A method for designing and selecting parameters of anti-disturbance controller for non-minimum phase system

Abstract

The present invention provides a method for designing an anti-disturbance controller for a non-minimum phase system and selecting parameters, and belongs to the technical field of chemical process control. It solves the technical problem that a non-minimum phase system is difficult to achieve a fast and stable response when facing parameter uncertainty and external disturbances. The technical solution is as follows: linearize the non-minimum phase system at the steady-state operating point, design a generalized Smith predictor to predict the output of the system, and implement feedback regulation through an anti-disturbance controller to generate an appropriate control signal; design a multi-objective gray wolf optimization algorithm to optimize the controller parameters. The beneficial effects of the present invention are as follows: the control method improves the dynamic response speed and steady-state accuracy of the system by predicting and compensating for disturbances in advance, and significantly improves the robustness and anti-disturbance performance of the system.

Description

Technical Field

[0001] The invention relates to the technical field of chemical process non-minimum phase system control, and in particular to a non-minimum phase system anti-disturbance controller design and parameter selection method. Background Art

[0002] In practical applications, the control problem of non-minimum phase systems is very critical, especially in chemical processes. For example, in some chemical reaction processes, there are complex dynamic relationships between the input of reactants and the output of products, which may cause control systems to face non-minimum phase challenges. In such a system, when trying to quickly reach the set production target, the non-minimum phase characteristics of the system may cause fluctuations in output or quality, thereby affecting the stability and economic benefits of the entire production line.

[0003] For example, in large-scale continuous flow chemical production lines, changes in raw materials, temperature fluctuations, or changes in other operating conditions may cause undesirable overshoots in the dynamic response of the system, which not only affects product quality but may also cause safety issues. In addition, existing control schemes such as PID controllers often find it difficult to simultaneously meet the requirements of fast response and high stability when dealing with this type of system, especially when the external environment and internal parameters change. The paper "Gain-scheduling trajectory control of a continuous stirred tank reactor" mentioned that when a continuous stirred tank reactor (CSTR) faces non-minimum phase characteristics, traditional control methods such as PID controllers cannot effectively deal with the control difficulties caused by parameter changes and environmental interference. The paper pointed out that in practical applications, PID controllers show obvious deficiencies in dealing with non-minimum phase problems in CSTRs, specifically in the difficulty of balancing overshoot and stability of the control system. Similarly, the paper "Active Disturbance Rejection Control for Non-Minimum Phase Systems under Plant Reconstruction" discusses in detail the complexity of non-minimum phase systems in the control process and emphasizes the limitations of existing control methods in dealing with such systems. The paper points out that traditional control methods often find it difficult to simultaneously meet the requirements of dynamic response and system stability when dealing with non-minimum phase systems, especially when the system is subject to external disturbances and internal parameter changes.

[0004] In summary, the control problem of non-minimum phase system is crucial in the practical application of chemical process, and the existing control methods such as PID controller have obvious shortcomings in dealing with such systems. Therefore, it is particularly important to develop an anti-disturbance controller design and parameter selection method for non-minimum phase system. Summary of the invention

[0005] The purpose of the present invention is to provide a method for designing an anti-disturbance controller for a non-minimum phase system and selecting parameters. The proposed method can effectively suppress the inevitable disturbances from the production process, improve the robustness of the system to disturbances, ensure the continuity of the chemical process and the consistency of the product, and at the same time improve production efficiency while ensuring safety, which has a direct positive impact on the sustainable development and competitiveness enhancement of the chemical industry.

[0006] The inventive idea of ​​the present invention is: the present invention provides a non-minimum phase system anti-disturbance controller design and parameter selection method, introduces Smith predictor and adaptive control strategy in the anti-disturbance control of the non-minimum phase system, and realizes the robust control of the system by optimizing the controller parameters. For a non-minimum phase system, its control model and the estimated value of the input signal are designed, and a generalized Smith predictor anti-disturbance control structure is proposed. By generating control signals and adjusting system parameters, the dynamic response and steady-state performance of the system are improved. By selecting reasonable SP model parameters and using feedback control signals to generate the target output of the system, it is ensured that the system achieves the expected control performance. An improved multi-objective gray wolf optimization algorithm is introduced to select controller parameters, further improving the anti-disturbance performance and robustness of the system. Whether the adjusted parameters meet the stability requirements is judged by time domain and frequency domain simulation.

