Voice coil motor fast reflector generalized internal model control method, device and medium

Abstract

The embodiment of the invention discloses a voice coil motor fast reflector generalized internal model control method and device and a medium, belongs to the technical field of voice coil motor fast reflectors, and solves the problem that the performance of a controller is reduced due to perturbation of system parameters. Comprising the following steps: analyzing input and output data of the voice coil motor fast reflector system to obtain a corresponding nominal model; and performing loop shaping on the nominal model through a pre-compensator, and performing loop shaping controller calculation based on a standard H-infinity control algorithm to obtain a performance controller. The additive uncertainty of the nominal model and the model under the real working condition is determined, and an upper bound function corresponding to the additive uncertainty is obtained through curve fitting. And performing iterative processing on the preset controller to obtain a robust controller. And based on the performance controller and the robust controller, solving to obtain an inner ring controller so as to realize generalized internal model control of the voice coil motor fast reflector.

Description

Technical Field

[0001] The present application relates to the technical field of voice coil motor fast reflector, and in particular to a generalized internal model control method, device and medium for a voice coil motor fast reflector. Background Art

[0002] The voice coil motor fast reflector is driven by four voice coil motors, and a flexible hinge is used as the internal support structure. Due to its large resonance characteristics and coupling between axes, it will be difficult for general control algorithms to meet the high-speed and high-precision control requirements of the voice coil motor fast reflector. In addition, it generally works on airborne or shipborne platforms. The atmospheric turbulence and platform vibration in the working environment will cause certain parameter perturbations of the voice coil motor fast reflector. Under such perturbations, if the controller does not have strong robust stability, the tracking performance of the system will be reduced or even the system will be unstable.

[0003] In the current VCM control technology, most of them use the traditional control method of PID control, which is based on the transfer function model of the VCM. It has a fixed controller structure and adjusts the controller parameters by drawing the root locus. It has the advantages of few parameters and easy to implement control structure. It is currently the most widely used control method in the industry.

[0004] However, the traditional PID control method is difficult to effectively suppress the influence of the resonant characteristics in the voice coil motor fast reflector, and the PID control method is designed based on a single-input single-output system. The voice coil motor fast reflector is a multi-input multi-output system. The PID control method is difficult to effectively solve the adverse effects of coupling on the system.

[0005] In addition, in traditional control structures, performance and robustness are contradictory indicators. If a controller has good performance, its robustness will not be very good. On the contrary, if a controller has good robustness, its performance will definitely not be good. Therefore, the industry often balances performance and robustness by adjusting the parameters of the controller so that both performance and robustness reach a moderate level. However, this parameter compromise method is difficult to meet the needs of high-speed and high-precision control in parameter perturbation systems, so it is still difficult to solve the problem of controller performance degradation caused by system parameter perturbation. Summary of the invention

[0006] The embodiments of the present application provide a generalized internal model control method, device and medium for a voice coil motor fast mirror, which are used to solve the following technical problem: the problem of controller performance degradation caused by system parameter perturbation in the prior art.

[0007] The present application embodiment adopts the following technical solutions:

[0008] The embodiment of the present application provides a generalized internal model control method for a voice coil motor fast reflector, including: analyzing and processing the input and output data of the voice coil motor fast reflector system acquired by using a Matlab identification toolbox to obtain a nominal model corresponding to the voice coil motor fast reflector. The nominal model is loop-shaped by a pre-distortion device, and based on the standard H ∞ The control algorithm solves the loop shaping controller to obtain the performance controller. The additive uncertainty of the nominal model and the model under the actual working condition is determined, and the upper bound function corresponding to the additive uncertainty is obtained through curve fitting. Based on the upper bound function, the preset controller is iterated through the DK iterative algorithm to obtain a robust controller. Based on the performance controller and the robust controller, the inner loop controller is solved to realize the generalized internal model control of the voice coil motor fast reflector.

[0009] The present application embodiment adopts H ∞ The loop forming method is used to design the performance controller of the voice coil motor fast reflector, which effectively suppresses the influence of the flexible mode in the system on the system tracking accuracy, and the design method based on multi-input and multi-output effectively suppresses the inter-axis coupling of the voice coil motor fast reflector. Secondly, the embodiment of the present application designs a μ-integrated robust controller, and implements it using the architecture of generalized internal model control, which achieves a balance between system performance and robust stability, and can meet the performance requirements of the voice coil motor fast reflector under complex working conditions such as vibration. The embodiment of the present application can improve the tracking accuracy and dynamic response speed of the voice coil motor fast reflector, and at the same time can suppress the influence of parameter perturbations on the system tracking accuracy and response speed. The stable and high-performance operation of the voice coil motor fast reflector is guaranteed.

