, the seismic data of the area of interest, you can get the using station seismic studies on the truck, shown in figa. In this case, after training, the neural network can describe the relationship between geological properties and seismic data. For example, a trained neural network can describe the function to calculate the total acoustic resistance on the basis of seismic data. In this example, the function described by the trained neural network can be applied to seismic data to obtain a complete acoustic impedance model for the region of interest.

The model can describe the geological properties for the entire area of interest. Additionally, the model can usually describe the geological property described in the logs of wells used to create the model. In this case, data logging chart wells can be used for training the neural network to create models of the geological properties associated with the data logging chart well. For example, data logging chart wells associated with full acoustic resistance, can be used for training the neural network to obtain models of acoustic resistance.

The operation can then be adjusted on the basis of the model (block 712) in a variety of ways. For example, the user may adjust the operation using the receiving controller 614 (6) based on the model. In another example, the data visualization module can display the model associated with the area of interest. The operation can then be corrected or to be performed on the basis of the display. For example, the user can create a new borehole trajectory for the operation based on the model of the complete acoustic impedance so that a new borehole trajectory crosses the place with full high acoustic impedance. In this example, high total acoustic resistance may indicate that the place has a higher porosity, which in turn may indicate that in this place there is an increased amount of oil. In another example, the user can use the full model of acoustic impedance to create the borehole trajectory for use in exploration.

Method (7) is shown in a particular order. However, it should be clear that part of the method can be performed simultaneously or in different sequences.

On Fig shows the block diagram of the operational sequence of the method of creating a neural network. The method can be performed using, for example, the system 6. Additionally, the method can describe the generation, described as block 708 7.

The method includes receiving multiple charts of seismic well logging using the initial seismic cube and the set of shifted seismic cubes (block 802), receiving data logging chart wells associated with the region of interest (block 804), creating a neural network that contains the data set of charts of seismic well logging and data logging chart well (block 806), and the training of the neural network using data from the logs of wells to create the weight matrix includes a set of weights, where the set of weights associated with many charts of seismic well logging (block 808).

Many charts seismic well logging data can be obtained (block 802) from various sources. Specifically, many charts seismic well logging data can be obtained using the initial seismic cube and the set of shifted seismic cubes. For example, as discussed above for 6, a diagram of a seismic hole logs can be obtained from seismic cube using the tool 608 modeling (Fig.6).

Data logging chart wells associated with the region of interest can be obtained (block 804) from a variety of sources. As discussed for figa-1D and 6, information on the object of the geological environment can provide the sensor (S) on the rig floor or other sources. Data logging chart wells can be sent directly to the tool 608 simulated the I (6) or forward on the modeling tool, at least one of the servers 606 (6). Data logging chart well then, in General, adopts the interface of the modeling tool. Data wireline logs of wells may be associated with a group of wells. For example, data logging chart wells may contain the logs full of acoustic impedance, where each wireline logs full acoustic resistance is associated with a different well. In this example, the data logging chart wells can also be represented as the output.

You can then create a neural network using multiple charts of seismic well logging and data logging chart well (block 806). Specifically, you can create a neural network, where many charts of seismic well logging used as the input level of the neural network, and data logging chart well used as the output level of the neural network. For example, each of multiple charts of seismic well logging may correspond to a neuron in a neural network.

The neural network can be trained using data from the logs of the borehole (block 808) in a variety of ways. More specifically, the neural network can be trained using data from the logs of wells to create in the owl matrix, includes numerous weight, where the neural network additionally includes a weight matrix. In this case, data logging chart wells can act as the stopping criterion for training the neural network. For example, the back-propagation algorithm error learning can be applied to neural networks, where neurons assign weights in a neural network to meet the stopping criterion. When the stopping criterion corresponds to the data logging chart wells assigned weight can combine the neurons associated with samples of a variety of charts seismic well logging for a close match log charts full acoustic resistance is included in the data logging chart well. The algorithm back-propagation learning errors usually assigns the weight by counting errors associated with the neurons in the neural network. In this case, the neuron weights can be adjusted to minimize the local error of each neuron in the neural network.

