Event triggering-based dung beetle optimization electromagnetic drive micromirror control method
By combining an event-triggered mechanism, a dung beetle optimization algorithm, and a sliding mode controller, the problems of overshoot and excessive settling time in electromagnetically driven micromirror systems are solved, achieving fast, accurate, and stable micromirror control and improving the system's robustness and dynamic tracking capability.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-12-15
- Publication Date
- 2026-03-31
AI Technical Summary
Existing electromagnetically driven micromirror systems suffer from excessively high overshoot and excessively long settling time, which affect the system's real-time performance and dynamic tracking capabilities, and also lack anti-interference capabilities.
A composite control method combining event-triggered dung beetle optimization algorithm with model predictive controller and sliding mode controller is adopted. By leveraging the time-triggered mechanism and the robustness of the sliding mode controller, the control strategy of the electromagnetically driven micromirror is optimized. The parameters of the composite controller are tuned using the dung beetle optimization algorithm to achieve fast, accurate and stable pointing control.
This improves the control precision and stability of the electromagnetically driven micromirror, reduces overshoot and settling time, and enhances the system's anti-interference capability and dynamic tracking performance.
Smart Images
Figure CN121763734A_ABST
Abstract
Description
Technical Field
[0001] This invention belongs to the field of control method technology, specifically relating to an event-triggered optimized electromagnetic drive micromirror control method for dung beetles. Background Technology
[0002] Micro-optomechanical systems (MOEMS) are a branch of MEMS. A major application of MOEMS is MEMS micromirrors. There are four main driving methods for micromirrors: electrothermal, piezoelectric, electrostatic, and electromagnetic. Electromagnetically driven scanning micromirrors are widely used by researchers due to their advantages such as high scanning frequency, large driving torque, and low energy consumption.
[0003] Electromagnetically driven MEMS systems cannot function without control systems. The performance of electromagnetically driven MEMS systems can be improved through control methods. In practical applications of micromirrors, in addition to dynamic and static performance, it is also necessary to consider issues such as system robustness, power consumption, and noise. Therefore, we propose an electromagnetically driven micromirror control method. Summary of the Invention
[0004] The purpose of this invention is to provide an event-triggered optimized electromagnetic drive micromirror control method for dung beetles, in order to solve the problems mentioned in the background art.
[0005] To achieve the above-mentioned technical objectives, the technical solution adopted by the present invention is as follows: An event-triggered optimized electromagnetic drive micromirror control method for dung beetles, comprising the following steps: Step 1: Taking the electromagnetically driven micromirror system as the object, design a model prediction controller with a time-triggered mechanism to match it. When the controller output error is greater than a threshold, the prediction time domain remains unchanged; conversely, when the output error is less than the threshold, the prediction time domain decreases. Step 2: Use the control output of the model predictive controller as the reference input of the sliding mode controller, and the sliding mode controller will adjust and compensate for it in real time. Step 3: The parameters of the composite controller are tuned using an improved dung beetle optimization algorithm. The control strategy combines a model predictive controller and a sliding mode controller to control the angle rotation of the micromirror, thereby achieving fast, accurate and stable pointing of the electromagnetically driven micromirror.
[0006] The model predictive controller in step one consists of at least an event trigger, a model predictive control optimizer, a control hold, and a sensor.
[0007] The sensor is used to collect real-time data of the micromirror system, and the real-time deflection angle of the micromirror is obtained through the PSD position sensor; The time trigger is used to monitor the system status and decide whether to start the model predictive control optimizer based on preset trigger conditions; The event-triggered mechanism uses a combination of time and state-based triggering conditions, namely: Only if a minimum time interval has elapsed since the last trigger time And system state error ,in It is the system state at the previous trigger moment. This is the current system state. The norm of a vector This is a preset positive trigger threshold.
[0008] The model predictive control optimizer is used to solve a finite-time optimization problem and generate the optimal control sequence when the system is triggered, with the current system state as the initial state. The objective function J of the model predictive controller is:
[0009] Where Q, W, and P are weight matrices, and N is the prediction time domain. and These represent the state and control variables in the prediction time domain, respectively.
[0010] The control hold is used to store and apply the optimal control sequence during non-trigger periods.
[0011] During non-trigger periods, the control continues to apply the model from the previous side to predict the first control quantity in the optimal control sequence calculated by the control optimizer. The networked control system is formed by connecting event triggers and sensors via a network, and connecting the model predictive control optimizer with the control holder and actuator via a network.
[0012] The sliding mode controller in step two consists of at least a sliding surface and a control law.
[0013] The sliding surface is designed as a linear combination of state variables. When the system state reaches this sliding surface, the system moves along a predetermined trajectory. The control law is designed as an exponential control law, which enables the system state to be reached and maintained on the sliding surface within a finite time.
[0014] The dung beetle optimization algorithm in step three includes at least the following behaviors: rolling ball behavior, dancing behavior, reproductive behavior, foraging behavior, and stealing behavior.
[0015] The initial population of the dung beetle optimization algorithm uses a chaotic mapping, which allows the initial individuals to better cover the entire solution space; The dung beetle optimization algorithm introduces nonlinear convergence factors or adaptive weights to replace the original linear update strategy in the search boundary and update strategy of behaviors such as rolling, reproduction and foraging, or integrates the search mechanism of other algorithms, so as to better coordinate the global search and local development capabilities.
[0016] In this embodiment, when designing a model predictive controller for an electromagnetically driven MEMS system, it was found that although the model predictive controller can demonstrate excellent control performance in many control systems due to its predictive capability and optimization efficiency, it exhibits excessively high overshoot and long settling time in the micromirror system. This affects the real-time performance of the system during control, hindering dynamic tracking and anti-interference capabilities. To address this issue, a composite controller combining sliding mode control is proposed on top of a single control strategy. By leveraging the strong robustness of the sliding mode controller to compensate for the model predictive controller's dependence on an accurate mathematical model, and by utilizing an event-triggered mechanism-based model predictive controller to predict future control characteristics, the chattering problem of the sliding mode controller is resolved. Attached Figure Description
[0017] The present invention can be further illustrated by the non-limiting embodiments given in the accompanying drawings.
[0018] Figure 1 This is a schematic diagram of the system framework of the present invention.
[0019] Figure 2 This is a schematic diagram of the event-triggered model prediction and sliding mode dual closed-loop controller of the present invention.
[0020] Figure 3 This is a flowchart of the improved dung beetle algorithm of the present invention. Detailed Implementation
[0021] To enable those skilled in the art to better understand the present invention, the technical solution of the present invention will be further described below in conjunction with the accompanying drawings and embodiments.
[0022] like Figure 1-3 As shown, an event-triggered optimized electromagnetic drive micromirror control method for dung beetles is described, with the following specific steps: Step 1: Taking the electromagnetically driven micromirror system as the object, design a model prediction controller with a time-triggered mechanism to match it. When the controller output error is greater than a threshold, the prediction time domain remains unchanged; conversely, when the output error is less than the threshold, the prediction time domain decreases. Step 2: Use the control output of the model predictive controller as the reference input of the sliding mode controller, and the sliding mode controller will adjust and compensate for it in real time. Step 3: The parameters of the composite controller are tuned using an improved dung beetle optimization algorithm. The control strategy combines a model predictive controller and a sliding mode controller to control the angle rotation of the micromirror, thereby achieving fast, accurate and stable pointing of the electromagnetically driven micromirror.
[0023] The model predictive controller in step one consists of at least an event trigger, a model predictive control optimizer, a control hold, and a sensor.
[0024] The sensor is used to collect real-time data of the micromirror system, and the real-time deflection angle of the micromirror is obtained through the PSD position sensor; The dynamic model of the micromirror is as follows:
[0025] in It is the angular acceleration of the micromirror. It is the rotational angular velocity of the micromirror. It is the deflection angle.
[0026] make , At this point, the dynamic model of the micromirror can be rewritten as:
[0027]
[0028] Where b is the damping coefficient, k is the spring constant of the torsion bar, and J is the moment of inertia.
[0029] Then, the J, b, k, and T parameters are calculated sequentially to obtain the model.
[0030] The time trigger is used to monitor the system status and decide whether to start the model predictive control optimizer based on preset trigger conditions; The event-triggered mechanism uses a combination of time and state-based triggering conditions, namely: Only if a minimum time interval has elapsed since the last trigger time And system state error ,in It is the system state at the previous trigger moment. This is the current system state. The norm of a vector This is a preset positive trigger threshold.
[0031] The model predictive control optimizer is used to solve a finite-time optimization problem and generate the optimal control sequence when the system is triggered, with the current system state as the initial state. The objective function J of the model predictive controller is:
[0032] Where Q, W, and P are weight matrices, and N is the prediction time domain. and These represent the state and control variables in the prediction time domain, respectively.
[0033] The control hold is used to store and apply the optimal control sequence during non-trigger periods.
[0034] During non-trigger periods, the control continues to apply the model from the previous side to predict the first control quantity in the optimal control sequence calculated by the control optimizer. The networked control system is formed by connecting event triggers and sensors via a network, and connecting the model predictive control optimizer with the control holder and actuator via a network.
[0035] The sliding mode controller in step two consists of at least a sliding surface and a control law.
[0036] The sliding surface is designed as a linear combination of state variables. When the system state reaches this sliding surface, the system moves along a predetermined trajectory. The control law is designed as an exponential control law, which enables the system state to be reached and maintained on the sliding surface within a finite time.
[0037] in Figure 2 The block diagram for event-triggered model prediction and sliding mode controller is shown below, with the event steps as follows: 1. Initialize the system prediction time domain to Np, control time domain to Nc, and trigger conditions. The weight matrices are Q, W, and P.
[0038] 2. The model predictive controller uses the current state Starting from this point, rolling optimization is performed to calculate the control sequence. And send it to the control retainer, the control retainer will The first control variable is applied to the sliding mode controller, and the system state begins to change.
[0039] 3. The sensor continuously measures the status. And send it to the event trigger, exceeding the minimum time. And state error The new control law is obtained by entering the model predictive controller. Otherwise, the control will maintain the previously calculated control value and continue to output.
[0040] 4. New control sequence The data is fed into the control hold, overwriting the old sequence. The hold then begins executing the first control input of the new control sequence. .
[0041] 5. Input the control signal into the sliding mode controller to obtain the control output. Composite control rate It acts on the controlled object.
[0042] 6. Switch timings and repeat step 3.
[0043] Design an event-triggered model prediction controller, and initialize the system prediction time domain as follows. Control time domain Triggering conditions The weight matrices are Q, W, and P.
[0044] Then, the optimization objective function of the model predictive controller is solved. The predicted output of the model predictive controller As a reference input for the sliding mode controller, where let ,but Then let , .
[0045] The switching function is now: in:
[0046]
[0047] Based on the exponential tendency law:
[0048] The derivative of the synovial surface is:
[0049] Substituting the derivative of the sliding surface into the reaching law, the governing law is solved as follows:
[0050] In summary, the control rates of the designed model prediction and sliding diaphragm dual closed-loop controller are:
[0051] like Figure 3 As shown, the improved dung beetle algorithm calculation module, in order to obtain the optimal composite controller, uses the improved dung beetle algorithm to tune the optimization parameters, sets the number of iterations, initializes the population according to the upper and lower bounds of the population, and the dung beetle optimization algorithm in step three includes at least: rolling ball behavior, dancing behavior, reproductive behavior, foraging behavior, and stealing behavior.
[0052] The initial population of the dung beetle optimization algorithm uses a chaotic mapping, which allows the initial individuals to better cover the entire solution space; The dung beetle optimization algorithm introduces nonlinear convergence factors or adaptive weights to replace the original linear update strategy in the search boundary and update strategy of behaviors such as rolling, reproduction and foraging, or integrates the search mechanism of other algorithms, so as to better coordinate the global search and local development capabilities.
[0053] The specific method is as follows: S1: The population is initialized using a Logistic-Tent composite chaotic system, and the upper bound, lower bound, and number of iterations are set. The initialization formula is as follows:
[0054] Where t represents the number of iterations. This represents the position information of the i-th dung beetle during the t-th iteration.
[0055] S2: Dung beetles use the sun for navigation. When there is light and no obstacles, they will roll a ball. They use a triangular walking strategy to improve this rolling behavior, as shown in the following formula:
[0056] in , , .
[0057] S3: When a dung beetle encounters an obstacle, it will dance to obtain a new route. The update formula is as follows:
[0058] in ,like 0 or or ,but Do not update, otherwise Update the previous version.
[0059] S4: Under suitable conditions, female dung beetles will choose appropriate egg-laying locations. The updated formula is as follows:
[0060]
[0061] in , Indicates the lower and upper boundaries of the spawning area. This represents the current local optimum. , Let Lb and Ub represent the maximum number of iterations, and Lb and Ub represent the lower and upper bounds of the optimization problem, respectively. The spawning region is dynamically changing, and its update formula is as follows:
[0062] in This represents the position information of the i-th egg in the t-th iteration. and They are two independent random vectors of size 1×dim.
[0063] S5: The formula for updating the optimal foraging area of the dung beetle is as follows:
[0064]
[0065] in Indicates the globally optimal position. and These represent the lower and upper bounds of the optimal foraging area, respectively.
[0066] The formula for updating the position of the dung beetle is as follows:
[0067] in It is a random number that follows a normal distribution. yes A random vector within a certain range.
[0068] S6: Some dung beetles will steal dung balls from other dung beetles. The thief's location is updated using the following formula:
[0069] Where g is a random vector of size 1×dim that follows a normal stitch, and S is a constant.
[0070] The fitness function in this application adopts a performance index ITAE used for evaluating the performance of control systems. The fitness function is as follows:
[0071] Where T is the maximum evaluation event. This is the difference between the expected output and the system output at event t.
[0072] In this embodiment, when designing a model predictive controller for an electromagnetically driven MEMS system, it was found that although the model predictive controller can demonstrate excellent control performance in many control systems due to its predictive capability and optimization efficiency, it exhibits excessively high overshoot and long settling time in the micromirror system. This affects the real-time performance of the system during control, hindering dynamic tracking and anti-interference capabilities. To address this issue, a composite controller combining sliding mode control is proposed on top of a single control strategy. The strong robustness of the sliding mode controller compensates for the model predictive controller's dependence on an accurate mathematical model, while the event-triggered model predictive controller predicts future control characteristics to solve the chattering problem of the sliding mode controller.
[0073] The above embodiments are merely illustrative of the principles and effects of the present invention and are not intended to limit the invention. Any person skilled in the art can modify or alter the above embodiments without departing from the spirit and scope of the present invention. Therefore, all equivalent modifications or alterations made by those skilled in the art without departing from the spirit and technical concept disclosed in the present invention should still be covered by the claims of the present invention.
Claims
1. An event-triggered based control method for electromagnetically actuated micro-mirror based on dung beetle optimization, characterized in that: The specific steps are as follows: Step one: taking the electromagnetic drive micro mirror system as the object, a model predictive controller with time trigger mechanism is designed to match it, wherein when the controller output error is greater than the threshold, the prediction time domain remains unchanged, otherwise when the output error is less than the threshold, the prediction time domain is reduced; Step two: the control output of the model predictive controller is taken as the reference input of the sliding mode controller, which is adjusted and compensated in real time by the sliding mode controller; Step three: the improved scarab optimization algorithm is used to set the parameters of the composite controller, wherein the control strategy combining the model predictive controller and the sliding mode controller is used to control the angular rotation of the micro mirror, so as to realize the rapid, accurate and stable pointing of the electromagnetic drive micro mirror.
2. The control method of the event-triggered optimization of the electromagnetic drive micro mirror based on the dung beetle according to claim 1, wherein: The model predictive controller in step one is composed of at least an event trigger, a model predictive control optimizer, a control keeper and a sensor.
3. The control method of the event-triggered based Mayetopterus optimized electromagnetic drive micro-mirror according to claim 2, characterized in that: The sensor is used to collect real-time data of the micro mirror system, and the real-time deflection angle of the micro mirror is obtained through the PSD position sensor; The time trigger is used to monitor the system state, and decide whether to start the model predictive control optimizer according to the preset trigger condition; The model predictive control optimizer is used to solve the finite time domain optimization problem with the current system state as the initial state when triggered, and generate the optimal control sequence; The control keeper is used to store and apply the optimal control sequence in the non-trigger period.
4. The control method of the event-triggered optimization of the electromagnetic drive micro mirror based on the dung beetle according to claim 1, wherein: The sliding mode controller in step two is composed of at least a sliding surface and a control law.
5. The control method of the event-triggered Scarabaeidae-optimized electromagnetic drive micromirror according to claim 4, characterized in that: The sliding surface is designed as a linear combination of state variables, and when the system state reaches this sliding surface, the system moves along the predetermined trajectory; The control law is designed as an exponential control law, which can make the system state reach and remain on the sliding surface in a limited time.
6. The control method of the event-triggered Scarabaeidae-optimized electromagnetic drive micromirror according to claim 1, characterized in that: The scarab optimization algorithm in step three includes at least: rolling ball behavior, dancing behavior, breeding behavior, foraging behavior and stealing behavior.
7. The control method of the event-triggered May Beetle-optimized electromagnetic drive micromirror according to claim 6, characterized in that: The initialization population of the scarab optimization algorithm uses chaotic mapping, and the initialization individual can better cover the entire solution space; The scarab optimization algorithm updates the search boundary and strategy of rolling ball, breeding and foraging behaviors, introduces a nonlinear convergence factor or adaptive weight to replace the original linear update strategy, or integrates the search mechanism of other algorithms, so as to better coordinate the global search and local development ability.