A method and apparatus for switching fuzzy system fault detection filtering in a network environment

Abstract

The application discloses a switching fuzzy system fault detection filtering method under a network environment, and comprises the following steps: S1, constructing a T-S fuzzy system model; S2, designing a switching event trigger mechanism with disturbance and a dynamic quantizer to effectively reduce communication overhead under limited network bandwidth; S3, referencing a weighted fault function, and designing a switching fuzzy fault detection filter; S4, constructing a filtering error system, and utilizing a residual error evaluation function to detect whether a system fault occurs; S5, utilizing Lyapunov stability theory to ensure that the filtering error system satisfies a preset performance index; and S6, utilizing a decoupling skill to obtain a collaborative design of a filter gain matrix, an event trigger matrix and a dynamic quantization parameter. The application further discloses a switching fuzzy system fault detection filtering device under the network environment. The application is used for ensuring that a system satisfies global consistent asymptotic stability and has performance, and can be widely applied to the field of automatic control technology.

Description

Technical Field

[0001] This invention relates to the field of automatic control technology, and in particular to a method and apparatus for fault detection and filtering of switching fuzzy systems in a network environment, applicable to nonlinear switching systems in bandwidth-constrained network environments. Background Technology

[0002] Existing fault detection filtering methods for fuzzy systems in handover mostly use residence time or average residence time frameworks. These methods inherently impose constraints on the system handover frequency, making it difficult to accurately characterize the dynamic behavior of fast-handling systems. Furthermore, in network environments, event-triggered mechanisms have been proposed and widely used to overcome the network resource waste caused by traditional periodic sampling. This mechanism only transmits data when a predefined trigger event is violated, significantly improving communication efficiency. However, in the context of handover, how to design a triggering strategy that can sensitively perceive system handover behavior remains an open problem. In addition, in digital communication networks, system state or measurement outputs inevitably need to be quantized before transmission. The resulting quantization errors affect fault detection performance. Adding to the complexity, event triggering, quantization effects, and handover dynamics are intertwined, greatly increasing the difficulty of system analysis and synthesis.

[0003] Therefore, under the constraints of the continuous dwell time switching mechanism and network environment, an effective fault detection filter for the switching fuzzy system is designed to ensure that the system satisfies global uniform asymptotic stability and has Performance is the key to overcoming current technological bottlenecks. Summary of the Invention

[0004] The purpose of this invention is to overcome the shortcomings of the aforementioned background technology and provide a method and apparatus for switching fuzzy system fault detection filtering in a network environment, enabling it to ensure that the system satisfies global uniform asymptotic stability and has performance.

[0005] This invention provides a fault detection filtering method for a switching fuzzy system in a network environment, comprising the following steps: S1, constructing a TS fuzzy system model under continuous dwell time switching; S2, designing a perturbation-based switching event triggering mechanism and a dynamic quantizer based on the TS fuzzy system model to effectively reduce communication overhead under limited network bandwidth; S3, designing a switching fuzzy fault detection filter in a network environment using a weighted fault function; S4, constructing a filtering error system using the fuzzy fault detection filter, and using a residual evaluation function to detect whether a system fault has occurred; S5, ensuring the stability of the filtering error system and satisfying preset conditions using Lyapunov stability theory. Performance metrics; S6. Utilize appropriate decoupling techniques to obtain a collaborative design method for filter gain matrix, event trigger matrix, and dynamic quantization parameters.

[0006] In the above technical solution, step S1 includes the following: S11, establishing a nonlinear system model containing fuzzy rules and switching signals, that is, for the i-th activation subsystem... The system rule is: if for and for ,So: in, , , , , , , and These represent the number of fuzzy rules, the fuzzy set, the premise variables, the state vector, the output vector, the fault vector to be detected, and the external disturbance, respectively. , , , The system matrix is ​​known; It is a signal indicating a change in the duration of residence. The number of subsystems; specifically, the process of step S1 is as follows: S111, give a switching signal. Assumption: Switching signal It is left-continuous, that is, except for It is the switching time, for , Established; S112, a switching signal is given. Definition: Assuming a persistent dwell time switching signal There are infinitely many lengths not less than The non-adjacent intervals, and these intervals are defined by a maximum length of . If the consecutive intervals are separated, then Known as switching signal Duration of stay For duration; S113, switching the fuzzy system model can be reorganized as follows: in, Represents a fuzzy function. express exist The degree of membership in, and ,as well as , .