Alternatively, the neural network can be trained using some other learning algorithms. Examples of other learning algorithms include algorithms evolutionary computation algorithm for statistical evaluation, the algorithms of complex intelligence. Usually neural the et is used to model correlations between the observed input and output data. In other words, the trained neural network can be used for logical inference function based on the observed data sets. In this case, the result of each learning algorithm may be different trained neural network, where each trained neural network describes the different functions for the same set of observed input and output data.

The way pig shown in a particular order. However, it should be clear that part of the method can be performed simultaneously or in a different order or sequence.

Figure 9 shows an example graphical representation of a set of shifted seismic cubes, as described in block 706 7 above. Here the graphical representation includes the initial seismic cube (902), shifted seismic cube (904) and trajectory (906) wells. For example, shifted seismic cubes (904) can be obtained by shifting the initial seismic cube (902). In this example, the frequency shifted seismic cubes (904) decreases with increasing distance from the initial seismic cube (902). Additionally, the borehole trajectory (906) may be associated with the data logging chart well. Graphical representation additionally includes sample (908) charts of seismic well logging. For example, charts of seismic logging SK is Agin can be based on seismic data of the seismic cube (904), emerging along the charts of seismic well logging.

Figure 10 which shows an example of a graphical representation (1000) seismic data received in block 702 7. Graphical representation (1000) includes seismic data relating to the area of interest. Graphical representation (1000) can be presented to the user in the display, as described above in block 702 7. Seismic data can be obtained for a region of interest from different seismic sources. Examples of seismic sources include, but are not limited to, the station seismic studies on the truck such as shown in figa, seismic charge or seismic research vessel.

Figure 11 which shows an example of a graphical representation (1100) model obtained in block 710 7. Graphical representation (1100) model includes obtained the logs (1102)associated with the data logging chart well, and clearing the logs (1104) on the basis of the model. In this example, clearing the logs (1104) can be compared with the received logging charts (1102) to determine the quality of the graphical representation (1100) model. More specifically, the quality of the model can be determined on the basis of the correlation coefficient racetracktaglib charts (1104) and received wireline logs (1102). For example, received the logs (1102) may be based on data from wireline logs full of acoustic impedance obtained in the borehole, and clearing the logs (1104) can be obtained based on the data of the complete acoustic impedance obtained from the full model of acoustic impedance. In this case, the correlation coefficient of logs can match the quality of the model the total acoustic resistance.

From the above description it should be clear that you can perform various modifications and changes in the preferred and alternative embodiments of implementation of the present invention without departing from its real essence. For example, the method can be performed in a different sequence, and the generated components can be integrated or separate.

This description is only illustrative and should not be construed as limiting. Volume forecasting properties of subsurface formations should be defined only by the following paragraphs of the claims. The term "comprising" in the claims means "including at least" as this emphasizes the inclusion of elements in the composition formula of the invention, as an open group. The use of the singular includes the presence of which also forms the plural, if they are not specifically excluded.

Although the forecasting properties of underground formations described in relation to a limited number of embodiments, a person skilled in the field of technology exploited by the invention, it should be clear that it is possible to create other embodiments of not departing from the scope of the forecasting properties of subsurface formations described in this document. Thus, the forecasting properties of subsurface formations should be limited only by the attached claims.

1. A method for predicting properties of subsurface formations drilling site containing phases in which:
receive seismic data for the region of interest;
get the initial seismic cube using the seismic data, where the initial seismic cube is a three dimensional representation of seismic data;
form the set of shifted seismic cubes in the region of interest, using the seismic data, and the offset parameter
each of the set of shifted seismic cubes is shifted from the initial seismic cube; and
moreover, through the shift parameter set the direction and limit, in which the initial seismic cube should be shifted;
form neuron the second network using the initial seismic cube, many of shifted seismic cubes and data logging chart wells; and
apply the neural network to the specified seismic data to obtain a model of the region of interest, and the model is configured for use in adjusting the operation of the drilling site.

2. The method according to claim 1, wherein the step of forming the neural network includes the steps are:
get many charts of seismic well logging using the initial seismic cube and the set of shifted seismic cubes;
receive data logging chart wells associated with the region of interest;
form a neural network that contains many charts of seismic well logging and data logging chart wells; and
make training a neural network using the data from the logs of wells to create the weight matrix, with the weight matrix contains a number of scales associated with multiple charts of seismic well logging.

3. The method according to claim 2, in which the weight matrix form using the algorithm of back-propagation learning errors and data logging chart well is used as the stopping criterion in the algorithm back-propagation learning errors.

