An intelligent actuator adaptive control method and system based on internet of things

Abstract

The application relates to an intelligent actuator adaptive control method and system based on the Internet of Things, and belongs to the technical field of automation and intelligent control. The method comprises the following steps: an intelligent actuator designed by mechatronics integrates an actuating mechanism driven by a servo motor, a multi-modal sensor group, a microprocessor and a communication unit; execution parameters are collected in real time and fed back to the microprocessor; operation state parameters and control instructions are encrypted and transmitted to a cloud server through the communication unit; the cloud server analyzes system logs through a mode clustering algorithm, identifies abnormal working conditions deviating from normal operation sequences, and transmits detection results to a terminal application; the terminal application receives multi-modal instruction input through a touch panel, maps the input into operation instructions based on a rule engine, generates a control strategy, and uploads the control strategy to the cloud server. The adaptive control of the intelligent actuator is realized, the working efficiency and precision of the actuator are improved, and remote monitoring and intelligent management are realized.

Description

Technical Field

[0001] This invention belongs to the field of automation and intelligent control technology, specifically relating to an adaptive control method and system for intelligent actuators based on the Internet of Things. Background Technology

[0002] Traditional actuators rely on local controllers based on PLCs or microcontrollers, using manual adjustments or preset parameters for control. This approach leads to the following problems:

[0003] Lack of remote monitoring: Operational data (such as location, temperature, current, etc.) cannot be obtained in real time, requiring on-site inspections, resulting in significant response delays;

[0004] Data silo effect: The lack of communication interfaces between local control systems and upper-level management platforms (such as insufficient support for Modbus and Profibus protocols) creates information gaps;

[0005] Limited flexibility: Adjusting parameters requires physical contact with the equipment, making it unable to adapt to dynamic operating conditions.

[0006] Traditional methods simplify the actuator into a linear system (amplification stage + integration stage), but actual parameters (such as friction coefficient and load inertia) change significantly during operation. Adjusting PID parameters through trial and error is time-consuming and difficult to handle sudden environmental changes. Under complex operating conditions, the system exhibits strong nonlinearity (such as valve core vibration) and coupling characteristics (such as multi-degree-of-freedom robotic arms), rendering traditional linear control strategies ineffective. Furthermore, early processors (such as 8-bit microcontrollers) struggle to run advanced algorithms (such as model predictive control), limiting real-time performance. Mechanical transmission backlash and motor inertia cause command response lag, exacerbating overshoot and oscillation.

[0007] Existing actuators mostly use analog signals (4-20mA) or a single bus protocol (such as RS485), making it difficult to seamlessly connect with industrial internet platforms (supporting OPC UA and MQTT); high-precision sensor data (such as 1 / 250 resolution positioning signals) are sensitive to network latency, and local devices lack data cleaning and feature extraction capabilities, requiring cloud processing, which increases response latency; while intelligent algorithms require a large amount of historical data for training, but traditional systems lack data acquisition mechanisms, and fixed-parameter PID cannot adapt to sudden load changes. Summary of the Invention

[0008] To address the aforementioned problems in the existing technology, this invention provides an adaptive control method and system for intelligent actuators based on the Internet of Things. The objective of this invention can be achieved through the following technical solutions:

[0009] An adaptive control method for intelligent actuators based on the Internet of Things (IoT) includes:

[0010] S1: The intelligent actuator, designed through mechatronics, integrates a servo motor-driven actuator, a multimodal sensor group, a microprocessor, and a communication unit; the multimodal sensor group collects execution parameters in real time and feeds them back to the microprocessor, which generates control commands based on the motor servo control model;

[0011] The motor servo control model identifies electromagnetic torque and mechanical motion parameters offline using the recursive least squares method, and uses model reference adaptive control to achieve online parameter dynamic updates.

[0012] S2: The operating status parameters and control commands are encrypted and transmitted to the cloud server through the communication unit; the cloud server uses a time series database to store data, analyzes system logs through a pattern clustering algorithm, identifies abnormal operating conditions that deviate from the normal operating sequence, and transmits the detection results to the terminal application;

[0013] S3: The terminal application receives multimodal command input through the touch panel, maps the input into operation commands based on the rule engine and generates control strategies, which are then uploaded to the cloud server. The cloud server integrates real-time status parameters and historical operating data, optimizes the control strategy through the fusion of dynamic feedforward compensation and feedback correction compensation, and sends it to the microprocessor via the MQTT protocol.

