CN108667673B - Nonlinear network control system fault detection method based on event trigger mechanism - Google Patents

Nonlinear network control system fault detection method based on event trigger mechanism Download PDF

Info

Publication number
CN108667673B
CN108667673B CN201810652991.XA CN201810652991A CN108667673B CN 108667673 B CN108667673 B CN 108667673B CN 201810652991 A CN201810652991 A CN 201810652991A CN 108667673 B CN108667673 B CN 108667673B
Authority
CN
China
Prior art keywords
fault
fuzzy
fault detection
network control
control system
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN201810652991.XA
Other languages
Chinese (zh)
Other versions
CN108667673A (en
Inventor
王迎春
郑龙飞
杨东升
庞萌萌
谷永强
王占山
会国涛
刘振伟
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Northeastern University China
Original Assignee
Northeastern University China
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Northeastern University China filed Critical Northeastern University China
Priority to CN201810652991.XA priority Critical patent/CN108667673B/en
Publication of CN108667673A publication Critical patent/CN108667673A/en
Application granted granted Critical
Publication of CN108667673B publication Critical patent/CN108667673B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
    • H04L41/06—Management of faults, events, alarms or notifications
    • H04L41/0631—Management of faults, events, alarms or notifications using root cause analysis; using analysis of correlation between notifications, alarms or events based on decision criteria, e.g. hierarchy, tree or time analysis
    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
    • G05B23/00—Testing or monitoring of control systems or parts thereof
    • G05B23/02—Electric testing or monitoring
    • G05B23/0205—Electric testing or monitoring by means of a monitoring system capable of detecting and responding to faults
    • G05B23/0218—Electric testing or monitoring by means of a monitoring system capable of detecting and responding to faults characterised by the fault detection method dealing with either existing or incipient faults
    • G05B23/0243—Electric testing or monitoring by means of a monitoring system capable of detecting and responding to faults characterised by the fault detection method dealing with either existing or incipient faults model based detection method, e.g. first-principles knowledge model
    • G05B23/0254—Electric testing or monitoring by means of a monitoring system capable of detecting and responding to faults characterised by the fault detection method dealing with either existing or incipient faults model based detection method, e.g. first-principles knowledge model based on a quantitative model, e.g. mathematical relationships between inputs and outputs; functions: observer, Kalman filter, residual calculation, Neural Networks
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
    • H04L41/06—Management of faults, events, alarms or notifications
    • H04L41/0695—Management of faults, events, alarms or notifications the faulty arrangement being the maintenance, administration or management system
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
    • H04L41/14—Network analysis or design
    • H04L41/145—Network analysis or design involving simulating, designing, planning or modelling of a network

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Physics & Mathematics (AREA)
  • Artificial Intelligence (AREA)
  • Evolutionary Computation (AREA)
  • Mathematical Physics (AREA)
  • General Physics & Mathematics (AREA)
  • Automation & Control Theory (AREA)
  • Data Exchanges In Wide-Area Networks (AREA)

Abstract

The invention provides a fault detection method of a nonlinear network control system based on an event trigger mechanism, and relates to the technical field of network system fault detection. Firstly, establishing a T-S fuzzy model of a nonlinear network control system, setting event trigger conditions, establishing a fuzzy fault detection filter model, establishing a fault weighting system and further establishing a fault detection system model; selecting a proper residual error evaluation function and a proper detection threshold value according to the fault detection system model, and detecting whether the fault of the nonlinear network control system occurs or not; and finally, further designing a parameter matrix and an event trigger matrix of the fault detection filter according to the stability of the fault detection system and the sufficient conditions of the fault detection filter. The fault detection method of the nonlinear network control system based on the event trigger mechanism greatly improves the robustness to external disturbance and communication delay, and the application of the event trigger mechanism can save limited network resources and computing resources.

