Linear motor driving system control method based on H-infinity robust baseline control and L1 adaptive compensation

Abstract

The invention relates to the technical field of intelligent compensation control, in particular to a linear motor driving system control method based on robust baseline control and adaptive compensation. The method comprises the following steps: establishing a linear motor state space model; a dynamic output feedback baseline controller is obtained by iteratively solving a Riccati equation set based on a disturbance suppression level index, a closed-loop matrix is extracted, and a baseline control signal is output; calculating a steady-state gain and filtering the reference input to generate a steady-state feed-forward control signal; estimating the total uncertainty on line through a state predictor, constructing a projection operator to decompose the total uncertainty into a matched component and an unmatched component, performing output equivalent mapping on the unmatched component, and performing band-limiting processing through a low-pass filter to synthesize a self-adaptive compensation signal; and after the three paths of signals are superposed, the linear motor is driven through amplitude saturation and change rate limitation. According to the method, the robustness and the online compensation capability are fused, and the tracking precision and the anti-interference robustness of the linear motor are remarkably improved.

Description

Technical Field

[0001] This invention relates to the field of intelligent compensation control technology, and in particular to a method based on Robust baseline control and An adaptive compensation control method for linear motor drive systems. Background Technology

[0002] Linear motor drive systems, due to their elimination of intermediate transmission links, have advantages such as fast response speed, high positioning accuracy, and simple mechanical structure, and are widely used in high-end equipment fields such as precision positioning platforms, semiconductor manufacturing equipment, high-speed CNC machine tools, and automated assembly lines.

[0003] However, linear motor drive systems still face numerous control challenges in practical engineering applications. First, linear motors employ direct drive, meaning load changes, parameter perturbations, and external disturbances directly affect the motor's mover, lacking the mechanical filtering effect of a transmission link, making the system more sensitive to uncertainties. Second, periodic disturbances unique to linear motors, such as thrust fluctuations, cogging effects, and end effects, as well as nonlinear factors like changes in guide rail friction and cable drag, all contribute to increased tracking errors and decreased control quality. Furthermore, the relatively low mechanical stiffness of linear motor systems makes them prone to structural resonance, requiring control algorithms with good band-limiting characteristics to avoid high-frequency excitation.

[0004] To address the aforementioned problems, existing technologies often employ PID control, Robust control or Adaptive control and other methods are employed. Among these, PID control, due to its simple structure and ease of implementation, has been widely used in industrial motion control. However, for controlled objects such as linear motors that are sensitive to disturbances and have large parameter variations, fixed-parameter PID controllers struggle to simultaneously meet the requirements of high dynamic response and high steady-state accuracy. When the system experiences disturbances such as thrust fluctuations or load changes, PID control often requires online parameter adjustments or combination with other compensation methods, increasing debugging complexity and making it difficult to maintain optimal control performance under all operating conditions. Robust control, through the design of weighting functions and the solution of the Riccati equation or linear matrix inequalities, can guarantee the stability and disturbance suppression capability of a closed-loop system under conditions of model uncertainty and external disturbances. However, Once the controller is designed, its structure and parameters are relatively fixed. When the system operating conditions change significantly (such as sudden load changes or changes in friction characteristics), it is difficult to make online adjustments to maintain optimal control performance. Adaptive control achieves rapid compensation for unknown dynamics and disturbances by constructing a state predictor to estimate system uncertainties online and using a low-pass filter to band-limit the compensation signal. However, its application alone... In adaptive control, the lack of robust constraints on the baseline closed loop may lead to high-frequency jitter in the control input, or even excite unmodeled dynamics of the system.

[0005] Therefore, there is an urgent need for a control method suitable for linear motor drive systems that can still take into account dynamic response, steady-state accuracy and robust stability, and achieve high-precision, stable and reliable tracking control even when there are adverse factors such as parameter changes, external disturbances and flexible modes. Summary of the Invention

[0006] To address the shortcomings of existing technologies, this invention proposes a method based on... Robust baseline control and An adaptive compensation control method for linear motor drive systems is proposed to achieve high-precision tracking control and robust disturbance suppression of linear motor drive systems.

