A multi-mode current predictive control method for hybrid excitation axial magnetic field permanent magnet motor

Abstract

The present invention discloses a multi-mode current prediction control method for a hybrid excitation axial magnetic field permanent magnet motor, which realizes an effective current distribution control strategy for the hybrid excitation axial magnetic field permanent magnet motor under variable working conditions, thereby adapting to the control requirements of variable operating conditions. Taking into account the different emphases of control requirements under different working conditions of electric vehicles, online distribution of reference current is achieved by adjusting the proportional relationship of copper loss. Combined with model prediction current control, the armature voltage vector and the excitation voltage vector are synthesized by mixed action, thereby acting together on the armature magnetic field and the excitation magnetic field, thereby effectively reducing the influence of the coupling of the armature magnetic field and the excitation magnetic field on the control performance of the motor. The control method proposed by the present invention fully considers the control requirements of the entire operating condition, not only improving the dynamic response capability of the system in variable working conditions, but also improving the steady-state performance of the system in steady-state operation.

Description

Technical Field

[0001] The invention relates to a multi-mode current prediction control strategy for a hybrid excitation axial magnetic field permanent magnet motor, belonging to the technical field of motor control. Background Art

[0002] Hybrid-excitation axial-field permanent magnet motors (HEXPMMs) have the advantages of short axial dimensions, compact structure, convenient heat dissipation, wide speed regulation range, and high power density, and thus have good application prospects in in-wheel direct-drive systems for electric vehicles. The actual operating conditions of electric vehicles are complex and diverse, including frequent starting and stopping, acceleration and deceleration, heavy-load climbing, and high-speed cruising. To meet the control requirements of different operating conditions, HEXPMMs generally need to have excellent drive performance, such as high power density, wide speed regulation range, high efficiency, and high reliability. Therefore, how to achieve effective distribution and coordinated control of armature current and excitation current under multiple operating conditions, solve the coupling problem of the excitation field, permanent magnetic field, and armature field of HEXPMMs, and improve the dynamics and efficiency of the drive system has become a hot topic in the field of electric vehicle drive research.

[0003] At present, there are two basic control strategies for hybrid excitation axial magnetic field permanent magnet motors. First, i d = 0 control is the most commonly used control strategy, featuring the simplest control structure and fast response time. Considering the application of DC excitation windings, minimum copper loss control was proposed to maintain the minimum copper loss per unit torque. While these control methods can meet the requirements of certain single operating conditions, they struggle to meet the requirements of various operating conditions with complex requirements. Furthermore, by switching control algorithms, different control strategies can be adopted under different operating conditions. However, due to the lack of coordination and over-control between various control strategies, the distribution of excitation current and armature current is discontinuous. This often leads to sudden changes in the motor's torque, speed, and efficiency, failing to fully leverage the advantages of balanced control across various control strategies and multiple performance objectives. Summary of the Invention

[0004] This invention aims to overcome the shortcomings of existing control technologies by proposing a multi-mode current predictive control strategy for a hybrid-excitation axial-field permanent magnet motor. Taking into account the varying control requirements for each operating condition, this strategy effectively distributes current across a wide range of operating conditions.

[0005] The technical solution adopted by the present invention is a multi-mode current prediction control method for a hybrid excitation axial magnetic field permanent magnet motor, which is specifically implemented according to the following steps:

[0006] Step 1: At the current time k, the bus voltage U is collected from the main control circuit of the hybrid excitation axial magnetic field permanent magnet motor. dc (k), excitation voltage Uf (k), a, b, c phase current i a (k), i b (k), i c (k) and the excitation current i f (k) Calculate the motor rotor position angle θ and speed n based on the signal collected by the motor encoder;

[0007] Step 2: Substitute the phase current i obtained in step 1 a (k), i b (k), i c (k) After Park transformation, the d-axis current i in the synchronous rotating coordinate system at time k is obtained d (k) and q-axis current i q (k);