[0007] In order to achieve the above-mentioned invention object, the technical solution adopted by the present invention is specifically: a method for designing and selecting parameters of a non-minimum phase system anti-disturbance controller, which includes the following steps:

[0008] Step S1: First, consider a non-minimum phase system. The actual controlled object is linearized into the following P(s) form at the steady-state operating point:

[0009]

[0010] Among them, a is a positive constant coefficient, n r (s) is the numerator polynomial of the system, d r (s) is the denominator polynomial of the system;

[0011] Step S2: Design a generalized Smith predictor structure and optimize its parameters, which is divided into the following steps:

[0012] Step S2.1: The generalized Smith predictor structure includes a minimum phase model and an all-pass filter APF(s), The input signal is the control input u(s), and the output signal is the minimum phase output The specific form is as follows:

[0013]

[0014] in, is an estimate of the positive constant coefficient a, is the numerator polynomial n of the system r The estimated value of (s), is the denominator polynomial d of the system r (s) estimated value;

[0015] The input signal of the all-pass filter APF(s) is the output signal of the minimum phase model The output signal is the estimated value of y(s) The specific form is as follows:

[0016]

[0017] Step S2.2: Smith predictor parameter selection, the generalized Smith predictor structure of claim 1 is equivalent to the two-degree-of-freedom equivalent structure shown below, in which P eq (s) is the equivalent controlled object, P eq The input signal of (s) is the control signal u(s), and the output signal is Y(s) after equivalent processing. Among them, Q(s) is the compensator, the input signal is Y(s), and the output signal after compensation is y(s). The specific form is as follows:

[0018]

[0019] Among them, H(s) is the post filter;

[0020] Step S2.3: Select a suitable parameter set. First, the minimum phase model The m uncertainty parameters of the all-pass filter APF(s) are evenly divided into q equal parts within their value range, so there are m q Candidate Smith prediction models and controlled object models. Select one of the Smith prediction model parameters to check whether it is within the expected bandwidth frequency (BW) for any possible controlled object Q i The magnitude of N(s) is smaller than m d ,Right now:

[0021]

[0022] Among them, Q j is the jth controlled object model, and P is the controlled object model set. If the selected Smith estimation model parameters satisfy whether the controlled object Q is within the expected bandwidth frequency BW i The absolute value of the logarithm of N(jω) is less than m d This condition can be used as a suitable parameter;

[0023] Step S2.4: Select the optimal parameters. According to the selection method of step S2.1, a suitable parameter set can be obtained. Step S2.3 will select the optimal parameters in the set. On the basis of satisfying the Smith prediction model set in step S2.2, a set of parameters with the smallest performance index I value is selected as the Smith prediction model parameters. The performance index I is designed as follows:

[0024]

[0025] Where: D(jω) is the template of the controlled object, D eq (jω) is the equivalent controlled object template, C(·) is the area of ​​the template in the Nichols diagram, M(ω) is the weight of each study frequency, Ω is the study frequency set, and n is the number of study frequencies;

[0026] Step S3: Design a quantitative feedback anti-disturbance controller. The controller structure is determined according to the second-order anti-disturbance controller, including a feedback controller and a pre-filter. The input signal of the feedback controller G(s) is the reference input signal r(s) and the estimated value y sp (s), the output signal is u(s), as follows:

[0027]

[0028] Among them, l 1 ,l 2 ,β 1 ,β 2 ,β 3 ,b 0 are the controller parameters that need to be optimized; is the estimated value of the output signal y(s); is the output signal of the minimum phase model;

[0029] The input signal of the prefilter is the reference input signal r(s), and the output signal includes the smoothed signal v 1 (t) and the differential v of the signal 2 (t).

