An active disturbance rejection control method and system for a cascaded extended state observer
By representing complex disturbances as polynomials through a cascaded extended state observer and building a linear extended state observer, the problem of accurate estimation under complex disturbances in the linear active disturbance rejection control method is solved, and better disturbance suppression and dynamic response effects are achieved.
Patent Information
- Application Number
- CN202311007876.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2023-08-10
- Publication Date
- 2026-11-17
- Estimated Expiration
- 2043-08-10
AI Technical Summary
Existing linear active disturbance rejection control methods cannot accurately estimate external disturbances when faced with complex disturbances, resulting in large overshoots or divergent response curves and poor control quality.
By employing a cascaded extended state observer, the complex disturbances of the controlled system are represented in polynomial form. A cascaded structure including multiple linear extended state observers is constructed, and a linear control law is designed based on the degree of the polynomial. The stability of the controller is analyzed using Lyapunov stability theory.
It significantly enhances the controlled system's ability to suppress disturbances and its dynamic response capability, effectively eliminating complex disturbances and improving the system's control quality.
Smart Images

Figure CN116859750B_ABST
Abstract
Description
Technical Field
[0001] This invention belongs to the field of controller design technology, and in particular relates to an active disturbance rejection control method and system for a cascaded extended state observer. Background Technology
[0002] In recent years, the rapid development of control technology and intelligent algorithms has led to the crucial role of Active Disturbance Rejection (ADRF) algorithms and controllers in resisting disturbances, suppressing noise, and achieving real-time control. Due to its advantages of low model dependence and strong disturbance rejection capability, ADRF is widely used in aerospace systems, electrical systems, and robotics systems. A classic ADRF controller consists of a tracking differentiator, a nonlinear state error feedback control law, and an extended state observer. The tracking differentiator is used for transient control, ensuring the accuracy of the tracking input while reducing system overshoot. The nonlinear state error feedback control law rationally calculates the error between the control system's reference and observed values, providing real-time feedback control and effectively improving control performance. The extended state observer, as the key component of the ADRF controller, predicts and compensates for the total system disturbance, enabling feedforward control, enhancing the system's anti-interference capability, and reducing the controller's dependence on the model. Therefore, ADRF controllers occupy a pivotal position in modern control methods.
[0003] Compared to the classic active disturbance rejection control algorithm, the linear active disturbance rejection control algorithm reduces the number of tuning parameters and expands the application range of active disturbance rejection control. However, in actual control systems with complex disturbances, the linear active disturbance rejection control method cannot accurately estimate the complex external disturbances of the system, and the system may experience large overshoot or divergence of response curves.
[0004] The inventors have discovered that, to further improve the dynamic and static performance of control systems and enhance their disturbance rejection capability, current research has introduced various control structures to optimize active disturbance rejection (ADR) algorithms. For example, the addition of a phase-locked loop (PLL) observer helps the ADR algorithm resist low-frequency disturbances; cascaded extended state observers suppress the influence of high-frequency noise on the system, improving the accuracy of total disturbance prediction, but cannot completely suppress complex disturbances; the proposed composite control method combining a reduced-level cascaded extended state observer and integral sliding mode control aims to eliminate total disturbances. While these methods improve the system's dynamic performance to some extent, none can completely eliminate the adverse effects of complex total disturbances on the system, resulting in poor system control quality. Therefore, improving the system's ability to resist complex disturbances and its performance in quickly tracking the reference input has become a key issue in ADR research. Summary of the Invention
[0005] To address the aforementioned problems, this invention proposes an active disturbance rejection control method and system based on a cascaded extended state observer. This invention provides an improved active disturbance rejection control method based on a cascaded extended state observer to eliminate complex disturbances. It uses a polynomial to represent the total complex disturbance of the controlled system and constructs a cascaded linear extended state observer based on the polynomial's degree, significantly enhancing the controlled system's ability to suppress disturbances and improving its dynamic response capability.
