Fuzzy robust control method for wind power system based on dynamic memory event triggering mechanism

Abstract

The application discloses a fuzzy wind power system robust control method based on a dynamic memory event triggering mechanism and belongs to the technical field of automatic control of wind power generation systems, which comprises the following steps: constructing a T-S fuzzy model of a multi-region DFIG wind power system considering random faults of a controller and state time delay; designing a proportional-integral type controller structure; introducing a memory type event triggering mechanism based on a dynamic threshold and a historical error variable; and constructing a robustness criterion by using a Lyapunov-Krasovskii functional and a linear matrix inequality method. The application realizes unified guarantee of system control performance, robustness and stability in a communication resource limited environment.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of automatic control of wind power generation systems, in particular to a fuzzy wind power system robust control method based on a dynamic memory event triggering mechanism. It is suitable for multi-area doubly-fed induction generator (DFIG) wind power systems. The method aims to improve the frequency stability of the wind power system and effectively improve the utilization efficiency of communication resources in the case of random controller failure and communication delay. BACKGROUND

[0002] With the wide access of wind power in modern power systems, the problem of system frequency stability is increasingly prominent. Due to the volatility of wind energy, the inertial response capability of the system is significantly reduced, especially in large-scale multi-area wind power systems, the frequency deviation phenomenon is more significant, which seriously threatens the safety and reliability of system operation. The traditional load frequency control (LFC) method usually relies on timed sampling and deterministic control strategy, although it can maintain system performance in the worst case, but it often causes a large amount of redundant data transmission, causing communication network congestion and waste of computing resources.

[0003] In order to alleviate the above problems, researchers have proposed an event-triggered control (ETC) method, which only updates data when the system state meets the triggering condition, thereby effectively reducing the communication burden. However, the static event triggering mechanism (METS) has poor adaptability when facing dramatic changes in system state, and it is difficult to cope with uncertain factors such as random controller failure or communication delay. Among them, random controller failure is an important non-deterministic factor that cannot be ignored in practical applications, common causes include component aging, communication channel failure, etc., which may cause abnormal actuator output.

[0004] Doubly-fed induction generator (DFIG) has become the mainstream model of modern wind power systems due to its good grid performance and energy capture capability. However, its corresponding mathematical model has high nonlinearity and coupling characteristics, such as valve restriction, load disturbance and frequency deviation, which bring many challenges to controller design.

[0005] Therefore, with the continuous development of smart grids, how to ensure control performance while improving system stability and control robustness, security, and saving communication resources has become an important direction of networked control system research. SUMMARY

[0006] The present application aims at solving the problems of insufficient frequency regulation ability, waste of communication resources and weak system robustness of the existing wind power system control method under the conditions of random failure of the controller, communication delay and nonlinear coupling, and provides a fuzzy wind power system robust control method based on a dynamic memory event triggering mechanism.

[0007] The present application is achieved by the following technical solutions:

[0008] The present application provides a fuzzy wind power system robust control method based on a dynamic memory event triggering mechanism, which refers to Figure 1 and comprises the following steps:

[0009] Step 1: establishing a T-S fuzzy model of a multi-region double-fed induction generator wind power system;

[0010] Step 2: designing a PI controller with random fault modeling;

[0011] Step 3: constructing a dynamic memory event triggering mechanism;

[0012] Step 4: constructing a new Lyapunov-Krasovskii functional set and deriving stability conditions.

[0013] As a further improvement of the present application, the step 1 is specifically as follows:

[0014] A T-S fuzzy model with the following fuzzy rules is established:

[0015] Rule : If is ··· and is , then:

[0016]

[0017] wherein a=1, 2,..., l, , ..., ..., represent fuzzy sets, and l represents the number of fuzzy reasoning rules.

[0018] The fuzzy membership function is defined as follows:

[0019]

[0020] wherein q represents In the fuzzy set The quantity relationship, i The number of fuzzy inference rules under the current rule, ) represents the change of steam flow into the steam turbine. The premise value is bounded between And the subset function Both are selected as: And .

