Permanent magnet synchronous motor current control method and system based on model predictive control

Abstract

The invention provides a permanent magnet synchronous motor current control method and system based on model predictive control, and aims to improve the dynamic response speed and control precision of a motor through an advanced control strategy. According to the method, a model prediction control technology is utilized, a mathematical model of the motor is constructed, and an optimization algorithm is designed, so that accurate prediction and control of the current of the motor are realized. An intelligent algorithm, such as a neural network or fuzzy logic, is particularly introduced to enhance the adaptability and robustness of a control strategy. Compared with a traditional control method, the method has the advantages that the calculation burden is reduced, the real-time performance of control is ensured, and the performance of the system facing parameter change and external disturbance is remarkably improved. The innovative control method is suitable for industrial application occasions with extremely high requirements on dynamic performance and control precision, and has important practical application value and wide market prospect.

Description

Technical Field

[0001] The present invention relates to the technical field of motor control, and in particular to a permanent magnet synchronous motor current control method and system based on model predictive control theory. Background Art

[0002] With the rapid development of modern industrial automation and intelligent manufacturing, the requirements for motor control performance are getting higher and higher. Especially in high-performance servo systems, permanent magnet synchronous motors (PMSMs) are widely used due to their high efficiency, high power density and good dynamic response characteristics. However, although traditional PMSM control methods, such as vector control and direct torque control, have proven their effectiveness in many applications, they still face certain limitations when dealing with complex system nonlinearities, parameter changes and external disturbances.

[0003] In practical applications, the control accuracy and stability of PMSM are affected by many factors, including uncertainty in motor parameters, load disturbances, and system nonlinearity. These problems may lead to current control errors, affecting the performance and life of the motor. In addition, with the increasing requirements for motor response speed, traditional control methods have also shown shortcomings in terms of rapidity and adaptability. Therefore, developing a PMSM current control method that can provide faster dynamic response and higher control accuracy is crucial to improving the performance of the entire drive system.

[0004] Model predictive control (MPC), as an emerging control strategy, has attracted attention due to its advantages in dealing with multi-variable, multi-constraint and nonlinear problems. MPC can achieve comprehensive optimization of system performance by predicting future behavior and optimizing control input. However, when applying MPC to PMSM current control, challenges such as computational complexity, real-time performance and control accuracy need to be addressed. In addition, how to combine intelligent algorithms to improve the adaptability and robustness of MPC is also one of the current research hotspots.

[0005] In summary, the present invention aims to propose a permanent magnet synchronous motor current control method based on model predictive control, which comprehensively considers the dynamic characteristics and control requirements of the motor, adopts advanced optimization algorithms and intelligent control strategies, in order to achieve the purpose of improving the control performance of PMSM. Summary of the invention

[0006] The technical solution of the present invention is a permanent magnet synchronous motor current control method based on model predictive control theory, which realizes precise control of the motor current through the following key steps:

[0007] Motor parameter identification ensures the accuracy and adaptability of the control strategy. First, the identification of stator resistance is achieved through the DC current injection method. This method measures the DC voltage and current of the stator winding when the motor is stationary, and directly calculates the resistance value by applying Ohm's law, providing basic parameters for motor control. The identification of stator inductance is carried out through the high-frequency sinusoidal signal injection method. This method estimates the inductance value by injecting a high-frequency AC signal into the stator winding of the motor and measuring the corresponding current response. This technology is particularly suitable for occasions where high accuracy of inductance parameters is required because it reduces the interference of other motor parameters on the measurement results. The identification of initial position and back electromotive force is also important for motor control. The initial position is usually determined indirectly by an encoder or through the zero crossing point of the back electromotive force, while the back electromotive force is obtained by measuring the open circuit voltage of the stator winding when the motor is unloaded and the speed is known, which directly reflects the electrical characteristics of the motor. Furthermore, the identification of flux linkage and rotor resistance requires the motor to operate under specific conditions. The flux linkage is usually estimated by measuring the magnetization curve of the permanent magnet and combining it with the motor design data, while the rotor resistance and leakage inductance are measured by current and voltage detection in periodic interrupts and comparison interrupts, providing key parameters for the control algorithm. The application of online identification methods, such as least squares method, recursive least squares method, extended Kalman filter, etc., can monitor and update the key parameters of the motor in real time during the operation of the motor, and adapt to actual working conditions such as temperature changes, aging effects and load fluctuations. In addition, the application of model predictive control (MPC) technology and neural network method has further improved the accuracy and efficiency of identification, especially when dealing with motor parameters with nonlinear characteristics.