[0006] To achieve the above objectives, the present invention is implemented through the following technical solution:
[0007] In a first aspect, the present invention provides an active disturbance rejection control method for a cascaded extended state observer, comprising:
[0008] The complex disturbances existing in the controlled system are expressed in the form of a polynomial, and the highest order of the polynomial is determined according to the control accuracy.
[0009] Construct a cascaded linear extended state observer consisting of multiple linear extended state observers to estimate polynomial perturbations of the corresponding order; the number of linear extended state observers is equal to the degree of the polynomial.
[0010] Based on the constructed cascaded linear extended state observer, a linear control law is designed; the designed linear control law is used to control and eliminate disturbances.
[0011] Furthermore, the coefficient of each term in the polynomial represents the output gain at different times.
[0012] Furthermore, assuming that the observer gains at corresponding positions in the multiple linearly extended state observers are equal, the general formula of the cascaded linearly extended state observer is derived.
[0013] Furthermore, the general form of the cascaded linearly extended state observer is:
[0014]
[0015] in, t is the difference between the state value predicted by the cascaded linear extended state observer and the actual state, where n is a constant and t is time; and These represent the observed values of the system output y(t) and the total disturbance, respectively. It is considered as the predicted value of the system's residual disturbance; is the predicted perturbation of the (n-1)th level extended state observer; b0 is the estimated value of the controller gain; u(t) is the input signal; The perturbation value predicted by the first linearly extended state observer; β i and β j It is the gain of the cascaded linear expansion state observer.
[0016] Furthermore, the gain of each level of the extended state observer is the same: β i =β1=2ω0,β j =β2=ω0 2 , where is the observer bandwidth.
[0017] Furthermore, the control law is:
[0018]
[0019] u0(t)=K p (x(t)-y(t))
[0020] Where u(t) is the input signal; u0(t) is the output signal; The observed values of the total disturbance are: x(t) = y(t) / K; the system input is x(t) and the system output is y(t); K is K. p This is the proportional gain.
[0021] Furthermore, the stability of the cascaded linear expansion state observer is analyzed using Lyapunov stability theory.
[0022] Secondly, the present invention also provides an active disturbance rejection control system for a cascaded extended state observer, comprising:
[0023] The complex disturbance processing module is configured to: represent the complex disturbances existing in the controlled system as a polynomial, and determine the highest order of the polynomial based on the control accuracy.
[0024] The cascaded linear extended state observer building module is configured to: build a cascaded linear extended state observer including multiple linear extended state observers to estimate polynomial perturbations of the corresponding order; the number of linear extended state observers is equal to the degree of the polynomial;
[0025] The control module is configured to: design a linear control law based on the constructed cascaded linear expansion state observer; and use the designed linear control law to control and eliminate disturbances.
[0026] Thirdly, the present invention also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the active disturbance rejection control method of the cascaded extended state observer described in the first aspect.
[0027] Fourthly, the present invention also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the steps of the active disturbance rejection control method of the cascaded extended state observer described in the first aspect.
[0028] Compared with the prior art, the beneficial effects of the present invention are as follows:
[0029] 1. This invention represents the complex disturbances in the controlled system as a polynomial. The highest order of the polynomial is determined according to the control accuracy. A cascaded linear extended state observer including multiple linear extended state observers is constructed based on the degree of the polynomial. The number of linear extended state observers determines the level of the improved active disturbance rejection control method and the controller. This effectively eliminates all disturbances in the controlled system whose order is not higher than the highest order, as well as disturbances resulting from the linear combination of all disturbances. This significantly enhances the controlled system's ability to suppress disturbances and improves the dynamic response capability of the controlled system.
[0030] 2. This invention derives and determines the general form of the cascaded linear extended state observer, and proves the stability of the controller using Lyapunov's first method. Attached Figure Description
[0031] The accompanying drawings, which form part of this embodiment, are used to provide a further understanding of this embodiment. The illustrative embodiments and their descriptions are used to explain this embodiment and do not constitute an improper limitation of this embodiment.