[0007] In the above technical solution, the specific process of step S2 is as follows: S21, in order to reduce the communication overhead under limited network bandwidth, a handover event triggering scheme with disturbance is designed, wherein, If the current event trigger time is [time], then the next trigger time is [time]. Determined by the following triggering conditions:

[0008] in, , , , , , For trigger matrix and Given an event trigger threshold parameter; S22, based on the event trigger communication protocol, a dynamic quantization scheme is used to further reduce data transmission, wherein the dynamic quantizer is represented as: in, It is a dynamic parameter. Represent a static quantizer that satisfies:

[0009] in, and These are the quantizer range and quantization error, respectively. Furthermore, satisfy , The variables to be designed.

[0010] In the above technical solution, step S3 includes the following: S31, assuming that the filter and the system model share the same fuzzy premise variables, then switching the fuzzy fault detection filter can be expressed as: in, Let be the filter state vector. The quantized signal for the actual input filter and The residual signal generated for the filter; These are the gain matrices of the filter to be designed. Specifically, the process of step S3 is as follows: S311. In a network environment, the event-triggered mechanism experiences transmission delays when transmitting data. Based on the time delay transformation method, and Then you can get Where, assuming ,but Further, we can obtain S312. According to the time-delay transform method, the switching fuzzy fault detection filter can be expressed as:

[0011] in, And in the interval The above and below equations are true: ; S313, Use a weighted fault function To improve the dynamic performance of the fault detection system, among which It is known a priori, and The state space is: in, Let be the state vector of the weighted system, and the initial state be... ; The matrix is ​​known.

[0012] In the above technical solution, the specific process of step S4 is as follows: S41, Define the filtering error vector. , and estimating residual signals The filtering error system is then:

[0013] in,

[0014] ; S42, Settings For residual evaluation function and threshold Under the condition of no faults and bounded energy disturbances The supremum of the boundary, in its specific form is:

[0015] therefore, The value exceeds the threshold This allows you to identify the fault.

[0016] In the above technical solution, the specific process of step S5 is as follows: S51. The following definition is given: For a known dwell time switching signal... and initial conditions Assuming There exists a type function It can ensure Established, and So when the disturbance At that time, the system is globally consistent and asymptotically stable; S52, Performance: For a given and any nonzero perturbation Under zero initial conditions, the filtering error system satisfies S53. The following lemmas 1, 2, 3, 4 and theorem 1 are given: Lemma 1, specifically: For a given , ,as well as The following integral inequalities hold.

[0017] in, ; Lemma 2, specifically: for and Then in the interval The above can be obtained Lemma 3, specifically: Assume The switching sequence is not known a priori, but can be obtained instantaneously. Therefore, when... and At that time, the first Each subsystem is activated with a time interval of length [missing value]. ,set up This represents the minimum dwell time for all consecutive handovers, and , For positive constants, in any length of Maximum number of handovers within the time interval Furthermore, suppose that in any interval The total number of handovers is Then we can get Lemma 4, specifically: for a given scalar , , , Suppose there exists a switching Lyapunov function. and meeting the following conditions function ,

[0018]

[0019]

[0020] This demonstrates that the system is globally consistent and asymptotically stable, and that the persistence time switching signal satisfies: Theorem 1, specifically: Consider scalars and Based on Lemma 4, the filtering error system satisfies globally uniform asymptotic stability and A sufficient condition for performance is: for ,exist , and scalar Event Trigger Matrix Dynamic quantization parameters Make the PDT switching signal satisfy The following inequalities hold

[0021]

[0022] in,

[0023]

[0024]

[0025] S54. Establish a Lyapunov function based on the filtering error system, and prove, based on function operations and Theorem 1, that the filtering error system satisfies globally uniform asymptotic stability and performance.