[0010] In one implementation of the present application, before the nominal model is loop-shaped by a precompensator, the method also includes: performing a bilinear transformation on the nominal model; presetting a target open-loop singular value based on the bilinear change, performing internal and external factorization on the nominal model obtained after the transformation, and obtaining a precompensator, so as to compensate the nominal model by the precompensator so that the singular value of the controlled object is shaped into a desired open-loop shape.

[0011] In one implementation of the present application, the low-frequency band gain corresponding to the preset target open-loop singular value is greater than the preset gain threshold; and, the high-frequency band gain attenuation speed corresponding to the preset target open-loop singular value is greater than the preset speed threshold; and, the slope at the cutoff frequency corresponding to the preset target open-loop singular value is -20dB.

[0012] In one implementation of the present application, the nominal model is loop shaped by a predistorter, and based on the standard H ∞The control algorithm solves the loop shaping controller to obtain a performance controller, which specifically includes: changing the controlled object model into a state space form to perform a normalized left coprime decomposition on the controlled object model; in the process of solving the loop shaping controller, obtaining an optimal controller based on a preset robust performance index function; combining the optimal controller with a precompensator to obtain a performance controller.

[0013] In one implementation of the present application, based on an upper bound function, a preset controller is iteratively processed by a DK iterative algorithm to obtain a robust controller, which specifically includes: determining parameter perturbations and external disturbances corresponding to the controlled object system to construct an uncertain controlled object model based on the parameter perturbations and external disturbances; transforming the uncertain controlled object system through a linear fractional transformation; based on the upper bound function and the uncertain controlled object representation obtained after the transformation, the preset controller is iteratively processed by a DK iterative algorithm to obtain a robust controller; wherein the uncertain controlled object representation obtained after the transformation includes an uncertain controlled object model and an error weighting function.

[0014] In one implementation of the present application, an inner-loop controller is obtained based on a performance controller and a robust controller, specifically including: performing a standard left coprime decomposition of the controlled object and the performance controller based on Youla parameterization to obtain an inner-loop controller expression; solving the inner-loop controller expression to obtain the inner-loop controller.

[0015] In one implementation of the present application, a standard left coprime decomposition is performed on the controlled object and the performance controller based on Youla parameterization to obtain an inner loop controller expression, specifically including: based on Youla parameterization and function:

[0016]

[0017] Get the derived function:

[0018]

[0019] Among them, K is the controller; and Q are both stable transfer functions with minimum phase; For the nominal model The left coprime decomposition of is the left coprime decomposition of the performance controller K0; the derived function is analyzed and processed to obtain the expression of the inner loop controller.

[0020] In one implementation of the present application, the derived function is analyzed and processed to obtain an inner loop controller expression, which specifically includes: shifting the derived function and extracting common factors to obtain:

[0021]

[0022] The inner loop controller expression is obtained as:

[0023]

[0024] Among them, K is the controller; and Q are both stable transfer functions with minimum phase; For the nominal model The left coprime decomposition of is the left coprime decomposition of the performance controller K0.

[0025] The embodiment of the present application provides a generalized internal model control device for a voice coil motor fast reflector, comprising: at least one processor; and a memory connected to the at least one processor in communication; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor, so that the at least one processor can: analyze and process the input and output data of the voice coil motor fast reflector system acquired by the Matlab identification toolbox to obtain a nominal model corresponding to the voice coil motor fast reflector; perform loop shaping on the nominal model by a precompensator, and perform loop shaping on the nominal model based on the standard H ∞ The control algorithm solves the loop shaping controller to obtain the performance controller; determines the additive uncertainty of the nominal model and the model under the actual working condition, and obtains the upper bound function corresponding to the additive uncertainty through curve fitting; based on the upper bound function, the preset controller is iteratively processed through the DK iterative algorithm to obtain the robust controller; based on the performance controller and the robust controller, the inner loop controller is solved to realize the generalized internal model control of the fast reflector of the voice coil motor.