[0032] Figure 1 This is a structural diagram of Embodiment 1 of the present invention;
[0033] Figure 2 This is a flowchart of Embodiment 1 of the present invention;
[0034] Figure 3 The four complex disturbances ω(t)(t, t) of the load in Embodiment 1 of the present invention are 2 t 3 t 4 );
[0035] Figure 4 The active disturbance rejection controller S1ADRC of Embodiment 1 of the present invention resists disturbances ω(t)(t, t) 2 t 3 t 4 The response curve of );
[0036] Figure 5 The active disturbance rejection controller S2ADRC of Embodiment 1 of the present invention resists disturbances ω(t)(t, t) 2 t 3 t 4 The response curve of );
[0037] Figure 6 The active disturbance rejection controller S3ADRC of Embodiment 1 of the present invention resists disturbances ω(t)(t, t) 2 t 3 t 4 The response curve of );
[0038] Among them, 31 is the response curve with input disturbance t; 32 is the response curve with input disturbance t. 2 The response curve; 33. Input disturbance is t 3 The response curve; 34. Input disturbance is t 4 Response curve of the first-order improved active disturbance rejection controller with disturbance t; 35. Response curve of the first-order improved active disturbance rejection controller with disturbance t; 36. Response curve of the first-order improved active disturbance rejection controller with disturbance t 2 Response curve of a first-order improved active disturbance rejection controller at time t; 37. Disturbance is t 3 Response curve of a first-order improved active disturbance rejection controller at time t; 38. Disturbance is t 4 39. Response curve of the first-stage improved active disturbance rejection controller with disturbance t; 310. Response curve of the second-stage improved active disturbance rejection controller with disturbance t. 2 The response curve of the improved secondary active disturbance rejection controller at time t, 311, disturbance is t 3 The response curve of the improved secondary active disturbance rejection controller at time t; 312, disturbance is t 4 Response curves of a two-stage improved active disturbance rejection controller with disturbance t; 313. Response curves of a three-stage improved active disturbance rejection controller with disturbance t; 314. Response curves of a three-stage improved active disturbance rejection controller with disturbance t. 2 Response curve of a three-level improved active disturbance rejection controller at time t; 315, disturbance is t 3 Response curve of a three-level improved active disturbance rejection controller at time t; 316, disturbance is t 4 The response curve of the three-level improved active disturbance rejection controller. Detailed Implementation
[0039] The present invention will be further described below with reference to the accompanying drawings and embodiments.
[0040] It should be noted that the following detailed descriptions are exemplary and intended to provide further explanation of this application. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains.
[0041] Example 1:
[0042] When implementing active disturbance rejection control (ADRC) for aerospace systems, motor systems, robotic systems, and engine systems, existing methods improve ADRC algorithms by introducing advanced control algorithms and changing the structure of ADRC controllers, thereby enhancing their disturbance rejection capability. However, the problem of incomplete disturbance elimination still exists, resulting in poor system control quality.
[0043] To address the aforementioned problems, this embodiment employs a cascaded extended state observer-based active disturbance rejection control method. It uses Taylor's formula to represent complex disturbances in the controlled system, and constructs a cascaded linear extended state observer based on the Taylor expansion of the disturbance. This significantly enhances the controlled system's ability to suppress disturbances and improves its dynamic response. The general formula of the cascaded linear extended state observer is derived, and the stability of the controller is proven using Lyapunov stability theory. The specific implementation steps of this embodiment are as follows:
[0044] S1. For a first-order controlled system, establish a mathematical model of the system's complex disturbances, express the complex disturbances in polynomial form, and determine the highest order n of the disturbances based on actual engineering conditions and control accuracy; optionally, the first-order controlled system can be an aerospace system, an electric motor system, a robot system, an engine system, or other systems with disturbances.
[0045] S2. Construct a cascaded linear extended state observer using the polynomial established in step S1, derive the general form of the cascaded linear extended state observer, accurately estimate the complex disturbance in step S1, and design a linear control law to eliminate the disturbance based on the total disturbance.