[0008] The establishment of mathematical models is the core link in the field of motor control: based on the physical characteristics of the motor, such as the law of electromagnetic induction and the basic laws of circuits, voltage equations and flux equations are established. These equations describe the electrical behavior of the motor, including the relationship between stator current, voltage and flux. Through these relationships, the state space equations of the motor can be derived, which express the dynamic characteristics of the motor in the form of a set of first-order differential equations. Since the mathematical model of the three-phase motor has a strong coupling problem in the original coordinate system, it is necessary to simplify the model through coordinate transformation. The Clarke transformation (3s / 2s) converts the variables in the three-phase stationary coordinate system into a two-phase stationary coordinate system, while the Park transformation (2s / 2r) further converts the variables in the two-phase stationary coordinate system into a synchronous rotating coordinate system (dq axis), thereby achieving decoupling and making the design of the control strategy similar to that of a DC motor, simplifying the complexity of the control algorithm. Under the dq axis, the voltage equation can clearly express the relationship between current and voltage, that is, and where u d and u q are the voltages of the dd and q axes, i d and iq is the corresponding current, R s is the stator resistance, ω e is the electrical angular velocity, L d and L p is the inductance of the d-axis and q-axis, ψ d and ψ q is the magnetic flux, ψ f is the magnetic flux generated by the permanent magnet. Electromagnetic torque is the key to the motor output torque and is related to the dq axis components of the current. The torque equation is usually expressed as Where T e is the electromagnetic torque, and p is the number of pole pairs of the motor. In addition, the motor's speed and position are related to the electromagnetic torque through the mechanical equations of motion, which take into account the motor's rotational momentum, load torque, and stick-belt damping.

[0009] Control strategy generation involves the following aspects: Design a model predictive controller that is able to predict the system behavior for a period of time in the future based on the mathematical model of the motor. The controller takes into account the dynamic characteristics and operating conditions of the motor and calculates the optimal control input within a given time horizon through an optimization algorithm. Define an objective function, which usually includes the minimization of current tracking error and the smoothness of the control input. This function reflects the main indicators of control performance, such as fast response and reduced current fluctuations. Incorporate the physical limitations of motor operation in the control strategy, such as voltage limit, current limit, and motor speed range. These constraints ensure that the motor operates under safe and efficient conditions. Use a multi-objective optimization algorithm, such as the non-dominated sorting genetic algorithm II (NSGA-II), to balance different performance indicators. This algorithm is able to handle multiple objectives in motor control, such as smoothness of torque output and current tracking performance. For the multi-objective optimization problem of the motor rotor structure, a hierarchical optimization strategy is adopted to optimize the high-sensitivity parameters first, and then optimize the low-sensitivity parameters based on a new finite element model (FEM). This approach effectively reduces the computational dimension and saves computational time. The controller configuration module is designed as a real-time optimization system that adjusts the control input based on the real-time feedback from the motor to cope with load changes and system dynamics. Through finite element analysis (FEA) and other testing methods, the optimized controller performance is verified to ensure that it meets the design requirements and performs well in actual applications.

[0010] Rolling horizon optimization is an efficient real-time optimization strategy suitable for optimal control of dynamic systems, especially for fields that require fast response, such as autonomous driving, aerospace, and robotics. This strategy uses a distributed optimization algorithm with finite time convergence, combined with the Lyapunov method to ensure the convergence of the algorithm, and uses real-time optimal control tools such as SCPToolbox, using iterative linear quadratic regulators and enhanced Lagrangian methods to handle path constraints. At each sampling moment, the boundaries and objectives of the optimization problem are updated through predictive control to achieve performance optimization within a limited time range in the future. Compared with traditional global optimization, rolling optimization can timely correct various complex situations in the control process and improve signal quality and system reliability.