[0021] Using the center average fuzzy operator, product inference and singleton fuzzy operator, the fuzzy model can be expressed as the following global model:

[0022]

[0023] Where, , is the state variable of the system, is the derivative of the state variable of the system, represents the external disturbance, and the control input signal is , , , is the parameter matrix, represents the output variable of the system, represents the output parameter matrix, represents the change of steam flow into the steam turbine, represents the fuzzy membership function under a fuzzy inference rule, l represents the number of fuzzy inference rules, =1,2,.., l，i represents the number of fuzzy inference rules under the current rule.

[0024] As a further improvement of the present application, the step 2 is specifically as follows:

[0025] In order to track the frequency change of the multi-area DFIG integrated WPS, the following PI controller is used:

[0026]

[0027] Where, , represents the controller matrix, and represents the length of the integral interval.

[0028] For , is redefined by the control input The control input of the actuator comes from the zero-order holder, and the holding interval is , is the communication delay, is the sampling period, is the maximum allowed communication delay.

[0029] The holding interval of the zero-order holder can be subdivided into multiple sub-intervals, such as wherein and denote the instances from the current sampling instance to the subsequent sampling instance , wherein and

[0030]

[0031] For , the time-varying delay in the sawtooth structure is given by The following are the constraints of

[0032]

[0033] Thus, the PI controller can be replaced by the following expression:

[0034] .

[0035] In addition, the frequency response of the multi-area wind power system integrated with the doubly-fed converter is further monitored and analyzed, which more realistically reproduces the uncertainty faults that may occur in the power system. By introducing a dynamic adaptive mechanism, the robustness and fault tolerance of the system when facing faults are effectively improved. At the same time, the PI controller with random faults is taken as part of the control strategy, thereby enhancing the control performance of the system under non-ideal working conditions. On this basis, combined with the fuzzy center balancer, the following input controller expression is constructed:

[0036] .

[0037] In order to enhance the robustness of the system, a random fault model is introduced into the PI controller. Through the simulation and modeling of random faults, the control system can more effectively adapt to various uncertain factors and fault situations that may occur in the actual operation process, thereby improving its stability and reliability in complex environments. The random fault model is as follows:

[0038]

[0039] wherein, ​​represents a random fault sampling value obeying the Beta distribution defined in the interval (0, 1). The Beta distribution is selected to generate random faults, which not only ensures that the generated random variable values are always in the effective range, but also enhances the flexibility and adaptability of the power system to uncertain factors. In the simulation environment of uncertain events such as controller component failures, the following fuzzy proportional integral (PI) control strategy is designed to ensure the stability and control performance of the system:

[0040] Rule 2: If belongs to … and belongs to , then the fuzzy controller can be represented by the following expression:

[0041]

[0042] wherein, represents the integral control gain, represents the proportional control gain, and .

[0043] In summary, the fuzzy proportional integral control strategy can be processed into the following form by fuzzification:

[0044]

[0045] wherein, represents a random fault sampling value, represents a time-varying delay, represents a fuzzy membership function under another fuzzy inference rule, and j represents the number of fuzzy inference rules under the current rule.

[0046] Substituting the above formula into the global model gives:

[0047]

[0048] As a further improvement of the present application, step 3 is specifically as follows:

[0049] In power systems, static event triggering is usually used as a general event trigger as follows:

[0050]

[0051] wherein, represents the sequence at the transmission instant. Meanwhile, is defined, wherein and are constants. In addition, represents the synchronous sampling interval of the system, represents the gain matrix.

[0052] If the system state satisfies the condition specified in the above equation, the following transmission event sequence will be used . Although the adoption of the standard event-triggering strategy (ETS) reduces the usage rate of the required communication bandwidth, unnecessary data transmission can still occur due to the triggering condition set by the event-triggering strategy, even if the system has become stable. Such redundant communication can still be triggered when the system state changes weakly or even remains static, which reduces the utilization efficiency of the communication resources and is not conducive to the implementation of efficient networked control.