[0026] In the above technical solution, the specific process of step S54 is as follows: S541. Constructing Lyapunov functions: ; S542, According to the Lyapunov function There exist positive real numbers. and satisfy:

[0027] in, and Let represent the minimum and maximum eigenvalues ​​of the matrix, respectively. The above equation ensures that the condition in Lemma 4 holds, and the condition in Theorem 1 also holds. This also ensures that the condition in Lemma 4 holds; S543, based on Lemma 1, the filtering error system satisfies the globally consistent asymptotic stability condition. hour, ,Right now,

[0028] in,

[0029]

[0030] According to the conditions in Theorem 1, we can obtain the above equation. Therefore, according to Lemma 4, the filtering error system satisfies globally uniform asymptotic stability; S544, for The filtering error system satisfies Performance, specifically

[0031] in, Using Lemma 2 and Lemma 3, we can obtain

[0032] in, .

[0033] In the above technical solution, the specific process of step S6 is as follows: S61, the following Lemma 5 and Theorem 2 are given: Lemma 5, specifically: for any symmetric and and for The following inequalities hold ; Theorem 2, specifically: Consider scalars , and Based on Lemma 4, the filtering error system satisfies globally uniform asymptotic stability and A sufficient condition for performance is: for ,exist Event Trigger Matrix , Filter gain matrix Dynamic quantization parameters This ensures that the PDT switching signal satisfies The following inequalities hold:

[0034]

[0035] in,

[0036]

[0037]

[0038]

[0039]

[0040]

[0041] And the desired fault detection filter gain matrix is ​​determined by the following formula: ; S62. Regarding the matrix inequality of Theorem 1 In Nonlinear coupling terms and To obtain the required filter gain matrix Event Trigger Matrix , and dynamic quantization parameters .

[0042] In the above technical solution, the specific process of step S62 is as follows: S621, Definition Eliminate the nonlinear coupling terms in Theorem 1 Then set the matrix , And multiply the inequalities in Theorem 1 by the left side. And right multiplication At the same time, using the lemma, we can obtain 5. ,in, It is an adjustable scalar parameter; according to Eliminating nonlinear coupling terms According to the matrix inequality in Theorem 2 That is, guarantee ; S622. Under the constraints of the inequalities in Theorem 2, use the Matlab toolbox to solve for the desired filter gain matrix. Event Trigger Matrix , scalar in dynamic quantization parameters .

[0043] The present invention also provides a switching fuzzy system fault detection filter device in a network environment, which has a computer program that can execute the switching fuzzy system fault detection filter method in a network environment.

[0044] The present invention provides a method and apparatus for fault detection filtering in a fuzzy system under network conditions, which has the following advantages: This invention, under the constraints of a persistent dwell time switching mechanism and network environment, ensures that the system satisfies global uniform asymptotic stability and possesses [specific characteristics] through an effective switching fuzzy system fault detection filter. Performance. This enables triggering strategies that can keenly perceive system switching behavior in a switching context. In digital communication networks, the impact of quantization processing errors in system state or measurement output before transmission on fault detection performance is relatively small. Furthermore, the difficulty of interweaving event triggering, quantization effects, and switching dynamics during system analysis and synthesis is greatly reduced. Attached Figure Description

[0045] Figure 1 This is a schematic diagram of the overall process of Embodiment 1 of the present invention for switching fuzzy system fault detection filtering method in a network environment; Figure 2 This is a system block diagram of the fault detection filter in Embodiment 3 of the present invention, which describes the method for switching fuzzy system fault detection filtering in a network environment. Figure 3 This is the persistent dwell time switching signal in Embodiment 3 of the present invention for the switching fuzzy system fault detection filtering method in a network environment. Schematic diagram; Figure 4 This is a schematic diagram of the event triggering of Embodiment 3 of the fuzzy system fault detection filtering method in a network environment according to the present invention; Figure 5 This is the fault detection signal in Embodiment 3 of the present invention for switching fuzzy system fault detection filtering methods in a network environment. Schematic diagram; Figure 6 The residual evaluation function of Embodiment 3 of the present invention for the fault detection filtering method of switching fuzzy systems in a network environment. and threshold Schematic diagram; Figure 7 This is Embodiment 3 of the present invention, which describes a method for switching fuzzy system fault detection filtering in a network environment. Performance indicators Schematic diagram of the curve; Figure 8 This is an architectural diagram of the fault detection filtering device for switching fuzzy systems in a network environment according to the present invention. Detailed Implementation

[0046] The present invention will now be described in further detail with reference to the accompanying drawings and embodiments, but these embodiments should not be construed as limiting the present invention.