[0011] Control signal smoothing processes the control signal through a moving average filter to reduce high-frequency noise and optimize signal quality. A moving average filter of length N achieves smoothing by calculating the average value of every N sampling points of the signal; while the Savitzky-Golay filter estimates the smoothed value of each data point using a polynomial fit within a fixed-size window. The window length is usually selected to be 11 and the polynomial order is 2 to balance the smoothing effect and computational complexity. In addition, for signal features that do not want to be smoothed, such as the sharp edges of the clock signal, the median filter can retain these features while removing unnecessary high-frequency fluctuations. In practical applications, such as aircraft control systems, by using continuous or smooth control signals, such as SigmoidFunction, control jitter can be avoided to ensure the smoothness and reliability of system operation. Through these methods, control signal smoothing not only improves the performance and efficiency of the motor, but also enhances the stability and responsiveness of the system.

[0012] The intelligent algorithm adaptation module enhances the adaptability of the motor control system by integrating advanced intelligent algorithms: The intelligent algorithm adaptation module uses advanced intelligent algorithms such as neural networks, fuzzy logic and genetic algorithms to adaptively optimize the motor control strategy. These algorithms can dynamically adjust the control parameters according to the real-time data and environmental changes of the motor during operation, thereby improving the control accuracy and system stability. As a powerful pattern recognition tool, neural networks can model the dynamic behavior of motors by learning a large amount of input and output data. In the control strategy, neural networks can predict the response of the motor under specific operating conditions, thereby achieving precise adjustment of the control input. The self-learning ability of neural networks enables them to adapt to the slow changes of motor parameters. Fuzzy logic provides an effective way to deal with uncertainty and ambiguity problems. In motor control, fuzzy logic can handle the uncertainty caused by sensor noise or model inaccuracy, optimize control instructions through fuzzy rule base, and enhance the adaptability of the system. Genetic algorithms, as a search optimization algorithm, can find the optimal or approximately optimal solution in a complex search space. In the motor control strategy, genetic algorithms can be used to optimize control parameters to meet multi-objective performance requirements, such as minimizing energy consumption and maximizing dynamic response speed. By comprehensively using neural networks, fuzzy logic and genetic algorithms, the intelligent algorithm adaptation module can not only improve the adaptability and robustness of the motor control system, but also maintain the stability and efficiency of the motor performance in the face of complex and changing working environments. By real-time monitoring of the motor's operating status and dynamically adjusting the control strategy, the module ensures that the motor can achieve the optimal control effect under various working conditions, significantly improving the performance of the entire drive system.

[0013] The real-time anti-debt processing module realizes dynamic adjustment of control parameters by accurately monitoring the operating status and performance indicators of the motor. This module uses advanced sensor technology to collect key parameters of the motor in real time, such as current, voltage, speed and temperature. Assume that the actual current of the motor during operation is I actual =8.5A, and the target current is set to I target =10A, the real-time feedback processing module will detect this deviation and immediately start the closed-loop control mechanism. The closed-loop control system calculates the deviation value by comparing the actual parameters with the preset target parameters, and uses advanced control algorithms, PID controllers, to adjust the control input. If the current deviation exceeds the predetermined threshold, the PID controller will calculate the necessary adjustment amount by increasing or decreasing the duty cycle of the PWM signal, such as from D PWM =70% adjusted to D PWM=75% to compensate for the deviation and ensure that the motor current responds quickly and accurately tracks the target value. The real-time anti-clearing processing module also includes monitoring of the motor operating environment, such as temperature changes or load fluctuations, which may affect the performance of the motor. The module will adaptively adjust the intelligent algorithm adaptation module according to these environmental changes and optimize the control strategy to maintain the best performance of the motor under different conditions. In a high temperature environment, the module will activate the overheating protection mechanism to prevent motor damage by reducing the load or adjusting the cooling system.

[0014] The performance evaluation module quantifies the current control effect through a series of key performance indicators to ensure that the control effect meets the design requirements. This module closely monitors the current tracking error. If the actual current I actual With the expected current I desired If the difference between the two exceeds the preset value Threshold = 0.1A, the system will record and trigger an alarm. At the same time, the control response time is evaluated, which should ideally be less than 5 milliseconds to ensure that the motor can quickly adapt to changes. Current stability is measured by the root mean square value, and the goal is to maintain a low level, such as RMSCurrent = 0.5A, to reduce losses. Finally, the energy efficiency evaluation ensures that the system operates at a high efficiency by calculating the ratio of energy consumption to output power, such as EnergyEfficiency. These comprehensive evaluation results provide a basis for continuous optimization of the system to ensure that the motor can achieve efficient and accurate current control under various operating conditions.