[0053] Therefore, the following equation is redefined , so as to more effectively save communication resources, and a dynamic event-triggering mechanism (DETM) is proposed, which includes a parameter function of an auxiliary dynamic variable. This method fully considers the problem of waste of communication resources that can be caused by controller failure. In order to achieve effective monitoring, the controller needs to access historical information to assist in judging the change of the system state. By grouping and accumulating the event-triggering data at each time point, and combining the auxiliary dynamic variable and the event state of the current time, the dynamic event-triggering mechanism determines and confirms the sending time of the next sampling signal according to this. This mechanism not only improves the communication efficiency, but also enhances the adaptability and robustness of the system when facing uncertain factors. The new triggering condition is as follows:

[0054]

[0055] wherein, represents the sequence at the transmission moment, represents the sequence at the next transmission moment, s is a constant, h represents the synchronous sampling interval of the system, min represents the minimum value operation, represents a diagonal matrix, represents a threshold parameter, and both represent a matrix, and are given constants, represents the number of recently transmitted data packets, is a weighting function, f represents the data packet number of transmission, ) represents a parameter function. , , , , , , .

[0056] In addition, in the equation: ,

[0057] simultaneously wherein is a matrix determined in the subsequent analysis, the threshold parameter .

[0058] Thus, for the threshold parameter the expression is as follows:

[0059]

[0060] wherein and are located in the range (0, 1) representing the highest and lowest values, while is a predetermined constant.

[0061] As a further improvement of the invention, the step 4 is specified as follows:

[0062] For a given scalar value , , , , the following matrix can be derived:

[0063] , , , , and an arbitrary matrix .

[0064] The above matrix and the controller gain , together satisfy the subsequent conditions, ensuring the asymptotic stability of the system. Some of the vectors are specified as follows:

[0065]

[0066] The matrix product and the multiple quadratic form in , are contained in .

[0067] Therefore, in order to verify the asymptotic stability of the system, the Lyapunov-Krasovskii functional is constructed as follows:

[0068]

[0069] wherein denotes the set of Lyapunov-Krasovskii functionals, and each sub-function corresponding to the current state item, the delay state item, the history derivative item and the event-triggered disturbance item respectively, denotes the current state item, denotes the delay state item, denotes the history derivative item, denotes the event-triggered disturbance item, and the overall derivative is expressed as:

[0070]

[0071] Here is a linear combination of the state vector, is an augmented matrix function constructed.

[0072] After the above derivation, in order to verify the stability of the evaluation system, under the premise of meeting the condition , based on the reasoning of the V function, the weighted integral term of the system state derivative can be estimated as a quadratic function of the state variable difference , , and the corresponding matrix of the function can satisfy the following LMI structure:

[0073]

[0074] Wherein: , , .

[0075] In addition, when considering the nonlinear inequality constraints suffered by the system under the event-triggered condition, the following relationship can be derived:

[0076]

[0077] Wherein, respectively contain the history sampling state and the current sampling state.

[0078] Under the condition of external disturbance , the system Lyapunov function derivative can finally satisfy:

[0079] .

[0080] As a further improvement of the present application, it also includes:

[0081]

[0082] If , then , so as to ensure the asymptotic stability of the entire wind power system.

[0083] Set the scalar parameter 、 and control parameters 、 with the introduction of positive definite matrices U, W, M, 、 On this basis, if there exists a controller gain matrix where (P, Q, R) and satisfy the following linear matrix inequality condition:

[0084] 1. All sub-items must satisfy ;

[0085] 2. Block matrix form ;

[0086] This shows that under the condition of disturbance , the system can still meet the set performance indicators , and maintain robust asymptotic stability under the action of event-triggered mechanism. The specific performance is: .

[0087] In order to facilitate the solution of the controller, we introduce a positive definite matrix , and reconstruct the controller gain: .

[0088] Based on the above verification process, if there exist design variables and positive definite matrices , the following LMI condition can be met:

[0089] 1.

[0090] 2.

[0091] If these series of conditions are met, the system's robust controller can be constructed by solving LMI. At the same time, this design can be replaced based on the following equivalent variables:

[0092] And on the basis of replacement, the extended matrix can be introduced. After left and right multiplication of a series of inequalities and linearization processing by applying Schur complement formula, the original controller design problem can be converted into a standard convex optimization problem, ensuring that the system realizes robust stability and performance objectives under the dynamic event-triggered strategy.