[0007] To achieve the above objectives, the present invention adopts the following technical solution:

[0008] This invention proposes a method based on Robust baseline control and An adaptive compensation control method for a linear motor drive system includes the following steps:

[0009] S1. Establish a linear time-invariant state-space model of the linear motor drive system, which is expressed in the following form:

[0010]

[0011] In the formula, Let be the system state vector. To control the input, For system output signals, For the system matrix, For the input matrix, This is the output matrix;

[0012] S2. Obtain the reference input signal And calculate the reference input signal. With system output signal Tracking error between A linear time-invariant state-space model based on S1 and a preset disturbance suppression level index. Construct a generalized controlled object; and through Iteratively solving the corresponding Riccati equations yields... Dynamic output feedback baseline controller ; connect the generalized controlled object with the baseline controller Connect the points to form a closed-loop system, and extract the nominal closed-loop state matrix of the object side from this closed-loop system as the closed-loop matrix. The tracking error Input to the baseline controller Output baseline control signal ;

[0013] S3, through the closed-loop matrix Input matrix and output matrix Calculate steady-state gain ; reference input signal After passing through the first low-pass filter After processing, the pre-filtered reference signal is obtained. Then utilize the aforementioned steady-state gain For the pre-filtered reference signal The signal is processed to generate a steady-state feedforward control signal. ;

[0014] S4, Based on the closed-loop matrix The total uncertainty of the system is estimated online using a state predictor. Based on matching input channel matrix Constructing a linear projection operator The total uncertainty Decomposed into matching components With mismatched components For the mismatched components Perform an equivalent mapping on the output to obtain a mapping mismatch signal. ; through the second low-pass filter For the matched components Mismatch signal with mapping Band-limiting processing is performed separately to obtain the corresponding matching compensation signal. With mismatch compensation signal The two are then combined into an adaptive compensation signal. ;in The mapping formula is:

[0015]

[0016]

[0017]

[0018] In the formula, To match the channel steady-state gain matrix The inverse matrix, Output the influence matrix for the mismatched components;

[0019] S5, Calculate the total control signal For the total control signal The amplitude saturation limit and rate of change limit are applied sequentially to generate the system input signal. The total control signal The calculation formula is:

[0020] .

[0021] Furthermore, in S2, Dynamic output feedback baseline controller The state space implementation is as follows:

[0022]

[0023]

[0024] In the formula, For the internal state vector of the baseline controller, For the baseline controller state matrix, Input matrix for baseline controller Baseline controller output matrix, This is the feedforward matrix for the baseline controller.

[0025] Furthermore, S4 specifically includes:

[0026] S401, Closed-loop matrix obtained based on S2 Establish a state predictor:

[0027]

[0028] In the formula, To predict the state vector, For the input control signal, For the total uncertainty of the system, For feedback gain, >0;

[0029] S402, in each sampling period Within, based on prediction error Update total uncertainty ,

[0030] ;

[0031] S403, First, based on the matching input channel matrix Constructing a linear projection operator :

[0032]

[0033] Then use the linear projection operator The total uncertainty Decomposed into matching components With mismatched components :

[0034]

[0035]

[0036]

[0037] In the formula, For the mismatched linear projection operator, It is the identity matrix;

[0038] S404, Regarding the mismatched components Perform an equivalent mapping on the output to obtain a mapping mismatch signal. ; through the second low-pass filter For the matched components Mismatch signal with mapping Band-limiting processing is performed separately to obtain the corresponding matching compensation signal. With mismatch compensation signal :

[0039]

[0040] ;

[0041] S405, Matching compensation signal With mismatch compensation signal Synthesized into an adaptive compensation signal :

[0042] .

[0043] Furthermore, in S401, the feedback gain The range of values ​​for is: .