[0008] Step 3: Compare the given speed with the speed obtained in step 1. The speed deviation is passed through the PI regulator to obtain the given torque T e * , a multi-mode current control strategy is used to allocate a constant current, that is, according to the given values ​​of motor speed n and acceleration α, the actual values ​​are compared, the objective functions corresponding to four typical operating conditions are selected, and the corresponding Lagrangian constraint equations are substituted to calculate the d-axis current reference value i dref , q-axis current reference value i qref , excitation current reference value i fref ;

[0009] Step 4: According to the d-axis current reference value i obtained in step 3 dref , q-axis current reference value i qref , excitation current reference value i fref , the d-axis current i obtained in step 2 d (k), q-axis current i q (k), the excitation current i obtained in step 1 f (k) The duty cycle t of the applied voltage vector is calculated using the discrete prediction model of the multi-vector model predictive current control ijz , t ab ;

[0010] Step 5: Duty cycle t obtained from step 4 ijz , t ab , synthesize multiple groups of virtual action voltage vectors, select the optimal virtual action voltage vector using the cost function, and output the corresponding duty cycle to the main power converter and the excitation power converter.

[0011] Furthermore, the specific steps of allocating the constant current using the multi-mode current control strategy in step 3 are as follows:

[0012] The low-speed range includes operating conditions such as heavy-load climbing and normal cruising. The output torque requirement in this range is high, and the motor speed does not exceed the rated speed. Therefore, the control system does not need to consider the inverter voltage limit. The excitation current is only used for magnetization control, and an extended Lagrange multiplier method is used to find the minimum value of the objective function under the torque constraint.

[0013] The operating conditions in the high-speed zone include medium- and high-speed climbing conditions and high-speed cruising conditions. Since the motor speed in this zone far exceeds the rated speed, the control system needs to use field weakening control to reduce the air gap flux, thereby widening the motor's speed regulation range. The extended Lagrange multiplier method is used to solve the minimum value of the objective function, which must simultaneously meet torque and voltage constraints.

[0014] The above equations are solved by referring to the solution method of the extended Lagrange multiplier method, that is, the partial derivative of each variable in the equation is solved and set to 0, and then the Newton iteration method is used to solve the partial derivative equation to obtain the optimal solution of each variable.

[0015] Furthermore, the extended Lagrange multiplier method is used to find the minimum value of the objective function under the torque constraint:

[0016]

[0017] Where λ is the Lagrange multiplier; T eref is the given torque; i d 、i q 、i f are d-axis current, q-axis current, and excitation current respectively; p n is the pole pair number; ψ f is the permanent magnet flux; L d , L q are d-axis inductance and q-axis inductance respectively; M f is the mutual inductance between the armature winding and the excitation winding; f(i d ,i q ,i f ) is the objective function designed according to the control requirements of different operating conditions.

[0018] Furthermore, the objective function needs to satisfy the constraints of torque and voltage at the same time:

[0019]

[0020] Where λ1 and λ2 are Lagrange multipliers; U dc is the DC bus voltage; ω e is the electrical angular velocity; f(i d ,i q ,i f ) is the objective function designed according to the control requirements of different operating conditions.

[0021] Furthermore, the objective function is designed as follows based on the control requirements of four typical working conditions:

[0022] (5) Heavy load climbing condition

[0023] Under this operating condition, the motor operates in the low-speed range, with large load torque and frequent changes. Therefore, by improving the utilization rate of the excitation current in the current distribution, the maximum output torque of the motor can be increased. More importantly, it is necessary to improve the rapid response capability of the current to reduce the impact of sudden load torque on the motor control performance. The objective function is designed as follows:

[0024]

[0025] Where R s is the armature winding resistance; R f is the resistance value of the excitation winding; T emax is the maximum output torque; T e * is the given torque; n is the motor speed sampling feedback value; n bref is the field weakening base speed; Δn is the absolute value of the difference between the given speed and the feedback speed; α is the acceleration of the motor speed. In this objective function, the copper loss ratio in the excitation winding decreases with the increase of torque. Increasing the distribution ratio of the excitation current in the current distribution increases the response speed of the excitation current, thereby improving the control system's ability to respond to torque.