[0046] S3. Analyze the stability of the cascaded linear expansion state observer described in step S2 using Lyapunov stability theory.
[0047] Optionally, in step S1, the first-order controlled system is:
[0048] y′(t)=-ay(t)+bu(t)+ω(t) (1)
[0049] Where y'(t) is the first derivative of y(t); u(t) and y(t) are the input and output signals, respectively; a is the output gain; b is the controller gain, which is an unknown quantity; ω(t) represents a complex disturbance, which can be transformed into ω(t) = a0 + a1t using polynomial fitting. 1 +a2t 2 +…+a n t n , where n is a constant and t is time.
[0050] Assuming b0 is an estimate of the controller gain b, formula (1) is transformed into the following form:
[0051]
[0052] Let f(t) be considered as the total disturbance, then equation (2) can be rewritten as follows:
[0053]
[0054] Among them, f(t)=-ay(t)+(b-b0)u(t)+ω(t).
[0055] Let the state variables x1(t) = y(t) and x2(t) = f(t), the state-space expression of formula (3) is rewritten as follows:
[0056]
[0057] Here, x1'(t) is the first derivative of x1(t), and f'(t) represents the first derivative of x2(t).
[0058] In step S1, the highest order of the perturbation polynomial can be determined from the perspective of actual engineering requirements and control accuracy, or solely from the perspective of actual engineering requirements or control accuracy, and t can be discarded. n . higher-order infinitesimals.
[0059] Optionally, in step S2, the specific steps for building the cascaded linear expansion state observer and designing the control law are as follows:
[0060] S2.1, Cascaded Linear Expansion State Observer:
[0061] The cascaded linear extended state observer includes: linear extended state observer LESO1, linear extended state observer LESO2... linear extended state observer LESOn, and the cascaded linear extended state observer is used to estimate polynomial perturbations of the corresponding order.
[0062] The linear extended state observer LESO1 and linear extended state observer LESO2 models are as follows:
[0063]
[0064]
[0065] in, This is the difference between the predicted output and the actual output of the linearly extended state observer LESO1; and These are the observed values of x1(t) and x2(t); β1 and β2 are the observer gains of the linearly extended state observer LESO1, β1 = 2ω0, β2 = ω0. 2 ω0 is the observer bandwidth. The difference between the system state predicted by the linearly extended state observer LESO2 and the actual state; It is the observed value of x1(t) and is related to They have the same function; β3 and β4 are the observer gains of the linearly extended state observer LESO2. To simplify the parameter design, β3 = β1, β4 = β2; z2 2(t) is considered as the predicted value of the system's residual disturbance.
[0066] The general form of the cascaded linearly extended state observer is derived as follows:
[0067]
[0068] in, t is the difference between the state value predicted by the cascaded linear extended state observer and the actual state, where n is a constant and t is time; and These represent the observed values of the system output y(t) and the total disturbance, respectively. It is considered as the predicted value of the system's residual disturbance; is the predicted perturbation of the (n-1)th level extended state observer; b0 is the estimated value of the controller gain; u(t) is the input signal; The perturbation value predicted by the first linearly extended state observer LESO1; β i and β j It is the gain of the cascaded linear expansion state observer.
[0069] The characteristic equation of the cascaded linear expansion state observer is:
[0070] s 2 +β1s+β2=(s+ω0) 2 =0 (8)
[0071] To reduce the difficulty of parameter tuning, the system bandwidth method is used to tune the parameters. The gain of each stage of the extended state observer is the same, β i =β1=2ω0,β j =β2=ω0 2 ω0 is the observer bandwidth.
[0072] S2.2, The control law is:
[0073]
[0074] u0(t)=K p (x(t)-y(t))
[0075] Where u(t) is the input signal; u0(t) is the output signal; The observed values of the total disturbance are: x(t) = y(t) / K; the system input is x(t) and the system output is y(t); K is K. p This is the proportional gain.
