A generalized estimator, a generalized anti-disturbance controller and a design method

Abstract

The invention relates to a generalized estimator, a generalized anti-disturbance controller and a design method, and belongs to the field of estimators. The generalized estimator comprises an inner-loop estimation controller and a nominal model module of a controlled object. The inner-loop estimation controller transmits a control signal to the nominal model module, and the nominal model module contains a nominal model of the controlled object. The generalized estimator has a unified structure and functions that can be mutually converted. The structure of the generalized estimator unifies typical anti-disturbance controllers into the same control framework to obtain a more universal generalized anti-disturbance controller, and the generalized anti-disturbance controller can design different control strategies according to the existing conditions of most systems.

Description

Technical Field

[0001] The present invention relates to the field of estimators, and in particular to a generalized estimator, a generalized anti-disturbance controller and a design method. Background Art

[0002] One of the major tasks in the design of electromechanical control systems is: what strategies should be adopted to suppress various disturbances. For example, 1) Disturbance Observer Based Control (DOBC), 2) Tornambè Controller (TC), 3) Active Disturbance Rejection Control (ADRC), 4) Desired Dynamic PID (DDE-PID), 5) Cascade Control (CC) and other control structures all use different control strategies to suppress various disturbances. The application of these disturbance rejection controllers has different prerequisites and disadvantages. 1) DOBC requires the inversion of the nominal model of the controlled object, and the design process is more cumbersome when the nominal object is not reversible and the relative order is greater than or equal to 2; 2) TC requires the measurable information of derivatives of all orders less than the relative order; 3) ADRC considers the nominal model of all objects to be integral series type, and fails to make full use of model information; 4) DDE-PID requires special processing of the differential link and fails to make full use of model information; 5) CC requires the measurable intermediate variables of the controlled object.

[0003] In addition, in actual engineering, it is often necessary to design an estimator based on the prior knowledge of the model, hardware conditions, the knowledge level of the engineer, and the structure selection. However, the design principles of the above-mentioned different estimators are quite different, and it is difficult for engineers to be familiar with their performance and design a suitable estimator. Therefore, it is urgent to propose a practical estimator that is unified in structure and can be converted into each other in function. This estimator should have the following characteristics: even in different situations, it should follow the same method in structure and design, and in terms of function, it can be adjusted according to different model information and estimation requirements. Summary of the invention

[0004] The purpose of the present invention is to provide a generalized estimator, a generalized anti-disturbance controller and a design method, so as to propose a generalized estimator with unified structure and mutually transformable functions and a universal anti-disturbance controller, which can design different control strategies according to the existing conditions of most systems.

[0005] To achieve the above object, the present invention provides the following solutions:

[0006] A generalized estimator, comprising: an inner loop estimation controller and a nominal model module of a controlled object;

[0007] The inner loop estimation controller is connected to the nominal model module of the controlled object;

[0008] The inner loop estimation controller is used to transmit a control signal to the nominal model module according to the input quantity and the virtual measured output quantity;

[0009] The nominal model module is used to input the control signal into the nominal model of the controlled object, obtain the controlled output vector and the virtual measured output quantity, and transmit the virtual measured output quantity to the inner loop estimation controller.

[0010] Optionally, the transfer function of the generalized estimator includes: the input quantity y to the controlled output vector y v The transfer function of the input quantity y to the virtual measured output quantity y c The transfer function of

[0011] The input quantity y to the controlled output vector y v The transfer function is In the formula, Represents the input y to the controlled output vector y v The transfer function, G 2 represents the second generalized object transfer matrix, E B represents the first controller transfer matrix, E F represents the second controller transfer matrix;

[0012] The input quantity y to the virtual measured output quantity y c The transfer function is In the formula, Represents the input quantity y to the virtual measurement output quantity y c The transfer function, G 1 Represents the first generalized object transfer matrix.

[0013] Optionally, when the nominal model of the controlled object is known and there is only one nominal model, the control signal u e is the estimated value of the inverse of the controlled object, and the generalized estimator is equivalent to a disturbance observer;

[0014] When the nominal model of the controlled object is known and there are two or more nominal models, the control signal u e is the estimated value of the inverse of the controlled object, the output of the nominal model other than the last level nominal model is the estimated value of the intermediate variable of the system, and the generalized estimator is equivalent to a controller of cascade structure;

[0015] When the nominal model of the controlled object is an integral series type, the output quantity and control signal u of the nominal model other than the last level nominal model e They are all estimated values ​​of the derivative information of the system, and the generalized estimator is equivalent to the extended state observer of the active disturbance rejection controller;

[0016] When the derivative information of the controlled object output is directly measurable, the output of the nominal model other than the last level nominal model is the measured value of the derivative information of the system, and the generalized estimator is equivalent to a Tornambe type robust controller.