[0015] The control instruction execution module receives the optimization control instruction output by the model predictive controller, and the current setting value I setpoint =9A, the module uses high-precision power electronic devices IGBT, by adjusting the switching state of these devices and the duty cycle of the PWM signal, from D PWM =50% increase to D PWM = 60%, to accurately adjust the motor current and ensure that the actual current I measured The built-in controller or DSP monitors the current feedback in real time, responds quickly to any deviations, and makes necessary adjustments to maintain the stability and efficiency of the motor under different operating conditions.

[0016] System optimization and iteration dynamically adjusts and improves the control strategy by continuously collecting performance feedback and monitoring system status. Specifically, the system uses key indicators provided by the performance evaluation module, such as current tracking error, response time, current stability, etc., to identify deficiencies in the control process. Then, combined with advanced technologies such as machine learning, the control parameters are carefully adjusted, such as optimizing the duty cycle of the PWM signal or adjusting the rules of the fuzzy logic controller to adapt to different working conditions and improve system efficiency. At the same time, the system also includes regular evaluation and calibration of hardware components to ensure that all devices can maintain optimal performance in long-term operation. The whole process is a repetitive dynamic optimization that ensures that the motor control system can continuously adapt to changing needs, improve control accuracy and system efficiency, and achieve self-improvement and continuous performance improvement.

[0017] The specific operation steps of the system are as follows:

[0018] Starting the motor parameter identification module is the first step in the control process. The precise parameters of the motor are obtained by real-time monitoring of the motor's operating status, such as voltage, current, and speed. These parameters include but are not limited to resistance, inductance, flux linkage, and rotor position, which are crucial to the accuracy of the subsequent control algorithm.

[0019] The state-space equation building module constructs a mathematical model of the motor based on its physical characteristics and electrical behavior. This model is usually expressed in the form of a set of first-order differential equations that describe the dynamic relationship between the motor current, voltage, and flux.

[0020] Configure the model predictive controller and set the optimization objectives and constraints. The optimization objectives may include minimizing the current tracking error and the smoothness of the control input, while the constraints may involve the voltage and current limits of the motor.

[0021] Run the rolling optimization solver module to calculate the optimal control instructions for the next control cycle based on the current and predicted system states. This process involves predicting future behavior and optimizing control inputs to achieve comprehensive optimization of system performance.

[0022] The control signal smoothing module processes the calculated control signal to reduce high-frequency noise and mutations, ensuring the smoothness of the control signal and the stability of the motor operation.

[0023] The intelligent algorithm adaptation module dynamically adjusts the control parameters based on real-time feedback and system performance evaluation. This may include neural networks, fuzzy logic or genetic algorithms to improve the adaptability and robustness of the control strategy.

[0024] The real-time feedback processing module adjusts the control strategy according to the feedback signals from the motor, such as the actual current and speed, to achieve closed-loop control. This helps improve the stability and responsiveness of the system.

[0025] The performance evaluation module conducts quantitative analysis on the current control effect, monitors and records key performance indicators, and ensures that the control effect reaches the expected goal.

[0026] The control execution module accurately applies the optimized control instructions to the motor, and achieves precise control of the motor current through high-precision power electronic devices, ensuring that the motor operates according to the predetermined performance indicators.

[0027] The above technical solution can achieve the following three key technical effects:

[0028] 1. Significantly improve control accuracy and response speed: This technical solution can predict and optimize the motor current behavior in the future through model predictive control theory. This forward-looking control strategy achieves precise regulation of current and ensures that the motor can quickly respond to control inputs under various working conditions, thereby significantly improving control accuracy and the dynamic response speed of the system.

[0029] 2. Enhance the robustness and adaptability of the system: Introducing intelligent algorithm adaptation modules, such as neural networks, fuzzy logic or genetic algorithms, enables the control system to adapt to changes in motor parameters and external disturbances, effectively improving the robustness of the system. This intelligent adaptive adjustment capability enables the motor control system to maintain stable performance in the face of complex and changing working environments.

[0030] 3. Reduce motor operation loss and improve energy efficiency: By smoothing the control signal, the motor loss caused by sudden changes in the control signal is reduced. At the same time, the optimized control strategy reduces the energy consumption of the motor during operation. This not only improves the efficiency of the motor, but also helps to reduce operating costs, achieve energy conservation and emission reduction, and is of great significance to environmental protection and sustainable development. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] Figure 1 This is a block diagram of the permanent magnet synchronous motor current control system of the present invention.