[0044] Furthermore, in S402, the total uncertainty... The update adopts a discrete update form:

[0045]

[0046]

[0047] In the formula, For sampling sequence number, The sampling period is This is the leakage coefficient, used to suppress long-term drift of the estimated value under the influence of noise. , The total uncertainty at the current moment, The total uncertainty at the previous moment, To constrain the projection operator, used to update the Constraints on a predefined bounded set Inside, and They are respectively The upper and lower bounds; It is a positive definite adaptive matrix.

[0048] Furthermore, the first low-pass filter The transfer function is in the form of:

[0049]

[0050] Second low-pass filter The transfer function is in the form of:

[0051]

[0052] In the formula, This is the cutoff frequency of the first low-pass filter. The damping ratio of the first low-pass filter. This is the cutoff frequency of the second low-pass filter. The damping ratio of the second low-pass filter. It is a Laplace complex variable.

[0053] Furthermore, in S5, the total control signal is controlled by a saturation function. Amplitude saturation limiting processing is performed; during the rate of change limiting processing, the absolute value of the slope of the constraint signal is less than or equal to a preset threshold.

[0054] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0055] (1) The present invention adopts The dynamic output feedback baseline controller serves as the foundation for system stability and disturbance rejection performance. It constructs a generalized controlled object based on a preset disturbance rejection level index and obtains the baseline controller by solving the algebraic Riccati equations. This ensures the closed-loop system maintains robust stability and the expected disturbance rejection capability under model parameter perturbations and external disturbances. Based on this, the steady-state gain is calculated using the nominal closed-loop state matrix extracted from the closed-loop system, and a steady-state feedforward control signal is generated based on reference pre-filtering, effectively reducing steady-state deviation. Simultaneously, it introduces... An adaptive compensation structure estimates the total uncertainty of the system online using a state predictor and generates an adaptive compensation signal, thereby achieving high-precision tracking in the linear motor drive system.

[0056] (2) To address the characteristics of linear motor drive systems where disturbances such as thrust fluctuations, cogging effects, and load changes directly affect the mover and lack mechanical filtering, this invention designs an uncertainty decomposition mechanism and an output equivalent mapping mechanism based on a projection operator. By constructing a linear matching projection operator, the total uncertainty is precisely decomposed into a matched component that can be directly compensated by the control input and an unmatched component that cannot be directly compensated. For the unmatched component, output equivalent mapping is used for processing, where the output influence matrix of the unmatched component directly calculates the impact of the unmatched disturbance on the system output, and then converts it to the control input through the inverse matrix of the steady-state gain matrix of the matching channel. Compared with the prior art, the mapping mechanism of this invention directly addresses the disturbance itself, and has stronger targeting and better suppression effect on disturbances such as thrust fluctuations acting on the mover, significantly enhancing the system's adaptability to complex disturbances and parameter perturbations.

[0057] (3) This invention introduces multiple band-limiting and bounded estimation mechanisms in the adaptive compensation channel, effectively suppressing high-frequency noise and unmodeled dynamic excitation, ensuring the engineering applicability of the method. At the frequency domain level, the matched component and the mapped mismatch signal are band-limited by a second-order low-pass filter and then synthesized into an adaptive compensation signal, ensuring that the compensation energy is concentrated within the effective bandwidth of the system. At the time domain level, the total uncertainty update adopts a discrete update law with leakage terms and constrained projection operators. The leakage coefficient suppresses the drift of the estimated value, and the constrained projection operator constrains the estimated value within a bounded set preset according to the physical characteristics of the driver. At the output level, the total control signal is sequentially subjected to amplitude saturation limiting and rate of change limiting processing, so that the control command is always within the physical allowable range of the actuator. The above mechanisms together ensure the stability, reliability and engineering feasibility of the control method, and have good promotion value in high-precision linear motor servo systems. Attached Figure Description

[0058] Figure 1 This is a schematic block diagram of the overall structure of the control method proposed in the embodiments of the present invention;

[0059] Figure 2 This is a graph showing the position tracking error using the PID control method.