[0076] Stability is the primary condition that a control system must satisfy. Therefore, the stability of all linearly extended state observers needs to be proven. Optionally, in step S3, the specific steps for proving stability are as follows:
[0077] S3.1 Prove the stability of the linearly extended state observer LESO1:
[0078] Let e1 be the tracking error of the linear extended state observer LESO1. According to the system's state equation and the constructor of the linear extended state observer LESO1, the error state equation of the linear extended state observer LESO1 is written as follows:
[0079]
[0080] Where, e1 = z1 - x, x = [x1 x2] T h = -ω e (t), ω e (t) is the differential form of x2(t);
[0081] According to Lyapunov stability theory analysis, when the gain of the extended state observer is positive, then D... e The eigenvalues are respectively
[0082]
[0083] And D e When the Hurwitz matrix is used, the controlled system is externally stable, meaning both the input and output are bounded. Therefore, once the observer gain is selected and the observer bandwidth ω0 is positive, the prediction error of the linearly extended state observer LESO1 is bounded for any bounded step size h.
[0084] S3.2 Prove the stability of the linearly extended state observer LESO2.
[0085] Assume e2 is the tracking error of the linearly extended state observer LESO2, and the error state equation is as follows:
[0086]
[0087] In the formula, e2 = z2 - x, Since the system is linear, under the condition that the linear extended state observer LESO1 is stable, the input state of the linear extended state observer LESO2 is bounded. Therefore, if the value of ω0 is positive, the linear extended state observer LESO2 is externally stable for any bounded quantity h.
[0088] S3.3, Prove the stability of the cascaded linear expansion state observer:
[0089] When extended to a cascaded linear extended state observer, the error state equation becomes as follows:
[0090]
[0091] in, e n z represents the state estimation error of the linearly extended state observer LESON. n For perturbation estimation of the linearly extended state observer LESON, z n-1 LESO (Linear Extended State Observer) (n-1) The perturbation estimation. When the selected observer gain is positive, D is... en The eigenvalues are all negative, satisfying the Lyapunov indirect method, D en It is BIBO (Bounded Input Bounded Output) stable. The entire system's input and output are bounded when all state observers are externally stable.
[0092] Example 2:
[0093] To fully verify the effectiveness of the method in Example 1 in eliminating complex disturbances, and to demonstrate that the cascaded extended state observer can resist disturbances whose highest order is no higher than its level, three methods, S1ADRC, S2ADRC, and S3ADRC, can be used to monitor the controlled object under conditions such as... Figure 3 The four external disturbances ω(t)(t, t) shown are 2 t 3 t 4 A comparative study of control quality under working conditions.
[0094] Optionally, four types of external disturbances ω(t)(t, t) can be selected. 2 t 3 t 4 According to ω(t)(t、t 2 t 3 Three external disturbances were cascaded into first-, second-, and third-level linear extended state observers to verify that the improved active disturbance rejection control method can accurately eliminate complex disturbances of order equal to or lower than the number of cascades n, ω(t)(t 4 This is used to demonstrate that the system becomes uncontrollable when the order of the disturbance exceeds the number of cascade observers. Next, an improved active disturbance rejection control method using cascade linear extended state observers is employed to control four types of disturbances. The specific steps are as follows:
[0095] A first-order controlled system with complex disturbances is:
[0096] y′(t)=-y(t)+4u(t)+ω(t) (1)
[0097] Where y'(t) is the first derivative of y(t); u(t) and y(t) are the input and output signals, respectively; ω(t)(t, t)2 t 3 t 4 ).
[0098] Let f(t) be considered as the total disturbance, then equation (1) can be rewritten as follows:
[0099]
[0100] Among them, f(t)=-y(t)+ω(t)(t, t 2 t 3 t 4 ).
[0101] Let the state variables x1(t) = y(t) and x2(t) = f(t), the state-space expression of formula (3) is rewritten as follows:
[0102]
[0103] Here, x1'(t) is the first derivative of x1(t), and f'(t) represents the first derivative of x2(t).