[0017] Optionally, the nominal model of the controlled object in the nominal model module is one or more;

[0018] When there are multiple nominal models of the controlled object, the multiple nominal models are connected in cascade.

[0019] A generalized disturbance rejection controller, comprising: an outer loop controller and the aforementioned generalized estimator;

[0020] The input end of the outer loop controller is connected to the output end of the nominal model module in the generalized estimator, and the output end of the outer loop controller is connected to the input end of the controlled object; the outer loop controller is used to obtain the control amount according to the reference signal and the controlled output vector, and transmit the control amount to the controlled object;

[0021] The output end of the controlled object is connected to the input end of the generalized estimator; the controlled object is used to generate an output quantity under the control of the control quantity and the interference of the disturbance quantity, and the output quantity is input into the generalized estimator as the input quantity of the generalized estimator;

[0022] The generalized estimator is used to obtain a controlled output vector according to an input quantity, and transmit the controlled output vector to an outer loop controller.

[0023] Optionally, the outer loop controller includes: a controller C, a first difference module, a second difference module and a generalized filter;

[0024] The input end of the controller C is connected to the output end of the nominal model module in the generalized estimator, and the output end of the controller C is connected to the first input end of the first difference module; the controller C is used to input a reference signal and output a tracking control component according to the reference signal and the virtual measurement output;

[0025] The second input end of the first difference module is connected to the output end of the generalized filter, and the output end of the first difference module is respectively connected to the first input end of the second difference module and the input end of the controlled object; the first difference module is used to subtract the disturbance estimation output by the generalized filter from the tracking control component to obtain the control amount, and transmit the control amount to the second difference module and the controlled object respectively;

[0026] The second input end of the second difference module is connected to the output end of the inner loop estimation controller in the generalized estimator, and the output end of the second difference module is connected to the input end of the generalized filter; the second difference module is used to transmit the control signal minus the control amount to the generalized filter.

[0027] Optionally, the control formula in the outer loop controller is:

[0028]

[0029] In the formula, u represents the control quantity, u 0 represents the tracking control component, represents the disturbance estimator, C F represents the feedforward controller, C B represents the feedback controller, Q represents the quality factor of the generalized filter, K Q represents the filter coefficient, s represents the differential operator, r represents the reference signal, y v Represents the virtual measurement output, y v :=(y v1 ,y v2 ,y v3 , ..., y v(m-1) ,y vm ) T ,y vi Represents the output of the m-(i-1)th nominal model, i=1,2,3,…,m-1,m; symbol: = is a definition symbol.

[0030] Optionally, the transfer function of the generalized anti-disturbance controller includes: a transfer function from a reference signal r to an input quantity y, and a transfer function from a disturbance quantity d to an input quantity y;

[0031] The transfer function from the reference signal r to the input quantity y is: In the formula, represents the transfer function from the reference signal r to the input quantity y, G p Represents the input-output relationship of the real controlled object, C BQ represents the feedback controller and the generalized filter vector, C BQ :=(C B Q), Represents the input y to the controlled output vector y v The transfer function of

[0032] The transfer function from the disturbance d to the input y is: In the formula, Represents the transfer function from the disturbance d to the input y.

[0033] A generalized anti-disturbance controller design method, the design method comprising:

[0034] Determine the nominal model of the controlled object;

[0035] According to the nominal model of the controlled object, the parameters of the inner loop estimation controller in the generalized estimator are adjusted;

[0036] Connect the input end of the tuned generalized estimator to the output interface of the controlled object, and connect the output end of the tuned generalized estimator to the input interface of the outer loop controller to complete the structural configuration of the generalized disturbance rejection controller;

[0037] The parameters of the outer loop controller are tuned to obtain a tuned generalized disturbance rejection controller.