[0093] It needs to be further explained that the technical features corresponding to the above options can be combined or replaced to form new technical solutions without conflict.​

[0094] Compared with the prior art, the present application has the beneficial effects that:

[0095] The present application is directed to a multi-area doubly-fed induction generator (DFIG) wind power system, and proposes a T-S fuzzy control modeling method combined with a random fault modeling mechanism and a dynamic memory event triggering mechanism (DMETS) of a PI controller. The method not only guarantees the robustness and fault tolerance of the system, but also significantly reduces the communication burden and improves the bandwidth utilization. Based on the stability criterion constructed by a new Lyapunov-Krasovskii functional family, the robust stability of the system under the constraints of random faults and event triggering is strictly proved. The numerical simulation results further verify that the control strategy proposed in the present application is superior to the traditional method in terms of frequency regulation performance, communication efficiency and fault adaptability, and has high engineering application value. The specific contributions include:

[0096] (1) A T-S fuzzy control modeling method suitable for a multi-area DFIG wind power system is proposed, which effectively deals with the nonlinear characteristics of the system and improves the practicality and adaptability of the model.

[0097] (2) A PI controller with random fault modeling is designed, which simulates the gain loss and signal interference that may occur in the actual operating environment of the controller, and enhances the robustness and fault tolerance of the system.

[0098] (3) A dynamic memory event triggering mechanism is proposed, which considers the current state of the system, the historical error and the dynamically adjusted triggering threshold, effectively reduces the communication frequency and improves the bandwidth utilization efficiency.

[0099] (4) A new Lyapunov-Krasovskii functional set is constructed, combined with the dynamic time delay and nonlinear coupling processing mechanism, which effectively reduces the conservatism of the stability criterion and strictly verifies the robust stability of the control system.

[0100] (5) Through numerical simulation verification, the control strategy proposed in the present application is superior to the traditional load frequency control (LFC) method in terms of frequency regulation effect, communication resource utilization and fault adaptability, and has broad engineering application prospects. BRIEF DESCRIPTION OF DRAWINGS

[0101] Figure 1 The present application is a fuzzy wind power system robust control method flow chart based on a dynamic memory event triggering mechanism;

[0102] Figure 2 The present application is a dynamic memory event triggering diagram for isolated regions in Example 1;

[0103] Figure 3Threshold parameter trigger behavior fluctuation chart in embodiment 2 of the present application;

[0104] Figure 4 Dynamic memory event trigger threshold dynamic characteristic schematic diagram in embodiment 2 of the present application;

[0105] Figure 5 Dynamic memory event trigger threshold dynamic characteristic schematic diagram in embodiment 2 of the present application when a disturbance is applied;

[0106] Figure 6 System trigger interval change with time when threshold is 0.15 in embodiment 2 of the present application;

[0107] Figure 7 System trigger interval change with time when threshold is 0.25 in embodiment 2 of the present application;

[0108] Figure 8 System trigger interval change with time when threshold is 0.35 in embodiment 2 of the present application;

[0109] Figure 9 System trigger interval change with time when threshold is 0.45 in embodiment 2 of the present application. DETAILED DESCRIPTION

[0110] The technical solutions of the present application will be described clearly and completely below in combination with the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the present application, rather than all the embodiments. The components of the embodiments of the present application described and shown in the accompanying drawings can be arranged and designed in various different configurations. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative labor fall within the scope of protection of the present application.

[0111] It should be noted that the defects of the above prior art solutions are the results obtained by the inventors after practice and careful study. Therefore, the discovery process of the above problems and the solutions proposed by the embodiments of the present application to the above problems should be the contributions made by the inventors to the present application in the process of invention and creation, and should not be understood as technical content known to those skilled in the art.

[0112] Embodiment 1: Single-area DFIG wind power system control based on event trigger mechanism

[0113] This embodiment designs a robust control strategy with random fault tolerance ability for isolated single-area wind power systems, mainly including:

[0114] 1. System structure:

[0115] Firstly, a standalone wind power region is constructed. The region contains a doubly-fed induction generator (DFIG) model and is not connected to other regions through tie lines, forming a local autonomous system structure. The following equations are defined:

[0116] Thus, we can get the single-region wind power system under isolated conditions:

[0117]

[0118] On this basis, the system uses a T-S fuzzy model to approximate the nonlinear dynamics. By designing the corresponding fuzzy rules, the regional frequency deviation, valve hysteresis, and power dynamics are linearized and expressed in the form of state space. The T-S fuzzy model with the following fuzzy rules is considered:

[0119] Rule : If is , then:

[0120]

[0121] Rule 2: If is , then:

[0122]

[0123] where is the state variable of the system, denotes external disturbances. The control input signal is . A, B, F are parameter matrices.