[0047] This invention provides a fault detection filtering technique for handover fuzzy systems in a network environment. Considering the complex handover behavior of the system, a persistent dwell time handover strategy is adopted to construct the handover TS fuzzy system model. To reduce communication overhead under limited network bandwidth, a handover event triggering mechanism with perturbation is proposed, and a dynamic quantizer is used to further reduce the transmission of redundant data. Furthermore, considering the security and reliability of the system in practical applications, a handover fuzzy fault detection filter is designed based on the event triggering mechanism and the quantizer. By utilizing the mode-dependent Lyapunov functional method, not only is the global uniform asymptotic stability of the filtering error system guaranteed, but also... Furthermore, it improved performance and provided a collaborative design method for filter gain matrix, event trigger matrix, and dynamic quantization parameters.

[0048] Example 1 See Figure 1 This embodiment describes a method for switching fuzzy system fault detection filtering in a network environment, comprising the following steps: S1. Construct a fuzzy system model of TS under continuous dwell time switching.

[0049] S2. Design a perturbation-based switching event triggering mechanism and a dynamic quantizer to effectively reduce communication overhead under limited network bandwidth.

[0050] S3. Using a weighted fault function, design a switching fuzzy fault detection filter in a network environment.

[0051] S4. Construct a filtering error system and use the residual evaluation function to detect whether a system fault has occurred.

[0052] S5. Using Lyapunov stability theory, ensure the stability of the filtering error system and meet the preset requirements. Performance metrics.

[0053] S6. Utilize appropriate decoupling techniques to obtain a collaborative design method for filter gain matrix, event trigger matrix, and dynamic quantization parameters.

[0054] Example 2 This embodiment is basically the same as Embodiment 1, except that: Furthermore, in step S1, the specific steps are as follows: S11. Establish a nonlinear system model that includes fuzzy rules and switching signals. That is, for the i-th activation subsystem... The system rules are as follows: if for and for ,So:

[0055] in, , , , , , , and These represent the number of fuzzy rules, the fuzzy set, the premise variables, the state vector, the output vector, the fault vector to be detected, and the external disturbance, respectively. , , , It is a known system matrix. It is a signal indicating a change in the duration of residence. It represents the number of subsystems.

[0056] Furthermore, in step S11, the specific steps are as follows: S111, Give switching signal Assumption: Switching signal It is left-continuous, that is, except for It is the switching time, for , Established.

[0057] S112, Give the switching signal Definition: Assuming a persistent dwell time switching signal There are infinitely many lengths not less than The non-adjacent intervals, and these intervals are defined by a maximum length of . If the consecutive intervals are separated, then Known as switching signal Duration of stay For duration.

[0058] S113. Switching the fuzzy system model can reorganize it as follows:

[0059] in, Represents a fuzzy function. express exist The degree of membership in, and ,as well as , .

[0060] Furthermore, in step S2, the specific steps are as follows: S21. To reduce communication overhead under limited network bandwidth, a handover event triggering scheme with perturbation is designed, characterized in that... If the current event trigger time is [time], then the next trigger time is [time]. Determined by the following triggering conditions:

[0061] in, , , , , , For trigger matrix and The threshold parameter for triggering a given event.

[0062] S22. Based on the event-triggered communication protocol, a dynamic quantization scheme is used to further reduce data transmission. The dynamic quantizer is represented as follows:

[0063] in, It is a dynamic parameter. Represent a static quantizer that satisfies:

[0064] in, and These are the quantizer range and quantization error, respectively. Furthermore, satisfy , The variables to be designed.

[0065] Furthermore, in step S3, the specific steps are as follows: S31. Assuming the filter and the system model share the same fuzzy premise variables, then switching the fuzzy fault detection filter can be expressed as:

[0066] in, Let be the filter state vector. The quantized signal for the actual input filter and The residual signal generated for the filter; These are the gain matrices of the filter to be designed.