[0060] Figure 3 This is a position tracking error curve diagram using the control method proposed in the embodiments of the present invention. Detailed Implementation

[0061] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0062] Example

[0063] refer to Figure 1 This embodiment proposes a method based on Robust baseline control and The adaptive compensation control method for linear motor drive systems can be applied to applications such as precision positioning platforms, high-speed linear feed mechanisms, and automated equipment. The implementation steps of this method are as follows:

[0064] S1. Based on the physical characteristics of the linear motor, establish a linear time-invariant state-space model of its drive system. The model is expressed as follows:

[0065]

[0066] In the formula, System state vector;

[0067] The control input is typically a torque or current command from the motor.

[0068] This is the system output signal, typically displacement or velocity;

[0069] For the system matrix, For the input matrix, This is the output matrix.

[0070] In this embodiment, the displacement of the moving part is taken. and mover velocity As system state variables, i.e., state vectors Considering the equivalent mass of the mover. Equivalent damping of guide rail Thrust constant Then the dynamic equation of the linear motor can be expressed as:

[0071]

[0072] In the formula, control input The corresponding thrust or current command from the driver. The mover displacement is selected as the system output, i.e. Rewriting the above dynamic equations in state-space form, we get:

[0073]

[0074]

[0075] therefore, , , .

[0076] The model established in this step is used to characterize the nominal dynamic characteristics of the linear motor drive system, where the system matrix... It describes the integral relationship between the mover displacement and velocity, as well as the dynamic characteristics of the velocity, with the input matrix... The output matrix reflects the influence of the control input on the mover acceleration. The system state is mapped to a measurable displacement output. This model serves as the basis for subsequent baseline controllers. The design, closed-loop matrix extraction, steady-state gain calculation, and state predictor construction provide a unified mathematical foundation. In practical applications, model parameters... , , It can be obtained through system identification experiments or by calculating based on parameters in the motor manual.

[0077] S2, Design Dynamic output feedback baseline controller Specifically, it includes the following sub-steps:

[0078] S201, Calculate the reference input signal With system output signal Tracking error between :

[0079]

[0080] S202, A linear time-invariant state-space model based on S1 and a preset disturbance suppression level index Construct a generalized controlled object. Specifically, select a sensitivity weighting function. and control quantity weighting function The physical controlled object and the weighting function are arranged according to... The standard combination of mixed sensitivity problems constitutes a generalized controlled object. The disturbance suppression level index Typically, a positive number less than 1 is chosen; in this embodiment, we take... =0.5, to ensure good disturbance suppression capability.

[0081] pass Iteratively solving the corresponding algebraic Riccati equations yields... Dynamic output feedback baseline controller state matrix Input matrix Output matrix and feedforward matrix The baseline controller The state space implementation is as follows:

[0082]

[0083]

[0084] In the formula, For the internal state vector of the baseline controller, for Time derivative,

[0085] S203, the generalized controlled object With baseline controller The connections form a closed-loop system, and the nominal closed-loop state matrix corresponding to the state of the controlled object is extracted from this closed-loop system as the closed-loop matrix. The closed-loop matrix The system is described in the baseline controller. The expected dynamic characteristics under the action provide a reference benchmark for subsequent steady-state feedforward and adaptive compensation.

[0086] S204, Tracking error Input to the baseline controller Output baseline control signal This baseline control signal ensures the system's basic robust stability and disturbance suppression capability.

[0087] S3. To further reduce the system's steady-state error and improve tracking accuracy, the input matrix is ​​used... and output matrix and the closed-loop matrix obtained by S2 Calculate steady-state gain :

[0088]

[0089] Considering that the reference input may contain abrupt or rapidly changing components, thereby exciting the system's high-frequency dynamics, a second-order first-low-pass filter is first employed. For the reference input signal Filtering is performed to obtain the pre-filtered reference signal. :

[0090]

[0091] In this embodiment, the first low-pass filter The transfer function is:

[0092]

[0093] In the formula, This is the cutoff frequency of the first low-pass filter, determined based on the system's closed-loop bandwidth, and is typically taken as 2-3 times the closed-loop bandwidth. The damping ratio of the first low-pass filter is taken as [value missing] in this embodiment. To achieve a good balance between amplitude-frequency and phase-frequency characteristics, It is a Laplace complex variable.

