Permanent magnet synchronous linear motor multi-step prediction thrust control method, device and medium based on block matrix optimization

Abstract

The application discloses a permanent magnet synchronous linear motor multi-step prediction thrust control method based on block matrix optimization, equipment and medium, and the method comprises the steps of establishing a thrust and flux linkage prediction model; constructing an average block matrix and a front dense and rear sparse block matrix; calculating the average absolute error of the thrust in the past sampling period; setting the upper limit and lower limit of the threshold; comparing the average absolute error of the thrust with the two, judging the operating condition of the system, switching the block matrix according to the operating condition, reconstructing the control voltage sequence by dimension reduction, inputting the control voltage sequence after dimension reduction and reconstruction into the thrust and flux linkage prediction model, calculating the predicted thrust and predicted stator flux linkage in the future multiple time steps, calculating the prediction error, and substituting the prediction error into the multi-step prediction value function to perform extreme value optimization, selecting the optimal control voltage vector that makes the value function minimum, and applying the optimal control voltage vector to the drive inverter to control the stator current and stator flux linkage of the permanent magnet synchronous linear motor. The application improves the dynamic response accuracy and steady-state robustness of the system.

Description

Technical Field

[0001] This invention relates to the field of linear servo control technology, specifically to a method, device, and medium for multi-step predictive thrust control of a permanent magnet synchronous linear motor based on block matrix optimization. Background Technology

[0002] Permanent magnet synchronous linear motors, by eliminating intermediate transmission links, offer advantages such as high thrust density, fast dynamic response, and high positioning accuracy. In cutting-edge fields such as high-end CNC machine tools and semiconductor manufacturing, linear servo systems must possess extremely high dynamic response rates to cope with frequent starts and stops and sudden changes in external loads. While traditional direct thrust control is simple in structure, it is prone to large thrust fluctuations and inconsistent switching frequencies. Model predictive thrust control, due to its robustness and ease of handling constraints, is widely used.

[0003] Compared to single-step prediction, multi-step predictive control (MMC) significantly enhances the system's anti-interference capability by extending the prediction time domain to anticipate future dynamics. However, traditional MMC employs an exhaustive search method for optimization, and its computational load increases exponentially with the prediction step size, leading to a severe deterioration in system real-time performance. Although existing algorithms can reduce some computational load, their complexity remains high, making it difficult to achieve optimal performance on hardware platforms with limited computing power. Summary of the Invention

[0004] This invention addresses the shortcomings of existing technologies by providing a method, device, and medium for multi-step predictive thrust control of permanent magnet synchronous linear motors based on block matrix optimization. It aims to solve the problems of exponential growth in computational complexity with prediction time, large computational load, poor real-time performance, and difficulty in responding quickly to disturbances in existing multi-step predictive thrust control methods for linear motors.

[0005] To achieve the above objectives, the present invention adopts the following technical solution:

[0006] A multi-step predictive thrust control method for a permanent magnet synchronous linear motor based on block matrix optimization includes the following steps: Step 1: Establish a thrust and flux prediction model for multiple future time steps in a synchronous rotating coordinate system; Step 2: Construct two types of block matrices to reduce the dimensionality space of multi-step predictive control, including an averaged block matrix for steady-state operation and a front-dense-back-sparse block matrix for transient response; Step 3: Calculate the thrust mean absolute error index over a specific sampling period in the past based on the sliding window algorithm; Step 4: Set the upper and lower threshold limits; compare the thrust mean absolute error index with the upper and lower threshold limits to determine the system's operating conditions, and adaptively switch the corresponding block matrix based on the operating condition judgment results; Step 5: Use a block matrix to perform dimensionality reduction and reconstruction of the control voltage sequence. Input the dimensionality-reconstructed control voltage sequence into the thrust and flux linkage prediction model to calculate the predicted thrust and predicted stator flux linkage in multiple future time steps. After calculating the prediction error, input it into the multi-step prediction value function for extreme value optimization. Select the optimal control voltage vector that minimizes the value function and apply this optimal control voltage vector to the drive inverter to control the stator current and stator flux linkage of the permanent magnet synchronous linear motor. Finally, achieve closed-loop tracking control of the electromagnetic thrust of the permanent magnet synchronous linear motor.