[0067] Furthermore, in step S31, the specific steps are as follows: S311. In a network environment, the event-triggered mechanism experiences data transmission delays. Based on the time delay transformation method, let... and Then you can get

[0068] Where, assuming ,but Further, we can obtain .

[0069] S312. According to the time-delay transform method, the switching fuzzy fault detection filter can be expressed as:

[0070] in, And in the interval The above and below equations are true: .

[0071] S313, Use a weighted fault function To improve the dynamic performance of the fault detection system, among which It is known a priori, and The state space is:

[0072] in, Let be the state vector of the weighted system, and the initial state be... ; The matrix is ​​known.

[0073] Furthermore, in step S4, the specific steps are as follows: S41. Define the filter error vector , and estimating residual signals The filtering error system is then:

[0074] in,

[0075] .

[0076] S42, Settings For residual evaluation function and threshold Under the condition of no faults and bounded energy disturbances The supremum of the boundary, in its specific form is:

[0077] therefore, The value exceeds the threshold This allows you to identify the fault.

[0078] Furthermore, in step S5, the specific steps are as follows: S51. The following definition is given: For a known dwell time switching signal... and initial conditions Assuming There exists a type function It can ensure Established, and So when the disturbance At that time, the system is globally consistent and asymptotically stable.

[0079] S52, Performance: For a given and any nonzero perturbation Under zero initial conditions, the filtering error system satisfies .

[0080] S53. The following lemmas 1, 2, 3, 4 and theorem 1 are given: Lemma 1, specifically: For a given , ,as well as The following integral inequalities hold.

[0081] in, .

[0082] Lemma 2, specifically: for and Then in the interval The above can be obtained .

[0083] Lemma 3 is as follows: Assumption The switching sequence is not known a priori, but it can be obtained instantaneously. Therefore, when... and At that time, the first Each subsystem is activated with a time interval of length [missing value]. .set up This represents the minimum dwell time for all consecutive handovers, and , It is a positive constant. In any length of... Maximum number of handovers within the time interval Furthermore, suppose that in any interval... The total number of handovers is Then we can get .

[0084] Lemma 4 is as follows: For a given scalar , , , Suppose there exists a switching Lyapunov function. and meeting the following conditions function ,

[0085]

[0086]

[0087] This demonstrates that the system is globally consistent and asymptotically stable. Furthermore, the persistence time switching signal satisfies: .

[0088] Theorem 1, specifically: Consider scalars and Based on Lemma 4, the filtering error system satisfies globally uniform asymptotic stability and A sufficient condition for performance is: for ,exist , and scalar Event Trigger Matrix Dynamic quantization parameters Make the PDT switching signal satisfy The following inequalities hold

[0089]

[0090] in,

[0091]

[0092]

[0093] .

[0094] S54. Establish a Lyapunov function based on the filtering error system, and prove, based on function operations and Theorem 1, that the filtering error system satisfies globally uniform asymptotic stability and performance.

[0095] Furthermore, in step S54, the specific steps are as follows: S541. Constructing Lyapunov functions: .

[0096] S542, According to the Lyapunov function There exist positive real numbers. and satisfy:

[0097] in, and Let represent the minimum and maximum eigenvalues ​​of the matrix, respectively. The above equation ensures that the condition in Lemma 4 holds, and the condition in Theorem 1... This also ensures that the condition in Lemma 4 holds.

[0098] S543. Based on Lemma 1, the filtering error system satisfies the globally uniform asymptotic stability condition. hour, ,Right now,

[0099] in,

[0100]

[0101] According to the conditions in Theorem 1, we can obtain the above equation. Therefore, according to Lemma 4, the filtering error system satisfies global uniform asymptotic stability.

[0102] S544, Regarding The filtering error system satisfies Performance, specifically

[0103] in, Using Lemma 2 and Lemma 3, we can obtain

[0104] in, .

[0105] Furthermore, in step S6, the specific steps are as follows: S61. Given the following Lemma 5 and Theorem 2: Lemma 5 is as follows: For any symmetric and and for The following inequalities hold .

[0106] Theorem 2, specifically: Consider scalars , and Based on Lemma 4, the filtering error system satisfies globally uniform asymptotic stability and A sufficient condition for performance is: for ,exist Event Trigger Matrix , Filter gain matrix Dynamic quantization parameters This ensures that the PDT switching signal satisfies The following inequalities hold:

[0107]

[0108] in,

[0109]

[0110]

[0111]

[0112]

[0113]

[0114] And the desired fault detection filter gain matrix is ​​determined by the following formula: .

