Method of detecting faulty units in dynamic system |
|
                        
IPC classes for russian patent Method of detecting faulty units in dynamic system (RU 2453898):
 Method of searching for faulty module in dynamic system / 2451319
Number of dynamic modules of a controlled system is determined, the reaction of a good system is recorded at an interval at control points and integral estimates of output signals of the system are determined. System signals are transmitted to the first inputs of multiplier units; an arithmetic mean value of moduli of signal time derivatives is transmitted to second inputs of the multiplier units; output signals of the multiplier units are transmitted to inputs of integration units; integration is completed at a certain moment in time; output signal estimates obtained from integration are recorded; simultaneously, integral estimates of signals of models are determined for each of the control points, obtained as a result of test deviations of parameters of each of the modules, for which a corresponding test deviation of parameters is introduced into each model for the dynamic system module. Deviations of integral estimates of signals of the model, obtained as a result of test deviations of parameters of corresponding modules, are determined. Standardised values of deviations of integral estimates of signals of the model are determined. Standardised values of deviations of integral estimates of signals are determined. Diagnostic features are determined and a faulty module is determined from the minimum value of the diagnostic feature.
 Method for distributed monitoring and adaptive control of multilevel system and apparatus for realising said method / 2450335
Method for distributed monitoring and adaptive control of a multilevel system is based on the adaptive control system principle. During normal operation, a distributed structure is viewed by a wide on-line field having low-resolution which is sufficient for detecting a local disturbance. Further, the on-line field is narrowed in the vicinity of the disturbance which is viewed in more detail and through fine analysis, the nature of what is happening in is determined (recognition) in a specific decision block. The apparatus, which is in form of a decision unit for determining the state of the multilevel system, operates in three modes: current monitoring, monitoring error estimation and training.
Process parameter transmitter having acceleration sensor / 2450311
Process parameter transmitter for use in a system for controlling or monitoring a production process has a transmitter housing and a process parameter sensor, having a sensor output signal associated with the process parameter. The accelerometer is connected to the transmitter and provides an accelerometer output signal associated with acceleration. The diagnostic circuit provides a diagnostic output signal as a function of the sensor output signal and the accelerometer output signal.
 Method of testing aircraft pedal system and device to this end / 2450310
Proposed device comprises aircraft pedal actuator, transducer to measure force applied to aircraft pedal to actuate it and to generate signal corresponding to force applied, device to measure deviation of component in response to actuation of aircraft pedal, and control unit to process signals from force transducer and deviation meters to generate output signals indicating aforesaid deviation depending on force applied to aircraft pedal.
 Method of searching for faults in dynamic unit in continuous system / 2450309
Reaction of a properly operating system to an input action is first recorded on an interval at control points at discrete moments in time; output signals of a model for each of the control points obtained as a result of trial deviations of parameters of all units are determined, for which trial deviation is successively introduced into each transfer function parameter for all units of the dynamic system and output signals of the system are found for the same input action; the resultant output signals for each of the control points and each of the trial deviations at discrete moments in time are picked up; deviations of signals of the model, which are obtained as a result of trial deviations of corresponding parameters of all structural units from the reaction of the properly operating system are determined; the system with nominal characteristics is replaced with the controlled system; a similar test signal is transmitted to the input of the system; signals of the controlled system for control points at discrete moments in time are determined; deviations of signals of the controlled system for control points at discrete moments in time from nominal values are determined; diagnostic features for each of the parameters are determined from the relationship; a faulty parameter is determined from the minimum value of the diagnostic feature.
 Method of searching for faulty module in discrete dynamic system / 2444774
Unlike the existing method of searching for a faulty module in a continuous dynamic system, the reaction of a time-discrete system know to be properly functioning is recorded for discrete diagnosis cycles with pitch in control points; integral estimates of output signals are determined, for which at the moment of transmitting a test signal, discrete integration of signals is simultaneously initiated with pitch in each of the control points by transmitting signals to inputs of multiplier units, and a discrete exponential signal to second inputs of the multiplier units; output signals of the multiplier units are transmitted to inputs of the discrete integration units; integration is completed at the control instant; the obtained estimates are recorded; integral estimates of model signals are determined for each control point, for which a sample deviation of the parameter of the discrete transfer function is successively entered into each unit of the system and integral estimates of output signals of the system obtained as a result of integrating output signal estimates for each control point are found and each of the sample deviations is recorded; a defective module of the discrete system is determined by the minimum of the diagnostic feature.