[0026] The embodiment of the present application provides a non-volatile computer storage medium storing computer executable instructions, wherein the computer executable instructions are configured to: analyze and process the input and output data of the voice coil motor fast reflector system acquired by using the Matlab identification toolbox to obtain a nominal model corresponding to the voice coil motor fast reflector; perform loop shaping on the nominal model by using a pre-compensator, and perform loop shaping on the nominal model based on the standard H ∞ The control algorithm solves the loop shaping controller to obtain the performance controller; determines the additive uncertainty of the nominal model and the model under the actual working condition, and obtains the upper bound function corresponding to the additive uncertainty through curve fitting; based on the upper bound function, the preset controller is iteratively processed through the DK iterative algorithm to obtain the robust controller; based on the performance controller and the robust controller, the inner loop controller is solved to realize the generalized internal model control of the fast reflector of the voice coil motor.

[0027] At least one of the above technical solutions adopted in the embodiment of the present application can achieve the following beneficial effects: ∞ The loop forming method is used to design the performance controller of the voice coil motor fast reflector, which effectively suppresses the influence of the flexible mode in the system on the system tracking accuracy, and the design method based on multi-input and multi-output effectively suppresses the inter-axis coupling of the voice coil motor fast reflector. Secondly, the embodiment of the present application designs a μ-integrated robust controller, and implements it using the architecture of generalized internal model control, which achieves a balance between system performance and robust stability, and can meet the performance requirements of the voice coil motor fast reflector under complex working conditions such as vibration. The embodiment of the present application can improve the tracking accuracy and dynamic response speed of the voice coil motor fast reflector, and at the same time can suppress the influence of parameter perturbations on the system tracking accuracy and response speed. The stable and high-performance operation of the voice coil motor fast reflector is guaranteed. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative labor. In the drawings:

[0029] Figure 1 A standard control structure diagram provided for an embodiment of the present application;

[0030] Figure 2 A schematic diagram of the parameterization of a controller Youla provided in an embodiment of the present application;

[0031] Figure 3 A generalized internal model control structure diagram provided in an embodiment of the present application;

[0032] Figure 4 A flow chart of a generalized internal model control method for a voice coil motor fast reflection mirror provided in an embodiment of the present application;

[0033] Figure 5 A schematic diagram of the structure of a performance controller K0 provided in an embodiment of the present application;

[0034] Figure 6 A structural diagram of an uncertainty feedback control system provided in an embodiment of the present application;

[0035] Figure 7 A schematic diagram of a linear fraction transformation provided in an embodiment of the present application;

[0036] Figure 8A schematic structural diagram of a generalized internal model control device for a voice coil motor fast reflection mirror provided in an embodiment of the present application.

[0037] Reference numerals:

[0038] 200 : a generalized internal model control device for a voice coil motor fast reflection mirror, 201 : a processor, and 202 : a memory. DETAILED DESCRIPTION

[0039] The embodiments of the present application provide a generalized internal model control method, device and medium for a voice coil motor fast reflection mirror.

[0040] In order to enable those skilled in the art to better understand the technical solutions in this application, the technical solutions in the embodiments of this application will be clearly and completely described below in conjunction with the drawings in the embodiments of this application. Obviously, the described embodiments are only part of the embodiments of this application, not all of them. Based on the embodiments of this specification, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of this application.

[0041] The technical solution proposed in the embodiments of the present application is described in detail below with reference to the accompanying drawings.

[0042] Under the standard control structure, system performance and robustness exist in a trade-off form, and it is impossible to ensure strong robustness of the system under the premise of good system performance. Therefore, under the constraints of the standard control structure, it is almost impossible to design a controller that can make the system have both high performance and strong robustness. Multi-objective complementary control breaks this limitation. It introduces an inner loop under the standard control structure to achieve the coexistence of high performance and strong robustness.