4. The method according to claim 2, in which when creating aroney network additionally reduce the size, at least one of the set of weights based on the relative importance of each at least one from a set of scales.

5. The method according to claim 1, in which the model corresponds to model the total acoustic resistance.

6. The method according to claim 1, wherein the operation is one operation selected from the group consisting of scouting operations, operations, drilling and extraction operations.

7. The method according to claim 1, in which a neural network describes a non-linear relationship between seismic data and geological properties of the area of interest.

8. The method according to claim 1, in which the direction is at least one selected from the group consisting of vertical direction and horizontal direction.

9. Forecasting system properties of subsurface formations drilling site that contains:
the data processing module, configured to receive seismic data for the region of interest;
block modeling, configured to:
the receipt of the initial seismic cube using the seismic data, where the initial seismic cube is a three dimensional representation of seismic data; and
shift the initial seismic cube to form the set of shifted seismic cubes in the region of interest on the basis of the decree is the R seismic data and shift parameter;
each of the set of shifted seismic cubes is shifted from the initial seismic cube; and
moreover, the shift parameter specifies the direction and the limit in which the initial seismic cube should be shifted;
a training module configured for:
the formation of the neural network using the initial seismic cube, the set of shifted seismic cubes and data logging chart wells; and
application of neural network to the specified seismic data to obtain a model of the region of interest, and the model is configured for use in adjusting the operation of the drilling site.

10. The system according to claim 9, in which:
block modeling is additionally configured to receive a set of charts of seismic well logging using the initial seismic cube and the set of shifted seismic cubes;
the data processing module is additionally configured to receive data logging chart wells associated with the region of interest; and
a training module is additionally configured to:
the formation of the neural network the neural network contains many charts of seismic well logging and data logging chart wells; and
neural network training using donnycarney charts wells for forming the weight matrix, thus the weight matrix contains a number of scales associated with multiple charts of seismic well logging.

11. The system of claim 10, in which the weight matrix is formed using the algorithm of back-propagation learning errors, and data logging chart well is used as the stopping criterion in the algorithm back-propagation learning errors.

12. The system of claim 10, in which the module is additionally configured to reduce the size of at least one of the set of weights based on the relative importance of each at least one from a set of scales.

13. The system according to claim 9, in which the model corresponds to model the total acoustic resistance.

14. The system according to claim 9, in which the operation is one operation selected from the group consisting of scouting operations, operations, drilling and extraction operations.

15. The system according to claim 9, in which the direction is at least one selected from the group consisting of vertical direction and horizontal direction.

16. Machine-readable medium having stored thereon instructions, which when executed on a computer cause the computer to perform a method for predicting properties of subsurface formations drilling site, and the instructions provide the ability to operate on the I:
the receipt of the initial seismic cube using the seismic data, where the initial seismic cube is a three-dimensional representation of the specified seismic data;
formation of set of shifted seismic cubes using seismic data in the area of interest, using the seismic data, and the offset parameter
each of the set of shifted seismic cubes is shifted from the initial seismic cube; and
moreover, the shift parameter specifies the direction and the limit in which the initial seismic cube should be shifted;
training a neural network containing the source of the seismic cube and the set of shifted seismic cubes, based on data from the logs of wells; and
application of neural networks to seismic data to obtain a model of the region of interest, the model configured for adjustment of operations the drilling site.

17. Machine-readable medium according to item 16, wherein the instructions additionally provide the possibility of functioning for:
getting multiple charts of seismic well logging using the initial seismic cube and the set of shifted seismic cubes;
receiving data logging chart wells is, associated with the region of interest;
training the neural network using data from the logs of wells for the formation of the weight matrix, with the weight matrix containing the set of weights associated with multiple charts of seismic well logging.

18. A machine-readable medium of 17, in which the weight matrix is formed using the algorithm of back-propagation learning errors, and data logging chart well is used as the stopping criterion in the algorithm back-propagation learning errors.

19. A machine-readable medium of 17, in which the creation of the neural network further comprises reducing the size of at least one of the set of weights based on the relative importance of each at least one from a set of scales.

20. Machine-readable medium according to clause 16, in which the direction is at least one selected from the group consisting of vertical direction and horizontal direction.



 

Same patents:

FIELD: information technology.