[0115] S62. Regarding the matrix inequality of Theorem 1 In Nonlinear coupling terms and To obtain the required filter gain matrix Event Trigger Matrix , and dynamic quantization parameters .

[0116] Furthermore, in step S62, the specific steps are as follows: S621, Definition Eliminate the nonlinear coupling terms in Theorem 1 Then set the matrix. , And multiply the inequalities in Theorem 1 by the left side. And right multiplication At the same time, using the lemma, we can obtain 5. ,in, This is an adjustable scalar parameter. According to... Eliminating nonlinear coupling terms According to the matrix inequality in Theorem 2 That is, guarantee .

[0117] S622. Under the constraints of the inequalities in Theorem 2, use the Matlab toolbox to solve for the desired filter gain matrix. Event Trigger Matrix , scalar in dynamic quantization parameters .

[0118] Example 3 The technical features of this embodiment will be described in detail and completely through MATLAB simulation experiments to verify the effectiveness of the design method proposed in this invention.

[0119] This embodiment provides a numerical simulation to verify the feasibility of switching the fault detection filter in a fuzzy system under a network environment. It also includes illustrations showing the technical effects achieved using the parameters calculated in this embodiment. Figure 2 As shown. As a benchmark example, we used a single-link robotic arm system:

[0120] in, , , , and These are the robot arm's angle, mass, moment of inertia, damping, and length, respectively. To apply the proposed filtering method, the robot arm system is transformed into a switched fuzzy system with a switching region and a fuzzy weighting function, whose parameters are:

[0121]

[0122]

[0123] Other matrices are

[0124]

[0125]

[0126]

[0127] Solving Theorem 2 yields:

[0128]

[0129]

[0130]

[0131]

[0132] Next, set The membership function is:

[0133] And the fault signal is:

[0134] Then, Figures 3-7 The simulation results are based on the initial conditions. and The images obtained are all schematic diagrams illustrating the technical effects of this embodiment.

[0135] Depend on Figure 3 Persistent dwell time switching signal Schematic diagram.

[0136] Depend on Figure 4 The number of triggered events and the event triggering interval show that the designed event triggering mechanism can effectively reduce the update frequency of the filter, thereby significantly reducing the consumption of communication resources in the constrained network.

[0137] Depend on Figure 5 Simulation results of the fault detection signal show that, when set Inside, After weighting model Filtered output The steady-state value is close to 6.

[0138] Depend on Figure 6 Residual evaluation function and threshold The simulation results shown in the diagram indicate that... , ,get Furthermore, with a detection time delay of 0.08 seconds, the designed fault detection filter can achieve a rapid response to fault signals.

[0139] Depend on Figure 7 Performance indicators The simulation results over time show that, It converged to 0.7661, which indicates No more than Therefore, the designed fault detection filter has good interference suppression capability.

[0140] It should be noted that the above are merely simulation examples of the present invention and are not intended to demonstrate the present invention.

[0141] Example 4 See Figure 8 This embodiment provides a fault detection filtering device for switching fuzzy systems in a network environment, comprising the following parts: Fuzzy System Construction Module: Constructs a TS fuzzy system model under continuous dwell time switching; Triggering mechanism building module: Design a switching event triggering mechanism with perturbation and a dynamic quantizer to effectively reduce communication overhead under limited network bandwidth; Fault detection filter module: Using a weighted fault function, a fuzzy fault detection filter is designed to switch in a network environment; Fault detection module for the function: Constructs a filtering error system and uses a residual evaluation function to detect whether a system fault has occurred; Filtering error stabilization module: Utilizing Lyapunov stability theory, this module ensures the stability of the filtering error system and meets preset requirements. Performance metrics; Decoupling Techniques Module: Utilizing appropriate decoupling techniques, this module provides a collaborative design method for filter gain matrix, event trigger matrix, and dynamic quantization parameters.

[0142] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.

[0143] The contents not described in detail in this specification are existing technologies known to those skilled in the art.