[0030]

[0031] in, l 1 ,l 2 ,β 1 ,β 2 ,β 3 ,b 0 ,a 0 are the controller parameters that need to be optimized;

[0032] Step S4: An improved adaptive multi-objective grey wolf algorithm (EMOGWO) is introduced to optimize the parameters of the feedback controller G(s) and the prefilter F(s) in step S3, which is divided into the following steps:

[0033] Step S4.1: Select the square of the system open-loop gain, robust stability, tracking performance, and anti-disturbance performance as the objective function, as follows:

[0034] f 1 (x) = k g 2 (11)

[0035] f 2 (x) = sta min +r min +d min (12)

[0036] where k g is the open-loop gain of the system; min is the minimum value of the robust stability constraint; r min is the minimum value of the tracking performance constraint; d min is the minimum value of the anti-interference performance constraint;

[0037] Step S4.2: In order to better adapt to the nonlinear and dynamic characteristics of the non-minimum phase system, a nonlinear adjustment strategy is introduced so that the parameters can be dynamically adjusted according to the current number of iterations, thereby improving the global search capability and adaptability of the algorithm. The details are as follows:

[0038]

[0039] Among them, a ini is the initial value, t is the current iteration number, and T is the control parameter. Through this nonlinear adjustment, a broad search can be performed in the early stages of the algorithm and a fine search can be performed in the later stages.

[0040] Step S4.3: In order to avoid premature convergence of the population, crowding distance sorting is introduced when updating individual positions to maintain population diversity. The details are as follows:

[0041]

[0042] Where d(i) is the crowding distance of individual i, M is the number of targets, and are the target values ​​of the front and back neighbors of individual i on target m, and are the maximum and minimum values ​​of the target m respectively;

[0043] Step S4.4: When selecting the alpha, beta, and delta alpha leaders, an adaptive selection strategy is introduced so that the selection probability of the individual is dynamically adjusted according to its fitness. The details are as follows:

[0044]

[0045] Among them, P i is the probability of selecting individual i as the leader, R i is the fitness of individual i, R mean is the average fitness of the population, and k is the adjustment parameter.

[0046] Step S5: Assign the optimal result obtained from the search to the parameters of the anti-disturbance controller, and perform simulation verification in the time domain to determine whether the system meets the stability requirements. If so, the design of an anti-disturbance controller for a non-minimum phase system is completed. If not, return to step S4 to search for better parameters.

[0047] Compared with the prior art, the present invention has the following beneficial effects:

[0048] (1) Aiming at the characteristics of non-minimum phase systems, the present invention proposes a generalized Smith predictor auto-disturbance rejection control structure, which significantly improves the anti-disturbance capability of the system. According to the proposed control method, the optimal parameter selection criterion is introduced to determine the predictor model, and the auto-disturbance rejection controller parameter tuning method is proposed under the equivalent model to ensure that the designed controller can achieve the expected performance indicators.

[0049] (2) In order to solve the problem that the parameters of the anti-disturbance controller are difficult to adjust, which is often time-consuming and labor-intensive, the present invention adopts an improved adaptive multi-objective grey wolf algorithm EMOGWO to optimize the parameters of the controller. Compared with the traditional parameter adjustment method, the automatic design method of the present invention is faster and easier to obtain excellent parameters, thereby ensuring the efficiency and stability of the system operation.

[0050] (3) The present invention combines advanced control theory with intelligent optimization algorithms to adapt to the complexity and dynamic changes of non-minimum phase systems in chemical processes, and provides a systematic and effective control strategy. In this way, the operating efficiency and product quality of chemical processes can be significantly improved, while reducing energy consumption and environmental impact.

[0051] (4) The method of the present invention can effectively suppress the inevitable disturbances from the production process and improve the robustness of the system to disturbances, thereby ensuring the continuity of the chemical process and the consistency of the product. At the same time, it can also improve production efficiency while ensuring safety, which has a direct positive impact on the sustainable development and competitiveness of the chemical industry. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] The accompanying drawings are used to provide further understanding of the present invention and constitute a part of the specification. They are used to explain the present invention together with the embodiments of the present invention and do not constitute a limitation of the present invention.

[0053] Figure 1 This is a structural diagram of a non-minimum phase system anti-disturbance controller design and parameter selection method provided in Example 1 of the present invention.

[0054] Figure 2 This is a flow chart of a method for designing an anti-disturbance controller for a non-minimum phase system and selecting parameters provided in Example 1 of the present invention.

[0055] Figure 3 This is an algorithm flow chart of an improved adaptive multi-objective grey wolf algorithm EMOGWO provided in Example 1 of the present invention.