[0038] Optionally, determining the nominal model of the controlled object specifically includes:

[0039] When the controlled object is a minimum phase electromechanical system with a relative order of 1 When , the nominal model is In the formula, G p1 Represents the input-output relationship of the electromechanical system with the minimum phase, G n1 Represents G p1 The corresponding nominal model transfer function, s represents the differential operator;

[0040] When the controlled object is a non-minimum phase electromechanical system with a relative order of 1 When , the nominal model is In the formula, G p2 Represents the input-output relationship of the electromechanical system with non-minimum phase, G n2 Represents G p2 The corresponding nominal model transfer function;

[0041] When the controlled object is a minimum phase electromechanical system with a relative order of 2 When , the nominal model is In the formula, G p3 Represents the input-output relationship of the electromechanical system with the minimum phase of relative order 2, G n3 Represents G p3 The corresponding nominal model transfer function;

[0042] When the controlled object is a cascade electromechanical system with a relative order of 2 and a minimum phase When , the nominal model is and In the formula, G p Represents the input-output relationship of the cascade electromechanical system with the minimum phase of relative order 2, G′ n1 represents the first-level nominal model transfer function of the cascade electromechanical system, G′ n2 represents the second-stage nominal model transfer function of the cascaded electromechanical system.

[0043] According to the specific embodiments provided by the present invention, the present invention discloses the following technical effects:

[0044] The present invention discloses a generalized estimator, a generalized anti-disturbance controller and a design method. The generalized estimator comprises an inner-loop estimation controller and a nominal model module of a controlled object. The inner-loop estimation controller transmits a control signal to the nominal model module. The nominal model module contains a nominal model of the controlled object. The generalized estimator has a unified structure and its functions can be mutually converted. The structure of the generalized estimator unifies typical anti-disturbance controllers into the same control framework to obtain a more generalized anti-disturbance controller. The generalized anti-disturbance controller can design different control strategies according to the existing conditions of most systems.

[0045] The generalized anti-disturbance controller in the present invention can easily derive estimated information such as inverse and intermediate variables of the controlled object according to the requirements of the feedback system. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0047] Figure 1 A structural diagram of a generalized estimator provided by an embodiment of the present invention;

[0048] Figure 2 A structural diagram of a generalized anti-disturbance controller provided by an embodiment of the present invention;

[0049] Figure 3 A flow chart of a generalized anti-disturbance controller design method provided by an embodiment of the present invention;

[0050] Figure 4 A comparison diagram of the anti-disturbance performance of an electromechanical system with a minimum phase of relative order 1 provided by an embodiment of the present invention under a traditional disturbance observer-based control and a generalized estimator-based disturbance observer control; Figure 4 (a) is the comparison effect diagram of the input y. Figure 4 (b) is the comparison effect diagram of the control amount u;

[0051] Figure 5 A comparative effect diagram of the anti-disturbance performance of a non-minimum phase electromechanical system with a relative order of 1 provided by an embodiment of the present invention under a traditional disturbance observer-based control and a generalized estimator-based disturbance observer control; Figure 5 (a) is the comparison effect diagram of the input y. Figure 5 (b) is the comparison effect diagram of the control amount u;

[0052] Figure 6A comparison diagram of the tracking performance and disturbance rejection performance of the traditional active disturbance rejection control and the active disturbance rejection control based on the generalized estimator provided in the embodiment of the present invention; Figure 6 (a) is the comparison effect diagram of the input y. Figure 6 (b) is the comparison effect diagram of the control amount u;

[0053] Figure 7 A comparison diagram of the tracking performance and anti-disturbance performance of a traditional Tornambe type robust controller and a Tornambe type robust controller based on a generalized estimator provided in an embodiment of the present invention; Figure 7 (a) is the comparison effect diagram of the input y. Figure 7 (b) is the comparison effect diagram of the control amount u;

[0054] Figure 8 A comparison diagram of the tracking performance and anti-disturbance performance of the traditional cascade control provided in the embodiment of the present invention and the cascade control based on the generalized estimator proposed in this patent; Figure 8 (a) is the comparison effect diagram of the input y. Figure 8 (b) in the figure is a comparison diagram of the control value u. DETAILED DESCRIPTION

[0055] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0056] The purpose of the present invention is to provide a generalized estimator, a generalized anti-disturbance controller and a design method, so as to propose a generalized estimator with unified structure and mutually transformable functions and a universal anti-disturbance controller, which can design different control strategies according to the existing conditions of most systems.

[0057] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0058] Example 1

[0059] After conducting a large amount of theoretical and applied research on various observers, the present invention deeply explored their essence and the internal connection between them, and proposed an estimator that is more general in structure and more diverse in function, named Generalized Estimating (GE). This estimator builds a bridge in structure, through which the functional transformation between different observers can be realized. Specifically, GE is a closed-loop estimator composed of a standard condition model and a controller. The standard condition model can be an exact model, a weak model or even a model-free model according to the researcher's prior knowledge of the model information.