[0104] S2. Based on the established polynomial complex disturbance, construct cascaded linear extended state observers S1ADRC, S2ADRC and S3ADRC to accurately estimate the complex disturbance in step S1, and design a linear control law to eliminate the disturbance based on the total disturbance.
[0105] The specific steps for building a cascaded linear extended state observer and designing the control law are as follows:
[0106] S2.1, Cascaded Linear Expansion State Observer:
[0107] The cascaded linear expansion state observers include: linear expansion state observer LESO1, linear expansion state observer LESO2, and linear expansion state observer LESO3.
[0108] The models for the linearly extended state observers LESO1, LESO2, and LESO3 are as follows:
[0109]
[0110]
[0111]
[0112] in, The difference between the predicted output and the actual output of LESO1. and These are the observed values of x1(t) and x2(t); β1 and β2 are the observer gains of the linearly extended state observer LESO1, β1 = 2ω0, β2 = ω0. 2 ω0 is the observer bandwidth. The difference between the system state predicted by LESO2 and the actual state. For LESO3 prediction error, It is the observed value of x1(t) and is related to They have the same function; β3 and β4 are the linear extended state observer LESO2 observer gains, and β5 and β6 are the linear extended state observer LESO2 observer gains. To simplify the parameter design, β6 = β3 = β1, β6 = β4 = β2. (t) is considered as the predicted value of the system's residual disturbance.
[0113] S2.2, the control laws are as follows:
[0114]
[0115] u0(t)=K p (x(t)-y(t))
[0116]
[0117] u0(t)=K p (x(t)-y(t))
[0118]
[0119] u0(t)=K p (x(t)-y(t))
[0120] In the formula, K p Here, u0(t) is the proportional gain, u0(t) is the output signal of the proportional control, and x(t) is the reference input.
[0121] S3. Using Lyapunov stability theory, analyze the stability of the cascaded linear extended state observers in step S2, and prove the stability of the linear extended state observers LESO1, LESO2, and LESO3.
[0122] The specific steps for proving stability are as follows:
[0123] S3.1 Prove the stability of the linearly extended state observer LESO1:
[0124] Let e1 be the tracking error of the linear extended state observer LESO1. According to the system's state equation and the constructor of the extended state observer LESO1, the error state equation of the linear extended state observer LESO1 is written as follows:
[0125]
[0126] Where, e1 = z1 - x,
[0127] According to Lyapunov's indirect method, when the gains of the extended state observers are all positive, then D... e The eigenvalues are as follows:
[0128]
[0129] And D e When the Hurwitz matrix is used, the controlled system is externally stable, meaning both the input and output are bounded. Therefore, once the observer gain is selected and the observer bandwidth ω0 is positive, the prediction error of the linearly extended state observer LESO1 is bounded for any bounded step size h.
[0130] S3.2 Prove the stability of the linearly extended state observer LESO2:
[0131] Assume e2 is the tracking error of the linearly extended state observer LESO2, and the error state equation is as follows:
[0132]
[0133] In the formula, e2 = z1 - x, Since the system is linear, under the condition that the linear extended state observer LESO1 is stable, the input state of the linear extended state observer LESO2 is bounded. Therefore, if the value of ω0 is positive, the linear extended state observer LESO2 is externally stable for any bounded quantity h.
[0134] S3.3 Prove the stability of the linearly extended state observer LESO3:
[0135] The error state equation for LESO3 is as follows:
[0136]
[0137] In the formula, e3 represents the state estimation error of the linearly extended state observer LESO3, and z3 is the perturbation estimate of the linearly extended state observer LESO3. When the selected observer gain is positive, D...e3 The eigenvalues are all negative, satisfying the Lyapunov indirect method, D e3 For BIBO stability. When all state observers are externally stable, the entire system's inputs and outputs are bounded.