[0014] S4: The microprocessor generates PWM modulation instructions according to the optimized control strategy to adjust the torque output characteristics of the servo motor. The feedback correction compensation amount is generated by the extended state observer to generate the optimal compensation voltage to shorten the load step response time.

[0015] Specifically, the multimodal sensor group described in S1 includes:

[0016] A high-precision encoder provides real-time feedback of speed and position signals; a thermocouple array detects the temperature rise of the motor and transmission components; the signals are amplified and filtered before being input into a microprocessor.

[0017] Specifically, the construction of the motor servo control model includes:

[0018] Construct a multi-source disturbance term, including a load mutation disturbance term and a parameter mismatch disturbance term;

[0019] A dynamic compensation strategy is adopted, which dynamically adjusts the proportional / integral coefficients by fusing incremental PID and fuzzy kernel;

[0020] The execution of the dynamic compensation strategy satisfies the following:

[0021] The feedforward compensation amount is adjusted in real time by the control strategy issued by the cloud server to pre-compensate for known disturbances;

[0022] The feedback correction amount dynamically adjusts the uncompensated disturbance based on real-time feedback of the execution parameters.

[0023] Specifically, the generation of the feedback correction amount is as follows:

[0024] By extending the state observer, the unmodeled dynamics are treated as a summation disturbance, and a state-space model is established to estimate the disturbance value in real time.

[0025] Based on the disturbance estimate, the optimal compensation voltage sequence is generated by model predictive control rolling optimization. The multi-source disturbance terms specifically include load change disturbance terms and parameter mismatch disturbance terms.

[0026] Specifically, the abnormal operating condition detection described in S2 includes:

[0027] The sliding window algorithm is used to segment and align real-time data according to time series.

[0028] When there is a time lag between real-time data and historical operating condition data, the missing points are filled in by interpolation algorithm;

[0029] Input the multi-dimensional dataset into the LSTM network to predict the perturbation type and trigger feedforward compensation.

[0030] Specifically, the execution of the rule engine described in S3 includes:

[0031] Multimodal command inputs are converted into structured data using a digital signal analyzer;

[0032] A three-tier rule base architecture is used for business logic management:

[0033] Basic rule base: stores parameter threshold rules and state machine-driven mode switching logic;

[0034] Scenario rule library: Binds rule combinations to specific working conditions and sets up a fast response mode;

[0035] Dynamic rule base: Enables online rule optimization and iteration through a dual-engine architecture.

[0036] Specifically, the optimization of the control strategy described in S3 includes:

[0037] The steps for generating the dynamic feedforward compensation amount include:

[0038] Compare the deviation between the actual output and the reference model output;

[0039] Based on the perturbation type predicted by the LSTM network, the feedforward gain matrix is ​​dynamically adjusted using the gradient descent method.

[0040] Steps for generating feedback correction compensation:

[0041] Based on the online parameter identification results, interpolation optimization is performed using a compensation lookup table;

[0042] The control strategy integration steps include:

[0043] Weights are assigned to the feedforward compensation and feedback correction compensation based on the operating condition confidence level.

[0044] The Lagrange multiplier method is used to ensure that the control sequence satisfies the physical constraints of the actuator.

[0045] Specifically, the weight allocation satisfies the following conditions: feedback correction compensation weight > 70% during steady-state operation, and feedforward compensation weight > 50% during load abrupt changes.

[0046] An IoT-based intelligent actuator adaptive control system includes:

[0047] Edge execution module: integrates servo motors, multimodal sensor groups and microprocessors for real-time data acquisition and local control command generation;

[0048] Cloud-based collaboration module: Includes a time-series database and anomaly detection unit, which aligns historical operating condition data with real-time data through pattern clustering algorithms;

[0049] Strategy optimization module: Deployed on a cloud server, it is configured with a dynamic feedforward compensator and an extended state observer, and outputs a control strategy with fused compensation.

[0050] Human-computer interaction module: includes a touch panel and a rule engine, which maps multimodal inputs into operation commands;

[0051] Among them, the edge execution module and the cloud collaboration module communicate through the encrypted MQTT protocol, and the output of the strategy optimization module directly drives the microprocessor to adjust the PWM modulation instructions.