[0060] See also Figure 1 A generalized estimator provided by an embodiment of the present invention includes: an inner loop estimation controller and a nominal model module of a controlled object. The inner loop estimation controller is connected to the nominal model module of the controlled object. The inner loop estimation controller is used to transmit a control signal to the nominal model module according to an input quantity and a virtual measured output quantity. The nominal model module is used to input the control signal into the nominal model of the controlled object, obtain a controlled output vector and a virtual measured output quantity, and transmit the virtual measured output quantity to the inner loop estimation controller.

[0061] Figure 1 middle, y c is the controlled output vector, y v Represents the virtual measurement output. The input and output relationship of GE can be expressed as

[0062]

[0063] Here, G and E represent the generalized plant transfer matrix and the controller transfer matrix, respectively. 1 represents the first generalized object transfer matrix, G 2 represents the second generalized object transfer matrix, represents the estimated value of the controlled output vector, represents the virtual measurement output estimate, u e is the control signal, represents the estimated value of the control signal, represents the estimated value of the input quantity, E B represents the first controller transfer matrix, E F represents the second controller transfer matrix. Generally

[0064] y c :=(y e1 ,y e2 ,y e3 ,…,y e(m-1) ,y em ,u e ) T

[0065] y v :=(y v1 ,y v2 ,y v3 ,…,y v(m-1) ,y vm ) T

[0066] Among them, y ei represents the output of the m-(i-1)th nominal model, i = 1, 2, 3, ..., m-1, m; the symbol : = is a definition symbol. At this time,

[0067]

[0068] Among them, T 1 and T 2 is the transfer matrix, T 1 Generally, it is a unit matrix, G ni represents the i-th nominal model. T 2 Generally, it is the following diagonal matrix

[0069]

[0070] At this time, G 2 Simplified to

[0071]

[0072] Therefore, GE can be described as

[0073]

[0074]

[0075] in, and y to y respectively v and c The transfer function of .

[0076] Exemplarily, when the nominal model of the controlled object is known and there is only one nominal model, the control signal u e is the estimated value of the inverse of the controlled object, and the generalized estimator is equivalent to a disturbance observer.

[0077] When the nominal model of the controlled object is known and there are two or more nominal models, the control signal u e is the estimated value of the inverse of the controlled object, the output of the nominal model other than the last level nominal model is the estimated value of the intermediate variable of the system, and the generalized estimator is equivalent to a cascade structure controller.

[0078] When the nominal model of the controlled object is an integral series type, the output quantity and control signal u of the nominal model other than the last level nominal model e They are all estimated values ​​of the derivative information of the system, and the generalized estimator is equivalent to the extended state observer of the active disturbance rejection controller.

[0079] When the derivative information of the controlled object output is directly measurable, the output of the nominal model other than the last level nominal model is the measured value of the derivative information of the system, and the generalized estimator is equivalent to a Tornambe type robust controller.

[0080] The nominal model of the controlled object in the nominal model module is one or more. When there are multiple nominal models of the controlled object, the multiple nominal models are connected in a cascade form.

[0081] Example 2

[0082] The embodiment of the present invention provides a generalized anti-disturbance controller, such as Figure 2 As shown, the generalized anti-disturbance controller includes: an outer loop controller and a generalized estimator of Example 1. The input end of the outer loop controller is connected to the output end of the nominal model module in the generalized estimator, and the output end of the outer loop controller is connected to the input end of the controlled object. The outer loop controller is used to obtain the control quantity according to the reference signal and the controlled output vector, and transmit the control quantity to the controlled object. The output end of the controlled object is connected to the input end of the generalized estimator, and the controlled object is used to generate an output quantity under the control of the control quantity and the interference of the disturbance quantity, and the output quantity is input into the generalized estimator as the input quantity of the generalized estimator. The generalized estimator is used to obtain the controlled output vector according to the input quantity, and transmit the controlled output vector to the outer loop controller.

[0083] In one example, the outer loop controller includes: a controller C, a first difference module, a second difference module and a generalized filter. The input end of the controller C is connected to the output end of the nominal model module in the generalized estimator, and the output end of the controller C is connected to the first input end of the first difference module. The controller C is used to input a reference signal and output a tracking control component according to the reference signal and the virtual measurement output. The second input end of the first difference module is connected to the output end of the generalized filter, and the output end of the first difference module is respectively connected to the first input end of the second difference module and the input end of the controlled object. The first difference module is used to subtract the disturbance estimation amount output by the generalized filter from the tracking control component to obtain the control amount, and transmit the control amount to the second difference module and the controlled object respectively. The second input end of the second difference module is connected to the output end of the inner loop estimation controller in the generalized estimator, and the output end of the second difference module is connected to the input end of the generalized filter. The second difference module is used to transmit the control signal to the generalized filter after subtracting the control amount.

