Method for controlling condition of multivariate object

FIELD: measurement technology.

SUBSTANCE: invention refers to the sphere of measurement technologies and can be used to control and analyse the condition of complex multivariate objects being the elements of the connection systems and automation. The result is achieved by the identification of abnormal changes of condition's characteristics of one of the complex of identical elements of multivariate objects during in functioning by setting the standard values and allowed deflections of the characteristics of the elements' condition of the multivariate object, measuring current values of the parametres under control, calculating the current values of the characteristics of the elements' condition of the multivariate object and comparing them with the given standard values and allowed deflections.

EFFECT: improvement of the operation and accuracy of the evaluation of the condition of the multivariate object.

3 dwg

The invention relates to measuring technique and can be used for control and analysis of complex multivariate objects which are elements of communication systems and automation.

There is a method of continuous passive control parameters phone lines implemented in the device type LST-1007, in which pre-check phone line for damage and unauthorized connections, specify the parameters to be monitored telephone line, measured parameters to be monitored, memorize them as standards, continuously measure and compare the current values of monitored parameters with a reference, generate an alarm signal when the discrepancy of the measured values with the reference, continue measurements at their coincidence with the reference values of monitored parameters.

The disadvantages of this method are the low accuracy of testing results due to the impact of external factors (climate, time, and other), than due to the low probability of making objective decisions about the state of the controlled object and the low economic efficiency of the control system, it limits the scope of application of the method.

Also known "Method of measurement of parameters of RLC-circuits", patent of Russian Federation №2100813, class G01R 27/26, 27.12.1997, the basis of the tion on the measurement of the duration of transient processes in the resistive-capacitive or resistive-inductive measuring circuit, at the entrance which serves disturbing voltage varying as a function of time, and measure its parameters at the exit of the investigated RLC-circuit.

Compared with the previous allows to obtain more complete information about the characteristics of the controlled object, the disadvantage of this method is the relative complexity caused by the need to generate special probing signals, narrow scope, due to the incompatibility of the way of measuring with the normal functioning of telephone lines and the impossibility of continuous measurement of their parameters, as well as low economic efficiency of the method.

The closest to the technical nature of the claimed is a method of evaluating the effectiveness of large systems comprising a large number of controlled parameters, RF patent No. 2210112 "Unified way Chernyakova / Petrushin to evaluate the effectiveness of large systems", class GO6F 17/00, Appl. 7.06.2001.

Prototype method is that pre-define a set of structural elements - representation of multivariate object (IGOS) in the form of a hierarchy of structural elements (sea), particular characteristics of the state that corresponds to each element of the sea, normative values corresponding to each particular show the Liu condition, the weights of importance corresponding to each private indicator of sea, and also in advance in a storage device (memory) of the terminal server write a program to calculate the particular parameters and, finally, preliminary in memory of the workstation engineer knowledge load information obtained in the course of a survey of experts of this field of knowledge, choose the method of calculation and run the procedure, select to switch the measured parameters, automatically read the information from the sensors through the converters and record it in the memory read information in the terminal server transform parameter values into corresponding digital data by using various special converters, remember digital data in the storage device, calculate private and General characteristics of the state of Uganda program for computing characteristics of state using terminal server, compare them with the predefined values, display and document the results of calculations and comparisons on the video monitor and the printer.

In comparison with analogues prototype method has a wider scope for both simple and multivariate objects.

The disadvantage of the prototype is relatively low precision current OC the NRI status of IGOS and efficiency identify reasons for changes in its condition prior to the onset of the critical values of monitored parameters, as well as identify the different behavior of the parameters of one of the sea relative to the others. This is due to the fact that the conclusion about the state of the controlled IGOS do when conducting full cycle control, the results of which indicate only the critical deterioration of the parameters.

The objective of the proposed technical solution is to develop a method of monitoring the status of IGOS, providing higher accuracy assessment IGOS; increased efficiency assessment IGOS by identifying trends in parameters before they reach critical values and identifying different (abnormal) behavior of parameters in one sea regarding the behavior of other parameters of the same sea.