[0007] To optimize the above technical solution, the specific measures also include: Furthermore, in step 1, the thrust and flux prediction model is specifically as follows: Predicting stator flux The expression is:

[0008] In the formula, To predict stator flux linkage; For permanent magnet flux linkage; The sampling period; The switching signal corresponding to the control voltage; Stator resistance; Stator current; Predicting electromagnetic thrust It is given by the following formula:

[0009] In the formula, To predict stator flux linkage; τ The distance between the magnetic poles; The current is the stator current along the d-axis.

[0010] Further, in step 2, the expression for the averaged block matrix is ​​as follows:

[0011] In the formula, It is an averaged block matrix; The expression for the front-dense-back-sparse block matrix is ​​as follows:

[0012] In the formula, It is a block matrix with dense front and sparse back.

[0013] Furthermore, step 3 specifically involves: The expression for the thrust mean absolute error index over the past W sampling periods is as follows:

[0014] In the formula, It is the thrust mean absolute error index. express kj The actual thrust value at that moment. express kj The predicted thrust value at any given time.

[0015] Furthermore, step 4 specifically involves: Set upper limit for threshold With threshold lower limit The current thrust mean absolute error index Compared with a threshold, when When, switch to the front-dense-back-sparse block matrix I2; when When the thrust mean absolute error index is within the upper limit of the threshold, switch to the averaged block matrix I1; With threshold lower limit During this period, the block matrix state of the previous period is maintained.

[0016] Furthermore, in step 5, the dimensionality reduction and reconstruction of the control voltage sequence using a block matrix specifically involves:

[0017] In the formula, It is a control voltage sequence. It is the control voltage sequence after dimensionality reduction and reconstruction. It is a block matrix. i =1 indicates that the averaged block matrix I1 is used. i =2 indicates that the matrix I2 is a block matrix with denser front and sparser back.

[0018] Further, in step 5, the step of inputting the dimension-reduced and reconstructed control voltage sequence into the thrust and flux linkage prediction model to calculate the predicted thrust and predicted stator flux linkage over multiple future time steps specifically involves: The switching signals corresponding to the control voltages in the dimensionality-reconstructed control voltage sequence In the input thrust and flux linkage prediction model, the predicted thrust and predicted stator flux linkage are calculated over multiple future time steps.

[0019] Furthermore, in step 5, the multi-step prediction value function is specifically as follows:

[0020] In the formula, It is a multi-step prediction value function. It is a control voltage sequence A linear mapping relationship between the column vector sets of flux linkage error and thrust error. It is the error quantity generated by the control algorithm in the column vector set of flux linkage error and thrust error. This is the weight coefficient matrix.

[0021] The present invention also proposes an electronic device, comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the multi-step predictive thrust control method for permanent magnet synchronous linear motors based on block matrix optimization as described above.

[0022] The present invention also proposes a computer-readable storage medium storing a computer program that enables a computer to execute the multi-step predictive thrust control method for permanent magnet synchronous linear motors based on block matrix optimization as described above.

[0023] The beneficial effects of this invention are: This invention addresses the technical bottleneck between multi-step prediction and long-term time-domain optimization and the limitation of hardware computing power.

[0024] First, by constructing an averaged block matrix and a front-dense-back-sparse block matrix, the dimensionality of the long time-domain control sequence is reduced. Without changing the overall control law of the system, the number of optimization traversals of the microprocessor is greatly reduced, the computational burden and time delay are reduced, and the problem of large step size predictive control being difficult to deploy in real time in actual hardware is solved.

[0025] Secondly, the designed thrust error-based sliding window adaptive switching mechanism, relying on dual-threshold hysteresis judgment logic, enables flexible scheduling between steady-state high computing efficiency mode and transient high tracking accuracy mode. This mechanism effectively shields the frequent false triggering phenomenon caused by steady-state high-frequency noise using moving average filtering, balancing low computing power consumption and strong anti-interference capability during system operation.