Description

Nonlinear network control system fault detection method based on event trigger mechanism
Technical Field
The invention relates to the technical field of network system fault detection, in particular to a nonlinear network control system fault detection method based on an event trigger mechanism.
Background
The network control system has gained wide attention in complex industrial control systems due to the advantages of low installation and maintenance cost, high safety and reliability, flexible communication structure and the like. In a network control system, sensors, actuators and controllers are interconnected by a shared communication network. With the increasing requirements of network control systems on safety, stability and high performance, the problem of fault detection for network control systems becomes an important research field.
Due to the introduction of a communication network and the characteristics of a network control system, new problems and challenges are inevitably brought to the network control system, such as communication delay, data packet loss, data misordering, limited bandwidth and the like. At present, most of research results of network control systems are design methods of controllers and filters aiming at the problems of time lag, packet loss, disorder and the like of the systems, and the problems of fault diagnosis of the network control systems are relatively few. In addition, most of research results on fault detection of network control systems are linear systems as research objects, but most of industrial systems and actual systems in life are nonlinear, so that the fault detection problem of the nonlinear network control system has very important theoretical research value and practical application prospect.
For the fault detection problem of the nonlinear network control system, a time-triggered method is mostly adopted, but in the actual working process, not all sampling data and measurement output need to be transmitted. Therefore, the time trigger easily causes the waste of limited network bandwidth, and further aggravates the occurrence of network induced delay and data packet loss. Event-triggered communication mechanisms have received considerable attention in order to reduce the transmission of "unnecessary" data in the network while ensuring the desired system performance. The basic idea of the event-triggered strategy is that the sampled data will only be transmitted when a pre-set threshold is met. The fault detection of the nonlinear network control system based on the event trigger communication mechanism can not only timely and accurately detect whether the fault occurs, but also save limited network resources and meet the development trend of fault detection.
Disclosure of Invention
Aiming at the defects of the prior art, the invention provides a fault detection method of a nonlinear network control system based on an event trigger mechanism, which realizes the detection of the fault of the nonlinear network control system.
The fault detection method of the nonlinear network control system based on the event trigger mechanism comprises the following steps:
step 1, carrying out modeling analysis on a nonlinear network control system with process faults, sensor faults and output disturbance by using a Takagi-Sugeno (namely T-S) fuzzy model method, and establishing a T-S fuzzy model of the nonlinear network control system;
the fuzzy rule used by the fuzzy model method is as follows:
Rule i：IF z1(t)is Mi1(z)and...and zp(t)is Mip(z)，THEN
Figure BDA0001704627830000021
wherein i is a fuzzy rule number, and z (t) ═ z1(t)，z2(t)，...，zp(t)]For the front-piece variable containing the state quantity information in the nonlinear network control system, p is the number of the front-piece variables, MijIs a fuzzy set, j is 1, 2, …, p, x (t) is a state variable of the nonlinear network control system, y (t) is a measurement output, omega (t) is an external disturbance, f (t) is a fault signal detected by a sensor, Ai，Di，Fi，Ci，Ei，GiA matrix of known suitable dimensions;
the established T-S fuzzy model of the nonlinear network control system is as follows:
Figure BDA0001704627830000022
where r is the number of fuzzy rules, wi(z (t)) is a membership function of the nonlinear network control system;
step 2, setting an event trigger condition, and determining whether the measurement output of the sensor should be transmitted to a filter according to the event trigger condition;
the set event trigger conditions are as follows:
Figure BDA0001704627830000023
where h is the sampling time interval of the sensor, ikh is the transmission time of the sampling signal of the sensor, ik+1h is the time at which the next sampled signal is transmitted, > 0 is the weighting matrix that needs to be designed, ek(ikh+jh)＝y(ikh+jh)-y(ikh) For threshold error, ε is the event trigger parameter, y (i)kh) The measured output of the sampled signal at a time on the sensor, y (i)kh + jh) is the current measurement output of the sensor;
sampling the signal y (i) at the previous momentkh) When the sampling signal is transmitted, the sampling signal is transmitted to the filter only when the current sampling signal meets the trigger condition;
step 3, establishing a fuzzy fault detection filter model by using a T-S fuzzy model method;