[0084] The control formula in the outer loop controller is:

[0085]

[0086] In the formula, u represents the control quantity, u 0 represents the tracking control component, represents the disturbance estimator, C F represents the feedforward controller, C B represents the feedback controller, Q represents the quality factor of the generalized filter, r represents the reference signal, y v Represents the virtual measurement output, y v :=(y v1 ,y v2 ,y v3 , ..., y v(m-1) ,y vm ) T ,y vi Represents the output of the m-(i-1)th nominal model, i=1,2,3,…,m-1,m; symbol: = is a definition symbol.

[0087] The transfer function of the generalized disturbance rejection controller includes not only the two transfer functions of the generalized estimator, but also the transfer function from the reference signal r to the input quantity y and the transfer function from the disturbance quantity d to the input quantity y.

[0088] The transfer function from the reference signal r to the input quantity y is In the formula, represents the transfer function from the reference signal r to the input quantity y, G p Represents the input-output relationship of the real controlled object, C BQ represents the feedback controller and the generalized filter vector, C BQ :=(C B Q), Represents the input y to the controlled output vector y v The transfer function of .

[0089] The transfer function from the disturbance d to the input y is: In the formula, Represents the transfer function from the disturbance d to the input y.

[0090] The generalized disturbance rejection controller provides a unified disturbance rejection control system design framework, which covers the control structures of 1) disturbance observer based control (DOBC), 2) Tornambè controller (TC), 3) active disturbance rejection control (ADRC), 4) expected dynamic PID (DDE-PID), 5) cascade control (CC), etc., so it is called the generalized disturbance rejection controller. The generalized disturbance rejection controller can easily derive the estimated information of the inverse and intermediate variables of the controlled object according to the needs of the feedback system.

[0091] Example 3

[0092] The embodiment of the present invention proposes a generalized anti-disturbance controller design method for the generalized anti-disturbance controller of embodiment 2, such as Figure 3 As shown, the design method includes the following steps:

[0093] Step S1, determining the nominal model of the controlled object.

[0094] The following example illustrates how to determine the nominal model of the controlled object:

[0095] When the controlled object is a minimum phase electromechanical system with a relative order of 1 When , the nominal model is In the formula, G p1 Represents the input-output relationship of the electromechanical system with the minimum phase, G n1 Represents G p1 The corresponding nominal model transfer function, s represents the differential operator;

[0096] When the controlled object is a non-minimum phase electromechanical system with a relative order of 1 When , the nominal model is In the formula, G p2 Represents the input-output relationship of the electromechanical system with non-minimum phase, G n2 Represents G p2 The corresponding nominal model transfer function;

[0097] When the controlled object is a minimum phase electromechanical system with a relative order of 2 When , the nominal model is In the formula, G p3 Represents the input-output relationship of the electromechanical system with the minimum phase of relative order 2, G n3 Represents G p3 The corresponding nominal model transfer function;

[0098] When the controlled object is a cascade electromechanical system with a relative order of 2 and a minimum phase When , the nominal model is and In the formula, G p Represents the input-output relationship of the cascade electromechanical system with the minimum phase of relative order 2, G′ n1 represents the first-level nominal model transfer function of the cascade electromechanical system, G′ n2 represents the second-stage nominal model transfer function of the cascaded electromechanical system.

[0099] Step S2, adjusting the parameters of the inner loop estimation controller in the generalized estimator according to the nominal model of the controlled object.

[0100] GE is the nominal model G of the cascade form of the inner loop estimation controller E and the controlled object n1 , G n2 To G nm A closed-loop system such as Figure 1 The GE structure diagram is shown in Figure 1. The parameters of the inner loop estimation controller E are adjusted so that the GE output y e1 Approximate the GE input y as quickly as possible. In principle, the controller E can be any controller. Here, a controller based on the expected dynamic proportional integral derivative (DDE-PID) is recommended.

[0101]

[0102] Among them, P, I, D i ,b,i=1,2,…,i=m-1 are the parameters to be adjusted, specifically

[0103]

[0104] A commonly used relationship definition is as follows

[0105]

[0106] Among them, ω e is the expected bandwidth of GE. Another commonly used relationship is

[0107]

[0108] Therefore, GE only has three parameters that need to be adjusted ω e ,γ,l. A simple engineering method is used to adjust these three parameters.