[0138] The response curve of S1ADRC against complex external disturbances is as follows: Figure 4 As shown. When the input is a step signal with an amplitude of 1 and the input disturbance is t, the S1ADRC response curve 35 is in a steady state. When the disturbance becomes a second-order t... 2 Third order t 3 and fourth order t 4 The response curves 36, 37 and 38 continuously increase monotonically, showing a divergent state, and the system's output curve cannot enter a stable state.
[0139] The response curve of S2ADRC against complex external disturbances is as follows: Figure 5 As shown. S2ADRC eliminates disturbances t, t 2 Regarding the impact on the system, response curves 39 and 310 are in a stable state. When the order of the disturbance increases to the third and fourth order, response curves 311 and 312 monotonically increase, indicating an out-of-control state.
[0140] The response curve of S3ADRC against complex external disturbances is as follows: Figure 6 As shown, S3ADRC can eliminate disturbances with a maximum order of third order, and response curves 313, 314, and 315 are all in a stable state. When the applied disturbance is a fourth order t 4 At that time, the response curve 316 monotonically increases and cannot reach a steady state.
[0141] Example 3:
[0142] This embodiment provides an active disturbance rejection control system for a cascaded extended state observer, including:
[0143] The complex disturbance processing module is configured to: represent the complex disturbances existing in the controlled system as a polynomial, and determine the highest order of the polynomial based on the control accuracy.
[0144] The cascaded linear extended state observer building module is configured to: build a cascaded linear extended state observer including multiple linear extended state observers to estimate polynomial perturbations of the corresponding order; the number of linear extended state observers is equal to the degree of the polynomial;
[0145] The control module is configured to: design a linear control law based on the constructed cascaded linear expansion state observer; and use the designed linear control law to control and eliminate disturbances.
[0146] The operating method of the system is the same as the active disturbance rejection control method of the cascaded extended state observer in Embodiment 1, and will not be repeated here.
[0147] Example 4:
[0148] This embodiment provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of the active disturbance rejection control method for the cascaded extended state observer described in Embodiment 1.
[0149] Example 5:
[0150] This embodiment provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, it implements the steps of the active disturbance rejection control method of the cascaded extended state observer described in Embodiment 1.
[0151] The above description is merely a preferred embodiment of this practice and is not intended to limit the scope of this practice. Various modifications and variations can be made to this practice by those skilled in the art. Any modifications, equivalent substitutions, or improvements made within the spirit and principles of this practice should be included within the protection scope of this practice.
Claims
1. A method for active disturbance rejection control of a cascaded extended state observer, characterized in that, include: The complex disturbances existing in the controlled system are expressed in the form of a polynomial, and the highest order of the polynomial is determined according to the control accuracy. The first-order controlled system is: + ω ( t ) y '( t )yes y ( t The first derivative of ) u ( t )and y ( t These are the input signal and the output signal, respectively. a It is the output gain; b It is the controller gain, which is an unknown quantity; ω ( t () represents a complex perturbation, which can be transformed into a polynomial fitting method. ω ( t )= a 0 + a 1 t 1 +a 2 t 2 +…+ a n t n ,in, n It is a constant. t For time; Assumption b 0 is the controller gain b The estimated value is transformed into the following form: set up f ( t This is considered as the total disturbance and rewritten as follows: in, f ( t )=- ay ( t )+( bb 0) u ( t )+ ω ( t ); Construct a cascaded linear extended state observer consisting of multiple linear extended state observers to estimate polynomial perturbations of the corresponding order; the number of linear extended state observers is equal to the degree of the polynomial. Assuming that the observer gains at corresponding positions in a plurality of linearly extended state observers are equal, the general formula of the cascaded linearly extended state observer is derived. The general form of the cascaded linear extended state observer is: in, This represents the difference between the state values predicted by the cascaded linearly extended state observer and the actual state. n It is a constant. t For time; and Representing system output respectively y ( t ) and the observed values of the total disturbance; It is considered as the predicted value of the system's residual disturbance; For the ( n -1) Predicted perturbations of the extended state observer; This is an estimate of the controller gain; For input signals; The perturbation value predicted by the first linearly extended state observer; β i and β j It is the gain of the cascaded linearly extended state observer; Based on the constructed cascaded linear extended state observer, design a linear control law; use the designed linear control law to control and eliminate disturbances. The control law is as follows: in, For input signals; For output signal; The observed value represents the total disturbance. Input for the system; For system output; The proportional gain is used to design a linear control law to eliminate the disturbance based on the total disturbance.