[0056] Figure 4 This is a curve chart of loop shaping analysis results of a quantitative feedback anti-disturbance controller provided in Example 1 of the present invention.

[0057] Figure 5 This is a time domain simulation analysis system tracking performance curve diagram provided in Example 1 of the present invention.

[0058] Figure 6 This is a time domain simulation analysis system anti-interference performance curve diagram provided in Example 1 of the present invention.

[0059] Figure 7 This is a curve chart of loop shaping analysis results of a quantitative feedback anti-disturbance controller provided in Example 2 of the present invention.

[0060] Figure 8 This is a time domain simulation analysis system tracking performance curve diagram provided in Example 2 of the present invention.

[0061] Fig. 9 This is a time domain simulation analysis system anti-interference performance curve diagram provided in Example 2 of the present invention. DETAILED DESCRIPTION

[0062] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. Of course, the specific embodiments described here are only used to explain the present invention and are not used to limit the present invention.

[0063] Example 1

[0064] See also Figure 1 to Figure 2 The technical solution provided by this embodiment is a method for designing and selecting parameters of an anti-disturbance controller for a non-minimum phase system, which introduces a Smith predictor and an adaptive control strategy into the anti-disturbance control of the non-minimum phase system, and realizes robust control of the system by optimizing the controller parameters. Figure 1 It is a structural diagram of the method proposed in this embodiment. For a non-minimum phase system, the control model and the estimated value of the input signal are designed, and a generalized Smith predictor anti-disturbance control structure is proposed. By generating control signals and adjusting system parameters, the dynamic response and steady-state performance of the system are improved. By selecting reasonable Smith predictor model parameters and using feedback control signals to generate the target output of the system, the system is ensured to achieve the expected control performance. An improved multi-objective gray wolf optimization algorithm is introduced to select controller parameters to further improve the anti-disturbance performance and robustness of the system.

[0065] Figure 2 The flowchart of the method proposed in this embodiment is shown, which is a distillation process anti-disturbance control method considering input saturation compensation, comprising the following steps:

[0066] Step S1: First, consider a non-minimum phase system. The actual controlled object is linearized into the following P(s) form at the steady-state operating point:

[0067]

[0068] Among them, a is a positive constant coefficient, n r (s) is the numerator polynomial of the system, d r (s) is the denominator polynomial of the system;

[0069] Step S2: Design a generalized Smith predictor structure and optimize its parameters, which is divided into the following steps:

[0070] Step S2.1: The generalized Smith predictor structure includes a minimum phase model and an all-pass filter APF(s), The input signal is the control input u(s), and the output signal is the minimum phase output The specific form is as follows:

[0071]

[0072] in, is an estimate of the positive constant coefficient a, is the numerator polynomial n of the system r The estimated value of (s), is the denominator polynomial d of the system r (s) estimated value;

[0073] The input signal of the all-pass filter APF(s) is the output signal of the minimum phase model The output signal is the estimated value of y(s) The specific form is as follows:

[0074]

[0075] Step S2.2: Smith predictor parameter selection, the generalized Smith predictor structure of claim 1 is equivalent to the two-degree-of-freedom equivalent structure shown below, in which P eq (s) is the equivalent controlled object, P eq The input signal of (s) is the control signal u(s), and the output signal is Y(s) after equivalent processing. Among them, Q(s) is the compensator, the input signal is Y(s), and the output signal after compensation is y(s). The specific form is as follows:

[0076]

[0077] Among them, H(s) is the post filter;

[0078] Step S2.3: Select a suitable parameter set. First, the minimum phase model The m uncertainty parameters of the all-pass filter APF(s) are evenly divided into q equal parts within their value range, so there are m q Candidate Smith prediction models and controlled object models. Select one of the Smith prediction model parameters to check whether it is within the expected bandwidth frequency BW for any possible controlled object Q i The magnitude of N(s) is smaller than m d ,Right now:

[0079]

[0080] Among them, Q j is the jth controlled object model, and P is the controlled object model set. If the selected Smith estimation model parameters satisfy whether the controlled object Q is within the expected bandwidth frequency BW i The absolute value of the logarithm of N(jω) is less than m d This condition can be used as a suitable parameter;