[0052] Specifically, the microprocessor runs a real-time operating system and supports a fusion control algorithm of incremental PID and fuzzy kernel;

[0053] The extended state observer is coupled with model predictive control, and the generated optimal compensation voltage sequence is directly input into the motor drive circuit.

[0054] The beneficial effects of this invention are as follows:

[0055] The IoT-based intelligent actuator adaptive control method provided by this invention achieves precise control of the actuator, significantly improving system stability and response speed. First, the integration of a multi-modal sensor group enables the system to collect and feedback execution parameters in real time, providing a reliable data foundation for real-time analysis by the microprocessor. Second, the introduction of a cloud server not only enables remote data storage and abnormal condition detection but also allows for optimization of control strategies based on real-time status parameters, further enhancing the system's intelligence. Finally, through PWM modulation and adjustment of the servo motor's torque output characteristics, precise control of the actuator's operating state is achieved, meeting control requirements under various complex operating conditions. Attached Figure Description

[0056] To facilitate understanding by those skilled in the art, the present invention will be further described below with reference to the accompanying drawings.

[0057] Figure 1 This is a schematic diagram of the structure of an IoT-based intelligent actuator adaptive control system according to the present invention. Detailed Implementation

[0058] To further illustrate the technical means and effects of the present invention in achieving its intended purpose, the following detailed description of the specific implementation methods, structures, features, and effects of the present invention, in conjunction with the accompanying drawings and preferred embodiments, is provided.

[0059] Please see Figure 1 An adaptive control method for intelligent actuators based on the Internet of Things includes:

[0060] S1: The intelligent actuator, designed through mechatronics, integrates a servo motor-driven actuator, a multimodal sensor group, a microprocessor, and a communication unit; the multimodal sensor group collects execution parameters in real time and feeds them back to the microprocessor, which generates control commands based on the motor servo control model;

[0061] The motor servo control model identifies electromagnetic torque and mechanical motion parameters offline using the recursive least squares method, and uses model reference adaptive control to achieve online parameter dynamic updates.

[0062] S2: The operating status parameters and control commands are encrypted and transmitted to the cloud server through the communication unit; the cloud server uses a time series database to store data, analyzes system logs through a pattern clustering algorithm, identifies abnormal operating conditions that deviate from the normal operating sequence, and transmits the detection results to the terminal application;

[0063] S3: The terminal application receives multimodal command input through the touch panel, maps the input into operation commands based on the rule engine and generates control strategies, which are then uploaded to the cloud server. The cloud server integrates real-time status parameters and historical operating data, optimizes the control strategy through the fusion of dynamic feedforward compensation and feedback correction compensation, and sends it to the microprocessor via the MQTT protocol.

[0064] S4: The microprocessor generates PWM modulation instructions according to the optimized control strategy to adjust the torque output characteristics of the servo motor. The feedback correction compensation amount is generated by the extended state observer to generate the optimal compensation voltage to shorten the load step response time.

[0065] Specifically, the multimodal sensor group in S1 includes a high-precision encoder that provides real-time feedback of rotational speed and position, and an integrated thermocouple for detecting motor overheating or abnormal temperature rise in transmission components. The signal is amplified and filtered before being input into the microprocessor.

[0066] In this embodiment, the servo motor drive uses a high-precision electrical actuator, which has the characteristics of fast dynamic response (microsecond-level adjustment) and adjustable torque output. The multi-modal sensor network integrates a high-precision encoder and a thermocouple array to realize the acquisition of multi-dimensional parameters such as position, speed, and temperature. The microprocessor runs a real-time operating system (RTOS) to support the fusion control of PID algorithm and load compensation.

[0067] Specifically, the motor servo control model in S1 identifies parameters offline based on electromagnetic torque and mechanical motion using the recursive least squares method, and achieves online parameter dynamic updates by combining model reference adaptive control; it also constructs multi-source disturbance terms; eliminates the multi-source disturbance terms through the dynamic compensation strategy, and dynamically adjusts the proportional / integral coefficients by fusing incremental PID and fuzzy kernel to eliminate steady-state errors.

[0068] In this embodiment, based on the static characteristics of the motor, the electromagnetic torque equation is simplified, retaining only core parameters such as stator resistance, rotor resistance, and stator / rotor inductance to reduce computational complexity. A pseudo-random M-sequence (such as generated by a linear feedback shift register) is used as the excitation signal, and the dynamic response of the motor is excited by a step voltage injection method to avoid voltage errors caused by dead time.