SUBSTANCE: in an ophthalmic-microsurgical computer local area network for vitreoretinal operations, formatting devices are in form of a radial-annular structure consisting of a single set of automated workstations (AWS), which synchronously or asynchronously functioning, processing, converting, transmitting, analysing, synthesising hierarchical structures of an artificial neural network: diagnosis AWS (DAWS), ophthalmic-microsurgical AWS (OMAWS), subsequent operation stages AWS (SOSAWS), component AWS (CAWS), surgeon's operating unit (SOUAWS), with opposite forward and reverse flow of information in between, where each AWS has at least one neural circuit, interconnected identification units (IU), an interpolation unit (INU), an extrapolation unit (EU), which are the neural network converting and transmitting elements (NNCTE), a decision unit (DU), which is the neural network analysis and synthesis element (NNASE).

EFFECT: simultaneous improvement of accuracy of determination and quality of identifying diagnoses, determining indications for conducting operations, high selectivity when conducting operations, accuracy in determining the sequence of operations, simulating operations, accuracy in choosing the anaesthetic method, accuracy of providing implants and expendable materials, optimisation of flow of information and necessities during vitreoretinal ophthalmic-microsurgical operations.

1 dwg

FIELD: physics.

SUBSTANCE: neuron simulation method is based on calculation of squares of Euclidean distance from the input vector to each of 2n vertices of a unit n-dimensional cube in weighting units, and multiplication of values inverse to these distance values with components of the target vector respectively, and then summation in an adder and conversion in the activation unit through an activation function.

EFFECT: possibility of simulating a neuron of any given Boolean function from a complete set of from n variables.

6 dwg, 1 tbl

FIELD: information technology.

SUBSTANCE: multilayer modular computer system has several layers, including a neural network layer, a transport layer and a processor layer, wherein the transport layer contains network controller-router modules, the processor layer contains processor modules, and all the said modules have multiple inputs and outputs connected to each other and connected to the inputs and outputs of the system. The processor modules train neural network domain modules.

EFFECT: high decision speed, possibility of grafting layers and modules in each layer during operation of the system with a complex task, high reliability of the computer system.

3 cl, 1 dwg

FIELD: information technologies.

SUBSTANCE: invention may be used for building of modular neural computers, which function in symmetrical system of residual classes. Stated neuron network comprises unit of neuron network of end ring of senior coefficient generation for generalised positional system of numeration, unit of polarity shift, unit of error detection, buses "with errors" and "without errors".

EFFECT: reduced hardware complexity.

3 dwg

FIELD: physics; computer engineering.

SUBSTANCE: present invention pertains to neurocomputers. The device has a unit for storing a binary input signal, a logic AND-OR circuit, internal memory unit, unit for generating the output string of codes, a generator of synchronising pulses, control unit, a unit for selecting duration and extracting information, analysis block and a corrector unit.

EFFECT: increased rate of operation, providing for the possibility of distinguishing change in state of processed signals, increased noise immunity, possibility of making super-complex neural networks, and simplification of design.

9 cl, 1 dwg

FIELD: modular neuro-computing systems.

SUBSTANCE: neuron network contains input layer of neurons, at inputs of which residuals of number being divided are received through system of modules, (n-1) neuron networks of finite ring for addition, (n-1) neuron networks of finite ring for multiplication, neuron network for expanding a tuple of numerical system of residues, and as output of neuron network for dividing numbers represented in system of residual classes are outputs of neuron network of finite ring for multiplication and output of neuron network for expansion of tuple of numerical system of residues.

EFFECT: expanded functional capabilities, increased speed of division, reduced volume of equipment.

1 dwg

FIELD: neuron-like computing structures, possible use as processor for high speed computer systems.

SUBSTANCE: device contains artificial neuron network composed of analog neurons, at least one controllable voltage block, a group of long neuron-like nonlinear communication units, each one of which contains serially connected circuit for synchronization and selection of radio impulse envelope, auto-generator with self-suppression circuit, a length of coaxial line, realizing functions of antenna, additional circuit for synchronization and selection of radio-impulse envelope.

EFFECT: increased information processing speed due to increased paralleling degree of computing processes.

2 dwg

FIELD: neuro-cybernetics, possible use in artificial neuron networks for solving various problems of logical processing of binary data.