[0043] Figure 1 A standard control structure diagram provided in the embodiment of the present application, such as Figure 1 As shown, r is the reference signal, e is the deviation signal, d is the interference signal, y is the output signal, K is the controller, and G is the controlled object. Assume that the controller K0 and the nominal model There exists a standard left coprime factorization:

[0044]

[0045] Tracking the reference input r, rejecting disturbances d and perturbations d in a standard feedback structure Δ All are completed by a controller. However, in actual control, this problem is often encountered. The controller designed has good performance when there is no parameter perturbation in the system. However, due to factors such as noise and vibration in the working environment, the system has parameter perturbations. becomes G. At this time, according to the nominal model The performance of the designed controller will be greatly reduced. This problem of controller performance degradation caused by the perturbation of the system's internal parameters is called the robust performance problem. When there is a large parameter perturbation in the system, how to design a controller to ensure the stability of the system is called the robust stability problem. The two can be collectively referred to as the robustness problem. At present, these two problems are widely existing in the industry. In the standard feedback structure, the controller performance problem and the robustness problem are contradictory problems.

[0046] In industry, engineers often need to adjust controller parameters to pursue a balance between controller performance and robustness. This "compromise" idea is currently the mainstream idea in the industry. However, improving robustness through this "compromise" idea comes at the expense of controller performance. This way of solving the problem is not optimal.

[0047] In order to solve the contradiction between performance and robustness, the embodiment of the present application proposes a generalized internal model control strategy, under which performance and robustness can be complemented rather than compromised. The generalized internal model control framework can be derived from the classical feedback control structure with the help of Youla parameterization. Figure 1 The Youla parameterization of the controller K is shown in formula (2):

[0048]

[0049] Figure 2 A schematic diagram of the parameterization of a controller Youla provided in an embodiment of the present application, Figure 1 The controller Youla in the parameterization is as follows Figure 2 As shown, K is given by and Q, among which and Q are both minimum phase stable transfer functions. For the nominal model The left coprime decomposition of is the left coprime decomposition of the controller K0, as shown in formula (3). ∞ , must meet

[0050]

[0051] Figure 3 A generalized internal model control structure diagram provided in the embodiment of the present application, by changing The position of the input signal can be obtained Figure 3 The multi-objective complementary control structure shown in Figure 2The structures shown are equivalent because the transfer function from the output signal y to the controlled variable u does not change.

[0052] exist Figure 3 When the output of Q is q=0, the structure is the same as Figure 1 The classical feedback control structure shown is completely equivalent, so the controller design method in the classical feedback control structure can be used to design the controller K0. Usually, the controller K0 is designed as a high-performance controller, and its main task is to ensure tracking accuracy and tracking bandwidth. Figure 3 K in the equation is a robust controller designed considering the model parameter perturbation Δ. Its function is to ensure the robust stability and robust performance of the system when there is a large parameter perturbation. When there is no parameter perturbation and disturbance in the system (d Δ =0, d = 0), the input f of the inner loop Q = 0, and the robust controller does not work. When there is parameter perturbation or interference in the system, the robust controller intervenes in the system to complete the control of d and d □ compensation control.

[0053]

[0054] Formula (4) is the relationship between the reference signal r and the perturbation d under the generalized internal model control structure. □ And the influence of disturbance d on the control variable u. By comparing formula (1), it can be seen that the generalized internal model control structure has achieved the separation of the two tasks of tracking and suppressing the influence of parameter perturbations on the system to a certain extent. The tracking task is completed by the performance controller K0, and the anti-disturbance and suppression of the influence of parameter perturbations are achieved through the design of the inner loop Q.

[0055] The generalized internal model control design in the embodiment of the present application can be divided into the following three steps:

[0056] (1) A performance controller K0 is designed based on the nominal model. The controller K0 can meet the performance requirements of the system and has a high closed-loop tracking bandwidth and strong anti-interference ability.

[0057] (2) A robust controller K is designed based on the model that takes uncertainty into consideration. The controller K has strong robustness and can effectively cope with the impact of uncertainties such as perturbations of system internal parameters.

[0058] (3) Based on the designed performance controller K0 and robust controller K, solve the inner loop controller Q.

[0059] Figure 4 A flow chart of a generalized internal model control method for a voice coil motor fast reflection mirror provided in an embodiment of the present application is as follows: Figure 4 As shown, the generalized internal model control method of the voice coil motor fast reflection mirror includes the following steps:

[0060] Step 101: Analyze and process the input and output data of the voice coil motor fast reflector system obtained by using the Matlab identification toolbox to obtain a nominal model corresponding to the voice coil motor fast reflector.