[0032] Figure 2 The flow chart of the permanent magnet synchronous motor current control method of the present invention is DETAILED DESCRIPTION

[0033] In a specific embodiment of the present invention, we describe in detail a permanent magnet synchronous motor current control method based on model predictive control, covering the entire process from real-time identification of motor parameters to precise execution of control instructions:

[0034] The following will be combined with the attached embodiment of the present invention Figure 1 and Figure 2, the technical solutions in the embodiments of the present invention are clearly and completely described. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0035] Motor parameter identification is the key to ensuring the accuracy and adaptability of the permanent magnet synchronous motor (PMSM) control strategy, involving the following specific numerical technical details: The DC current injection method is usually used. When the motor is stationary, the DC voltage and current of the stator winding are measured, and the resistance value is calculated using Ohm's law. Where V rms is the RMS voltage across the winding, I rms is the root mean square current flowing through the winding. It is carried out through the high-frequency sinusoidal signal injection method, injecting a sinusoidal voltage of known amplitude and frequency into the stator winding of the motor, measuring the corresponding current response, and using the definition of inductance to estimate the inductance value. This method is suitable for occasions with high requirements for inductance parameter accuracy. The initial position is usually determined indirectly by an encoder or through the zero crossing point of the back electromotive force. The back electromotive force is obtained by measuring the open circuit voltage of the stator winding when the motor is unloaded and the speed is known. The flux linkage is estimated by measuring the magnetization curve of the permanent magnet and combining it with the motor design data. The rotor resistance and leakage inductance are measured by current and voltage detection in periodic interrupts and comparison interrupts. Online identification methods include least squares method, recursive least squares method, extended Kalman filtering, etc. These algorithms can handle noise interference, extract useful signal features, and realize real-time estimation of parameters. Combined with model predictive control technology, online identification of motor parameters can be realized to further improve the accuracy and efficiency of identification. Artificial neural network is used for parameter identification, which has good nonlinear approximation ability and is particularly suitable for identifying motor parameters with nonlinear characteristics. By injecting high-frequency current during motor operation and observing its effect on motor performance, the inductance parameters of the motor can be identified. This method has very high identification accuracy. In actual simulation, after 1 second of identification, the q-axis inductance is stabilized at 18.45mH, which is very close to the standard value of 18.5mH, and the identification error is less than 3%.

[0036] The establishment of mathematical models is the core link in the field of motor control: based on the physical characteristics of the motor, such as the law of electromagnetic induction and the basic laws of circuits, voltage equations and flux equations are established. These equations describe the electrical behavior of the motor, including the relationship between stator current, voltage and flux. Through these relationships, the state space equations of the motor can be derived, which express the dynamic characteristics of the motor in the form of a set of first-order differential equations. Since the mathematical model of the three-phase motor has a strong coupling problem in the original coordinate system, it is necessary to simplify the model through coordinate transformation. The Clarke transformation (3s / 2s) converts the variables in the three-phase stationary coordinate system into a two-phase stationary coordinate system, while the Park transformation (2s / 2r) further converts the variables in the two-phase stationary coordinate system into a synchronous rotating coordinate system (dq axis), thereby achieving decoupling and making the design of the control strategy similar to that of a DC motor, simplifying the complexity of the control algorithm. Under the dq axis, the voltage equation can clearly express the relationship between current and voltage, that is, and where u d and u q are the voltages of the dd and q axes, i d and i q is the corresponding current, R s is the stator resistance, ω e is the electrical angular velocity, L d and L q is the inductance of the d-axis and q-axis, ψ d and ψ q is the magnetic flux, ψ f is the magnetic flux generated by the permanent magnet. Electromagnetic torque is the key to the motor output torque and is related to the dq axis components of the current. The torque equation is usually expressed as Where T e is the electromagnetic torque, and p is the number of pole pairs of the motor. The motor's maximum speed and position are related to the electromagnetic torque through the mechanical equations of motion, which take into account the motor's rotational momentum, load torque, and viscous damping.