[0109] 1) ω e Set as the cutoff frequency of the controlled object;

[0110] 2) Select γ according to the control system requirements;

[0111] 3) Choose a sufficiently large l;

[0112] 4) When the control requirements are not met, reduce l until the system is critically stable;

[0113] 5) If in this ω e If l does not meet the control requirements, increase ω e And repeat steps 3) and 4);

[0114] 6) Repeat step 5) until the control requirements are met.

[0115] Then, GE will have the following properties

[0116] 1. When the nominal model of the controlled object is known and there is only one nominal model, ue is the estimated value of the inverse of the controlled object. At this time, the generalized estimator is equivalent to a disturbance observer, and the generalized estimator estimates the inverse of the nominal model in a simple and indirect inversion method.

[0117] 2. When the nominal model of the controlled object is known and there are two or more nominal models, u e is the estimated value of the inverse of the controlled object, y e2 To e(m-1) ,y em is the estimated value of the intermediate variable of the system. At this time, the generalized estimator is equivalent to the control of the cascade structure, and the generalized estimator estimates the intermediate variables by a simple and indirect measurement method.

[0118] 3. When the nominal model of the controlled object is an integral series type, that is, G n1 =G n2 =... =G nm =1 / s, the generalized estimator is equivalent to the extended state observer of the active disturbance rejection controller, y e2 To e(m-1) ,y em 、u e is an estimate of the derivative information of the system.

[0119] 4. When the derivative information of the controlled object output is directly measurable, the generalized estimator is equivalent to a Tornambe type robust controller, y e2 To e(m-1) ,y em is a measure of the derivative information of the system.

[0120] Through the above analysis, it is not difficult to see that the generalized estimator provides a unified design framework for different control methods (control based on disturbance observer, cascade control, active disturbance rejection controller, Tornambe type robust controller), and the tuning method is simple and the physical meaning is clear. Under this framework, the generalized estimator can realize equivalent control strategies.

[0121] Step S3, connecting the input end of the tuned generalized estimator to the output interface of the controlled object, and connecting the output end of the tuned generalized estimator to the input interface of the outer loop controller, to complete the structural configuration of the generalized anti-disturbance controller.

[0122] Interface connection: Configure the generalized disturbance rejection controller (GDRC) structure. Connect the input y of the tuned generalized estimator to the output interface of the controlled system, and connect the output u of the generalized estimator to the output interface of the controlled system. e ,y e1 To e(m-1) ,y em Connect with the corresponding interface of the controller to complete the generalized anti-disturbance controller structure configuration, such as Figure 2 shown.

[0123] Step S4, tuning the parameters of the outer loop controller to obtain a tuned generalized anti-disturbance controller.

[0124] Adjust the parameters of the generalized disturbance rejection controller. Design based on the system robustness, disturbance rejection, and tracking performance Figure 2 The structure of the controller C and the generalized filter Q and their parameters are adjusted, where the input of C is the reference signal r and the feedback signal y v .

[0125]

[0126] and are the transfer functions from r and d to y, respectively, expressed as

[0127]

[0128]

[0129] in,

[0130] C:=(C F -C B )

[0131] C BQ :=(C B Q)

[0132] In principle, the controller C can be any controller. Here, we recommend the design method of the inner loop estimation controller E, that is, the DDE-PID controller. Its tuning process is also consistent with E and will not be repeated here.

[0133]

[0134] Among them, K Q is the filter coefficient.

[0135] The present invention designs the controller of GE based on the basic idea of ​​expected dynamic parameterization. The closed-loop characteristics of GE are configured using DDE PID, and the closed-loop dynamics are finally adjusted to a more standardized expected dynamic. The characteristics of GE on this expected dynamic are easier to analyze, and this method based on expected dynamic parameterization is easy to adjust and has clear physical meaning.

[0136] The generalized estimator of the inner loop of the generalized disturbance rejection controller is designed by the expected dynamic PID, so the estimation performance of the generalized estimator can be intuitively evaluated based on the gap with the expected dynamics. The convergence speed of the generalized estimator of the inner loop of the generalized disturbance rejection controller can be intuitively adjusted by the expected dynamic PID.

[0137] The following is combined with Figures 4 to 8 The specific implementation mode and working process of the present invention are further described.

[0138] For a relative order 1 minimum phase electromechanical system The nominal model is obtained as Using the nominal model, an inner loop estimation controller E is designed to make its dynamic characteristics as fast as possible. According to the system robustness, anti-disturbance and tracking performance requirements, the controller C and generalized filter Q are designed. Figure 4 The figure shows the comparative effect of the traditional disturbance observer-based control and the generalized estimator-based disturbance observer control proposed in this patent in terms of anti-disturbance performance.