[0061] In one embodiment of the present application, the present application embodiment first performs system identification on the voice coil motor fast reflector, and uses the Matlab identification toolbox to obtain a high-precision nominal model of the voice coil motor fast reflector according to the input and output data of the system.

[0062] Specifically, first, it is necessary to collect the input and output data of the voice coil motor fast reflector system. These data can be obtained through experimental measurement to ensure the accuracy and completeness of the data. Using mathematical tools such as Matlab and the system identification toolbox, a mathematical model of the voice coil motor fast reflector is constructed based on the collected data. This step usually includes selecting a suitable model structure, such as a transfer function, a state space model, etc., and using optimization algorithms such as the least squares method and genetic algorithm to fit the data to obtain a high-precision nominal model.

[0063] Step 102: loop shaping the nominal model through the pre-distortion device, and based on the standard H ∞ The control algorithm solves the loop shaping controller to obtain a performance controller.

[0064] In one embodiment of the present application, a bilinear transformation is performed on the nominal model. Based on the bilinear change, a target open-loop singular value is preset, and the nominal model obtained after the transformation is internally and externally factored to obtain a precompensator, so that the nominal model is compensated by the precompensator, so that the singular value of the controlled object is shaped into a desired open-loop shape.

[0065] Specifically, in the embodiment of the present application, the performance controller adopts H based on the nominal model. ∞ Loop shaping design method, H ∞ The loop shaping control algorithm can design the cutoff frequency of the system by designing the shape of the open loop and ensure the stability of the system in the frequency band where the amplitude decays rapidly.

[0066] The compensator is used to compensate the nominal model, and a precompensator W1 is designed to make the controlled object G s The singular values ​​of are shaped into the desired open-loop shape, where

[0067] Figure 5 A schematic diagram of the structure of a performance controller K0 provided in an embodiment of the present application is shown in FIG. Figure 5 As shown, in the design of controller K0, the standard H ∞ The control algorithm solves it and When designing a circuit, consider designing it to have a large low-frequency gain and a small high-frequency gain.

[0068] Furthermore, in the design of the predistorter, firstly, a bilinear transformation as shown in formula (5) is used to convert becomes R, where The frequency characteristic curve of R is approximately the same between the frequencies ω1 and ω2.

[0069]

[0070] in, is the original untransformed operator; S is the new operator after transformation; ω1 is the approximate starting frequency; ω2 is the approximate cutoff frequency.

[0071] And adopt bilinear change to assume that the ideal open-loop singular value is G d , factorize R internally and externally.

[0072] R=N1M1 -1 , (6)

[0073] Among them, N1 is the full-pass transfer matrix, M1 -1 is the external factor of R. T Do inner and outer factorization:

[0074] M1 T =N2M2 -1 , (7)

[0075] Among them, N2 is the full-pass transfer matrix, M2 -1 For M1 T The external factor of the predistortion device W1 can be obtained by formula (8).

[0076] W1=IM2 -T G d , (8)

[0077] Here, I is the identity matrix whose main diagonal is all 1.

[0078] In one embodiment of the present application, the low-frequency gain corresponding to the preset target open-loop singular value is greater than the preset gain threshold. Also, the high-frequency gain attenuation speed corresponding to the preset target open-loop singular value is greater than the preset speed threshold. Also, the slope at the cutoff frequency corresponding to the preset target open-loop singular value is -20 dB.

[0079] In one embodiment of the present application, in solving the performance controller K0, in order to meet the requirements of high steady-state accuracy and high dynamic characteristics, the ideal open-loop shape G dIt is required to have a high gain in the low frequency band, a rapid gain attenuation in the high frequency band, and a slope of -20dB at the cutoff frequency. d The selection form is shown in formula (9):

[0080]

[0081] Among them, k is the proportional coefficient. The selection of k needs to be combined with the actual controlled object. Since the control voltage input of VCA-FSM is constrained, the value of k needs to be reasonably selected based on experiments.

[0082] In one embodiment of the present application, the controlled object model is changed into a state space form to perform a normalized left coprime decomposition on the controlled object model. In the process of solving the loop shaping controller, an optimal controller is obtained based on a preset robust performance index function. The optimal controller is combined with the precompensator to obtain the performance controller.