[0094] Through the steady-state gain For the pre-filtered reference signal The signal is processed to obtain the signal before filtering. Generate steady-state feedforward control signal :

[0095]

[0096] Steady-state feedforward control signal With reference input signal The trend of change is consistent with that of tracking error, which can effectively reduce the steady-state component in tracking error.

[0097] S4, Design The adaptive compensator estimates and compensates for the total uncertainty of the system online, specifically including the following sub-steps:

[0098] S401, Closed-loop matrix obtained based on S2 Establish a state predictor:

[0099]

[0100] In the formula, To predict the state vector, The control signal input to the state predictor is used in the discrete implementation to avoid the predictor and control signal from overlapping and forming an algebraic loop. It can be taken as the total control signal at the previous sampling time. This is the feedback gain, used to accelerate convergence. >0; To match the input channel matrix, the preferred option is to select the single-input, single-output case. .

[0101] S402, in each sampling period Within, based on prediction error For total uncertainty Update the process. To ensure the stability and boundedness of the estimation process, a discrete update form with leakage terms and projection operators is adopted:

[0102]

[0103] In the formula, For sampling sequence number, The sampling period is taken in this embodiment. , Leakage coefficient, In this embodiment, we take , As a positive definite adaptive matrix, this embodiment takes... , The total uncertainty at the current moment, The total uncertainty at the previous moment, To constrain the projection operator, used to update the Constraints on a predefined bounded set Within this framework, ensure that the estimated values ​​comply with the system's physical constraints.

[0104] The predefined bounded set Based on the physical constraints of the linear motor driver, in this embodiment, it is taken as a set of upper and lower bound constraints based on elements:

[0105]

[0106] In the formula, and They are respectively The upper and lower bounds are taken in this embodiment. , .

[0107] S403, First, based on the matching input channel matrix Constructing a linear projection operator :

[0108]

[0109] Its function is to project any vector onto the matching input channel. On the column space.

[0110] Then use the linear projection operator The total uncertainty Decomposed into matching components With mismatched components :

[0111]

[0112]

[0113]

[0114] In the formula, For the mismatched linear projection operator, For identity matrix; matching components lie in In the column space, disturbances that can be directly compensated by control inputs are represented; mismatched components lie in Within the orthogonal complement space, it cannot be directly compensated by the control input and needs to be processed through subsequent equivalent mapping.

[0115] S404, First, process the mismatched components. Perform an equivalent mapping on the output to obtain a mapping mismatch signal. ;

[0116]

[0117]

[0118]

[0119] In the formula, To match the channel steady-state gain matrix The inverse matrix, Output the influence matrix for the mismatched components;

[0120] Then pass through a second-order low-pass filter For the matched components Mismatch signal with mapping Band-limiting processing is performed separately to obtain the corresponding matching compensation signal. With mismatch compensation signal :

[0121]

[0122] ;

[0123] Second low-pass filter The transfer function is in the form of:

[0124]

[0125] In the formula, The cutoff frequency of the second low-pass filter is determined based on the system closed-loop bandwidth and the frequency range of the unmodeled dynamics. It is usually taken as 3-5 times the closed-loop bandwidth to ensure a fast response within the effective frequency band of the compensation signal, while suppressing high-frequency noise. The damping ratio of the second low-pass filter is taken in this embodiment. To obtain good dynamic response characteristics.

[0126] S405, Matching compensation signal With mismatch compensation signal Synthesized into an adaptive compensation signal :

[0127] .