the fuzzy rule used for modeling the fuzzy fault detection filter by using the T-S fuzzy model method is as follows:
Rulej：IF z1(ikh)is Nj1 and...and zp(ikh)is Njp，THEN
Figure BDA0001704627830000031
wherein x isf(t) is the state vector of the fuzzy fault detection filter,
Figure BDA0001704627830000037
as the true input to the blurring filter, zf(t) is a residual signal, Afj，Bfj，CfjAnd DfjFor the gain matrix of the fuzzy filter to be designed, j is the fuzzy rule number, NjiTo blur the set, z (i)kh)＝[z1(ikh)，z2(ikh)，...，zp(ikh)]Detecting a front-piece variable of a filter for the fuzzy fault;
the model of the established fuzzy fault detection filter is as follows:
Figure BDA0001704627830000032
wherein, wj(z(ikh) Is a membership function of the fault detection filter;
step 4, establishing a fault weighting system capable of improving the design freedom of the fault detection system;
the fault weighting system is shown in the following formula:
Figure BDA0001704627830000033
wherein f(s) is a fault signal, W(s) is a weighting matrix,
Figure BDA0001704627830000034
is a weighted fault signal;
the state space form of the fault signal f(s) and the weighting matrix w(s) is shown as follows:
Figure BDA0001704627830000035
wherein x isw(t) is a state space vector, fw(t) is the weighted fault signal, Aw，Bw，CwAnd DwIs a constant matrix;
step 5, establishing a fault detection system model according to a T-S fuzzy model of the nonlinear network control system, an event trigger condition, a T-S fuzzy model of a filter and a fault weighting matrix;
the established fault detection system model is shown as the following formula:
Figure BDA0001704627830000036
wherein the content of the first and second substances,
Figure BDA0001704627830000041
in order to detect the systematic residual error for a fault,
Figure BDA0001704627830000042
Figure BDA0001704627830000043
Figure BDA0001704627830000044
Figure BDA0001704627830000045
step 6, selecting a proper residual error evaluation function and a proper detection threshold value according to the fault detection system model, and detecting whether the fault of the nonlinear network control system occurs or not by comparing the values of the residual error evaluation function and the detection threshold value;
the residual evaluation function H (z)f) And a detection threshold JthAs shown in the following equation:
Figure BDA0001704627830000046
based on the residual evaluation function and the detection threshold, whether the fault of the nonlinear network control system occurs is judged according to the following relation:
Figure BDA0001704627830000047
when the residual evaluation function is larger than the detection threshold, the nonlinear network control system has a fault, and the fault detection system gives an alarm; otherwise, the nonlinear network control system works normally without alarming;
step 7, constructing a fuzzy Lyapunov function, obtaining sufficient conditions of the stability of the fault detection system and the existence of the fault detection filter by utilizing a Lyapunov stability theory, a related theory and a linear matrix inequality, and further designing a parameter matrix A of the fault detection filterfj，Bfj，CfjAnd DfjAnd an event trigger matrix Φ.
According to the technical scheme, the invention has the beneficial effects that: the method for detecting the fault of the nonlinear network control system based on the event trigger mechanism simultaneously considers the influences of process faults, sensor faults and external disturbance on the network control system in the modeling process of the nonlinear network control system. In the fault diagnosis process, the problems of communication delay, data packet loss, data disorder and the like in a network system are considered and solved. The front-piece variable of the filter fuzzy model is asynchronous with the front-piece variable of the network control system fuzzy model, so that the flexibility of filter design can be improved, and the cost is reduced; the application of the fuzzy Lyapunov function reduces the conservatism of the system. Compared with the prior art, the event trigger fuzzy H designed by the invention∞The filter carries out fault detection, on one hand, the technology not only greatly improves the fault sensitivity of the nonlinear network control system, but also has stronger robustness to external disturbance and data packet loss, and can effectively solve the fault detection problem of the nonlinear network control system; on the other hand, the introduction of the event trigger communication mechanism can effectively reduce the use of network bandwidth, save limited network resources and simultaneously save computing resources.
Drawings
Fig. 1 is a flowchart of a method for detecting a fault of a nonlinear network control system based on an event trigger mechanism according to an embodiment of the present invention;
FIG. 2 is a diagram illustrating a residual signal according to an embodiment of the present invention;
FIG. 3 is a schematic diagram of a residual evaluation function and a detection threshold varying with time according to an embodiment of the present invention;
FIG. 4 is a schematic diagram of the residual evaluation function with time variation when there is a fault and no fault according to an embodiment of the present invention;