[0037] When designing the control strategy for a permanent magnet synchronous motor, a model predictive controller (MPC) is developed that uses a mathematical model of the motor to predict the system behavior within 1 to 3 seconds in the future and uses an optimization algorithm to calculate the current tracking error E that can be minimized within a specific time range. I =∑(I ref -I actual ) 2The controller takes into account the dynamic characteristics of the motor, with a resistance of 0.5 ohms and an inductance of 0.01 Henry, and physical constraints are incorporated into the design, including a maximum voltage of 100 volts, a maximum current of 100 amperes, and a speed of no more than 3000 rpm, to ensure that the motor operates under safe and efficient conditions. Furthermore, we use the non-dominated sorting speed algorithm II (NSGA-II) for multi-objective optimization, balancing the current tracking performance and control input smoothness, such as a voltage change rate of no more than 0.5 volt-second. For the motor rotor structure, a hierarchical optimization strategy is implemented, with high-sensitivity parameters such as rotor mass of 0.05 kg m2 being optimized first, and then low-sensitivity parameters are optimized based on a new finite element model (FEM), reducing the amount of calculation by about 20%. The controller is configured as a real-time optimization system that is able to adjust the control input based on real-time feedback from the motor, such as load fluctuations of ±10%. Finally, the performance of the optimized controller is verified by finite element analysis (FEA), ensuring that the design performs well in practical applications while meeting the requirements of accuracy of ±1% and efficiency of more than 90%.

[0038] The rolling horizon optimization converts the nonlinear filtering problem into a constrained optimization problem by solving an optimization problem within a limited time range at each time step, effectively reducing the amount of calculation required to solve the optimization problem and significantly improving the accuracy of state estimation. In the specific implementation, the rolling horizon optimization first defines a cost function. For example, in the problem of aircraft attitude estimation, the cost function may include deviation terms from the reference attitude and deviation terms between the control input and the predetermined trajectory. These deviation terms are usually penalized using weight matrices Q and R in the following form: Among them, x and u represent the current state and control input respectively, and x ref and u ref are their reference values. The key advantage of the rolling horizon optimization is that it can explicitly handle system constraints, such as the normalization constraint of quaternions in attitude estimation. By introducing these constraints in the optimization process, the search space can be narrowed and the search efficiency and accuracy can be improved. This method meets the real-time requirements by applying the Gauss-Newton iteration method in the time domain window to solve the optimal state estimate. In practical applications, the rolling horizon optimization method can not only improve the accuracy of nonlinear target tracking, but also is insensitive to both initial errors and system errors. In the radar target tracking problem, the rolling horizon estimation (MHE) transforms the problem into an optimization problem in a finite time domain by introducing the arrival cost function, which effectively reduces the computational burden and improves the accuracy of the estimation.

[0039] Control signal smoothing processes the control signal through a moving average filter to reduce high-frequency noise and optimize signal quality. A moving average filter of length N achieves smoothing by calculating the average value of every N sampling points of the signal; while the Savitzky-Golay filter estimates the smoothed value of each data point using a polynomial fit within a fixed-size window. The window length is usually selected to be 11 and the polynomial order is 2 to balance the smoothing effect and computational complexity. In addition, for signal features that do not want to be smoothed, such as the sharp edges of the clock signal, the median filter can retain these features while removing unnecessary high-frequency fluctuations. In practical applications, such as aircraft control systems, by using continuous or smooth control signals, such as SigmoidFunction, control jitter can be avoided to ensure the smoothness and reliability of system operation. Through these methods, control signal smoothing not only improves the performance and efficiency of the motor, but also enhances the stability and responsiveness of the system.

[0040] The intelligent algorithm adaptation module can learn the current response of the motor under different loads. The current response of the motor under rated load is I rated = 10A, and in overload conditions the motor current response may reach I overload = 12A. The neural network learns the dynamic behavior of the motor through this data and predicts the response under specific operating conditions, achieving precise adjustment of the control input and adjusting the duty cycle of the PWM signal to D PWM =70% to accommodate overload conditions. The analog logic handles the uncertainty caused by sensor noise or model inaccuracy. For example, when the current deviation reaches ΔI = 0.5A, the rate of change is ΔI rate =0.1A / s, fuzzy logic optimizes control instructions through a fuzzy rule base to enhance the system's adaptability to these minor changes. Genetic algorithms are used in control strategies to optimize control parameters to meet multi-objective performance requirements, for example, minimizing energy consumption while ensuring motor response speed. Assume that the genetic algorithm finds a set of control parameters during the iteration process. The motor reaches a steady state within t=0.01s, while reducing energy consumption by η=5%. Through the comprehensive application of these intelligent algorithms, the intelligent algorithm adaptation module not only improves the adaptability of the motor control system, but also ensures the stability and efficiency of the motor performance in complex and changing working environments. In a high temperature environment, assuming that the resistance increases by R increase =0.2Ω, the neural network can predict this change and adjust the control strategy in time, the analog logic can handle the fluctuations in sensor readings caused by temperature changes, and the central transmission algorithm can find the optimal control parameters under these conditions to ensure that the motor runs at the highest efficiency.