[0081] Step S2.4: Select the optimal parameters. According to the selection method of step S2.1, a suitable parameter set can be obtained. Step S2.3 will select the optimal parameters in the set. On the basis of satisfying the Smith prediction model set in step S2.2, a set of parameters with the smallest performance index I value is selected as the Smith prediction model parameters. The performance index I is designed as follows:

[0082]

[0083] Where: D(jω) is the template of the controlled object, D eq (jω) is the equivalent controlled object template, C(·) is the area of ​​the template in the Nichols diagram, M(ω) is the weight of each study frequency, Ω is the study frequency set, and n is the number of study frequencies;

[0084] Step S3: Design a quantitative feedback anti-disturbance controller. The controller structure is determined according to the second-order anti-disturbance controller, including a feedback controller and a pre-filter. The input signal of the feedback controller G(s) is the reference input signal r(s) and the estimated value y sp (s), the output signal is u(s), as follows:

[0085]

[0086] Among them, l 1 ,l 2 ,β 1 ,β 2 ,β 3 ,b 0 are the controller parameters that need to be optimized; is the estimated value of the output signal y(s); is the output signal of the minimum phase model;

[0087] The input signal of the prefilter is the reference input signal r(s), and the output signal includes the smoothed signal v 1 (t) and the differential v of the signal 2 (t).

[0088]

[0089] in, l 1 ,l 2 ,β 1 ,β 2 ,β 3 ,b 0 ,a 0 are the controller parameters that need to be optimized;

[0090] Step S4: An improved adaptive multi-objective grey wolf algorithm EMOGWO is introduced to optimize the parameters of the feedback controller G(s) and the prefilter F(s) in step S3, which is divided into the following steps:

[0091] Step S4.1: Select the square of the system open-loop gain, robust stability, tracking performance, and anti-disturbance performance as the objective function, as follows:

[0092] f 1 (x) = k g 2 (11)

[0093] f 2 (x) = sta min +r min +d min (12)

[0094] where k g is the open-loop gain of the system; min is the minimum value of the robust stability constraint; r min is the minimum value of the tracking performance constraint; d min is the minimum value of the anti-interference performance constraint;

[0095] Step S4.2: In order to better adapt to the nonlinear and dynamic characteristics of the non-minimum phase system, a nonlinear adjustment strategy is introduced so that the parameters can be dynamically adjusted according to the current number of iterations, thereby improving the global search capability and adaptability of the algorithm. The details are as follows:

[0096]

[0097] Among them, a ini is the initial value, t is the current iteration number, and T is the control parameter. Through this nonlinear adjustment, a broad search can be performed in the early stages of the algorithm and a fine search can be performed in the later stages.

[0098] Step S4.3: In order to avoid premature convergence of the population, crowding distance sorting is introduced when updating individual positions to maintain population diversity. The details are as follows:

[0099]

[0100] Where d(i) is the crowding distance of individual i, M is the number of targets, and are the target values ​​of the front and back neighbors of individual i on target m, and are the maximum and minimum values ​​of the target m respectively;

[0101] Step S4.4: When selecting the alpha, beta, and delta alpha leaders, an adaptive selection strategy is introduced so that the selection probability of the individual is dynamically adjusted according to its fitness. The details are as follows:

[0102]

[0103] Among them, P i is the probability of selecting individual i as the leader, R i is the fitness of individual i, R mean is the average fitness of the population, and k is the adjustment parameter.

[0104] Step S5: Assign the optimal result obtained from the search to the parameters of the anti-disturbance controller, and perform simulation verification in the time domain to determine whether the system meets the stability requirements. If so, the design of an anti-disturbance controller for a non-minimum phase system is completed. If not, return to step S4 to search for better parameters.

[0105] Example 2

[0106] See also Figure 5 and Figure 6 For the Continuous Stirred Tank Reactor (CSTR) system, the control method of this embodiment is used to conduct simulation experiments. First, the system model is linearized to obtain the transfer function Then, the linearized model is controlled using the control structure and parameter selection method proposed in this embodiment to verify the effectiveness and superiority of the control strategy.