[0026] Finally, the strategy of this invention can maximize the degree of freedom of control within the limited underlying computing power. Without compromising control accuracy due to dimensionality reduction, it effectively suppresses steady-state thrust pulsation and speed pulsation. This allows the linear motor to significantly suppress speed drops when dealing with sudden load disturbances and recover to steady state at an extremely fast speed, thus comprehensively improving the dynamic response accuracy and steady-state robustness of the system. Attached Figure Description

[0027] Figure 1 This is a block diagram illustrating the principle of the control method of the present invention; Figure 2 This is a waveform diagram of the full-load noise floor in an embodiment of the present invention; Figure 3 This is a schematic diagram of the block matrix adaptive switching strategy provided in an embodiment of the present invention; Figure 4 A comparison chart of thrust curves; Figure 5 This is a comparison chart of the computation time of the method of the present invention and the conventional method with a control step size of 3; Figure 6 This is a comparison chart of the computation time of the method of the present invention and the conventional method with a control step size of 6. Detailed Implementation

[0028] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of this application.

[0029] Example 1 This invention proposes a multi-step predictive thrust control method for permanent magnet synchronous linear motors based on block matrix optimization. The principle of this method is as follows: Figure 1 As shown, it includes the following steps: Step 1: Read the actual current and position of the permanent magnet synchronous linear motor through sensors, and establish a thrust and flux prediction model for multiple future time steps in a synchronous rotating coordinate system; The dynamic model of the permanent magnet synchronous linear motor in the direction of linear motion is as follows:

[0030] In the formula, F e Electromagnetic thrust; M The mass of the moving part; F L For load thrust; B m denoted as the coefficient of viscous friction, v as the velocity, and t as the time.

[0031] The specific thrust and flux prediction model is as follows: Predicting stator flux The expression is:

[0032] In the formula, To predict stator flux linkage; For permanent magnet flux linkage; The sampling period; The switching signal corresponding to the control voltage; Stator resistance; Stator current; Predicting electromagnetic thrust It is given by the following formula:

[0033] In the formula, To predict stator flux linkage; τ The distance between the magnetic poles; The current is the stator current along the d-axis.

[0034] After discretizing the thrust and flux prediction model, the flux error and thrust error at the next time step are obtained:

[0035]

[0036] In the formula, for k+1 The flux linkage error at any given time; for k+1 Thrust error at any given moment; for k+1 The target reference flux linkage value at any given time; for k+1 The predicted flux linkage value at time 1; for k+1 The target reference thrust value at any given time; for k+1 The predicted thrust value at any given time.

[0037] To achieve rolling time-domain optimization, the thrust error and stator flux amplitude tracking error terms are substituted together to construct a traditional multi-step predictive control value function:

[0038] In the formula, Q This is the weight coefficient matrix. It is a set of column vectors containing flux linkage error and thrust error. This represents the traditional multi-step predictive control value function. This indicates the control step size for multi-step prediction.

[0039] Step 2: Construct two types of block matrices to reduce the dimensionality of multi-step predictive control space, including an averaged block matrix for steady-state operation and a front-dense-back-sparse block matrix for transient response.

[0040] Since the length of the control time domain determines the control voltage sequence ΔU Increasing the prediction step size significantly increases the computational complexity of the aforementioned value function, given the degrees of freedom. Furthermore, based on the characteristics of model predictive control, each control cycle only outputs... ΔU The first control variable in the process can be approximated for the long-term control variable and precisely optimized for the control variable at the near time. This effectively reduces the computational burden while ensuring the current control accuracy. Therefore, a block matrix is ​​introduced to reconstruct the dimension of the control increment.

[0041] To control step size N c Taking multi-step prediction of 6 as an example, let adjacent control variables be the same, that is Δu 1 =Δu 2 , Δu 3 =Δ u 4 , Δu 5 =Δu 6. The control voltage sequence can then be converted into the form shown below:

[0042] in, I 1 represents the averaged block matrix. This reduces the original 6-dimensional control voltage sequence degrees of freedom to 3 dimensions.