[0139] For a non-minimum phase electromechanical system with relative order 1 Using the nominal model, an inner loop estimation controller E is designed to make its dynamic characteristics as fast as possible. According to the system robustness, anti-disturbance and tracking performance requirements, the controller C and generalized filter Q are designed. Figure 5 The figure shows the comparative effect of the traditional disturbance observer-based control and the generalized estimator-based disturbance observer control proposed in this patent in terms of anti-disturbance performance.

[0140] For a relative order 2 minimum phase electromechanical system Using the nominal model, an inner loop estimation controller E is designed to make its dynamic characteristics as fast as possible. According to the system robustness, anti-disturbance and tracking performance requirements, the controller C and generalized filter Q are designed. Figure 6 The figure shows the comparative effect of traditional active disturbance rejection control and active disturbance rejection control based on generalized estimator proposed in this patent in terms of tracking performance and disturbance rejection performance.

[0141] For a relative order 2 minimum phase electromechanical system Using the nominal model, an inner loop estimation controller E is designed to make its dynamic characteristics as fast as possible. According to the system robustness, anti-disturbance and tracking performance requirements, the controller C and generalized filter Q are designed. Figure 7 The figure shows the comparative effect of the traditional Tornambe type robust controller and the Tornambe type robust controller based on the generalized estimator proposed in this patent in terms of tracking performance and anti-disturbance performance.

[0142] For a relative order 2 minimum phase cascade electromechanical system The nominal model is obtained as Using the nominal model, an inner loop estimation controller E is designed to make its dynamic characteristics as fast as possible. According to the system robustness, anti-disturbance and tracking performance requirements, the controller C and generalized filter Q are designed. Figure 8 The figure shows the comparative effect of traditional cascade control and the cascade control based on generalized estimator proposed in this patent in terms of tracking performance and anti-disturbance performance.

[0143] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the various embodiments can be referenced to each other.

[0144] The principles and implementation methods of the present invention are described in this article using specific examples. The description of the above embodiments is only used to help understand the method and core idea of ​​the present invention. At the same time, for those skilled in the art, according to the idea of ​​the present invention, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting the present invention.

Claims

1. A generalized estimator, It is characterized in that The generalized estimator includes: an inner loop estimation controller and a nominal model module of the controlled object; The inner loop estimation controller is connected to the nominal model module of the controlled object; The inner loop estimation controller is used to transmit a control signal to the nominal model module according to the input quantity and the virtual measured output quantity; The nominal model module is used to input the control signal into the nominal model of the controlled object, obtain the controlled output vector and the virtual measured output, and transmit the virtual measured output to the inner loop estimation controller; When the nominal model of the controlled object is known and there is only one nominal model, the control signal u e is the estimated value of the inverse of the controlled object, and the generalized estimator is equivalent to a disturbance observer; When the nominal model of the controlled object is known and there are two or more nominal models, the control signal u e is the estimated value of the inverse of the controlled object, the output of the nominal model other than the last level nominal model is the estimated value of the intermediate variable of the system, and the generalized estimator is equivalent to a controller of cascade structure; When the nominal model of the controlled object is an integral series type, the output quantity and control signal u of the nominal model other than the last level nominal model e They are all estimated values ​​of the derivative information of the system, and the generalized estimator is equivalent to the extended state observer of the active disturbance rejection controller; When the derivative information of the controlled object output is directly measurable, the output of the nominal model other than the last level nominal model is the measured value of the derivative information of the system, and the generalized estimator is equivalent to a Tornambe type robust controller.

2. The generalized estimator according to claim 1, It is characterized in that The transfer function of the generalized estimator includes: the input quantity y to the controlled output vector y v The transfer function of the input quantity y to the virtual measured output quantity y c The transfer function of The input quantity y to the controlled output vector y v The transfer function is In the formula, Represents the input y to the controlled output vector y v The transfer function, G 2 represents the second generalized object transfer matrix, E B represents the first controller transfer matrix, E F represents the second controller transfer matrix; The input quantity y to the virtual measured output quantity y c The transfer function is In the formula, Represents the input quantity y to the virtual measurement output quantity y c The transfer function, G 1 Represents the first generalized object transfer matrix.

3. The generalized estimator according to claim 1, It is characterized in that The nominal model of the controlled object in the nominal model module is one or more; When there are multiple nominal models of the controlled object, the multiple nominal models are connected in cascade.