[0107] Figure 5 It shows that the system tracks a set reference signal (1 mol / l) from t=0 without external interference. Although non-minimum phase systems usually have response delays and initial reverse responses (zero points in the right half plane), the system shown in the figure can respond quickly in the initial stage and reach the target value almost in a short time (about 0.5 hours). This shows that the control method adopted in this embodiment can effectively cope with the complex dynamic characteristics of non-minimum phase systems and achieve rapid response. The system eventually stabilizes near the target value with almost no obvious steady-state error. Even if there are slight fluctuations during the initial response, the system can quickly recover and stabilize at the target value, showing the high accuracy and stability of the control method in non-minimum phase systems. Despite slight fluctuations, there is no significant overshoot, which shows that the control method can effectively suppress overshoot and improve the stability of the system. This is particularly important when dealing with non-minimum phase systems because such systems are more prone to overshoot problems.

[0108] Figure 6The figure shows that the system is disturbed by external factors at time t=1, and its anti-interference and recovery capabilities are evaluated. It can be seen from the figure that when the system is disturbed by external factors, although the output value fluctuates greatly, it can be quickly adjusted and restored to the target value. This shows that the control method used has strong anti-interference ability in non-minimum phase systems and can quickly identify and correct the impact of external interference on the system. After the interference disappears, the system output stabilizes near the target value without continuous oscillation or fluctuation, indicating that the control method can recover and maintain the stability of the system after processing the interference. This characteristic is particularly important for non-minimum phase systems because their dynamic response is more complex and stability is more difficult to guarantee.

[0109] Figure 5 and Figure 6 It shows that the control method used in this embodiment can provide fast, accurate and stable control effects when processing non-minimum phase systems, and is suitable for complex control scenarios that require high precision and strong anti-interference capabilities.

[0110] Example 3

[0111] See also Figure 7 and Figure 8 , for a chemical process non-minimum phase system, a simulation experiment was carried out using the control method of this embodiment. First, the system model was linearized to obtain the transfer function Then, the linearized model is controlled using the control structure and parameter selection method proposed in this embodiment to verify the effectiveness and superiority of the control strategy.

[0112] Figure 8 The dynamic response of the system under undisturbed conditions is demonstrated. It can be observed that the output increases rapidly and quickly reaches the set steady-state value within a short period of time after the system is started. This fast response shows that the control structure can effectively handle the initial instability of the non-minimum phase system and exhibits good transient response performance. In addition, the system remains stable after reaching the steady-state value without obvious oscillation or overshoot, which reflects the high efficiency and stability of the control structure in steady-state regulation.

[0113] Fig. 9 It shows the dynamic response of the system under disturbance conditions. It can be clearly seen in the figure that the response of the system in the initial stage is Figure 8 Similarly, the steady-state value is quickly reached. However, when the system is disturbed for about 1 hour, the output deviates briefly but quickly returns to the set steady-state value. This phenomenon shows that the control structure not only performs well under initial conditions, but also has strong anti-interference ability under external disturbances and can quickly restore the system to a stable state. This fast disturbance suppression capability is particularly important for non-minimum phase systems, because such systems are usually very sensitive to disturbances.

[0114] pass Figure 8 and Fig. 9 It can be clearly seen that the control structure used has significant advantages in dealing with non-minimum phase systems. This superior performance not only improves the overall stability and reliability of the system, but also enhances the system's adaptability in complex environments, providing an efficient and reliable solution for the control of non-minimum phase systems.