SUBSTANCE: method for realization of logical nonequivalence function by neuron with two inputs is based on multiplication of input signals with corresponding weight coefficients and summing them, after that the total is transformed in activation block firstly by quadratic transfer function, and then by threshold function at neuron output.

EFFECT: realization by one neuron of first order of logical nonequivalence function of two variables.

5 dwg, 1 tbl

FIELD: computer engineering, possible use in modular neuro-computer systems.

SUBSTANCE: in accordance to invention, neuron network contains input layer, neuron nets of finite ring for determining errors syndrome, memory block for storing constants, neuron nets for computing correct result and OR element for determining whether an error is present.

EFFECT: increased error correction speed, decreased amount of equipment, expanded functional capabilities.

1 dwg, 3 tbl

Neuron-like element // 2295769

FIELD: cybernetics, possible use as a cell for neuron networks.

SUBSTANCE: neuron-like element may be used for realization on its basis of neuron network for solving problems of estimation of functioning of complicated open systems, estimation of degree of optimality of various solutions by ensuring possible construction of model of researched system, both hierarchical and recurrent, with consideration of varying original and working condition of its elements and variants of their functioning, during modeling taking into consideration the level of self-sufficiency of neuron-like elements, susceptibility to effect of external signals, type and errors of setting of their parameters and parameters of input signals, and also provision of given precision of self-teaching of neuron network. Device contains input block, block for setting and normalizing weight coefficients, block for computing parameters of input signals, adder, signals share limiter, block for computing input part of condition, block for setting internal state, block for computing internal part of distance, block for counting distance, memory block, analyzer of state change value, block for determining precision of self-teaching of neuron network, block of determined dependencies, switch, output block, control block, random numbers generator.

EFFECT: creation of neuron-like element.

2 cl, 1 dwg

FIELD: physics.

SUBSTANCE: method of obtaining measurement results of an underground layer through a signal propagating through the layer comprises steps on which: a signal is received after propagation through the layer, the unmatched form of the wave is rejected, and the first event of the received propagated signal is determined, wherein the determining step includes: receiving the wave form which is the propagated signal, which includes the received propagated signal, which includes its first event, determining for a point along the wave form a likelihood function which represents the likelihood that corresponding parts of the waveform before and after the said point presents different signals, wherein the said step for determining likelihood involves applying a statistical function for determining the point along the wave form on which the likelihood function is maximum, and the position of the said point along the wave form is roughly determined, which is used at the step for determining likelihood. The determined point is given as the first event using a localisation function. Said method is realised using the corresponding device.

EFFECT: simple procedure for taking acoustic or electromagnetic measurements and high speed of obtaining required data.

23 cl, 16 dwg

FIELD: mining.

SUBSTANCE: in method to process seismic data using discrete wavelet transform, including provision of seismic data in the form of set of seismic roads, each of initial seismic roads represented as vector of counts is exposed to discrete wavelet transform (M iterations) with subsequent decay (decomposition) of initial seismic signal into layers of detail dl(n) with various power and frequency characteristics; each of specified layers of detail dl(n) is analysed by their target value with account of solved seismic task, afterwards valuable separate layers of detail are selected in wavelet transform of initial seismic signal to build their partial sums for further processing and interpretation of seismic data.

EFFECT: making it possible to analyse seismic data with localisation of signal features in wavelet-frequency area with improved quality of separation of signal component features in space-time coordinates, with division of wave fields into separate components and increased signal-noise ratio.

5 dwg

FIELD: physics.

SUBSTANCE: method of extraction and/or removal of the concerned signal, including: initiation of acoustic waves using well-logging system having, at least, one transmitting device and, at least, one receiver; automatic filtering of the concerned signal from acoustic signal data received with, at least, one receiver, through calculation of the first coherence measure, including the concerned signal, and displaying of the first coherence measure to time domain, and self-filtered acoustic signal data processing based on coherence measure. Specified method is realised in number of versions using corresponding devices.

EFFECT: improved quality of registered signal interpretation.

53 cl, 6 dwg

FIELD: oil and gas industry.