[0124] The fuzzy membership function can be defined as follows:

[0125] where , , thus the following formula can be derived:

[0126]

[0127] where . Thus the following equation can be derived:

[0128]

[0129] In summary, the following equation is obtained:

[0130]

[0131] 2. Design of Proportional-integral Controller with Random Faults:

[0132] In the actual operation of power systems, controllers may encounter various random faults, such as sensor failure, actuator gain loss, signal interference, and hardware failure. These random problems can significantly affect the stability and performance of the system, especially in critical infrastructure such as wind power systems, where ensuring safe and stable operation is particularly important. Therefore, this paper designs a control model considering random faults, which introduces random parameters subject to Beta distribution to characterize the random drift of actuator gain, random switching or local interruption of communication channels, and other failure conditions, thereby enhancing the robustness and fault tolerance of the controller in complex operating environments. The controller input form is as follows:

[0133]

[0134] where is a random fault sampling data, with Beta distribution between (0, 1), which can better grasp the distribution of random faults of the controller.

[0135] 3. Analysis of Function Stability:

[0136] After constructing the complete system model and controller model, the following Lyapunov-Krasovskii functional is constructed to analyze the stability of the system: , each sub-item considers the current state, delay state, derivative integral, and event error, and uses integral inequality and lemma tools to derive the system derivative to satisfy:

[0137]

[0138] Then, the stability analysis conditions are converted to linear matrix inequality form, and MATLAB + YALMIP tools are used to solve, obtaining the controller gain and that satisfies the robustness constraint. Analyzing the stability of isolated region systems, although the system tends to be stable, there are still slight disturbances due to controller faults.

[0139] 4. State Response in Isolated Region State:

[0140] In power systems, static event triggering is commonly used as a general event trigger as follows:

[0141]

[0142] where represents the sequence at the transmission instant. At the same time, define where and are constants. Furthermore, denotes the synchronous sampling interval of the system, denotes the gain matrix.

[0143] If the system state satisfies the condition specified in the static event-triggering, the following transmission event will be activated . The adoption of the standard ETS reduces the usage rate of the required communication bandwidth. However, even if the system is stable, unnecessary data transmission defined in the static event-triggering can still occur through the ETS.

[0144] Therefore, the is redefined, thereby more effectively saving the communication resources, and a dynamic event-triggering mechanism (DETM) containing an auxiliary dynamic variable is proposed. This method fully considers the problem of waste of communication resources that may be caused by controller failure. To achieve effective monitoring, the controller needs to access historical information to assist in judging the change of the system state. By grouping and accumulating the event-triggering data at each time point, and combining the auxiliary dynamic variable with the event state of the current time, the dynamic event-triggering mechanism determines and confirms the sending time of the next sampling signal accordingly. This mechanism not only improves the communication efficiency, but also enhances the adaptability and robustness of the system when facing uncertain factors. The new triggering condition is as follows:

[0145]

[0146] wherein, denotes the sequence of the transmission moment, denotes the sequence of the next transmission moment, s is a constant, h denotes the synchronous sampling interval of the system, min denotes the minimum value operation, denotes a diagonal matrix, denotes a threshold parameter, and both denote a matrix, and are both given constants, denotes the number of the most recently transmitted data packets, is a weighting function, f denotes the number of the transmitted data packets, denotes a parameter function. , , , , , , .

[0147] In addition, in the formula: ,

[0148] Meanwhile wherein is a matrix determined in the subsequent analysis, threshold parameter .

[0149] Thus for threshold parameter the expression is as follows:

[0150]

[0151] wherein and are located in the range of (0, 1) to represent the highest and lowest values of , and is a predetermined constant.

[0152] Figure 2 Finally, numerical simulation is performed and the results are shown in The state response has faster convergence speed and smaller steady-state error under the DMETS mechanism. Compared with the traditional ETS and DETS triggering mechanisms, DMETS greatly reduces the number of communications, verifying the superiority of the proposed method in isolated area control.