Claims

1. A method for switching fuzzy system fault detection filtering in a network environment, characterized in that: Includes the following steps: S1. Construct a fuzzy system model for TS under continuous dwell time switching; S2. Based on the TS fuzzy system model, a perturbation-based switching event triggering mechanism and dynamic quantizer are designed to effectively reduce communication overhead under limited network bandwidth. S3. Using a weighted fault function, design a switching fuzzy fault detection filter in a network environment; S4. Construct a filtering error system using a fuzzy fault detection filter, and use a residual evaluation function to detect whether a system fault has occurred. S5. Using Lyapunov stability theory, ensure the stability of the filtering error system and meet the preset requirements. Performance metrics; S6. Utilize appropriate decoupling techniques to obtain a collaborative design method for filter gain matrix, event trigger matrix, and dynamic quantization parameters.

2. The method for switching fuzzy system fault detection filtering in a network environment according to claim 1, characterized in that: Step S1 includes the following: S11. Establish a nonlinear system model that includes fuzzy rules and switching signals, i.e., for the i-th activation subsystem... The system rules are as follows: if for and for ,So: in, , , , , , , and These represent the number of fuzzy rules, the fuzzy set, the premise variables, the state vector, the output vector, the fault vector to be detected, and the external disturbance, respectively. , , , The system matrix is ​​known; It is a signal indicating a change in the duration of residence. This refers to the number of subsystems. Specifically, the process of step S1 is as follows: S111, Give switching signal Assumption: Switching signal It is left-continuous, that is, except for It is the switching time, for , Established; S112, Give the switching signal Definition: Assuming a persistent dwell time switching signal There are infinitely many lengths not less than The non-adjacent intervals, and these intervals are defined by a maximum length of . If the consecutive intervals are separated, then Known as switching signal Duration of stay Duration; S113. Switching the fuzzy system model can reorganize it as follows: in, Represents a fuzzy function. express exist The degree of membership in, and ,as well as , .

3. The method for switching fuzzy system fault detection filtering in a network environment according to claim 2, characterized in that: The specific process of step S2 is as follows: S21. To reduce communication overhead under limited network bandwidth, a handover event triggering scheme with perturbation is designed, wherein, If the current event trigger time is [time], then the next trigger time is [time]. Determined by the following triggering conditions: in, , , , , , For trigger matrix and For a given event trigger threshold parameter; S22. Based on the event-triggered communication protocol, a dynamic quantization scheme is used to further reduce data transmission. The dynamic quantizer is represented as follows: in, It is a dynamic parameter. Represent a static quantizer that satisfies: in, and These are the quantizer range and quantization error, respectively. In addition, satisfy , The variables to be designed.

4. The method for switching fuzzy system fault detection filtering in a network environment according to claim 3, characterized in that: Step S3 includes the following: S31. Assuming the filter and the system model share the same fuzzy premise variables, then switching the fuzzy fault detection filter can be expressed as: in, Let be the filter state vector. The quantized signal for the actual input filter and The residual signal generated for the filter; These are the gain matrices of the filter to be designed. Specifically, the process of step S3 is as follows: S311. In a network environment, the event-triggered mechanism experiences data transmission delays. Based on the time delay transformation method, and Then you can get Where, assuming ,but Further, we can obtain ; S312. According to the time-delay transform method, the switching fuzzy fault detection filter can be expressed as: in, And in the interval The above and below equations are true: ； S313, Use a weighted fault function To improve the dynamic performance of the fault detection system, among which It is known a priori, and The state space is: in, Let be the state vector of the weighted system, and the initial state be... ; The matrix is ​​known.

5. The method for switching fuzzy system fault detection filtering in a network environment according to claim 4, characterized in that: The specific process of step S4 is as follows: S41. Define the filter error vector , and estimating residual signals The filtering error system is then: in, ； S42, Settings For residual evaluation function and threshold Under the condition of no faults and bounded energy disturbances The supremum of the boundary, in its specific form is: therefore, The value exceeds the threshold This allows you to identify the fault.