[0043] To improve transient response during parameter abrupt changes, a block matrix with a denser front and a sparser back was designed. This matrix retains the accuracy of the control cycle closest to the current time, ensuring good transient response. I It is represented by 2, and its form is as follows:

[0044] Step 3: To accurately identify the current dynamic and steady-state operating state of the system and match a suitable block matrix, the thrust mean absolute error index is calculated based on the sliding window algorithm over a specific sampling period. This index, calculated by averaging the errors over the past W sampling periods, acts as a filter and accurately characterizes the recent overall tracking trend of the system.

[0045] The expression for the thrust mean absolute error index over the past W sampling periods is as follows:

[0046] In the formula, It is the thrust mean absolute error index. express kj The actual thrust value at that moment. express kj The predicted thrust value at any given time.

[0047] Step 4: Set the upper and lower threshold limits; compare the thrust mean absolute error index with the upper and lower threshold limits to determine the system's operating condition, and adaptively switch the corresponding block matrix based on the operating condition judgment result.

[0048] Based on the steady-state ripple characteristic test under full load conditions, the experimental waveform is as follows: Figure 2 As shown, a safety margin is reserved and a lower limit for the threshold is set. At the same time, the upper limit of the threshold is set. Set to twice the lower limit of the threshold.

[0049] The flowchart of the block matrix switching logic is as follows: Figure 3 As shown. The current thrust mean absolute error index... Compared with a threshold, when When, switch to the front-dense-back-sparse block matrix I2; when When the thrust mean absolute error index is within the upper limit of the threshold, switch to the averaged block matrix I1; With threshold lower limit During this period, the block matrix state of the previous period is maintained.

[0050] Step 5: Use a block matrix to perform dimensionality reduction and reconstruction of the control voltage sequence. Input the dimensionality-reconstructed control voltage sequence into the thrust and flux linkage prediction model to calculate the predicted thrust and predicted stator flux linkage in multiple future time steps. After calculating the prediction error, input it into the multi-step prediction value function for extreme value optimization. Select the optimal control voltage vector that minimizes the value function and apply this optimal control voltage vector to the drive inverter to control the stator current and stator flux linkage of the permanent magnet synchronous linear motor. Finally, achieve closed-loop tracking control of the electromagnetic thrust of the permanent magnet synchronous linear motor.

[0051] The specific steps for dimensionality reduction and reconstruction of the control voltage sequence using a block matrix are as follows:

[0052] In the formula, It is a control voltage sequence. It is the control voltage sequence after dimensionality reduction and reconstruction. It is a block matrix. i =1 indicates that the averaged block matrix I1 is used. i =2 indicates that the matrix I2 is a block matrix with denser front and sparser back.

[0053] The reconstructed control voltage sequence is input into the thrust and flux linkage prediction model to calculate the predicted thrust and predicted stator flux linkage over multiple future time steps. The switching signals corresponding to the control voltages in the dimensionality-reconstructed control voltage sequence In the input thrust and flux linkage prediction model, the predicted thrust and predicted stator flux linkage are calculated over multiple future time steps.

[0054] The traditional multi-step predictive control value function Convert to ΔU The new value function form for the variable:

[0055] In the formula, For the new value function, It is a control voltage sequence A linear mapping relationship between the column vector sets of flux linkage error and thrust error. It is the error quantity generated by the control algorithm in the column vector set of flux linkage error and thrust error. This is the weight coefficient matrix. This transformation significantly reduces the dimensionality of the solution without changing the overall control law of the system.

[0056] The process of calculating the prediction error and then inputting it into the multi-step prediction value function for extreme value optimization is as follows: Substitute the prediction error into the new value function. Perform extreme value optimization.

[0057] Example 2 This invention proposes an electronic device, including: a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the multi-step predictive thrust control method for permanent magnet synchronous linear motors based on block matrix optimization as described in Embodiment 1.

[0058] Example 3 This invention proposes a computer-readable storage medium storing a computer program that enables a computer to execute the multi-step predictive thrust control method for a permanent magnet synchronous linear motor based on block matrix optimization as described in Embodiment 1.

[0059] To verify the effectiveness of the control method of this invention, the following comparative simulation experiment was conducted. In MATLAB / Simulink, the linear motor was set to start from rest and accelerate to 2 m / s. The sampling frequency of the algorithm module was 2 kHz, and an external load of 100 N was suddenly applied to the linear motor at 0.2 s.