[0069] The offline identification objective function is:

[0070] ,

[0071] in, R s For stator resistance, L s For stator inductance, R r L is the rotor resistance. r For rotor inductance; Given the input voltage, current and other observed data vectors, the Butterworth filter equation is solved by improving the Euler numerical solution method to directly obtain the signal derivative information and reduce discretization error.

[0072] The online parameter dynamic update method is as follows: an ideal motor model is constructed based on the offline identification results, and a reference speed is output. The parameters of the actual motor model are updated online through RLS, and the actual speed is output. The parameters are dynamically adjusted according to the error between the reference speed and the actual speed.

[0073] Through incremental PID dynamic parameter tuning, the discretized equation is:

[0074] ,

[0075] The conditional integration method is adopted, and the integration term is frozen when the error exceeds the threshold to prevent actuator saturation.

[0076] Specifically, the multi-source disturbance terms include load mutation disturbance terms and parameter mismatch disturbance terms.

[0077] Specifically, the dynamic compensation strategy adopts a combination of feedforward compensation and feedback correction based on the characteristics of the multi-source disturbance terms. The feedforward compensation part is adjusted in real time through the control strategy sent by the cloud server received by the microprocessor to pre-compensate known disturbance terms. The feedback correction part dynamically adjusts the compensation amount for incompletely compensated disturbance terms based on the real-time feedback of the execution parameters.

[0078] Specifically, the compensation amount of the feedback correction is optimized by the rolling optimization of the extended state observer to generate the optimal compensation voltage, thereby shortening the load step response time.

[0079] Specifically, the abnormal operating condition detection method in S2 clusters the system logs by pattern to identify abnormal operating states that deviate from the normal operating sequence.

[0080] Specifically, the terminal application obtains multimodal command input through the touch panel parameter setting interface, including manually inputting control parameters and selecting a preset control mode, and obtains operation commands by performing condition-action mapping on the multimodal command input based on the rule engine;

[0081] The rule engine uses a rule base to store predefined business logic rules. It uses a digital signal parser to convert the multimodal instruction input into structured data. The business logic rules include threshold rules for manually input control parameters and switching logic for preset control modes. The engine matches the business logic rules in the rule base with the structured data and converts the matched rules into specific control strategies.

[0082] In this embodiment, a three-level rule base architecture is adopted to realize the modular management of business logic. The basic rule base is used to store core business logic rules, define parameter boundary conditions through threshold rules, and set mode switching rules based on state machine condition judgment. The scenario rule base is for rule combinations for specific working conditions and sets a fast response mode. The dynamic rule base realizes online optimization of rules and realizes rule iteration through a dual-engine architecture.

[0083] Specifically, the optimization method of the cloud server in S3, which combines the real-time status parameters of the actuator to optimize the control strategy, is as follows:

[0084] The cloud server integrates the real-time status parameters of the actuators with historical operating data, aligns the real-time data of the multimodal sensor group with the historical operating condition data in the cloud time series database, and dynamically adjusts the feedforward compensation amount by comparing the difference between the actual output and the reference model output to generate a dynamic feedforward compensation amount. The cloud server uses a compensation lookup table and combines the online parameter identification results to perform interpolation optimization to obtain the feedback correction compensation amount. The control strategy sent by the terminal application is fused with the feedforward compensation amount and the feedback correction compensation amount to obtain the optimal control quantity sequence, which is then sent to the microprocessor at the edge via the MQTT protocol.

[0085] In this embodiment, the cloud server uses timestamp synchronization technology to time-align the real-time data of the multimodal sensor group (encoder, thermocouple, etc.) with the historical operating condition data in the time-series database. A sliding window algorithm is used to segment the real-time data, ensuring consistency in dimensions such as sampling frequency and timestamp offset. For example, when there is a time lag between real-time rotational speed data and historical temperature data, an interpolation algorithm is used to fill in the missing points, forming a time-aligned multi-dimensional dataset.

[0086] Dynamic feedforward compensation is achieved based on the following steps:

[0087] Reference model comparison: The cloud server has a built-in ideal motor model (electromagnetic torque equation identified offline) that compares the actual output (speed, torque) with the output of the reference model in real time and calculates the deviation.