[0153] Embodiment 2: This embodiment optimizes and further reforms the previous isolated area wind power system to perform collaborative control and anti-disturbance performance analysis of a multi-area interconnected DFIG wind power system. This embodiment expands to a wind power system including two areas, each area being configured with a DFIG power generation unit, power being transmitted between the areas through a tie line, frequency synchronization and power balance needing to be achieved, belonging to a typical distributed wide-area power system structure. On this basis, a unified fuzzy modeling method is used, the state variables of each area including frequency deviation , power transmission error and internal control variables. In the modeling process, fuzzy rules are used to divide the working interval to obtain a T-S fuzzy system coupled with the areas.

[0154] Thereafter, the controller continues the PI structure in Embodiment 1, introduces the DMETS triggering mechanism, ensures that the controller is updated only when the system state changes significantly, avoids redundant signal transmission, and the unified triggering condition is: wherein the threshold function has a time-varying memory characteristic.

[0155] On the basis of the multi-area system, a Lyapunov-Krasovskii functional is further constructed, and the system stability conditions are derived under the factors of controller failure, communication time delay and external disturbance. Subsequently, the obtained conditions are equivalent to the solvable linear matrix inequality (LMI) form, which is used to analyze whether the system meets the robust performance constraint as shown below:

[0156]

[0157] After completing the basic system establishment, the parameters in the following table are set to design the model simulation.

[0158] Table 1 Simulation parameters of wind power system

[0159]

[0160] After numerical simulation using the parameters listed in Table 1, the steady-state operation of the two-area interconnected system (WPS) can be obtained. During the dynamic operation process, sustained low-amplitude oscillation occurs, which is mainly due to the small signal disturbance received by the terminal node. This type of oscillation has typical decay period characteristics, which are as follows:

[0161] 1. The disturbance variable presents an alternating decay law, and its amplitude gradually converges as the system approaches a steady-state equilibrium.

[0162] 2. After entering the steady state, the disturbance amplitude obeys a uniform statistical distribution, which confirms that an equilibrium energy exchange mechanism has been established between the interconnected areas.

[0163] On this basis, the DMETS-based dynamic characteristic analysis is shown in Figure 4 . The research data show that the adjustment mechanism is significantly related to the system behavior shown in Figure 3 . Specifically, the parameter matrix of the nonlinear NCS that satisfies the two fuzzy rules can be represented as follows:

[0164] 1. When the system response shows significant fluctuations, will tend to threshold value, at which time the event frequency significantly increases;

[0165] 2. In the process of system stabilization and response error convergence to zero, then gradually approaches reference value, and the triggering event is correspondingly reduced.

[0166] When the simulation of the system is completed, this embodiment studies the difference between dynamic memory event triggering and traditional triggering in the multi-area wind power system under the T-S fuzzy environment, Figure 4The experimental results of DMETS triggering threshold dynamic characteristics are shown, wherein, Figure 4 (a) is the change of triggering interval during operation, Figure 4 (b) is the evolution of triggering threshold over time, which reflects the coordination adjustment ability of the mechanism in the dynamic response stage and its advantage of maintaining system consistency. Table 2 gives the comparison of relative communication resource utilization of different triggering modes.

[0167] Table 2 Comparison of results of different event triggering modes

[0168]

[0169] As can be seen from Table 2, there are significant differences in relative communication resource utilization among ETS, METS, DETS and DMETS four technologies, among which various ETS technologies have obvious effect in reducing the use rate of NCCs, especially DMETS and METS technologies, which perform particularly outstanding in improving communication efficiency.

[0170] In order to further verify the robustness of the controller under disturbance conditions, the operation process of DMETS mechanism is tested under the condition of step disturbance applied to the system, and the results are shown in Figure 5 , wherein, Figure 5 (a) shows the change of DMETS triggering interval under disturbance, Figure 5 (b) shows the corresponding triggering threshold dynamic evolution over time, reflecting that the mechanism can quickly adjust the triggering frequency at the initial stage of disturbance, and gradually prolong the triggering interval at the system recovery stage, so as to maintain the stability and consistency of the system.