[0041] The real-time anti-debt processing module realizes dynamic adjustment of control parameters by accurately monitoring the operating status and performance indicators of the motor. This module uses advanced sensor technology to collect key parameters of the motor in real time, such as current, voltage, speed and temperature. Assume that the actual current of the motor during operation is I actual =8.5A, and the target current is set to I target =10A, the real-time feedback processing module will detect this deviation and immediately start the closed-loop control mechanism. The closed-loop control system calculates the deviation value by comparing the actual parameters with the preset target parameters, and uses advanced control algorithms, PID controllers, to adjust the control input. If the current deviation exceeds the predetermined threshold, the PID controller will calculate the necessary adjustment amount by increasing or decreasing the duty cycle of the PWM signal, such as from D PWM =70% adjusted to D PWM =75% to compensate for the deviation and ensure that the motor current responds quickly and accurately tracks the target value. The real-time feedback processing module also includes monitoring of the motor operating environment, such as temperature changes or load fluctuations, which may affect the performance of the motor. The module will adaptively adjust the intelligent algorithm adaptation module according to these environmental changes and optimize the control strategy to maintain the best performance of the motor under different conditions. In a high temperature environment, the module will activate the overheating protection mechanism to prevent motor damage by reducing the load or adjusting the cooling system.

[0042] The performance evaluation module quantifies the current control effect through a series of key performance indicators to ensure that the control effect meets the design requirements. This module closely monitors the current tracking error. If the actual current I actual With the expected current I desired If the difference between the current and the output power exceeds the preset value Threshold = 0.1A, the system will record and trigger an alarm. At the same time, the control response time is evaluated, which should ideally be less than 5 milliseconds to ensure that the motor can quickly adapt to changes. Current stability is measured by the root mean square value, and the goal is to maintain a low level, such as RMSCurrent = 0.5A, to reduce losses. Energy efficiency evaluation ensures that the system operates at a high efficiency by calculating the ratio of energy consumption to output power, such as EnergyEfficiency. These comprehensive evaluation results provide a basis for continuous optimization of the system to ensure that the motor can achieve efficient and accurate current control under various operating conditions.

[0043] The control instruction execution module receives the optimization control instruction output by the model predictive controller, and the current setting value I setpoint =9A. This module uses high-precision power electronic devices IGBTs to adjust the switching state of these devices and the duty cycle of the PWM signal to PWM =50% increase to D PWM= 60%, to accurately adjust the motor current and ensure that the actual current I measured The built-in controller or DSP monitors the current feedback in real time, responds quickly to any deviations, and makes necessary adjustments to maintain the stability and efficiency of the motor under different operating conditions.

[0044] System optimization and iteration dynamically adjusts and improves the control strategy by continuously collecting performance feedback and monitoring system status. Specifically, the system uses key indicators provided by the performance evaluation module, such as current tracking error, response time, current stability, etc., to identify deficiencies in the control process. Then, combined with advanced technologies such as machine learning, the control parameters are carefully adjusted, such as optimizing the duty cycle of the PWM signal or adjusting the rules of the fuzzy logic controller to adapt to different working conditions and improve system efficiency. At the same time, the system also includes regular evaluation and calibration of hardware components to ensure that all devices can maintain optimal performance in long-term operation. The whole process is a repetitive dynamic optimization that ensures that the motor control system can continuously adapt to changing needs, improve control accuracy and system efficiency, and achieve self-improvement and continuous performance improvement.