fig. 5 is a schematic diagram illustrating a relationship between a release time and a release interval of an event trigger policy according to an embodiment of the present invention.
Detailed Description
The following detailed description of embodiments of the present invention is provided in connection with the accompanying drawings and examples. The following examples are intended to illustrate the invention but are not intended to limit the scope of the invention.
In this embodiment, the fault detection method for the nonlinear network control system based on the event trigger mechanism of the present invention is used to perform fault detection on the nonlinear network control system.
The method for detecting the fault of the nonlinear network control system based on the event trigger mechanism, as shown in fig. 1, includes the following steps:
step 1: modeling and analyzing a nonlinear network control system with process faults, sensor faults and output disturbance by using a Takagi-Sugeno (T-S) fuzzy model method to establish a T-S fuzzy model of the nonlinear network control system;
the fuzzy rule used by the fuzzy model method is as follows:
Rule i：IF z1(t)is Mi1(z)and...and zp(t)is Mip(z)，THEN
Figure BDA0001704627830000051
wherein i is a fuzzy rule number, and z (t) ═ z1(t)，z2(t)，...，zp(t)]For the front-piece variable containing the state quantity information in the nonlinear network control system, p is the number of the front-piece variables, MijJ is 1, 2, …, p, x (t) is a state variable of the nonlinear network control system,y (t) is the measurement output, ω (t) is the external disturbance, f (t) is the fault signal detected by the application sensor, Ai，Di，Fi，Ci，Ei，GiA matrix of known suitable dimensions;
the established T-S fuzzy model of the nonlinear network control system is as follows:
Figure BDA0001704627830000061
where r is the number of fuzzy rules, wi(z (t)) is a membership function of the nonlinear network control system;
the T-S fuzzy model is a nonlinear network control system model including process faults, sensor faults and external disturbances, and is a more extensive form. Order fault signal
Figure BDA0001704627830000063
Wherein f isp(t) and fs(t) represents process and sensor faults, respectively; definition Fi＝[F pi0| and Gi＝[0 Gsi]Wherein F ispiAnd GsiRespectively, a coefficient matrix for the fault. When F is presenti＝[Fpi 0]，Gi＝[0 0]Then, a fault signal f (T) in the established T-S fuzzy model represents a process fault; when F is presenti＝[0 0]，Gi＝[0 Gsi]And (3) establishing a fault signal f (T) in the T-S fuzzy model to represent the sensor fault. The fault signals in this embodiment include sensor faults and process faults.
Step 2: setting an event trigger condition, and determining whether the measurement output of the sensor should be transmitted to the filter according to the event trigger condition;
the set event trigger conditions are as follows:
Figure BDA0001704627830000062
whereinH is the sampling interval of the sensor, ikh is the transmission time of the sampling signal of the sensor, ik+1h is the time at which the next sampled signal is transmitted, > 0 is the weighting matrix that needs to be designed, ek(ikh+jh)＝y(ikh+jh)-y(ikh) For threshold error, ε is the event trigger parameter, y (i)kh) The measured output of the sampled signal at a time on the sensor, y (i)kh + jh) is the current measurement output of the sensor;
sampling the signal y (i) at the previous momentkh) When the sampling signal is transmitted, the sampling signal is transmitted to the filter only when the current sampling signal meets the trigger condition;
applying event trigger conditions to the fault detection process, the next trigger moment is related not only to the selected trigger parameter, but also to the measurement output of the sampler transmitted at the previous moment. The use of the event trigger mechanism can effectively reduce data transmission and save limited network resources. At the same time, the event trigger condition only needs to calculate the threshold value, i.e. ε y, at each trigger momentT(ikh)Φy(ikh) At a time interval (i)kh，ik+1h]The event trigger mechanism saves network resources and computing resources at the same time.
And step 3: considering the influence of an event triggering mechanism and network induced delay, establishing a fuzzy fault detection filter model by using a T-S fuzzy model method;
the fuzzy rule used for modeling the fuzzy fault detection filter by using the T-S fuzzy model method is as follows:
Rule j：IF z1(ikh)is Nj1 and...and zp(ikh)is Njp，THEN
Figure BDA0001704627830000071
wherein x isf(t) is the state vector of the fuzzy fault detection filter,
Figure BDA0001704627830000076
as the true input to the blurring filter, zf(t) is a residual signal, Afj，Bfj，CfjAnd DfjFor the gain matrix of the fuzzy filter to be designed, j is the fuzzy rule number, NjiTo blur the set, z (i)kh)＝[z1(ikh)，z2(ikh)，...，zp(ikh)]Detecting a front-piece variable of a filter for the fuzzy fault;
the established fuzzy fault detection filter model is as follows:
Figure BDA0001704627830000072
wherein, wj(z(ikh) Is a membership function of the fault detection filter;