[0083] Specifically, the controlled object model G of the system s Written in state space form, let the matrix L make A s +LC s If stable, G s Perform a normalized left coprime factorization:

[0084]

[0085] Among them, (M s N s )satisfy:

[0086] M s (jω)M s * (jω)+N s (jω)N s * (jω)=I, (11)

[0087] And L = -YCs*, we can get:

[0088]

[0089] Where Y ≥ 0 is the stable solution of the following Riccati equation:

[0090] A s Y+YA s * -YC s * C s Y+B s B s *=0. (13)

[0091] In the controller solution process, by solving γ min Can make K ∞ is the optimal controller, but the optimal controller does not necessarily exist, so most people choose to solve the suboptimal controller:

[0092]

[0093] Among them, λ max (YQ) is the largest eigenvalue of YQ, and Q is the solution of the following Lyapunov equation:

[0094] Q(A s -YC s * C s )+(A s -YC s * C s ) * Q+C s * C s =0. (15)

[0095] Select γ>γ min , the solution can be obtained as the suboptimal controller K ∞ :

[0096]

[0097] in,

[0098]

[0099] The controller K ∞ And the designed compensator W1 is combined to obtain the final loop shaping performance controller K0 as shown in formula (18).

[0100] K0=W1K ∞ . (18)

[0101] Step 103: Determine the additive uncertainty of the nominal model and the model under the actual working condition, and obtain the upper bound function corresponding to the additive uncertainty through curve fitting.

[0102] In one embodiment of the present application, under given input conditions, a nominal model is used for simulation or calculation to obtain a series of output values, and under the same input conditions, the output values ​​of the real system are measured or recorded. For each input point, the difference between the nominal output and the real output is calculated, that is, the additive uncertainty is obtained. According to the characteristics of the data and the requirements of the problem, a suitable fitting function is selected, and the parameters of the fitting function are determined using the least squares method, nonlinear least squares method or other optimization methods, and the goodness of fit indicators (such as mean square error MSE, root mean square error RMSE, etc.) are used to evaluate the fitting effect.

[0103] Furthermore, the function obtained by curve fitting describes the average value or trend of the additive uncertainty. However, in order to design a robust controller, it is usually necessary to know the upper bound of the additive uncertainty, that is, a value larger than the actual uncertainty, to ensure that the controller remains stable even in the worst case. For example, if the fitting function itself has an upper bound, the maximum value or upper bound can be directly used as the upper bound function; in order to estimate the uncertainty more conservatively, a safety margin can be added to the fitting function f(x); if the additive uncertainty changes greatly with time or input conditions, you can consider using a dynamic upper bound function, that is, a function that changes with time or input conditions.

[0104] Step 104: Based on the upper bound function, the preset controller is iteratively processed by the DK iterative algorithm to obtain a robust controller.

[0105] In one embodiment of the present application, the parameter perturbation and external disturbance corresponding to the controlled object system are determined to construct an uncertain controlled object model based on the parameter perturbation and the external disturbance. The uncertain controlled object system is transformed by a linear fractional transformation. Based on the upper bound function and the uncertain controlled object representation obtained after the transformation, the preset controller is iteratively processed by the DK iterative algorithm to obtain a robust controller. The uncertain controlled object representation obtained after the transformation includes an uncertain controlled object model and an error weighting function.

[0106] Specifically, this embodiment considers the upper bound of parameter perturbation uncertainty and adopts the μ synthesis method to design the robust controller K. μ synthesis can keep the system stable under parameter perturbation and external disturbance, and the control performance will not be greatly reduced, which can effectively reduce the adverse effects of parameter perturbation on the controller.

[0107] Specifically:

[0108]

[0109] Figure 6 A structural diagram of an uncertainty feedback control system provided in an embodiment of the present application, Figure 6The feedback control system shown can be expressed in the form of (Linear Fraction Transformation, LFT), where W p is the weighting function of the error, and M is the controlled object model considering uncertainty.

[0110] Figure 7 A schematic diagram of a linear fraction transformation provided in an embodiment of the present application is shown in FIG. Figure 7 As shown, in LFT form:

[0111]

[0112] like Figure 6 For the LFT shown in Figure 2, we need to find a stable controller K that minimizes the μ norm in equation (20):

[0113]

[0114] in,

[0115] F l (P,K)=P 11 +P 12 K(IP 22 K) -1 P 21 (twenty one)

[0116] The μ-synthesized robust controller solution problem is solved using DK iteration and can be solved using the dksyn function in Matlab.