[0045] The specific operation steps of the system are as follows:

[0046] Starting the motor parameter identification module is the first step in the control process. The precise parameters of the motor are obtained by real-time monitoring of the motor's operating status, such as voltage, current, and speed. These parameters include but are not limited to resistance, inductance, flux linkage, and rotor position, which are crucial to the accuracy of the subsequent control algorithm.

[0047] The state-space equation building module constructs a mathematical model of the motor based on its physical characteristics and electrical behavior. This model is usually expressed in the form of a set of first-order differential equations that describe the dynamic relationship between the motor current, voltage, and flux.

[0048] Configure the model predictive controller and set the optimization objectives and constraints. The optimization objectives may include minimizing the current tracking error and the smoothness of the control input, while the constraints may involve the voltage and current limits of the motor.

[0049] Run the rolling optimization solver module to calculate the optimal control instructions for the next control cycle based on the current and predicted system states. This process involves predicting future behavior and optimizing control inputs to achieve comprehensive optimization of system performance.

[0050] The control signal smoothing module processes the calculated control signal to reduce high-frequency noise and mutations, ensuring the smoothness of the control signal and the stability of the motor operation.

[0051] The intelligent algorithm adaptation module dynamically adjusts the control parameters based on real-time feedback and system performance evaluation. This may include neural networks, fuzzy logic or genetic algorithms to improve the adaptability and robustness of the control strategy.

[0052] The real-time feedback processing module adjusts the control strategy according to the feedback signals from the motor, such as the actual current and speed, to achieve closed-loop control. This helps improve the stability and responsiveness of the system.

[0053] The performance evaluation module conducts quantitative analysis on the current control effect, monitors and records key performance indicators, and ensures that the control effect reaches the expected goal.

[0054] The control execution module accurately applies the optimized control instructions to the motor, and achieves precise control of the motor current through high-precision power electronic devices, ensuring that the motor operates according to the predetermined performance indicators.

[0055] The above shows and describes the basic principles, main features and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited by the above embodiments. The above embodiments and descriptions are only preferred examples of the present invention and are not intended to limit the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, which fall within the scope of the present invention. The scope of protection of the present invention is defined by the attached claims and their equivalents.

Claims

1. A permanent magnet synchronous motor current control method based on model predictive control, characterized in that: It includes the following modules: motor parameter identification module, which is used to identify motor parameters in real time; state space equation establishment module, which is used to build the mathematical model of the motor; model predictive controller configuration module, which is used to design the control strategy, including the objective function and constraints; Rolling optimization solution module, used to solve the optimal control command in real time; control signal smoothing module, used to optimize the control signal to reduce motor loss; Intelligent algorithm adaptation module, used to adaptively adjust control parameters; The real-time feedback processing module is used to adjust the control strategy according to the feedback signal; the performance evaluation module is used to evaluate the current control performance; and the control execution module is used to implement the control instructions on the motor.

2. The method according to claim 1, characterized in that The motor parameter identification module uses the motor operation data to perform parameter estimation.

3. The method according to claim 1, characterized in that The state-space equation building module builds a model based on the physical characteristics of the motor.

4. The method according to claim 1, characterized in that The model predictive controller configuration module adopts a multi-objective optimization algorithm.

5. The method according to claim 1, characterized in that The rolling optimization solution module adopts an optimization algorithm within a limited time range.

6. The method according to claim 1, characterized in that The control signal smoothing module reduces the high frequency components of the control signal through filtering technology.

7. The method according to claim 1, characterized in that The intelligent algorithm adaptation module includes a neural network, fuzzy logic or a genetic algorithm.

8. The method according to claim 1, characterized in that The real-time feedback processing module implements closed-loop control to improve the stability of the system.

9. The method according to claim 1, characterized in that: The performance evaluation module performs quantitative analysis on the current control effect through preset performance indicators.

10. A step of implementing the method according to any one of claims 1 to 9, comprising: Start the motor parameter identification module to obtain the motor parameters in real time; Use state-space equations to build modules and establish a mathematical model of the motor; Configure the model predictive controller and set the optimization objectives and constraints; Run the rolling optimization solution module to calculate the control instructions for the next control cycle; Processing the control signal through a control signal smoothing module; Adjust control parameters according to intelligent algorithm adaptation module; Through the real-time feedback processing module, the control strategy is adjusted according to the motor feedback; Monitor and record current control performance using the performance evaluation module; Execute control instructions to the motor and achieve precise control of current through the control execution module.

Images