The claimed method expands the Arsenal of tools for this purpose.

This objective is achieved in that in the known method of monitoring the status of multivariable object (IGOS), namely, that pre-specify a set of N≥2 structural elements (sea) and M≥N parameters describing their condition, measure and memorize the measured parameters of the sea, determines the status of IGOS in the process of functioning, the results of which are documented and take a decision about the status of IGOS, in addition to N identical sea pre-set the mill is Artie values of their parameters P ^sm. Also set the allowable deviation Δ^apprparameters from the standard values and the maximum allowable differencebetween the maximum and minimum values for each of the m - controlled parameters. To characterize the status of IGOS reads the previously stored values of parametersand P_micompare them with each other and in each group the same type of sea compute the difference (deviation) Δ_mithe measured values of the m-th parameter from its default value of. Read previously stored extreme valueandcompare them and the results of the comparison determine the difference Δ_mijbetween the maximumand minimumlevels measured values of the m-th parameter. Then compare the calculated characteristics

Δ_miand Δ_mijwith their predefined valuesand. After that, the measured parameters and calculated characteristics of the document. Ifandthe status of IGOS accept as healthy and repeat the cycle of measurements. Ifand/or the status of IGOS characterized as defective and additionally document the information about the parameters and characteristics of the i-th structural element.

Thanks to the new essential features in the way the possibility of detecting changes in the monitored parameters in the range of valid values, and, in addition, detection of abnormal changes of the controlled parameters of one or more sea on the background state similar parameters of other similar sea than is achieved in the proposed method improve the accuracy, efficiency current state assessment sea and the probability of making an objective decision about the status of IGOS.

The use of advanced condition monitoring sea to complete the full cycle of monitoring the status of all IGOS provides the economic efficiency of the control system.

The claimed method is illustrated by drawings on which is shown:

figure 1 is a drawing illustrating multivariate object and his group of similar structural elements;

figure 2 - drawings explaining the process of implementation of the method.

figure 3 is a drawing illustrating the results of comparative step-by-step simulation of the prototype method and the inventive method.

The accuracy of the current estimate IGOS known methods are limited by the following factors: the value of Kon is controlled parameters of the structural elements of the object in the process of their functioning can vary within acceptable values, defined specifications for IGOS. The parameters controlled the sea can have a deflection not to exceed the allowable values, however, indicate a trend of worsening or abnormal condition of one or more of the same sea. In this regard, the state IGOS will be assessed by known methods as acceptable as long as the value of the parameter monitored item will not be released outside of valid values.

For most options for building multivariate objects, you can make the assumption that similar structural elements of the object are exposed equally to the influence of external disturbing factors. Detection of abnormal changes in the parameters of one of the sea within groups of the same type indicates additional perturbing effects. The proposed method allows for early detection of abnormal behavior settings both private and group of the same sea, thereby to provide an objective assessment of the status of IGOS and timely measures to prevent irreversible consequences in the work of IGOS.

The implementation of the claimed method can be seen on the example of IGOS, which is a communication node (CA) network connecting lines of different types: wire line using fiber-optic cable (1₁-1 N1) or symmetric pair (1₁-1_N2), and wireless (e.g., microwave) radio link (1₁-1_N3) (group I, II, III in figure 1). For each group of similar structural elements of the specified set of controlled parameters

(P₁...P_M1; P₁...P_M2P₁...P_M3determining the operating characteristics or the characteristics of protection against unauthorized impacts. For example, for subscriber lines using a symmetric pair of operational parameters are the impedance and capacitance of the line, the mutual influence lines in multi-pair cable, the insulation resistance, the throughput of the line (digital line), line voltage, capacity zvonkovoy chain (for analog phone line connection) and other parameters that need to be monitored during operation of the lines to determine their condition and health.