[0117] Step 105: Based on the performance controller and the robust controller, an inner loop controller is obtained to realize generalized internal model control of the fast mirror of the voice coil motor.

[0118] In one embodiment of the present application, a standard left coprime decomposition is performed on the controlled object and the performance controller based on Youla parameterization to obtain an inner loop controller expression. The inner loop controller expression is solved to obtain the inner loop controller solution.

[0119] Specifically, in Figure 3 In the multi-objective complementary control structure shown in Figure 1, the inner loop controller Q is mainly used to improve the robustness of the system. According to the design principle of multi-objective complementary control, it is based on the identified nominal model. The performance controller K0 and the robust controller K are solved.

[0120] Furthermore, according to the Youla parameterization principle of the controller and equations (2) and (3), we can continue to deduce:

[0121]

[0122] Get the derived function:

[0123]

[0124] Among them, K is the controller; and Q are both stable transfer functions with minimum phase; For the nominal model The left coprime decomposition of is the left coprime decomposition of the performance controller K0;

[0125] The derived function is analyzed and processed to obtain the inner loop controller expression.

[0126] Furthermore, the derived function is transposed and the common factors are extracted to obtain:

[0127]

[0128] The inner loop controller expression is obtained as:

[0129]

[0130] Among them, K is the controller; and Q are both stable transfer functions with minimum phase; For the nominal model The left coprime decomposition of is the left coprime decomposition of the performance controller K0.

[0131] The above method can realize the design of generalized internal model control strategy for fast response of voice coil motor. The controller architecture finally solved is as follows: Figure 3 shown.

[0132] Figure 8 This is a schematic diagram of the structure of a generalized internal model control device for a voice coil motor fast reflection mirror provided in an embodiment of the present application. Figure 8 As shown, a generalized internal model control device 200 for a voice coil motor rapid reflector includes: at least one processor 201; and a memory 202 in communication with the at least one processor 201; wherein the memory 202 stores instructions executable by the at least one processor 201, and the instructions are executed by the at least one processor 201, so that the at least one processor 201 can: analyze and process the input and output data of the voice coil motor rapid reflector system acquired through the Matlab identification toolbox to obtain a nominal model corresponding to the voice coil motor rapid reflector; perform loop shaping on the nominal model through a pre-compensator, and based on the standard H ∞The control algorithm solves the loop shaping controller to obtain the performance controller; determines the additive uncertainty of the nominal model and the model under the actual working condition, and obtains the upper bound function corresponding to the additive uncertainty through curve fitting; based on the upper bound function, the preset controller is iteratively processed through the DK iterative algorithm to obtain the robust controller; based on the performance controller and the robust controller, the inner loop controller is solved to realize the generalized internal model control of the fast reflector of the voice coil motor.

[0133] The embodiment of the present application provides a non-volatile computer storage medium storing computer executable instructions, wherein the computer executable instructions are configured to: analyze and process the input and output data of the voice coil motor fast reflector system acquired by using the Matlab identification toolbox to obtain a nominal model corresponding to the voice coil motor fast reflector; perform loop shaping on the nominal model by using a pre-compensator, and perform loop shaping on the nominal model based on the standard H ∞ The control algorithm solves the loop shaping controller to obtain the performance controller; determines the additive uncertainty of the nominal model and the model under the actual working condition, and obtains the upper bound function corresponding to the additive uncertainty through curve fitting; based on the upper bound function, the preset controller is iteratively processed through the DK iterative algorithm to obtain the robust controller; based on the performance controller and the robust controller, the inner loop controller is solved to realize the generalized internal model control of the fast reflector of the voice coil motor.

[0134] Each embodiment in this application is described in a progressive manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the device, equipment, and non-volatile computer storage medium embodiments, since they are basically similar to the method embodiments, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

[0135] The above is only an embodiment of the present application and is not intended to limit the present application. For those skilled in the art, the embodiments of the present application may have various changes and variations. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the various embodiments of the present application.