the fault detection filter is used to generate a residual signal, which is important in fault detection systems and is used to determine whether a fault has occurred in the current network control system. Unlike the traditional parallel distributed compensation strategy, the front-end variable z (i) of the fault detection filterkh) Is not equal to the network control system's predecessor variable z (t). Therefore, the fault detection filter adopting the asynchronous front-element variable can improve the flexibility of design and reduce the design cost in practical application.
And 4, step 4: establishing a fault weighting system capable of improving the design freedom of a fault detection system;
the fault weighting system is shown in the following formula:
Figure BDA0001704627830000073
wherein f(s) is a fault signal, W(s) is a weighting matrix,
Figure BDA0001704627830000074
is a weighted fault signal;
the state space form of the fault signal f(s) and the weighting matrix w(s) is shown as follows:
Figure BDA0001704627830000075
wherein x isw(t) is a state space vector, fw(t) is the weighted fault signal, Aw，Bw，CwAnd DwIs a matrix of known constants;
step 5, establishing a fault detection system model according to a T-S fuzzy model of the nonlinear network control system, an event trigger condition, a T-S fuzzy model of a filter and a fault weighting matrix;
the established fault detection system model is shown as the following formula:
Figure BDA0001704627830000081
wherein the content of the first and second substances,
Figure BDA0001704627830000082
in order to detect the systematic residual error for a fault,
Figure BDA0001704627830000083
Figure BDA0001704627830000084
Figure BDA0001704627830000085
Figure BDA0001704627830000086
in the fault detection system, network induced delays, process faults, sensor faults and external disturbances are simultaneously included. In practical application, the faults and the disturbances are most likely to exist simultaneously, and the network-induced delay is inevitable during the network transmission process of data.
Step 6, selecting a proper residual error evaluation function and a proper detection threshold value according to the fault detection system model, and detecting whether the fault of the nonlinear network control system occurs or not by comparing the values of the residual error evaluation function and the detection threshold value;
residual evaluation function H (z)f) And a detection threshold JthAs shown in the following equation:
Figure BDA0001704627830000087
based on the residual evaluation function and the detection threshold, whether the fault of the nonlinear network control system occurs is judged according to the following relation:
Figure BDA0001704627830000088
when the residual evaluation function is larger than the detection threshold, the nonlinear network control system has a fault, and the fault detection system gives an alarm; otherwise, the nonlinear network control system is normal and does not give an alarm. In the selection process of the detection threshold, the maximum value of the residual error evaluation function under the condition of no fault is selected as the detection threshold, namely when only disturbance exists in the system, no matter how large the disturbance exists, the fault detection system does not give an alarm. Thus, the fault detection system can accurately detect faults in the environment with disturbance.
And 7: constructing a fuzzy Lyapuonv function, obtaining sufficient conditions of the stability of a fault detection system and the existence of a fault detection filter by utilizing a Lyapunov stability theory, a related lemma and a linear matrix inequality, and further designing parameters and an event trigger matrix of the fault detection filter;
step 7.1: constructing a fuzzy Lyapunov function V (t), and deriving the fuzzy Lyapunov function V (t) with respect to time to obtain stable fault detection system and fault detection filter memoryUnder sufficient conditions, i.e. 1) when
Figure BDA0001704627830000097
The filter residual error system is progressively stable; 2) under the zero initial condition, the filter residual error signal satisfies
Figure BDA0001704627830000098
Wherein gamma > 0 is H∞The level of the attenuation is such that,
Figure BDA0001704627830000091
the constructed fuzzy Lyapunov function is as follows:
Figure BDA0001704627830000092
wherein the content of the first and second substances,
Figure BDA0001704627830000093
for the fault detection problem of the nonlinear network control system, a known positive real number phi is assumedk(k ═ 1.., r) satisfies
Figure BDA0001704627830000099
At a given constant γ, τ1，τ3ε and the known filter gain matrix Afj，Bfj，Cfj，DfjIn the case of (2), the fault detection system is asymptotically stable and satisfies H∞Performance, if and only if there is a symmetric matrix Z1 and a symmetric positive definite matrix Pk＞0，Φ＞0，M＞0，Rκ＞0，
Figure BDA0001704627830000094
So that the following inequality is true,
Figure BDA0001704627830000095
Figure BDA0001704627830000096
wherein the content of the first and second substances,
Figure BDA0001704627830000101
Figure BDA0001704627830000102
H1＝[I 0 0]，
Figure BDA0001704627830000103
Ξ23＝(3-m)(R2-S2)，
Figure BDA0001704627830000104
Ξ25＝(m-2)(R3-S3)，
Figure BDA0001704627830000105
Ξ33＝Q1-M-R1-R2，Ξ34＝Q3+(3-m)S2+(m-2)R2，Ξ44＝Q2-Q1-R2-R3，Ξ45＝(3-m)R3+(m-2)S3-Q3，Ξ55＝-Q2-R3，Ξ66＝(ε-1)Φ，
Figure BDA0001704627830000106
Ξ77＝-γ2I，Ξ88＝-γ2I，D1i＝[Di Fi]，E1i＝[Ei Gi]，
Figure BDA0001704627830000107
Figure BDA0001704627830000108