[0128] This adaptive compensation signal combines the compensation effects for both matched and mismatched disturbances, and can further improve the tracking accuracy and disturbance rejection capability of the system based on baseline control.

[0129] S5. Combine the above three control signals and calculate the total control signal. :

[0130]

[0131] To prevent actuator saturation and sudden changes in control signals, the total control signal is... Post-processing:

[0132] Through saturation function For the total control signal Amplitude saturation limiting is performed to generate a saturated-limited signal. :

[0133]

[0134] Subsequently, the signal after saturation limitation was analyzed. The rate of change is limited to constrain the rate of change of the signal after saturation, ensuring that the absolute value of its slope is less than or equal to a preset threshold. To obtain safe and feasible system input signals In this embodiment, the thrust amplitude is set to a lower limit of -10N and an upper limit of 10N based on the physical limitations of the actuator, and the threshold value is used. .

[0135] System input signal The signal is sent to a power amplifier to drive the motor, thereby achieving high-precision position or speed tracking control of the linear motor system.

[0136] To verify the effectiveness of this invention, a comparative experiment was conducted on a linear motor experimental platform. The experimental platform consisted of a linear motor, a grating displacement sensor (0.1µm resolution), a servo driver, and a real-time controller (sampling period...). It consists of ) . The reference input is a sine wave with an amplitude of 10mm and a frequency of 1Hz.

[0137] Position tracking control experiments were conducted using both the PID control method and the control method proposed in this embodiment of the invention. The parameters of the PID control method were tuned, including the proportional gain... Integral gain Differential gain The parameters of the control method of the present invention are set according to the above embodiments.

[0138] Figure 2 To show the tracking error curve when using PID control, from Figure 2 As can be seen, the maximum absolute tracking error is 0.007072 mm (approximately 7.072 µm), and the root mean square error (RMS) is 0.000698 mm (approximately 0.698 µm). This indicates that the tracking error under PID control exhibits significant transient spikes and fluctuations during the dynamic change phase.

[0139] Figure 3 The tracking error curve is obtained when using the control method proposed in this invention, from... Figure 3 It can be seen that the maximum absolute error is reduced to 0.003675 mm (approximately 3.675 µm), and the root mean square error is reduced to 0.000365 mm (approximately 0.365 µm). Compared with the PID control method, the method proposed in this invention reduces the peak error by approximately 48.0% and the root mean square error by approximately 47.7%.

[0140] Further analysis Figure 3 The error curves shown also demonstrate that the control method of this invention can rapidly suppress transient errors during the dynamic response phase, significantly reduce error fluctuations during the steady-state phase, and exhibit a smooth response without obvious high-frequency oscillations. This indicates that the invention will... Robust stability of baseline control and The adaptive compensation and rapid disturbance suppression capabilities are organically combined to effectively improve the tracking accuracy and robustness of the linear motor drive system under uncertain and disturbed conditions.

[0141] The specific embodiments of the present invention are provided to enable those skilled in the art to understand or implement the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention.

[0142] It should be understood that the present invention is not limited to the content already described above, and various modifications and changes can be made without departing from its scope. The scope of the present invention is limited only by the appended claims.