[0171] Figures 6-9 The change of system triggering interval over time under different threshold conditions is shown, corresponding to the experimental results of 0.15, 0.25, 0.35 and 0.45 respectively. Figure 6 Threshold value is 0.15: threshold value is the lowest, triggering condition is the most strict (more easy to trigger), triggering interval is the shortest, communication frequency is the highest, initial triggering is intensive, and there is almost no long interval section. Figure 7 Threshold value is 0.25: threshold value is increased, triggering interval is slightly prolonged, communication frequency is relatively Figure 6 decreased, initial triggering is still intensive but sparse in middle and late stages. Figure 8 Threshold value is 0.35: threshold value is further increased, system triggering condition is more relaxed, triggering interval is significantly prolonged, and communication frequency is further reduced. Figure 9The medium threshold value is 0.45: the threshold value is the highest, the triggering condition is the most relaxed, the triggering interval is the longest, the communication frequency is the lowest, and the triggering is only concentrated in the stage of rapid change of system state. The DMETS mechanism does not affect the system response speed and output stability while significantly reducing the communication frequency, and has stronger adaptability to interference changes, fully embodying the practicability and robustness of the proposed control method.

[0172] The above detailed description is a detailed description of the present application, and cannot be considered as limiting the specific embodiments of the present application to these descriptions. For ordinary skilled persons in the technical field to which the present application belongs, some simple deductions and substitutions can be made without departing from the concept of the present application, and all of them should be considered as belonging to the protection scope of the present application.

Claims

1. A fuzzy robust control method for wind power system based on dynamic memory event triggering mechanism, characterized in that, The method comprises the following steps: Step 1, establishing a T-S fuzzy model of a multi-area doubly-fed induction generator wind power system; Step 2, designing a PI controller with random fault modeling; the PI controller is expressed by the following formula: wherein represents a control input signal, represents an integral control gain, represents a proportional control gain, and , represents a random fault sample value, represents a time-varying delay, represents a fuzzy membership function under another fuzzy inference rule, and j represents the number of the fuzzy inference rule under the current rule. Step 3, constructing a dynamic memory event trigger mechanism, wherein the event trigger data at each time point is grouped and accumulated, and combined with auxiliary dynamic variables and the event state of the current time, determining and confirming the sending time of the next sampling signal; Step 4, construct a new set of Lyapunov-Krasovskii functionals, derive stability condition; the set of Lyapunov-Krasovskii functionals is as follows: wherein, denotes a set of Lyapunov-Krasovskii functionals, each sub-function corresponds to a current state term, a delayed state term, a history derivative term and an event-triggered disturbance term, respectively, denotes a current state term, denotes a delayed state term, denotes a history derivative term, denotes an event-triggered disturbance term.

2. The fuzzy robust control method of wind power system based on dynamic memory event trigger mechanism according to claim 1, characterized in that, The T-S fuzzy model of the multi-area doubly-fed induction generator wind power system comprises: The fuzzy model is expressed as follows by using a center average fuzzy device, product reasoning and a singleton fuzzy device Global model: where, , is the state variable of the system, is the derivative of the state variable of the system, represents external disturbance, the control input signal is , , , is the parameter matrix, , , respectively represent the determinant of the corresponding parameter matrix, represents the system output variable, represents the output parameter matrix, represents the fuzzy rule corresponding to the change of steam flow into the steam turbine, represents the physical quantity of the actual steam flow change, represents the fuzzy membership function under a fuzzy inference rule, l represents the number of fuzzy inference rules, ，i represents the number of fuzzy inference rules under the current rule.

3. The fuzzy robust control method of wind power system based on dynamic memory event trigger mechanism according to claim 1, characterized in that, The dynamic memory event triggering mechanism comprises: The triggering condition is: wherein, denotes the sequence of transmission instants, denotes the sequence of next transmission instants, s is a constant, h denotes the synchronization sampling interval of the system, min denotes the min operation, denotes a diagonal matrix, denotes a threshold parameter, and both denote a matrix, and are both given constants, denotes the number of recently transmitted data packets, , is a weighting function, f denotes the number of transmitted data packets, denotes a parameter function.

4. The fuzzy robust control method of wind power system based on dynamic memory event trigger mechanism according to claim 2, characterized in that, The derivation Stability conditions include: The stability condition is set as wherein, is the decay performance index.

Images