Claims

1. A generalized internal model control method for a voice coil motor fast mirror, characterized in that: The method comprises: The input and output data of the voice coil motor fast reflector system are analyzed and processed by Matlab identification toolbox to obtain the nominal model corresponding to the voice coil motor fast reflector. The nominal model is loop-shaped by the predistorter and based on the standard H ∞ The control algorithm solves the loop shaping controller to obtain the performance controller; Determine the additive uncertainty of the nominal model and the model under the actual working condition, and obtain the upper bound function corresponding to the additive uncertainty by curve fitting; Based on the upper bound function, the preset controller is iteratively processed by a DK iterative algorithm to obtain a robust controller; Based on the performance controller and the robust controller, an inner loop controller is solved to realize generalized internal model control of the voice coil motor fast mirror.

2. A generalized internal model control method for a voice coil motor fast mirror according to claim 1, characterized in that: Before performing loop shaping on the nominal model by the predistorter, the method further includes: performing a bilinear transformation on the nominal model; Based on the bilinear variation preset target open-loop singular value, the nominal model obtained after the transformation is internally and externally factored to obtain the precompensator, so as to compensate the nominal model through the precompensator so that the singular value of the controlled object is shaped into a desired open-loop shape.

3. A generalized internal model control method for a voice coil motor fast mirror according to claim 2, characterized in that: The low-frequency band gain corresponding to the preset target open-loop singular value is greater than a preset gain threshold; And, the high-frequency band gain attenuation speed corresponding to the preset target open-loop singular value is greater than a preset speed threshold; And, the slope at the cutoff frequency corresponding to the preset target open-loop singular value is -20 dB.

4. A generalized internal model control method for a voice coil motor fast mirror according to claim 1, characterized in that: The nominal model is loop-shaped by the predistorter and based on the standard H ∞ The control algorithm solves the loop shaping controller to obtain the performance controller, which includes: Changing the controlled plant model into a state space form to perform a normalized left coprime decomposition on the controlled plant model; In the process of solving the loop shaping controller, the optimal controller is obtained based on the preset robust performance index function; The optimal controller is combined with the predistorter to obtain the performance controller.

5. A generalized internal model control method for a voice coil motor fast mirror according to claim 1, characterized in that: The method of iteratively processing the preset controller by using the DK iterative algorithm based on the upper bound function to obtain a robust controller specifically includes: Determining parameter perturbations and external disturbances corresponding to the controlled object system, so as to construct an uncertain controlled object model based on the parameter perturbations and the external disturbances; Transforming the uncertain controlled object system in the form of a linear fractional transformation; Based on the upper bound function and the uncertainty controlled object representation obtained after transformation, the preset controller is iteratively processed by the DK iterative algorithm to obtain the robust controller; The representation form of the uncertain controlled object obtained after the transformation includes an uncertain controlled object model and an error weighting function.

6. A generalized internal model control method for a voice coil motor fast mirror according to claim 1, characterized in that: The inner loop controller is obtained based on the performance controller and the robust controller, specifically comprising: Based on Youla parameterization, the controlled object and the performance controller are subjected to standard left coprime decomposition to obtain the inner loop controller expression. The inner loop controller expression is solved to obtain the inner loop controller.

7. A generalized internal model control method for a voice coil motor fast mirror according to claim 6, characterized in that: The controlled object and the performance controller are subjected to standard left coprime decomposition based on Youla parameterization to obtain the inner loop controller expression, which specifically includes: Based on Youla parameterization and functions: Get the derived function: Among them, K is the controller; and Q are both stable transfer functions with minimum phase; For the nominal model The left coprime decomposition of is the left coprime decomposition of the performance controller K0; The derived function is analyzed and processed to obtain the inner loop controller expression.

8. A generalized internal model control method for a voice coil motor fast mirror according to claim 7, characterized in that: The analyzing and processing the derived function to obtain the inner loop controller expression specifically includes: The derived function is transposed and the common factors are extracted to obtain: The inner loop controller expression is obtained as follows: Among them, K is the controller; and Q are both stable transfer functions with minimum phase; For the nominal model The left coprime decomposition of is the left coprime decomposition of the performance controller K0.

9. A generalized internal model control device for a voice coil motor fast reflector, characterized in that: The device comprises a memory for storing computer program instructions and a processor for executing the program instructions, wherein when the computer program instructions are executed by the processor, the device is triggered to execute the method according to any one of claims 1 to 8.

10. A non-volatile computer storage medium storing computer executable instructions, characterized in that: The computer executable instructions can execute the method according to any one of claims 1 to 8.

Images