[0088] Disturbance prediction: Combining historical operating data (load mutation pattern library), the LSTM network is used to predict the types of disturbances that may occur in the future; if historical data shows that there is a sudden temperature rise before a certain load mutation, feedforward compensation is triggered in advance.

[0089] Compensation Quantity Generation: Based on the deviation and predicted disturbance, the feedforward gain matrix is ​​dynamically adjusted using the gradient descent method to generate a feedforward compensation quantity adapted to the current operating conditions. This process supports multivariate coupled compensation, such as simultaneously compensating for torque fluctuations and mechanical clearance changes caused by thermal expansion.

[0090] Interpolation optimization for feedback correction compensation:

[0091] By extending the state observer, unmodeled dynamics (such as sudden changes in drive shaft friction) are treated as "total disturbances", and the disturbance values ​​are estimated in real time through the state-space model;

[0092] Based on disturbance estimates, model predictive control (MPC) is used to roll-optimize the compensation voltage sequence, thereby shortening the load step response time. For example, during motor acceleration, the ESO predicts the risk of current overshoot and injects reverse compensation voltage in advance to suppress oscillation.

[0093] In this embodiment, the cloud server integrates the original control strategy (speed setpoint), dynamic feedforward compensation, and feedback correction compensation issued by the terminal application:

[0094] Weights are assigned based on the confidence level of different operating conditions. During steady-state operation, feedback correction is the primary method (weight > 70%), while feedforward compensation is increased to over 50% during load abrupt changes. The Lagrange multiplier method is used to ensure that the fused control sequence meets the physical limitations of the actuator (maximum current, temperature threshold). In the offline stage, the motor is excited by a pseudo-random M-sequence to identify core parameters. In the online stage, model reference adaptive control (MRAC) is used to dynamically update parameters to avoid parameter drift caused by temperature rise.

[0095] An IoT-based intelligent actuator adaptive control system includes:

[0096] Edge execution module: integrates servo motors, multimodal sensor groups and microprocessors for real-time data acquisition and local control command generation;

[0097] Cloud-based collaboration module: Includes a time-series database and anomaly detection unit, which aligns historical operating condition data with real-time data through pattern clustering algorithms;

[0098] Strategy optimization module: Deployed on a cloud server, it is configured with a dynamic feedforward compensator and an extended state observer, and outputs a control strategy with fused compensation.

[0099] Human-computer interaction module: includes a touch panel and a rule engine, which maps multimodal inputs into operation commands;

[0100] Among them, the edge execution module and the cloud collaboration module communicate through the encrypted MQTT protocol, and the output of the strategy optimization module directly drives the microprocessor to adjust the PWM modulation instructions.

[0101] Specifically, the microprocessor runs a real-time operating system and supports a fusion control algorithm of incremental PID and fuzzy kernel;

[0102] The extended state observer is coupled with model predictive control, and the generated optimal compensation voltage sequence is directly input into the motor drive circuit.

[0103] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any way. Although the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art can make some modifications or alterations to the above-disclosed technical content to create equivalent embodiments without departing from the scope of the present invention. Any simple modifications, equivalent changes and alterations made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the scope of the present invention.

Claims

1. An adaptive control method for intelligent actuators based on the Internet of Things, characterized in that, include: S1: The intelligent actuator, designed through mechatronics, integrates a servo motor-driven actuator, a multimodal sensor group, a microprocessor, and a communication unit; the multimodal sensor group collects execution parameters in real time and feeds them back to the microprocessor, which generates control commands based on the motor servo control model; The motor servo control model identifies electromagnetic torque and mechanical motion parameters offline using the recursive least squares method, and uses model reference adaptive control to achieve online parameter dynamic updates. S2: The operating status parameters and control commands are encrypted and transmitted to the cloud server through the communication unit; the cloud server uses a time series database to store data, analyzes system logs through a pattern clustering algorithm, identifies abnormal operating conditions that deviate from the normal operating sequence, and transmits the detection results to the terminal application; S3: The terminal application receives multimodal command input through the touch panel, maps the input into operation commands based on the rule engine and generates control strategies, which are then uploaded to the cloud server. The cloud server integrates real-time status parameters and historical operating data, optimizes the control strategy through the fusion of dynamic feedforward compensation and feedback correction compensation, and sends it to the microprocessor via the MQTT protocol. S4: The microprocessor generates PWM modulation instructions according to the optimized control strategy to adjust the torque output characteristics of the servo motor. The feedback correction compensation amount is generated by the extended state observer to generate the optimal compensation voltage to shorten the load step response time.