Claims

1. A method based on Robust baseline control and An adaptive compensation linear motor drive system control method, characterized in that, Includes the following steps: S1. Establish a linear time-invariant state-space model of the linear motor drive system, which is expressed in the following form: In the formula, Let be the system state vector. To control the input, For system output signals, For the system matrix, For the input matrix, This is the output matrix; S2. Obtain the reference input signal And calculate the reference input signal. With system output signal Tracking error between A linear time-invariant state-space model based on S1 and a preset disturbance suppression level index. Construct a generalized controlled object; and through Iteratively solving the corresponding Riccati equations yields... Dynamic output feedback baseline controller ; The generalized controlled object is compared with the baseline controller. Connect the points to form a closed-loop system, and extract the nominal closed-loop state matrix of the object side from this closed-loop system as the closed-loop matrix. The tracking error Input to the baseline controller Output baseline control signal ; S3, through the closed-loop matrix Input matrix and output matrix Calculate steady-state gain ; reference input signal After passing through the first low-pass filter After processing, the pre-filtered reference signal is obtained. Then utilize the aforementioned steady-state gain For the pre-filtered reference signal The signal is processed to generate a steady-state feedforward control signal. ; S4, Based on the closed-loop matrix The total uncertainty of the system is estimated online using a state predictor. Based on matching input channel matrix Constructing a linear projection operator The total uncertainty Decomposed into matching components With mismatched components ; For the mismatched components Perform an equivalent mapping on the output to obtain a mapping mismatch signal. ; through the second low-pass filter For the matching components Mismatch signal with mapping Band-limiting processing is performed separately to obtain the corresponding matching compensation signal. With mismatch compensation signal The two are then combined into an adaptive compensation signal. ;in The mapping formula is: In the formula, To match the channel steady-state gain matrix The inverse matrix, Output the influence matrix for the mismatched components; S5, Calculate the total control signal For the total control signal The amplitude saturation limit and rate of change limit are applied sequentially to generate the system input signal. The total control signal The calculation formula is: 。 2. Based on claim 1 Robust baseline control and An adaptive compensation linear motor drive system control method, characterized in that, In S2, Dynamic output feedback baseline controller The state space implementation is as follows: In the formula, For the internal state vector of the baseline controller, For the baseline controller state matrix, Input matrix for baseline controller Baseline controller output matrix, This is the feedforward matrix for the baseline controller.

3. Based on claim 1 Robust baseline control and An adaptive compensation linear motor drive system control method, characterized in that, S4 specifically includes: S401, Closed-loop matrix obtained based on S2 Establish a state predictor: In the formula, To predict the state vector, For the input control signal, For the total uncertainty of the system, For feedback gain, >0; S402, in each sampling period Within, based on prediction error Update total uncertainty , ; S403, First, based on the matching input channel matrix Constructing a linear projection operator : Then use the linear projection operator The total uncertainty Decomposed into matching components With mismatched components : In the formula, For the mismatched linear projection operator, It is the identity matrix; S404, Regarding the mismatched components Perform an equivalent mapping on the output to obtain a mapping mismatch signal. ; through the second low-pass filter For the matching components Mismatch signal with mapping Band-limiting processing is performed separately to obtain the corresponding matching compensation signal. With mismatch compensation signal : ； S405, Matching compensation signal With mismatch compensation signal Synthesized into an adaptive compensation signal : 。 4. Based on claim 3 Robust baseline control and An adaptive compensation linear motor drive system control method, characterized in that, In S401, the feedback gain The range of values ​​for is: .

5. Based on claim 3 Robust baseline control and An adaptive compensation linear motor drive system control method, characterized in that, In S402, the total uncertainty The update adopts a discrete update form: In the formula, For sampling sequence number, The sampling period is This is the leakage coefficient, used to suppress long-term drift of the estimated value under the influence of noise. , The total uncertainty at the current moment, The total uncertainty of the previous moment, To constrain the projection operator, used to update the Constraints on a predefined bounded set Inside, and They are respectively The upper and lower bounds; It is a positive definite adaptive matrix.

6. Based on claim 1 Robust baseline control and An adaptive compensation linear motor drive system control method, characterized in that, First low-pass filter The transfer function is in the form of: Second low-pass filter The transfer function is in the form of: In the formula, This is the cutoff frequency of the first low-pass filter. The damping ratio of the first low-pass filter. This is the cutoff frequency of the second low-pass filter. The damping ratio of the second low-pass filter. It is a Laplace complex variable.

7. Based on claim 1 Robust baseline control and An adaptive compensation linear motor drive system control method, characterized in that, In S5, the total control signal is controlled by a saturation function. Amplitude saturation limiting processing is performed; during the rate of change limiting processing, the absolute value of the slope of the constraint signal is less than or equal to a preset threshold.

Images