Figure BDA0001704627830000109
Ψ22＝diag{-R1，-R2，-R3，-εΦ，-I}。
step 7.2: designing a parameter matrix and an event trigger matrix phi of the fault detection filter;
definition of
Figure BDA00017046278300001010
Wherein the content of the first and second substances,
Figure BDA00017046278300001011
the inverse matrix is further defined as follows:
Figure BDA00017046278300001012
by J2Multiplying diag { J, I, I, I, I, I, I, I, I } by formula (10) in front of
Figure BDA00017046278300001015
After multiplying by (10), then define
Xk＝U1k，
Figure BDA00017046278300001013
Figure BDA00017046278300001014
Thereby obtaining
Figure BDA0001704627830000111
Wherein the content of the first and second substances,
Figure BDA0001704627830000112
Figure BDA0001704627830000113
Figure BDA0001704627830000114
Figure BDA0001704627830000115
Υ66＝(ε-1)Φ，
Figure BDA0001704627830000116
Υ77＝Υ88＝diag{-γ2I，-γ2I}，
Figure BDA0001704627830000117
Figure BDA0001704627830000118
Figure BDA0001704627830000119
H3＝[0 I]，
F22＝diag{-R1 -R2 -R3- ε Φ -I }. Positive real number phik(k ═ 1.., r) satisfies
Figure BDA00017046278300001110
τ1，τ3ε, h, γ are known constants, Z2Is a symmetric matrix, U1l＞0，U1k＞0，Xk＞0，V＞0，Y＞0，M＞0，RkMore than 0(k is 1, 2, 3) is a symmetrical positive definite matrix,
Figure BDA00017046278300001111
since U (k) > 0, according to Schur supplement theory,
Xk-Y＞0 (13)
according to the formula (11), the following is obtained:
Figure BDA00017046278300001112
in the same way as the step 7.1, the following products are obtained:
Figure BDA0001704627830000121
it follows that if conditions (12) - (14) have a feasible solution, the fault detection filter can ensure that the fault detection system is asymptotically stable and satisfies H∞Performance;
define from
Figure BDA0001704627830000122
To zf(t) a mapping operator of
Figure BDA0001704627830000123
Further obtain
Figure BDA0001704627830000124
Wherein
Figure BDA0001704627830000125
Replacing the corresponding matrix in (11) with the matrix in (15) to obtain
Figure BDA0001704627830000126
Thus, H∞Fault detection filterParameters of the wave filter are
Figure BDA0001704627830000127
To date, the design of event-triggered fault detection filters has been completed.
Step 7.3 provides sufficient conditions for jointly designing the event trigger matrix phi and the fuzzy fault detection filter, and if feasible solutions exist in Linear Matrix Inequalities (LMIs) (12) to (14), the fault detection filter parameter Afj，Bfj，Cfj，DfjAnd the event trigger matrix Φ is obtained. And solving the linear matrix inequality through an LMI toolkit in the MATLAB to obtain the parameters and the event trigger matrix of the fault detection filter.
The parameter settings of the nonlinear network control system adopted in the present embodiment are as follows:
Figure BDA0001704627830000128
C1＝[1 0]，C2＝[1 0]，E1＝0.5，E2＝0.5，G1＝0.1，G2＝0.1。
the membership function of the fuzzy device is selected as:
Figure BDA0001704627830000129
w2(z(t))＝1-w1(z (t)); the parameters of the fault weighting system are set as: a. thew＝0.1，Bw＝0.25，Cw＝0.2，Dw＝0.65；
The external disturbance signal ω (t) is set to:
Figure BDA00017046278300001210
meanwhile, the fault signal is as follows:
Figure BDA0001704627830000131
let the sampling interval h equal to 10ms, τ1＝0.002，τ3＝0.2，φ1＝φ20.1, 0.1. By calculation, H∞The minimum value of the attenuation level γ is 0.6501. Using the LMI toolbox in MATLAB, let γ equal to 1, the corresponding event-triggered weighting matrix Φ equal to 3.7474 is obtained. Let initial state x (0) be [ 00 ]]T，xf(0)＝[0 0]T，xw(0) In this embodiment, the residual signal of the failure detection system is shown in fig. 2, and the changes over time of the residual evaluation function and the detection threshold are shown in fig. 3, and the detection threshold is selected for the failure detection system
Figure BDA0001704627830000132
The simulation result shows that the simulation results show that,
Figure BDA0001704627830000133
it follows that the fault was detected 0.8s after the occurrence.
The present embodiment also provides a graph of the change over time of the residual evaluation function at fault and no fault as shown in fig. 4, which demonstrates that the residual signal not only can detect whether a fault occurs, but also can distinguish the influence of the fault and disturbance on the system. The present embodiment also provides the release time and the release interval of the event trigger strategy as shown in fig. 5, and it can be seen from the figure that within 40s of the simulation time, the event trigger is triggered only 179 times, and the number of triggers is significantly reduced compared to 2000 times of time trigger. When a fault occurs, event triggering is significantly increased, and the reliability of fault detection is ensured. Therefore, the designed fault detection filter can not only detect faults in time, but also effectively reduce the use of limited bandwidth.
Finally, it should be noted that: the above examples are only intended to illustrate the technical solution of the present invention, but not to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some or all of the technical features may be equivalently replaced; such modifications and substitutions do not depart from the spirit of the corresponding technical solutions and scope of the present invention as defined in the appended claims.