2. The method according to claim 1, characterized in that, The multimodal sensor group described in S1 includes: A high-precision encoder provides real-time feedback of speed and position signals; a thermocouple array detects the temperature rise of the motor and transmission components; the signals are amplified and filtered before being input into a microprocessor.

3. The method according to claim 1, characterized in that, The construction of the motor servo control model includes: Construct a multi-source disturbance term, including a load mutation disturbance term and a parameter mismatch disturbance term; A dynamic compensation strategy is adopted, which dynamically adjusts the proportional / integral coefficients by fusing incremental PID and fuzzy kernel; The execution of the dynamic compensation strategy satisfies the following: The feedforward compensation amount is adjusted in real time by the control strategy issued by the cloud server to pre-compensate for known disturbances; The feedback correction amount dynamically adjusts the uncompensated disturbance based on real-time feedback of the execution parameters.

4. The method according to claim 3, characterized in that, The specific method for generating the feedback correction amount is as follows: By extending the state observer, the unmodeled dynamics are treated as a summation disturbance, and a state-space model is established to estimate the disturbance value in real time. Based on the disturbance estimate, the optimal compensation voltage sequence is generated by model predictive control rolling optimization. The multi-source disturbance terms specifically include load change disturbance terms and parameter mismatch disturbance terms.

5. The method according to claim 3, characterized in that, The abnormal operating condition detection described in S2 includes: The sliding window algorithm is used to segment and align real-time data according to time series. When there is a time lag between real-time data and historical operating condition data, the missing points are filled in by interpolation algorithm; Input the multi-dimensional dataset into the LSTM network to predict the perturbation type and trigger feedforward compensation.

6. The method according to claim 3, characterized in that, The execution of the rule engine described in S3 includes: Multimodal command inputs are converted into structured data using a digital signal analyzer; A three-tier rule base architecture is used for business logic management: Basic rule base: stores parameter threshold rules and state machine-driven mode switching logic; Scenario rule library: Binds rule combinations to specific working conditions and sets up a fast response mode; Dynamic rule base: Enables online rule optimization and iteration through a dual-engine architecture.

7. The method according to claim 1, characterized in that, The optimization of the control strategy described in S3 includes: The steps for generating the dynamic feedforward compensation amount include: Compare the deviation between the actual output and the reference model output; Based on the perturbation type predicted by the LSTM network, the feedforward gain matrix is ​​dynamically adjusted using the gradient descent method. Steps for generating feedback correction compensation: Based on the online parameter identification results, interpolation optimization is performed using a compensation lookup table; The control strategy integration steps include: Weights are assigned to the feedforward compensation and feedback correction compensation based on the operating condition confidence level. The Lagrange multiplier method is used to ensure that the control sequence satisfies the physical constraints of the actuator.

8. The method according to claim 7, characterized in that, The weight allocation satisfies the following conditions: feedback correction compensation weight > 70% during steady-state operation, and feedforward compensation weight > 50% during load abrupt changes.

9. An Internet of Things-based intelligent actuator adaptive control system for executing the method described in any one of claims 1-8, characterized in that, include: Edge execution module: integrates servo motors, multimodal sensor groups and microprocessors for real-time data acquisition and local control command generation; Cloud-based collaboration module: Includes a time-series database and anomaly detection unit, which aligns historical operating condition data with real-time data through pattern clustering algorithms; Strategy optimization module: Deployed on a cloud server, it is configured with a dynamic feedforward compensator and an extended state observer, and outputs a control strategy with fused compensation. Human-computer interaction module: includes a touch panel and a rule engine, which maps multimodal inputs into operation commands; Among them, the edge execution module and the cloud collaboration module communicate through the encrypted MQTT protocol, and the output of the strategy optimization module directly drives the microprocessor to adjust the PWM modulation instructions.

10. The system according to claim 9, characterized in that, The microprocessor runs a real-time operating system and supports a fusion control algorithm of incremental PID and fuzzy kernel. The extended state observer is coupled with model predictive control, and the generated optimal compensation voltage sequence is directly input into the motor drive circuit.

Images