[0060] To verify that the proposed adaptive switching block matrix optimization strategy outperforms traditional multi-step prediction in both computation time and response accuracy, two sets of simulation comparison experiments were designed, controlling the step size N respectively. c =6 and N c A comprehensive comparison was conducted with traditional multi-step prediction algorithms with a step count of 3, while also verifying the impact of the step count on control accuracy and computational efficiency. The parameters of the permanent magnet synchronous linear motor used in the simulation are shown in Table 1.

[0061] Table 1. Parameters and Simulation Conditions of Permanent Magnet Synchronous Linear Motor

[0062] like Figure 4The figure shows a comparison of electromagnetic thrust response waveforms under different control strategies. Observing the magnified region during the startup phase reveals that the traditional multi-step model predicts thrust control with a large thrust overshoot at startup, accompanied by continuous and severe oscillations. The thrust rise trajectory of the algorithm implemented in this invention is very smooth, achieving a rapid startup with near-zero overshoot. When the system encounters a sudden load change at 0.2s, the traditional control strategy exhibits large transient fluctuations and slow recovery, while the algorithm implemented in this invention responds quickly and transitions smoothly, significantly suppressing steady-state thrust pulsations. These experimental results demonstrate that the method of this invention effectively improves the dynamic response accuracy and disturbance rejection robustness of the system while maintaining low computational power.

[0063] like Figure 5 As shown, observe Figure 5 In the scatter plot of (a), the single-cycle computation time of the traditional algorithm is mainly concentrated around 0.25ms, while the computation time of the algorithm implemented in this invention is significantly reduced to about 0.20ms. Combined with... Figure 5 As can be seen from the percentage curve in (b), the block matrix optimization method significantly reduces the computation time compared to the traditional method throughout the entire running cycle, effectively saving the underlying hardware computing power.

[0064] like Figure 6 As shown, by Figure 6 As shown in (a), the computation time of the traditional algorithm increases sharply with the increase of the control step size. Especially during the transient moment of 0.2s load change, the computational burden of the traditional model increases sharply, forming a significant computational power spike, with the scatter distribution exceeding 0.5ms, exhibiting extremely poor computational stability. The algorithm implemented in this invention, after adopting block matrix dimensionality reduction, maintains a stable scatter distribution of computation time in a low range, effectively suppressing the surge in computational power under extreme conditions. Figure 6 Figure (b) further demonstrates that at the 0.2s transient change point, the computation time of the algorithm implemented in this invention is reduced compared to the traditional strategy.

[0065] In the embodiments disclosed in this application, a computer storage medium may be a tangible medium that may contain or store programs for use by or in conjunction with an instruction execution system, apparatus, or device. The computer storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of computer storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, and portable compact disc read-only memory (CD). ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0066] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed in this application can be implemented in electronic hardware or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0067] The above are merely preferred embodiments of the present invention. The scope of protection of the present invention is not limited to the above embodiments. All technical solutions falling within the scope of the present invention's concept are within the scope of protection of the present invention. It should be noted that for those skilled in the art, any improvements and modifications made without departing from the principles of the present invention should be considered within the scope of protection of the present invention.

Claims

1. A multi-step predictive thrust control method for a permanent magnet synchronous linear motor based on block matrix optimization, characterized in that, Includes the following steps: Step 1: Establish a thrust and flux prediction model for multiple future time steps in a synchronous rotating coordinate system; Step 2: Construct two types of block matrices to reduce the dimensionality space of multi-step predictive control, including an averaged block matrix for steady-state operation and a front-dense-back-sparse block matrix for transient response; Step 3: Calculate the thrust mean absolute error index over a specific sampling period in the past based on the sliding window algorithm; Step 4: Set the upper and lower threshold limits; The thrust mean absolute error index is compared with the upper and lower thresholds to determine the system's operating condition, and the corresponding block matrix is ​​adaptively switched based on the operating condition judgment result. Step 5: Use a block matrix to perform dimensionality reduction and reconstruction of the control voltage sequence. Input the dimensionality-reconstructed control voltage sequence into the thrust and flux linkage prediction model to calculate the predicted thrust and predicted stator flux linkage in multiple future time steps. After calculating the prediction error, input it into the multi-step prediction value function for extreme value optimization. Select the optimal control voltage vector that minimizes the value function and apply this optimal control voltage vector to the drive inverter to control the stator current and stator flux linkage of the permanent magnet synchronous linear motor. Finally, achieve closed-loop tracking control of the electromagnetic thrust of the permanent magnet synchronous linear motor.