 Determination method for dynamic parameters of marine movement mathematical model / 2442718
FIELD: ship navigation. SUBSTANCE: invention refers to ship navigation and can be used for forecasting the ship movements in the course of maneuvering. The fore and backward points are conditionally used. The fore and backwards points are located on the centerline plane of the ship. On a real time basis the coordinates of the fore and backward points are measured. Measurement of the coordinates is fulfilled with the help of the static shear stress receivers and with differential corrections. On the basis of the coordinate measurement results the current values of kinematic movement parameters are determined: linear speeds of the fore F (υf) and backward A (υa) points and their longitudinal (υfx, υax) and lateral (υfy, υay) components in the moving coordinates ZX0Y connected with the ship; longitudinal centre of the rotation (x0) in the moving coordinates ZX0Y connected with the ship; projection of the linear speed vector in the centre of gravity on the y axis 0Y (υy); linear speed of the ship centre of gravity (υ); curvature of the gravity path (R); angular rate of the ship (ω). The obtained results are used for calculation of the current values of the dynamic parameters of the marine movement mathematical model. On the basis of the mathematical model computer modeling is performed in order to forecast the ship movements in the course of maneuvering. EFFECT: improvement of the accuracy of forecasting of the ship movements in the course of maneuvering on the basis of an adequate mathematical model of its travel. 3 cl, 1 dwg  Method of operating industrial plant, and industrial plant control system / 2440596
Proposed method controlling certain number of plant operating parameters and process component parameters and stored in memory unit. Note here that fatigue index inherent in current state of component fatigue is defined. Note also that forecast fatigue is defined. Besides, component with maximum forecast fatigue is identified as drive component, while for multiple preset changes of states, defined is drive component forecast fatigue. Mind that proceeding from certain forecast fatigue values, one of state changes is selected and initiated.
 Method to search for faulty block in dynamic system / 2439648
Response of an admittedly faultless system is registered at a control interval in control points, several integral estimates are determined for output signals of the system for various integration parameters, the produced integral estimates of output signals are registered; several deviations are determined in integral estimates of model signals for each of control points received as a result of trial deviations of block parameters, for this purpose a trial deviation is introduced alternately in each unit of a dynamic model; deviations of model signal integral estimates are determined, produced as a result of trial deviations of structural block parameters; rated values are determined for deviation of integral estimates of model signals, produced as a result of trial deviations of appropriate block parameters; a system with rated characteristics is substituted with a controlled one, integral estimates of controlled system signals are determined for control points and several integration parameters, deviations are determined for integral estimates of controlled system signals for control points from rated values, rated values are determined for deviations of integral estimates of controlled system signals.
 Method to search for faulty block in continuous dynamic system / 2439647
As opposed to the available method of searching for a faulty block in a continuous dynamic system, elements of topological links are determined for each block included into the composition of the system for each control point Pji, j=1, …, k; j=1, …, m, elements Pji are determined from multiple values {-1,0,1}, the value -1 is determined, if the sign of signal transfer from the output of the i block to the j control point is negative, the value 0 is determined, if transfer of a signal from the output of the i block to the j control point is not available, the value 1 is determined, if the signal of signal transfer from the output of the i block to the j control point is positive, rated values are determined for elements of the vector of topological links for each block, diagnostic criteria are calculated, and using minimum value of a diagnostic criterion, the defect is determined.