2. The active disturbance rejection control method for a cascaded extended state observer as described in claim 1, characterized in that, The coefficient of each term in the polynomial represents the output gain at different times.
3. The active disturbance rejection control method for a cascaded extended state observer as described in claim 1, characterized in that, The gain of each extended state observer is the same: β i = β 1=2 ω 0, β j = β 2 =ω 0 2 , ω 0 represents the observer bandwidth.
4. The active disturbance rejection control method for a cascaded extended state observer as described in claim 1, characterized in that, The stability of the cascaded linear expansion state observer is analyzed using Lyapunov stability theory.
5. A cascaded extended state observer's active disturbance rejection control system, characterized in that, include: The complex disturbance processing module is configured to: represent the complex disturbances existing in the controlled system as a polynomial, and determine the highest order of the polynomial based on the control accuracy. The first-order controlled system is: + ω ( t ) y '( t )yes y ( t The first derivative of ) u ( t )and y ( t These are the input signal and the output signal, respectively. a It is the output gain; b It is the controller gain, which is an unknown quantity; ω ( t () represents a complex perturbation, which can be transformed into a polynomial fitting method. ω ( t )= a 0 + a 1 t 1 +a 2 t 2 +…+ a n t n ,in, n It is a constant. t For time; Assumption b 0 is the controller gain b The estimated value is transformed into the following form: set up f ( t This is considered as the total disturbance and rewritten as follows: in, f ( t )=- ay ( t )+( bb 0) u ( t )+ ω ( t ); The cascaded linear extended state observer building module is configured to: build a cascaded linear extended state observer including multiple linear extended state observers to estimate polynomial perturbations of the corresponding order; the number of linear extended state observers is equal to the degree of the polynomial; Assuming that the observer gains at corresponding positions in a plurality of linearly extended state observers are equal, the general formula of the cascaded linearly extended state observer is derived. The general form of the cascaded linear extended state observer is: in, This represents the difference between the state values predicted by the cascaded linearly extended state observer and the actual state. n It is a constant. t For time; and Representing system output respectively y ( t ) and the observed values of the total disturbance; It is considered as the predicted value of the system's residual disturbance; For the ( n -1) Predicted perturbations of the extended state observer; This is an estimate of the controller gain; For input signals; The perturbation value predicted by the first linearly extended state observer; β i and β j It is the gain of the cascaded linearly extended state observer; The control module is configured to: design a linear control law based on the constructed cascaded linear expansion state observer; and use the designed linear control law to control and eliminate disturbances. The control law is as follows: in, For input signals; For output signal; The observed value represents the total disturbance. Input for the system; For system output; The proportional gain is used to design a linear control law to eliminate the disturbance based on the total disturbance.
6. A computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by the processor, the program implements the steps of the active disturbance rejection control method for the cascaded extended state observer as described in any one of claims 1-4.
7. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the steps of the active disturbance rejection control method for the cascaded extended state observer as described in any one of claims 1-4.
Citation Information
Patent Citations
Trajectory boosting segment state deviation parsing and calculating method under action of disturbance gravitation
CN105184109A
Magnetically suspended control moment gyroscope framework angular rate servo system based on harmonic reducer
CN107992110A
Early warning and monitoring method and device for combustion oscillation phenomenon
CN113267291A
Active disturbance rejection control method of multivariable internal model structure based on photovoltaic grid-connected inverter
CN115437246A