SUBSTANCE: there are suggested a method and a device for identification of a value and inverse velocity of a wave in a bed, located around a well. The conception of fresnel volume is applied for kinematics tomography. The fresnel volume reflects the way of an acoustic wave propagation in the vicinity of a well. The implementation of the fresnel volume for data of acoustic survey facilitates a stable solution of an inverse problem and possibility of practical realisation of three dimensional (3-D) tomography. The solution of the inverse problem is achieved by the method of the iterative inverse projection.

EFFECT: facilitates a stable solution of the problem of identification of value and inverse velocity of a wave in a bed, located around a well, and also facilitates a practical realisation of a three dimensional tomography.

42 cl, 6 dwg

FIELD: invention refers to the mode assigned for definition of the danger of the water flow with slight depth of occurrence by using seismic data.

SUBSTANCE: seismic data may be processed for increasing its stratigraphical settlement by way of subsampling of seismic data on less then two millisecond interval. The fulfillment of the stratigraphical analysis relatively to seismic data and valuation of seismic characteristics of the seismic data may be used for choosing the control field. Before summing up with taking into consideration the form of the impulse they apply inversion to the seismic data in the chosen control field for receiving the model of elastic deforming which includes the speed of the longitudinal waves and the speed of the transversal waves. Then they define the danger of the water flow from the slight depth of occurrence using the model of elastic deforming by comparison the speed of the longitudinal waves with the speed of the transversal waves.

EFFECT: defines danger of water flow.

26 cl, 8 dwg

The invention relates to Geophysics

The invention relates to logging technology for measuring physical properties of underground formations, in particular to a method of logging and system using acoustic waves

FIELD: oil and gas industry.

SUBSTANCE: there created is a device and method of measuring the parameters characterising the rock bed in oil well with device for probe field generation in the rock bed zone and device that generates the flow through the zone in presence of probe field additionally including sensors sensitive to changes in the zone, note that sensor response indicates the values of fluid quantity and changes in hydrocarbons phases in the bed. The changes can be performed before the flow influences the measurement zone and after the flow appearance through the measurement zone.

EFFECT: advancement and improvement of devices and methods of defining the bed characteristics with the use of the flow created in the bed.

25 cl, 5 dwg

FIELD: physics.

SUBSTANCE: multi-position batcher-collector device consists of a sealed collector in form of a square made from transparent organic glass or carbonate below, on the outer side of which batching elements are attached. According to the invention, the batching elements have a movable cylinder which moves along the a fixed rod inside the cylinder, said rod having a longitudinal through hole for sucking liquid through a vacuum inside the collector where an excess collected dose of liquid is collected for washing the batching elements.

EFFECT: possibility of fast establishment of the required dose of liquid without replacing batching devices, cutting time for establishing dosage and high accuracy of positioning.

3 dwg

FIELD: mining.

SUBSTANCE: device includes spacer assembly in the form of elastic rings for measurement of radial deformations of the well, measuring assembly in the form of recorders of deformations of rings, recording system connected to the measuring assembly, and adjustment device in the form of a bar the axis of which is parallel to the well axis. Device is equipped with measuring device of axial deformations of the well, which is installed on the end face of adjustment device and made in the form of elastic ring and ring deformation recorder connected to the equipment; one of the ring diameters coincides with axis of positioning device; one side of ring is rigidly connected to the adjustment device, and diametrically opposite side of ring is designed for interaction with the well end face. One of diameters of each elastic ring for measurement of radial deformations of the well coincides with the axis of adjustment device; one side of each ring is rigidly fixed on the adjustment device, and diametrically opposite side of each ring is movable relative to adjustment device; at that, planes of rings are turned relative to each other in compliance with directions of measured deformations of the well wall, and the number of rings and their location along adjustment device corresponds to the number and location of deformation recording points throughout the well length.

EFFECT: enlarging functional capabilities of the device owing to providing the synchronous recording of seismic waves with axial and radial deformations of the well.

3 cl, 1 dwg

FIELD: mining.

SUBSTANCE: bed gas pressure gradient is measured in a payout bed of a well. A depth sample of bed gas is taken from this payout bed. Density of bed gas is determined according to the pressure gradient, as well as its weight in the selected sample according to sampler's volume. The depth sample is processed to release gas components from it. Using the produced processing data, weight of components in the taken sample is identified. Based on the difference of weights referred to volume of gas components it is decided on content of condensate in bed gas.

EFFECT: improved accuracy of survey.

3 cl, 1 tbl, 2 dwg

Up!