|
FIELD: process engineering. SUBSTANCE: invention relates to controlling and diagnosing ACS and its components. Unlike know methods, m trial deviations are recorded as total amount of considered single and multiple faults of units to be loaded in turn into parameters of every unit for single defects or into parameters of combinations of units for multiple defects. Minimum of diagnostic indicators is used to define availability of single of multiple defect. EFFECT: expanded testing performances. 1 dwg
The invention relates to the field of control and diagnostics of automatic control systems and their elements. There is a method of diagnosing dynamic parts of control systems (patent RF №2110828, MKI6G05B 23/02, 1998), based on the integration of the output signal of the block with a weight of e-αtwhere α is a real constant. The disadvantage of this method is that its application for control of multiple units of a control system of an arbitrary structure leads to the necessity of integration of input and output signals of each of the controlled block. The closest technical solution (prototype) is a way of finding a bad block in a dynamic system (a positive decision from 12.07.2010, for a patent on the invention under the application №2009123999/08(033242), MKI6G05 23/02, 2010). The disadvantage of this method is that it provides only a single structural defects. Technical problem on which this invention is directed, is to expand the functionality of the method for finding one or more bad blocks (multiple defects) in a dynamic system with arbitrary connection. This object is achieved in that pre-register response known good f j(t), j=1, 2,...., k on the interval t∈[0, TTo] k control points, and define the integral evaluation of the output signals fj(α), j=1, ..., k system, which at the time of the test signal at the input of the system with nominal characteristics simultaneously begin the integration of the control signals in each of the k control points with weights e-αtwhere α= replace the system with a nominal characteristics controlled at the input of the system serves a similar test signal x(t), define the integral evaluation of the signals of the controlled system for the k control points Fj(α), j=1, ..., k for the parameter of integration α, determine the deviation of the integral of the estimated signals of the controlled system for the k control points from the nominal values ∆ Fj(α)=Fj(α)-Fj(α), j=1, ..., k, define the normalized Delta values of the integral estimates of the signals of the controlled system from the relationship determine diagnostic characteristics of ratio the minimum values of the diagnostic sign determines the ordinal number of the defective block or combination of blocks. The essence of the method consists in the following. The method is based on the use of pilot deviations of the model parameters of a continuous dynamical system. Using the vector interpretation of expression (3), we write it as follows where φi(α) is the angle between the normalized vector (a vector of unit length) of the variance of the integral estimates of the signals of the object Thus, the normalized diagnostic feature (3) represents the value of the square of the sine of an angle formed in the k-dimensional space (where k is the number of control points) of the normalized vectors of integral evaluations trial variance of the signal model and the variance of the integral of the estimated signals of the diagnostic object. Trial deviation parameter combination corresponding structural units that minimizes the value of the diagnostic sign (3), indicates the presence of a defect or a combination of defective blocks. The region of possible values of the diagnostic characteristic lies in the interval [0, 1]. Thus, the proposed method of searching for bad blocks is to be executed the following operations. 1. As dynamic systems consider a system consisting of randomly connected dynamic blocks, with the number of single and multiple defects of blocks m. 2. Pre-determine the testing time TTo≥TPPwhere TPP- time transition system. Transition time estimate for the nominal values of the parameters of the dynamic system. 3. Define the parameter integral transforms of the signals from the relation 4. Record the number of control points k. 5. Pre-determine the normalized vectors 6. Serves the test signal x(t) (unit step, linearly increasing, rectangular pulse and so on) to the input of the control system with a nominal characteristics. Fundamental limitations on the type of input test the impact of proposed method does not. 7. Record the response of the system fj(α), j=1, 2,..., k on the interval t∈[0, TTo] k control points and determine the integral OC the NCI output signals F j(α), j=1, ..., k system. To do this, at the time of the test signal to the input of the control system with the nominal characteristics simultaneously begin the integration of the control signals in each of the k control points with weights e-αtwhere 8. Define the integral evaluation signal models for each of the k control points obtained in the course of the trial deviations of each of the m single and multiple defects of blocks, which in turn, for each combination of parameters of different structural units of the dynamic system is administered trial deviation of these parameters, the transfer function and perform paragraphs 6 and 7 for the same test signal x(t). The resulting integration of the evaluation of the output signals for each of the k control points and each of the m test of variance Pji(α), j=1, ..., k; i=1, ..., m register. 9. Define the deviation integral OC the knock signal model, the resulting trial variance parameters of one or more structural units Δji(α)=Pjt(α)-Fj(α), j=1, ..., k; i=1, ..., m. 10. Define the normalized variance of the integral of the estimated signal model, the resulting trial variance parameters of one or more units according to the formula:
11. Replace the system with a nominal characteristics controlled. At the input of the system serves a similar test signal x(t). 12. Define the integral evaluation of the signals of the controlled system for the k control points Fj(α), j=1, ..., k, carrying out the operations described in paragraphs 6 and 7 with respect to the controlled system. 13. Determine the deviation of the integral of the estimated signals of the controlled system for the k control points from the nominal values ∆ Fj(α)=Fj(α)-Fj(α), j=1, ..., k. 14. Calculate the normalized values of the variance of the integral of the estimated signals of the controlled system by the formula:
15. Calculate the diagnostic signs of a defective block, or several blocks according to the formula (3). 16. The minimum values of the diagnostic characteristic to determine a defective unit or defective blocks. Consider the implementation of the proposed method of finding the inverse defect for the system, structural diagram of which is presented on the figure (see Fig. Block diagram of the diagnostic object). Transfer function block:
nominal parameter values: T1'=5; T1"=1; K2=1; T2=1; K3=1; T3=5 S. we Define variants (m=7) pilot deviations in the form of reduction of the gain (k1, ..., k3each dynamic block and combinations of blocks 10%: k1=0.9 (i=1; k2=0.9 (i=2); k3=0.9 (i=3); k1=0.9 and k2=0.9 (i=4); k2=0.9 and k3=0.9 (i=5); k1=0.9 and k3=0.9 (i=6); (k1=0.9, k2=0.9 and k3=0.9 (i=7). When searching for multiple defect in the form of deviation of the gain by 20% k1=0.8, k2=0.8 and k3=0.8 (multiple defect No. 7) in the first, second and third link in the supply speed of the test input signal of unit amplitude and the integral of the estimated signals for parameter α=0.5 and TTo=10, using three control points located at the outputs of blocks using a trial deviation to a value of 10%, the obtained values of diagnostic signs by the formula (3): J1=0.9262; J2=0.08897; J3=8552; J4=0,4849; J5=0.398; J6=07402; J7=0.03559. The analysis values of the diagnostic signs shows that multiple defect in p the moat, the second and third structural units of the controlled system is correct. It should be noted that the method is operable at larger values of the test deviations (10-40%). The limitation on the amount of trial deviation is the need to preserve the stability of models with a trial deviations. Search multiple defects according to the proposed method in relation to the diagnostic object, represented in the figure, consists of the following operations: 1. Record the number of controlled single and multiple defects m=7. 2. By analyzing the graphs of the nominal transient characteristics, determine the transition time of the system. For this example, the transition process is TPP=8 C. Record the time of control TTo≥TPP. For this example, TTo=10 S. 3. Define the parameter signal integration 4. Fixed reference point on the outputs of blocks: k=3. 5. Pre-find elements of vectors ΔPi(α) variance of the integral of the estimated signal model, the resulting trial deviations of all controlled single and multiple defects. The value trial variance is chosen equal to 10%. 6. Find inormirovanny the vectors 7. Replace the system with a nominal characteristics controlled, which entered deviations of the first, second and third blocks from the nominal 20%. At the input of the system serves a similar test signal x(t). 8. Determine the deviation of the integral of the estimated signals of the controlled system for the three control points from the nominal values ∆ Fj(α)=Fj(α)-Fj(α), j=1, 2, 3. 9. Calculate the normalized values of the variance of the integral of the estimated signals of the controlled system 10. Calculate the diagnostic signs of faulty units by the formula (3): J1=0.9262; J2=0.08897; J3=8552; J4=0.4849; J5=0.398; J6=0.7402; J7=0.03559, where J1indicates a defect in the first block, J2accordingly indicates a defect in the second, J3indicates a defect in the third, J4indicates defects in the first and second blocks, J5- defects in the second and third blocks, J6- defects in the first and third blocks, a J7respectively defects in the first, second and third blocks. 11. The minimum value of the automotive technician is a visitor sign determine the combination of multiple defect (in this case - No. 7). Modeling of processes of searching multiple defects in other cases its manifestations for the diagnostic object, with the same parameter integral transforms α and unit step input signal gives the following values of diagnostic features. In the presence of defects in the blocks # 1 and # 3 (in the form of reduced parameters k1and k320%multiple defect No. 6): J1=0.9973; J2=0.9474; J3=896; J4=0.9661; J5=0.8994; J6=0.1254; J7=0.7995. In the presence of defects in the blocks # 2 and # 3 (in the form of reduced parameters k2and k320%defect No. 5): J1=0.3599; J2=0.2114; J3=2875; J4=0.9719; J5=0.001142; J6=0.6733; J7=0.2007. In the presence of defects in the blocks # 1 and # 2 (in the form of reduced parameters k1and k220%defect No. 4): J1=0.7204; J2=0.7441; J3=0.7929; J4=0.009458; J5=0.9985; J6=0.9989; J7=0.7533. We show that this method is efficient for searching a single structural defects. If there is a defect in the unit 3 (in the form of reducing the parameter k320%defect No. 3): J1=0.07426; J2=0.7469; J3=0; J4=0.8629; J5=0.2574; J6=0.5945; J7=0.7014. If there is a defect in the unit 2 (in the form of reducing the parameter k220%defect No. 2): J1=0.7842; J2=0; J3=0.747; J4=0.6549; J5=0.2397; J =0.8593; J7=0.05451. If there is a defect in the unit 1 (in the form of reducing the parameter k120%defect No. 1): J1=0; J2=0.7841; J3=0.07425; J4=0.8032; J5=0.3313; J6=0.8379; J7=0.8003. The minimum value of the diagnostic sign in all cases correctly indicates bad blocks. The way of finding bad blocks in a dynamic system, based on the fact that record the number of dynamic elements in the system, determine the testing time TTo≥TPPdetermine the parameter integral transforms of the signals from the relation
|