Claims (3)

1. A fault detection method of a nonlinear network control system based on an event trigger mechanism is characterized in that: the method comprises the following steps:
step 1, carrying out modeling analysis on a nonlinear network control system with process faults, sensor faults and output disturbance by using a T-S fuzzy model method, and establishing a T-S fuzzy model of the nonlinear network control system;
the fuzzy rule used by the fuzzy model method is as follows:
Rule i:IF z1(t)is Mi1(z)and…and zp(t)is Mip(z)，THEN
Figure FDA0003292575550000011
wherein i is a fuzzy rule number, and z (t) ═ z1(t),z2(t),...,zp(t)]For the front-piece variable containing the state quantity information in the nonlinear network control system, p is the number of the front-piece variables, MimFor fuzzy set, m is 1, 2, …, p, x (t) is state variable of nonlinear network control system, y (t) is measurement output, w (t) is external disturbance, f (t) is fault signal detected by sensor, Ai,Di,Fi,Ci,Ei,GiA matrix of known suitable dimensions;
the established T-S fuzzy model of the nonlinear network control system is as follows:
Figure FDA0003292575550000012
where r is the number of fuzzy rules, wi(z (t)) is a membership function of the nonlinear network control system;
step 2, setting an event trigger condition, and determining whether the measurement output of the sensor should be transmitted to a filter according to the event trigger condition;
the set event trigger conditions are as follows:
Figure FDA0003292575550000013
where h is the sampling time interval of the sensor, ikh is the transmission time of the sampling signal of the sensor, ik+1h is the time at which the next sampled signal is transmitted, Φ>0 is the event trigger matrix that needs to be designed, ek(ikh+lh)＝y(ikh+lh)-y(ikh) For threshold error, ε is the event trigger parameter, y (i)kh) The measured output of the sampled signal at a time on the sensor, y (i)kh + lh) is the current measurement output of the sensor;
sampling the signal y (i) at the previous momentkh) When the sampling signal is transmitted, the sampling signal is transmitted to the filter only when the current sampling signal meets the trigger condition;
step 3, establishing a fuzzy fault detection filter model by using a T-S fuzzy model method;
the fuzzy rule used for modeling the fuzzy fault detection filter by using the T-S fuzzy model method is as follows:
Rule j:IF z1(ikh)is Nj1 and…and zp(ikh)is Njp，THEN
Figure FDA0003292575550000021
wherein x isf(t) is the state vector of the fuzzy fault detection filter,
Figure FDA0003292575550000025
as the true input to the blurring filter, zf(t) is a residueDifference signal, Afj,Bfj,CfjAnd DfjFor the gain matrix of the fuzzy filter to be designed, j is the fuzzy rule number, NjmTo blur the set, z (i)kh)＝[z1(ikh),z2(ikh),...,zp(ikh)]Detecting a front-piece variable of a filter for the fuzzy fault;
the model of the established fuzzy fault detection filter is as follows:
Figure FDA0003292575550000022
wherein, wj(z(ikh) Is a membership function of the fault detection filter;
step 4, establishing a fault weighting system capable of improving the design freedom of the fault detection system;
step 5, establishing a fault detection system model according to a T-S fuzzy model of the nonlinear network control system, an event trigger condition, a T-S fuzzy model of a filter and a fault weighting matrix;
step 6, selecting a proper residual error evaluation function and a proper detection threshold value according to the fault detection system model, and detecting whether the fault of the nonlinear network control system occurs or not by comparing the values of the residual error evaluation function and the detection threshold value;
when the residual evaluation function is larger than the detection threshold, the nonlinear network control system has a fault, and the fault detection system gives an alarm; otherwise, the nonlinear network control system works normally without alarming;
step 7, constructing a fuzzy Lyapunov function, obtaining sufficient conditions of the stability of the fault detection system and the existence of the fault detection filter by utilizing a Lyapunov stability theory, a related theory and a linear matrix inequality, and further designing a gain matrix A of the fault detection filterfj,Bfj,CfjAnd DfjAnd an event trigger matrix Φ.
2. The method for detecting the fault of the nonlinear network control system based on the event trigger mechanism according to claim 1, wherein: step 4, the fault weighting system is shown as the following formula:
Figure FDA0003292575550000023
wherein f(s) is a fault signal, W(s) is a weighting matrix,
Figure FDA0003292575550000024
is a weighted fault signal;
the state space form of the weighting matrix w(s) is shown by the following equation:
Figure FDA0003292575550000031
wherein x isw(t) is a state space vector, fw(t) is the weighted fault signal, Aw，Bw，CwAnd DwIs a constant matrix.
3. The method for detecting the fault of the nonlinear network control system based on the event trigger mechanism according to claim 2, wherein: the fault detection system model established in the step 5 is shown as the following formula:
Figure FDA0003292575550000032
wherein the content of the first and second substances,
Figure FDA0003292575550000033
in order to detect the systematic residual error for a fault,
Figure FDA0003292575550000034
Figure FDA0003292575550000035
Figure FDA0003292575550000036
Figure FDA0003292575550000037
CN201810652991.XA 2018-06-22 2018-06-22 Nonlinear network control system fault detection method based on event trigger mechanism Active CN108667673B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201810652991.XA CN108667673B (en) 2018-06-22 2018-06-22 Nonlinear network control system fault detection method based on event trigger mechanism