For each m-th parameter of the set M - parameters of one of the groups of the same sea, depending on the objectives and modalities of IGOS set the standard (nominal) values of the parameters(defined in operational and technical documentation), and the allowable deviation values of each of the parametersfrom the norm and acceptable unwinding diameter the values of the m-th i.e. the marginal value of the difference between the deviations of the same parameter of the i-th Δ_iand the j-th Δ_jsimilar sea:. Valid values areandexpect minimizing the probability of making wrong decisions about healthy state IGOS for a given probability of erroneous decision on a failed state IGOS. Moreover, when setting these parameters must comply with the condition. The calculation is made according to the formulas of the control of the borders for cards Shewhart (GOST R 50779.42-99) or according to the decision-making criteria the Neyman-Pearson (Levin BYR Theoretical foundations of statistical radio engineering. "Owls. radio" 1975, 392 C.). These source data record in a memory of the control system.

During the operation of the MVE monitoring system provides continuous (or periodic with a given time interval control Δt_tomeasurements of current parameter values. Measurements of controlled parameters is performed using the corresponding parameter of the probes. Evaluation of the m-th parameter P_mdetermine the sample P_m(Δt_u), which is formed as a result of multiple samples during the measurement interval ∆ T_u. The duration of this interval is chosen recognize who I am from the desired averaging interval time values of P _m. This is because the objective of monitoring the status of the sea and all IGOS can affect short-term exceeding of the permissible values of monitored parameters and characteristics under the influence of external random factors (electromagnetic discharge, power surges and the like). Such random factors can affect both individual sea, and all IGOS. In this regard, the values of ∆ T_uyou can define more accurately only experimentally. The sample size n (number of samples) P_m(Δt_udetermine the required measurement accuracy and sensitivity of the control system to changes in the monitored parameters. The measured values of the controlled parameters P_mrecording in a memory of the control system.

To calculate the characteristics of the sea state read from the memory pre-recorded values ofand P_micompare them with each other and calculate the difference

Δ_mifor each i-th sea for each m-th parameter using a comparator (e.g., comparator). Similarly, to calculate the values Δ_mijread from the memory extreme values ofandfor each m-th parameter, compare them and calculate the difference Δ_mijusing a comparator. The village is e this read from the memory pre-recorded values of ,and the calculated values Δ_mi, Δ_mijand compare them using a comparator. Determining the status of IGOS by value Δ_mijcan be simplified by measuring the correlation coefficient R_mijvalues of the m-th parameter P_miand P_mjfor the i-th and j-th homogeneous sea with torrelamata. Varieties and peculiarities of application of torrelamata known (see, for example. Metrology and elektroradioizdeliya in telecommunication systems: a Textbook for universities / Vieeo [and others]. - M.: Higher. HQ., 2001. - 383 S.).

Obtained after scaling Δ_miand Δ_mij(ΔR_mij) and compare them with valid valuesanddata are the starting point for making decisions about the status of IGOS. It is possible different States of the sea and the reaction control system, presented in figure 2.

During the operation of the structural elements of each of the same type sea values of the m-th parameter can take individual values of P_m1P_m2P_mNin the General case, different from the standardand make random fluctuations within tolerancethis is Δ_mijalso naprawiam valid values for (see figa). In this case, IGOS operating normally, and the control system performing continuous (or periodic) measurement of current values of parameters, evaluates the status of IGOS as normal.

On figb presents the situation when the measured values of the controlled parameters of one or more sea have a tendency to deterioration and out of the region of permissible values under the influence of destabilizing the system perturbation (for example, temperature increases, the moisture of the sea environment and so on). In this case, the MPO continues to function normally whenand. At the same time, the control system detects when increasing Δ_mthe deterioration in operating conditions IGOS, and whenthe control system will indicate the output parameters IGOS for normative values.

On FIGU presents the situation when the nature of changes in the m-th parameter of one of the same type sea is significantly different from the changes of this parameter for other similar sea. In this case, when reachingand even when values of the m-th parameter of the other sea limits, i.ethe control system detects an abnormal condition of one of the sea and will indicate rejected the E. of its characteristics from the set of values to achieve the level exceeding. This behavior is one of the sea may occur due to any destabilizing factors, affecting only the i-th element of the group of the same sea. For example, damage to the line with the subsequent decrease in the insulation resistance of the cable. At the same time may be short random jumps parameter values caused by non-systemic causes are averaged (the measurement interval ∆ T_uthat is greater than the average duration of fixed short-term spikes parameters) and practically does not affect the overall assessment of the state of IGOS. If the control system is used correlated, then it can be used directly to determine the parameter Δ_mijreduction of the correlation coefficient ΔR_mijvalues of the m-th parameter of the i-th sea relative to the rest of the same sea.

On Figg presents the situation when the value of the m-th parameter of one of the same type sea abruptly changed and continues random fluctuations within the normal range without any tendencies to further change. This behavior of the object parameter may indicate, for example, unauthorized connection to the line of any technical device, leading to an abrupt single parameter change. Ifthen barwith found the abnormality in accordance with a known method of control is not possible. In this case, such an abnormality is detected at. The sensitivity of the control system for anomalous single speed jumps is determined by the value ofand efficiency of their detection - inspection interval ∆ T_to. Timely detection of such characteristic can prevent unauthorized access to protected resources IGOS.

The possibility of obtaining a positive effect when using the proposed method was confirmed by comparative step-by-step modeling and comparison of the prototype method and the inventive method. The comparison results are presented in figure 3.

Technique step-by-step simulation was as follows. Asked the time interval ∆ T_tobetween i-th and (i+1)-th cycles of monitoring the status of IGOS. For each of the situations depicted on the respective figb, 2B, 2G, was executed cycles of the control method prototype and declared the way up until the control system has not been identified unacceptable deviation of the controlled parameters. For the situation illustrated in figb, prototype method and the claimed method allows equally to detect a fault when the achievement inequality. Figure 3 the fact of detection of unacceptable deviations denoted by EIT is om "+". Upon detecting such a condition of IGOS in the above example corresponds to the time t_K3=3Δt_to(see figb, figure 3). In the situation illustrated in figv, the claimed method allows to detect anomalous behavior of one parameter of the same sea as early as the second cycle of control when reaching conditions(see figv, time). At the same time when using the prototype method such abnormality can only be detected in the third cycle of the control at time(see figv), i.e. if the condition. In the situation illustrated in Figg, prototype method does not allow to detect a fault of the i-th sea state IGOS mistaken intact, and the claimed method allows to detect a fault whenafter the second cycle of control (see timeon Figg, figure 3).

Thus, the inventive method during 3 cycles of control identified all three types of deviations of the controlled parameter, while in the method - prototype only two. Moreover, the detection time in the method-prototype exceeds the time spent on the detection of similar deviations when using the inventive method. By proactively detecting the structure trend of deterioration in operating conditions IGOS may timely elimination of the causes of failure and increase the speed of detection and accuracy assessment IGOS, therefore, it becomes possible to achieve the technical result.

Method of condition monitoring multi-parameter object, namely, that pre-specify a set of N≥2 controlled structural elements of the object and M≥N parameters describing their condition, measure and memorize the measured control parameters of the structural elements of the multi-parameter object, define the characteristics of multivariable state of the object during its operation, the results of which are documented and take a decision on the status of multiparameter object, characterized in that it further for N identical structural elements of the multi-parameter object pre-set standard values of their parameters P^cmpermissible deviations Δ^apprparameters from the standard values and the maximum allowable differencebetween the maximumand minimumlevels measured m-th parameter, where m=1,2,...,M, the i-th and j-th similar structural elements, where i=1,2,...N and j=1,2,...N, and i≠j, and to determine the characteristics of the state of multivariable object, reads the previously stored values of parametersand P_{mi compare them with each other and calculate the difference Δimeasured values of the m-th parameter Pmifor all N the same type of structural elements from its standard valueread previously stored extreme valueandcompare them and the results of the comparison determine the difference
Δmijbetween the maximumand minimumlevels measured values of the m-th parameter, compare the calculated characteristics Δmiand Δmijwith their predefined valuesandand after documentation of measured parameters and calculated characteristics of the state of multivariable object accept as normal whenandand repeat the cycle parameter measurements, and whenand/orthe state of the object is characterized as abnormal and additionally document the information about the parameters and characteristics of the i-th structural element.}