2. The multi-step predictive thrust control method for permanent magnet synchronous linear motors based on block matrix optimization as described in claim 1, characterized in that, In step 1, the thrust and flux prediction model is specifically as follows: Predicting stator flux The expression is: In the formula, To predict stator flux linkage; For permanent magnet flux linkage; The sampling period; The switching signal corresponding to the control voltage; Stator resistance; Stator current; Predicting electromagnetic thrust It is given by the following formula: In the formula, To predict stator flux linkage; τ The distance between the magnetic poles; The current is the stator current along the d-axis.

3. The multi-step predictive thrust control method for permanent magnet synchronous linear motors based on block matrix optimization as described in claim 1, characterized in that, In step 2, the expression for the averaged block matrix is ​​as follows: In the formula, It is an averaged block matrix; The expression for the front-dense-back-sparse block matrix is ​​as follows: In the formula, It is a block matrix with dense front and sparse back.

4. The multi-step predictive thrust control method for permanent magnet synchronous linear motors based on block matrix optimization as described in claim 1, characterized in that, Step 3 specifically involves: The expression for the thrust mean absolute error index over the past W sampling periods is as follows: In the formula, It is the thrust mean absolute error index. express kj The actual thrust value at that moment. express kj The predicted thrust value at any given time.

5. The multi-step predictive thrust control method for permanent magnet synchronous linear motors based on block matrix optimization as described in claim 1, characterized in that, Step 4 is as follows: Set upper limit for threshold With threshold lower limit The current thrust mean absolute error index Compared with a threshold, when When, switch to the front-dense-back-sparse block matrix I2; when When the thrust mean absolute error index is within the upper limit of the threshold, switch to the averaged block matrix I1; With threshold lower limit During this period, the block matrix state of the previous period is maintained.

6. The multi-step predictive thrust control method for permanent magnet synchronous linear motors based on block matrix optimization as described in claim 1, characterized in that, In step 5, the specific steps of using a block matrix to perform dimensionality reduction and reconstruction of the control voltage sequence are as follows: In the formula, It is a control voltage sequence. It is the control voltage sequence after dimensionality reduction and reconstruction. It is a block matrix. i =1 indicates that the averaged block matrix I1 is used. i =2 indicates that the matrix I2 is a block matrix with denser front and sparser back.

7. The multi-step predictive thrust control method for permanent magnet synchronous linear motors based on block matrix optimization as described in claim 1, characterized in that, In step 5, the step of inputting the dimension-reduced and reconstructed control voltage sequence into the thrust and flux linkage prediction model to calculate the predicted thrust and predicted stator flux linkage over multiple future time steps specifically involves: The switching signals corresponding to the control voltages in the dimensionality-reconstructed control voltage sequence In the input thrust and flux linkage prediction model, the predicted thrust and predicted stator flux linkage are calculated over multiple future time steps.

8. The multi-step predictive thrust control method for permanent magnet synchronous linear motors based on block matrix optimization as described in claim 1, characterized in that, In step 5, the multi-step prediction value function is specifically as follows: In the formula, It is a multi-step prediction value function. It is a control voltage sequence A linear mapping relationship between the column vector sets of flux linkage error and thrust error. It is the amount of error generated by the control algorithm in the column vector set of flux linkage error and thrust error. This is the weight coefficient matrix.

9. An electronic device, characterized in that, include: The memory, the processor, and the computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, it implements the multi-step predictive thrust control method for permanent magnet synchronous linear motors based on block matrix optimization as described in any one of claims 1-8.

10. A computer-readable storage medium storing a computer program, characterized in that, The computer program causes the computer to execute the multi-step predictive thrust control method for permanent magnet synchronous linear motors based on block matrix optimization as described in any one of claims 1-8.

Images