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201810652991.XA CN108667673B (en) 2018-06-22 2018-06-22 Nonlinear network control system fault detection method based on event trigger mechanism

Publications (2)

Publication Number Publication Date
CN108667673A CN108667673A (en) 2018-10-16
CN108667673B true CN108667673B (en) 2022-02-22

Family

ID=63772886

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201810652991.XA Active CN108667673B (en) 2018-06-22 2018-06-22 Nonlinear network control system fault detection method based on event trigger mechanism

Country Status (1)

Country Link
CN (1) CN108667673B (en)

Families Citing this family (19)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110045716B (en) * 2019-04-18 2020-08-11 中广核工程有限公司 Method and system for detecting and diagnosing early fault of closed-loop control system
CN110161882B (en) * 2019-06-12 2020-09-18 江南大学 Fault detection method of networked system based on event trigger mechanism
CN110673611B (en) * 2019-10-21 2021-06-08 武汉理工大学 Under-actuated unmanned ship control method based on event triggering scheme and T-S fuzzy system
CN110703742B (en) * 2019-11-08 2022-11-18 哈尔滨工业大学 Event-driven and output quantization-based fault detection method for unmanned surface vehicle system
CN111884217B (en) * 2020-07-30 2022-10-14 海南电网有限责任公司海口供电局 Single-machine infinite electric power system optimization control method based on T-S model
CN112327616B (en) * 2020-10-19 2022-09-16 江苏大学 Network control system controller design method based on event triggering
CN112350664B (en) * 2020-10-27 2021-09-24 电子科技大学 Limited frequency fault detection method based on event trigger strategy
CN112882391B (en) * 2021-01-26 2022-05-27 四川大学 Double-end event triggered nonlinear control method
CN113110363B (en) * 2021-05-24 2023-03-31 齐齐哈尔大学 Method for designing non-fragile fuzzy filter of network system based on event-driven strategy
CN113325822B (en) * 2021-05-25 2022-02-01 四川大学 Network control system fault detection method based on dynamic event trigger mechanism and sensor nonlinearity
CN113641104A (en) * 2021-08-23 2021-11-12 江南大学 Limited frequency domain fault detection method for tank reactor under dynamic event triggering
CN113848847B (en) * 2021-08-30 2024-06-04 北京工业大学 Nonlinear control system fault detection method based on T-S fuzzy model
CN113687596B (en) * 2021-08-31 2024-03-01 杭州电子科技大学 Modern wharf cargo management system fault detection method
CN113985197B (en) * 2021-10-18 2024-01-09 杭州电子科技大学 Event triggering asynchronous detection method for equipment faults of water service system
CN114035548B (en) * 2021-11-14 2024-03-26 北京工业大学 Fault detection method of T-S fuzzy control system based on kernel characterization
CN114545907B (en) * 2022-03-15 2023-12-19 中南大学 Fault detection method of flight control system based on filter
CN115225381B (en) * 2022-07-19 2023-05-12 海南大学 Asynchronous fault detection filter design method
CN115328142B (en) * 2022-08-26 2023-09-15 电子科技大学 Fault detection method for networked unmanned ship under replay attack
CN116527060B (en) * 2023-05-29 2024-01-05 北京理工大学 Information compression and anomaly detection method based on event trigger sampling

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103838140A (en) * 2014-01-27 2014-06-04 张旭 Weak nonlinear network control method based on direct inverse control algorithm
CN103941725A (en) * 2014-04-24 2014-07-23 淮海工学院 Fault diagnosis method of nonlinear network control system
CN107966908A (en) * 2018-01-17 2018-04-27 重庆大学 The fuzzy control method of non-linear truck-trailer systems based on event trigger mechanism

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102375442A (en) * 2010-08-23 2012-03-14 同济大学 Real-time on-line control system and method for miscellaneous nonlinear system

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103838140A (en) * 2014-01-27 2014-06-04 张旭 Weak nonlinear network control method based on direct inverse control algorithm
CN103941725A (en) * 2014-04-24 2014-07-23 淮海工学院 Fault diagnosis method of nonlinear network control system
CN107966908A (en) * 2018-01-17 2018-04-27 重庆大学 The fuzzy control method of non-linear truck-trailer systems based on event trigger mechanism

Also Published As

Publication number Publication date
CN108667673A (en) 2018-10-16

Similar Documents

Publication Publication Date Title
CN108667673B (en) Nonlinear network control system fault detection method based on event trigger mechanism
Cheng et al. An asynchronous operation approach to event-triggered control for fuzzy Markovian jump systems with general switching policies
Li et al. Event-based fault-tolerant control for networked control systems applied to aircraft engine system
Wang et al. Actuator fault diagnosis: an adaptive observer-based technique
CN110908364B (en) Fault detection method based on robust interval estimation
WO2017087440A1 (en) Anomaly fusion on temporal casuality graphs
Zhao et al. Distributed event-triggered state estimation and fault detection of nonlinear stochastic systems
CN112180899B (en) State estimation method of system under intermittent anomaly measurement detection
Atitallah et al. Event‐triggered fault detection for networked control systems subject to packet dropout
CN108972553A (en) A kind of space manipulator fault detection method based on particle filter algorithm
Song et al. An event-triggered approach to robust fault detection for nonlinear uncertain Markovian jump systems with time-varying delays
CN114046456B (en) Corrosion evaluation method and system for fusing fuzzy reasoning and neural network
CN115719294A (en) Indoor pedestrian flow evacuation control method and system, electronic device and medium
CN109309593B (en) Fault detection method of networked system based on Round-Robin protocol
Tehrani et al. Dynamic neural network-based estimator for fault diagnosis in reaction wheel actuator of satellite attitude control system
CN105718733B (en) Fault prediction method based on fuzzy nearness and particle filter
CN113988210A (en) Method and device for restoring distorted data of structure monitoring sensor network and storage medium
Gao et al. Quantitative analysis of incipient fault detectability for time-varying stochastic systems based on weighted moving average approach
CN105652795B (en) 3PTT-2R series-parallel numerical control machine tool servo system fault prediction device and method based on residual observer
De Oliveira et al. Energy-to-peak reduced order filtering for continuous-time markov jump linear systems with partial information on the jump parameter
CN116558406A (en) GNSS-accelerometer integrated bridge deformation monitoring abrupt fault detection method based on state domain
CN115718426A (en) Event-triggered STF fault detection method for satellite attitude control system
CN113625677A (en) Nonlinear system fault detection and estimation method and device based on adaptive iterative learning algorithm
CN117221075B (en) Discrete networking system fault detection method based on self-adaptive event trigger mechanism
CN113236506B (en) Industrial time delay system fault detection method based on filtering

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant