Model system for self-closed-loop optimization based on embedded algorithms
By adopting a self-closed-loop optimization model system based on embedded algorithms in the control system, the problem of insufficient adaptability and accuracy of traditional linear control systems in complex environments is solved, and the system's high adaptability and robustness is achieved, cost and risk of relying on professionals is reduced, and the level of production intelligence is improved.
Patent Information
- Application Number
- CN202411650932.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-11-19
- Publication Date
- 2025-04-04
AI Technical Summary
In complex and changeable working environments, traditional linear control systems are difficult to improve adaptability and accuracy, and common improvement methods increase system cost and design difficulty.
A self-closed-loop optimization model system based on embedded algorithms is adopted. The system includes a main control platform, an actuator and a position sensor. The main control platform consists of an embedded hardware platform, a data acquisition module, a central processing unit and a storage module. It has a built-in embedded PID controller and an automatic optimization module to support a deep reinforcement learning framework.
Through the combination of embedded microprocessor and PID control algorithm, the system significantly improves the adaptability and robustness of the system, can maintain good control effects under uncertain operating conditions, reduce the dependence on professional experience, control the overall cost, and improve the automation and intelligence of the production process.
Smart Images
Figure CN119758799A_ABST
Abstract
Description
Technical Field
[0001] The invention belongs to the technical field of automatic control, and specifically refers to a self-closed-loop optimization model system based on an embedded algorithm. Background Art
[0002] Smart manufacturing has become an important trend in the transformation of the global manufacturing industry. As one of the core technologies of smart manufacturing, automatic control technology has developed rapidly in recent years. In particular, linear control systems are widely used in various precision control situations due to their simple structure and easy implementation. Traditional linear control systems rely on preset paths for control, lack adaptability and flexibility, and are difficult to achieve ideal results in complex and changing working environments.
[0003] At present, in order to improve the adaptability and accuracy of control systems, common practices include: introducing advanced sensor technology to improve control performance by increasing external feedback. The introduction of additional hardware (such as advanced sensors) not only increases the cost of the system, but may also cause the entire system to fail due to hardware failure; applying more complex nonlinear control strategies, such as sliding mode control and fuzzy logic control. Although these methods can improve the control effect to a certain extent, they also increase the design difficulty and cost of the system. Secondly, although nonlinear control strategies can theoretically provide better control effects, due to the complexity of the algorithm itself, the actual control effect is often not as expected; the proportional-integral-derivative (PID) controller based on empirical design can effectively handle certain dynamically changing situations, but it is still difficult to achieve optimal control in complex environments. Summary of the invention
[0004] The technical problem to be solved by the present invention is how to improve the adaptability and accuracy of the control system when facing a complex and changeable working environment.
[0005] To solve the above problems, the technical solution adopted by the present invention is as follows: the model system of self-closed-loop optimization based on embedded algorithm proposed in the present invention includes a main control platform, an actuator and a position sensor, the main control platform is electrically connected to the actuator and the position sensor respectively, the main control platform includes an embedded hardware platform, a data acquisition module, a central processing unit (CPU) and a storage module, the embedded hardware platform is electrically connected to the actuator, the embedded hardware platform is composed of a main controller, an embedded PID controller is arranged in the main controller, the data acquisition module is electrically connected to the position sensor and the embedded hardware platform respectively, and the storage module and the embedded hardware platform are electrically connected to the central processing unit (CPU).
[0006] Furthermore, the main controller is an ARM Cortex-M series microcontroller, and the embedded PID controller is an STM32 series MCU single-chip microcomputer.
[0007] Furthermore, the embedded PID controller calculates the control quantity u(t)=Kpδ(t)+Ki∫δ(τ)dτ+Kddδ(t) / dt applied to the actuator at the next moment according to the PID formula, wherein δ(t) is the deviation between the current measured value of the actuator and the reference value, Kp is the proportional gain, Ki is the integral gain, and Kd is the differential gain.
[0008] Furthermore, the data acquisition module includes a sensor and a data interface. The sensor is a high-precision MEMS sensor. The data interface is designed to be compatible with multiple standard protocols including SPI / I2C, so as to access different types of sensors.
[0009] Furthermore, the central processing unit (CPU) is configured with a main frequency of ≥1 GHz and is responsible for running the core algorithm and coordinating the work of various components.
[0010] Furthermore, the storage module uses a Flash memory with a capacity of ≥8MB to store model parameters, historical data and configuration files.
[0011] Furthermore, the main control platform also includes a communication module, which supports multiple wireless transmission modes such as Wi-Fi, BLE and Zigbee, and supports data exchange with other devices or cloud platforms.
[0012] Furthermore, the main control platform also includes an automatic optimization module, which integrates an embedded algorithm, is developed based on a deep reinforcement learning framework, and dynamically adjusts the model structure and hyperparameters according to different application scenarios.
[0013] Furthermore, the communication module and the automatic optimization module are both connected to a central processing unit (CPU) via electrical signals.
[0014] Furthermore, the actuator selects different types of motors (DC motor, stepper motor) according to specific application scenarios, and the position sensor is used to monitor the actual displacement of the target object in real time and feed back information to the main controller.
[0015] The beneficial effects achieved by the present invention using the above structure are as follows:
[0016] 1. The self-closed-loop optimization model system based on embedded algorithm proposed in this scheme greatly improves the adaptability and robustness of the system through the combination of embedded microprocessor and PID control algorithm, and can maintain good control effect under various uncertain working conditions.
[0017] 2. The self-closed-loop optimization model system based on embedded algorithms proposed in this solution adopts a design concept that combines software and hardware. It can achieve ideal performance improvement without adding a large number of expensive peripherals, effectively controlling the overall cost.
[0018] 3. The self-closed-loop optimization model system based on embedded algorithms proposed in this solution reduces the reliance on the experience of professionals, reduces the probability of errors caused by human factors, and also saves a lot of time and human resources.
[0019] 4. The self-closed-loop optimization model system based on embedded algorithms proposed in this solution improves the automation and intelligence level of the entire production process, and helps promote technological innovation and development in related industries. BRIEF DESCRIPTION OF THE DRAWINGS
[0020] Figure 1 It is a schematic diagram of the overall system of the present invention;
[0021] Figure 2 It is a schematic diagram of the overall operation flow of the present invention.
[0022] The accompanying drawings are used to provide further understanding of the present invention and constitute a part of the specification. They are used to explain the present invention together with the embodiments of the present invention and do not constitute a limitation of the present invention. DETAILED DESCRIPTION
[0023] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments; based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.
[0024] like Figure 1 As shown, the model system of self-closed-loop optimization based on embedded algorithm proposed in the present invention includes a main control platform, an actuator and a position sensor. The main control platform is electrically connected to the actuator and the position sensor respectively. The actuator selects different types of motors (DC motor, stepper motor) according to specific application scenarios. The position sensor is used to monitor the actual displacement of the target object in real time and feed back information to the main controller.
[0025] The main control platform includes an embedded hardware platform, a data acquisition module, a central processing unit (CPU), a storage module, a communication module and an automatic optimization module. The embedded hardware platform is electrically connected to the actuator. The embedded hardware platform is composed of a main controller, which is equipped with an embedded PID controller. The data acquisition module is electrically connected to the position sensor and the embedded hardware platform respectively. The storage module, the embedded hardware platform, the communication module and the automatic optimization module are all electrically connected to the central processing unit (CPU).
[0026] The main controller uses an ARM Cortex-M series microcontroller with the characteristics of low power consumption and high performance. The embedded PID controller uses an STM32 series MCU microcontroller. The embedded PID controller calculates the control quantity u(t)=Kpδ(t)+Ki∫δ(τ)dτ+Kddδ(t) / dt applied to the actuator at the next moment according to the PID formula, where δ(t) is the deviation between the current measured value of the actuator and the reference value, Kp is the proportional gain, Ki is the integral gain, and Kd is the differential gain.
[0027] The data acquisition module includes sensors and data interfaces. The sensors are high-precision MEMS sensors to ensure the accuracy and stability of the data. The data interface is designed to be compatible with multiple standard protocols including SPI / I2C to facilitate access to different types of sensors for collecting environmental information and system status.
[0028] The configuration frequency of the central processing unit (CPU) is ≥1GHz to ensure the efficient execution of the algorithm. It is responsible for running the core algorithm and coordinating the work of various components. The storage module uses Flash memory with a capacity of ≥8MB to meet the storage requirements of large amounts of data and save model parameters, historical data and configuration files.
[0029] The communication module supports multiple wireless transmission methods such as Wi-Fi, BLE and Zigbee, which allows for convenient and flexible interaction with the outside world and data exchange with other devices or cloud platforms. The automatic optimization module integrates embedded algorithms and is developed based on a deep reinforcement learning framework. It can dynamically adjust the model structure and hyperparameters according to different application scenarios.
[0030] like Figure 2 As shown, the specific workflow is:
[0031] S1: After the system is started, the data acquisition module begins to monitor the surrounding environment and its own status, generates raw data streams, and the central processing unit (CPU) coordinates the work of the embedded hardware platform and the data acquisition module;
[0032] S2: The embedded hardware platform receives data collected by the data acquisition module, and calculates the control amount to be applied to the actuator at the next moment according to the PID formula through the embedded PID controller;
[0033] S3: The embedded PID controller outputs the parameters to the actuator to control the operation of the actuator;
[0034] S4: The data acquisition module collects the position sensor change data and inputs it into the embedded hardware platform;
[0035] S5: Calculate and optimize the control quantity on the actuator through the embedded PID controller, and repeat the self-closed-loop iterative optimization operation;
[0036] S6: The storage module saves model parameters and historical data for recording and adjustment;
[0037] S7: The communication module uploads important information to the cloud for backup and receives new instructions or data from the outside.
[0038] The present invention and its embodiments are described above, and such description is not restrictive. The drawings show only one embodiment of the present invention, and the actual structure is not limited thereto. In short, if ordinary technicians in the field are inspired by it, without departing from the purpose of the invention, they can design a structure and embodiment similar to the technical solution without creativity, which should belong to the protection scope of the present invention.
Claims
1. A model system based on self-closed-loop optimization of embedded algorithms, characterized by: It includes a main control platform, an actuator and a position sensor. The main control platform is electrically connected to the actuator and the position sensor respectively. The main control platform includes an embedded hardware platform, a data acquisition module, a central processing unit (CPU) and a storage module. The embedded hardware platform is electrically connected to the actuator. The embedded hardware platform is composed of a main controller. An embedded PID controller is arranged in the main controller. The data acquisition module is electrically connected to the position sensor and the embedded hardware platform respectively. The storage module and the embedded hardware platform are electrically connected to the central processing unit (CPU).
2. The model system for self-closed-loop optimization based on embedded algorithm according to claim 1, characterized in that: The main controller uses an ARM Cortex-M series microcontroller, and the embedded PID controller uses an STM32 series MCU single-chip microcomputer.
3. The model system for self-closed-loop optimization based on embedded algorithm according to claim 2, characterized in that: The embedded PID controller calculates the control quantity u(t)=Kpδ(t)+Ki∫δ(τ)dτ+Kddδ(t) / dt applied to the actuator at the next moment according to the PID formula, where δ(t) is the deviation between the current measured value of the actuator and the reference value, Kp is the proportional gain, Ki is the integral gain, and Kd is the differential gain.
4. The model system for self-closed-loop optimization based on embedded algorithm according to claim 3, characterized in that: The data acquisition module includes a sensor and a data interface. The sensor is a high-precision MEMS sensor. The data interface is designed to be compatible with multiple standard protocols including SPI / I2C, so as to access different types of sensors.
5. The model system for self-closed-loop optimization based on embedded algorithm according to claim 4, characterized in that: The central processing unit (CPU) is configured with a main frequency of ≥1 GHz and is responsible for running the core algorithm and coordinating the work of various components.
6. The model system for self-closed-loop optimization based on embedded algorithm according to claim 5, characterized in that: The storage module uses a Flash memory with a capacity of ≥8MB to store model parameters, historical data and configuration files.
7. The model system for self-closed-loop optimization based on embedded algorithm according to claim 6, characterized in that: The main control platform also includes a communication module, which supports multiple wireless transmission modes such as Wi-Fi, BLE and Zigbee, and supports data exchange with other devices or cloud platforms.
8. The model system for self-closed-loop optimization based on embedded algorithm according to claim 7, characterized in that: The main control platform also includes an automatic optimization module, which integrates an embedded algorithm, is developed based on a deep reinforcement learning framework, and dynamically adjusts the model structure and hyperparameters according to different application scenarios.
9. The model system for self-closed-loop optimization based on embedded algorithm according to claim 8, characterized in that: The communication module and the automatic optimization module are both connected to the central processing unit (CPU) via electrical signals.
10. The model system for self-closed-loop optimization based on embedded algorithm according to claim 9, characterized in that: The actuator selects different types of motors (DC motor, stepper motor) according to specific application scenarios, and the position sensor is used to monitor the actual displacement of the target object in real time and feed back information to the main controller.