[0115] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for designing and selecting parameters of a disturbance rejection controller for a non-minimum phase system, characterized in that: The following steps are involved: Step S1: First, consider a non-minimum phase system. The actual controlled object is linearized into the following P(s) form at the steady-state operating point: Among them, a is a positive constant coefficient, n r (s) is the numerator polynomial of the system, d r (s) is the denominator polynomial of the system; Step S2: design a generalized Smith predictor structure and optimize its parameters; The step S2 comprises the following steps: Step S2.1: The generalized Smith predictor structure includes a minimum phase model and an all-pass filter APF(s), The input signal is the control input u(s), and the output signal is the minimum phase output The specific form is as follows: in, is an estimate of the positive constant coefficient a, is the numerator polynomial n of the system r The estimated value of (s), is the denominator polynomial d of the system r (s) estimated value; The input signal of the all-pass filter APF(s) is the output signal of the minimum phase model The output signal is the estimated value of y(s) The specific form is as follows: Step S2.2: Smith predictor parameter selection: the generalized Smith predictor structure is equivalent to a two-degree-of-freedom equivalent structure, in which P eq (s) is the equivalent controlled object, P eq The input signal of (s) is the control signal u(s), and the output signal is Y(s) after equivalent processing. Among them, Q(s) is the compensator, the input signal is Y(s), and the output signal after compensation is y(s). The specific form is as follows: Among them, H(s) is the post filter; Step S2.3: Select a suitable parameter set, first transform the minimum phase model The m uncertainty parameters of the all-pass filter APF(s) are evenly divided into q equal parts within their value range, so there are m q candidate Smith prediction models and controlled object models, select one of the Smith prediction model parameters to check whether it is within the expected bandwidth frequency BW for any controlled object Q i The magnitude of N(s) is smaller than m d ,Right now: Among them, Q j is the jth controlled object model, P is the controlled object model set, if the selected Smith estimation model parameters satisfy whether it is within the expected bandwidth frequency BW for any controlled object Q i The absolute value of the logarithm of N(jω) is less than m d This condition is then used as a suitable parameter; Step S2.4: Select the optimal parameters. According to the selection method of step S2.1, a suitable parameter set is obtained. Step S2.3 selects the optimal parameters in the set. On the basis of satisfying the Smith prediction model set in step S2.2, a set of parameters with the smallest performance index I value is selected as the Smith prediction model parameters. The performance index I is designed as follows: Where: D(jω) is the template of the controlled object, D eq (jω) is the equivalent controlled object template, C(·) is the area of ​​the template in the Nichols diagram, M(ω) is the weight of each study frequency, Ω is the study frequency set, and n is the number of study frequencies; Step S3: Design a quantitative feedback anti-disturbance controller. The controller structure is determined according to the second-order anti-disturbance controller, including a feedback controller and a pre-filter. The input signal of the feedback controller G(s) is the reference input signal r(s) and the estimated value y sp (s), the output signal is u(s), as follows: Among them, l1, l2, β1, β2, β3, b0 are the controller parameters that need to be optimized; is the estimated value of the output signal y(s); is the output signal of the minimum phase model; The input signal of the prefilter is the reference input signal r(s), and the output signal includes the smoothed signal v1(t) and the differential of the signal v2(t); in, l1, l2, β1, β2, β3, b0, a0 are the controller parameters that need to be optimized; Step S4: introducing an improved adaptive multi-objective grey wolf algorithm EMOGWO to optimize the parameters of the feedback controller G(s) and the pre-filter F(s) in step S3; Step S5: Assign the optimal result obtained from the search to the parameters of the anti-disturbance controller, and perform simulation verification in the time domain to determine whether the system meets the stability requirements. If so, the design of an anti-disturbance controller for a non-minimum phase system is completed. If not, return to step S4 to search for better parameters.

2. The method for designing and selecting parameters of a non-minimum phase system anti-disturbance controller according to claim 1, characterized in that: The step S4 comprises the following steps: Step S4.1: Select the square of the system open-loop gain, robust stability, tracking performance, and anti-disturbance performance as the objective function, as follows: f1(x)=k g 2 (11) f2(x)=sta min +r min +d min (12) where k g is the open-loop gain of the system; min is the minimum value of the robust stability constraint; r min is the minimum value of the tracking performance constraint; d min is the minimum value of the anti-interference performance constraint; Step S4.2: Introduce a nonlinear adjustment strategy so that the parameters are dynamically adjusted according to the current number of iterations to improve the global search capability and adaptability of the algorithm, as follows: Among them, a ini is the initial value, t is the current iteration number, and T is the control parameter; Step S4.3: In order to avoid premature convergence of the population, crowding distance sorting is introduced when updating individual positions to maintain population diversity, as follows: Where d(i) is the crowding distance of individual i, M is the number of targets, and are the target values ​​of the front and back neighbors of individual i on target m, and are the maximum and minimum values ​​of the target m respectively; Step S4.4: When selecting the alpha, beta, and delta alpha alpha alpha alpha alpha beta alpha ... Among them, P i is the probability of selecting individual i as the leader, R i is the fitness of individual i, R mean is the average fitness of the population, and k is the adjustment parameter.

Images