6. The method for switching fuzzy system fault detection filtering in a network environment according to claim 5, characterized in that: The specific process of step S5 is as follows: S51. The following definition is given: For a known dwell time switching signal... and initial conditions Assuming There exists a type function It can ensure Established, and So when the disturbance At that time, the system is globally consistent and asymptotically stable; S52, Performance: For a given and any nonzero perturbation Under zero initial conditions, the filtering error system satisfies ; S53. The following lemmas 1, 2, 3, 4 and theorem 1 are given: Lemma 1, specifically: For a given , ,as well as The following integral inequalities hold. in, ； Lemma 2, specifically: for and Then in the interval The above can be obtained ; Lemma 3 is as follows: Assumption The switching sequence is not known a priori, but can be obtained instantaneously. Therefore, when... and At that time, the first Each subsystem is activated with a time interval of length [missing value]. ,set up This represents the minimum dwell time for all consecutive handovers, and , For positive constants, in any length of Maximum number of handovers within the time interval Furthermore, suppose that in any interval The total number of handovers is Then we can get ； Lemma 4 is as follows: For a given scalar , , , Suppose there exists a switching Lyapunov function. and meeting the following conditions function , This demonstrates that the system is globally consistent and asymptotically stable, and that the persistence time switching signal satisfies: ; Theorem 1, specifically: Consider scalars and Based on Lemma 4, the filtering error system satisfies globally uniform asymptotic stability and A sufficient condition for performance is: for ,exist , and scalar Event Trigger Matrix Dynamic quantization parameters Make the PDT switching signal satisfy The following inequalities hold in, ； S54. Establish a Lyapunov function based on the filtering error system, and prove, based on function operations and Theorem 1, that the filtering error system satisfies globally uniform asymptotic stability and performance.

7. The method for switching fuzzy system fault detection filtering in a network environment according to claim 6, characterized in that: The specific process of step S54 is as follows: S541. Constructing Lyapunov functions: ； S542, According to the Lyapunov function There exist positive real numbers. and satisfy: in, and Let represent the minimum and maximum eigenvalues ​​of the matrix, respectively. The above equation ensures that the condition in Lemma 4 holds, and the condition in Theorem 1 also holds. This also ensures that the condition in Lemma 4 holds; S543. Based on Lemma 1, the filtering error system satisfies the globally uniform asymptotic stability condition. hour, ,Right now, in, According to the conditions in Theorem 1, we can obtain the above equation. Therefore, according to Lemma 4, the filtering error system satisfies global uniform asymptotic stability; S544, Regarding The filtering error system satisfies Performance, specifically in, Using Lemma 2 and Lemma 3, we can obtain in, .

8. The method for switching fuzzy system fault detection filtering in a network environment according to claim 7, characterized in that: The specific process of step S6 is as follows: S61. Given the following Lemma 5 and Theorem 2: Lemma 5 is as follows: For any symmetric and and for The following inequalities hold ； Theorem 2, specifically: Consider scalars , and Based on Lemma 4, the filtering error system satisfies globally uniform asymptotic stability and A sufficient condition for performance is: for ,exist Event Trigger Matrix , Filter gain matrix Dynamic quantization parameters This ensures that the PDT switching signal satisfies The following inequalities hold: in, And the desired fault detection filter gain matrix is ​​determined by the following formula: ; S62. Regarding the matrix inequality of Theorem 1 In Nonlinear coupling terms and To obtain the required filter gain matrix Event Trigger Matrix , and dynamic quantization parameters .

9. The method for switching fuzzy system fault detection filtering in a network environment according to claim 8, characterized in that: The specific process of step S62 is as follows: S621, Definition Eliminate the nonlinear coupling terms in Theorem 1 Then set the matrix , And multiply the inequalities in Theorem 1 by the left side. And right multiplication At the same time, using the lemma, we can obtain 5. ,in, It is an adjustable scalar parameter; according to Eliminating nonlinear coupling terms According to the matrix inequality in Theorem 2 That is, guarantee ; S622. Under the constraints of the inequalities in Theorem 2, use the Matlab toolbox to solve for the desired filter gain matrix. Event Trigger Matrix , scalar in dynamic quantization parameters .

10. A fault detection filter device for switching fuzzy systems in a network environment, comprising a computer program, characterized in that: The computer program is capable of executing the switching fuzzy system fault detection filtering method in a network environment as described in any one of claims 1 to 9.

Images