4. A generalized disturbance rejection controller, It is characterized in that The generalized disturbance rejection controller comprises: an outer loop controller and the generalized estimator according to any one of claims 1 to 3; The input end of the outer loop controller is connected to the output end of the nominal model module in the generalized estimator, and the output end of the outer loop controller is connected to the input end of the controlled object; the outer loop controller is used to obtain the control amount according to the reference signal and the controlled output vector, and transmit the control amount to the controlled object; The output end of the controlled object is connected to the input end of the generalized estimator; the controlled object is used to generate an output quantity under the control of the control quantity and the interference of the disturbance quantity, and the output quantity is input into the generalized estimator as the input quantity of the generalized estimator; The generalized estimator is used to obtain a controlled output vector according to an input quantity, and transmit the controlled output vector to an outer loop controller.

5. The generalized disturbance rejection controller according to claim 4, It is characterized in that The outer loop controller comprises: a controller C, a first difference module, a second difference module and a generalized filter; The input end of the controller C is connected to the output end of the nominal model module in the generalized estimator, and the output end of the controller C is connected to the first input end of the first difference module; the controller C is used to input a reference signal and output a tracking control component according to the reference signal and the virtual measurement output; The second input end of the first difference module is connected to the output end of the generalized filter, and the output end of the first difference module is respectively connected to the first input end of the second difference module and the input end of the controlled object; the first difference module is used to subtract the disturbance estimation output by the generalized filter from the tracking control component to obtain the control amount, and transmit the control amount to the second difference module and the controlled object respectively; The second input end of the second difference module is connected to the output end of the inner loop estimation controller in the generalized estimator, and the output end of the second difference module is connected to the input end of the generalized filter; the second difference module is used to transmit the control signal minus the control amount to the generalized filter.

6. The generalized disturbance rejection controller according to claim 5, It is characterized in that The control formula in the outer loop controller is: In the formula, u represents the control quantity, u 0 represents the tracking control component, represents the disturbance estimator, C F represents the feedforward controller, C B represents the feedback controller, Q represents the quality factor of the generalized filter, K Q represents the filter coefficient, s represents the differential operator, r represents the reference signal, y v Represents the virtual measurement output, y v :=(y v1 ,y v2 ,y v3 , ..., y v(m-1) ,y vm ) T ,y vi Represents the output of the m-(i-1)th nominal model, i=1,2,3,…,m-1,m; symbol: = is a definition symbol.

7. The generalized disturbance rejection controller according to claim 6, It is characterized in that The transfer function of the generalized anti-disturbance controller includes: a transfer function from a reference signal r to an input quantity y, and a transfer function from a disturbance quantity d to an input quantity y; The transfer function from the reference signal r to the input quantity y is: In the formula, represents the transfer function from the reference signal r to the input quantity y, G p Represents the input-output relationship of the real controlled object, C BQ represents the feedback controller and the generalized filter vector, C BQ :=(C B Q), Represents the input y to the controlled output vector y v The transfer function of The transfer function from the disturbance d to the input y is: In the formula, Represents the transfer function from the disturbance d to the input y.

8. A generalized disturbance rejection controller design method, It is characterized in that The design method comprises: Determine the nominal model of the controlled object; According to the nominal model of the controlled object, the parameters of the inner loop estimation controller in the generalized estimator are adjusted; Connect the input end of the tuned generalized estimator to the output interface of the controlled object, and connect the output end of the tuned generalized estimator to the input interface of the outer loop controller to complete the structural configuration of the generalized disturbance rejection controller; Adjust the parameters of the outer loop controller to obtain a tuned generalized disturbance rejection controller; Determining the nominal model of the controlled object specifically includes: When the controlled object is a minimum phase electromechanical system with a relative order of 1 When , the nominal model is In the formula, G p1 Represents the input-output relationship of the electromechanical system with the minimum phase, G n1 Represents G p1 The corresponding nominal model transfer function, s represents the differential operator; When the controlled object is a non-minimum phase electromechanical system with a relative order of 1 When , the nominal model is In the formula, G p2 Represents the input-output relationship of the electromechanical system with non-minimum phase, G n2 Represents G p2 The corresponding nominal model transfer function; When the controlled object is a minimum phase electromechanical system with a relative order of 2 When , the nominal model is = In the formula, G p3 Represents the input-output relationship of the electromechanical system with the minimum phase of relative order 2, G n3 Represents G p3 The corresponding nominal model transfer function; When the controlled object is a cascade electromechanical system with a relative order of 2 and a minimum phase When , the nominal model is and In the formula, G p Represents the input-output relationship of the cascade electromechanical system with the minimum phase of relative order 2, G n ' 1 represents the first-level nominal model transfer function of the cascade electromechanical system, G n ' 2 represents the second-stage nominal model transfer function of the cascaded electromechanical system.

Images