[0026] (6) Normal cruising conditions

[0027] This operating condition often involves speed changes such as acceleration and deceleration, starting and stopping. Therefore, it is necessary to increase the starting torque to improve the motor's response speed to sudden speed changes. The overall control environment of this operating condition is relatively stable, and it is necessary to maintain minimum copper loss to improve the motor's steady-state performance. The objective function is designed as follows:

[0028]

[0029] Where n ref is a given speed. In this objective function, when the motor speed changes, the proportion of copper loss in the armature winding increases with the increase of speed. The utilization rate of the excitation current is improved to increase the starting torque of the motor and the responsiveness of the control system. At the same time, when the speed tends to be stable, the objective function will distribute the current according to the proportion of minimum copper loss to keep the motor running with the lowest loss.

[0030] (7) Medium and high speed climbing conditions

[0031] This operating condition is in the medium-high speed range, and the control system needs to adopt weak magnetic control to widen the motor's speed range. However, this will also lead to a decrease in starting torque and a slowdown in speed response. Therefore, under weak magnetic control, it is necessary to reduce the utilization rate of the excitation current and increase the utilization rate of the d-axis current. While ensuring the speed range, increasing the motor's starting torque can improve the utilization rate of the reluctance torque. The control system can respond quickly to sudden changes in the motor's speed to reduce the impact of sudden changes in speed on control performance. The objective function is designed as follows:

[0032]

[0033] In this objective function, the copper loss rate of the excitation winding increases with the increase of speed or torque; the field weakening control reduces the utilization rate of the excitation current, thereby improving the control system's responsiveness to speed and output torque;

[0034] (8) High-speed cruising conditions

[0035] This operating condition is in the high-speed range, and the control system needs to perform deep field weakening control to obtain a wider speed range, that is, to improve the utilization rate of the excitation current. However, the most important control goal is to keep the copper loss of the motor to a minimum during stable cruising, to ensure high stability and low loss of the motor under high-speed cruising conditions. The objective function is designed as follows:

[0036]

[0037] In this objective function, the proportion of copper loss in the armature winding increases with the increase of speed, reducing the proportion of armature current and improving the utilization rate of excitation current, thereby increasing the depth of the weak magnetic strategy to widen the speed range. When the speed tends to be stable, the objective function distributes the current according to the proportion of minimum copper loss to ensure that the motor runs with minimum loss.

[0038] Furthermore, the state space equation of the discrete prediction model of the multi-vector model predictive current control is expressed as:

[0039]

[0040] Where x(k) is the state vector, u(k) is the input vector, y(k) is the output vector, C is the identity matrix, and h(k) is the disturbance term.

[0041] The matrices and vectors in the above formula are as follows:

[0042]

[0043] Where, L f is the self-inductance of the excitation winding; T s is the sampling time; ud 、u q 、u f They are d-axis voltage, q-axis voltage, and excitation voltage respectively.

[0044] The cost function J is constructed as follows:

[0045] min{J}=|i d (k+2)-i dref |+|i q (k+2)-i qref |+|i f (k+2)-i fref |

[0046] Where i d (k+2), i q (k+2), i f (k+2) are the predicted values ​​of d-axis current, q-axis current and excitation current at time k+2 respectively; i dref 、i qref 、i fref They are the d-axis current given value, q-axis current given value, and excitation current given value respectively.

[0047] The present invention is also characterized in that:

[0048] The specific steps of the multi-mode current control strategy of the hybrid excitation axial magnetic field permanent magnet motor in step 3 are:

[0049] The low-speed range includes operating conditions such as heavy-load climbing and normal cruising. The output torque requirement in this range is high, and the motor speed does not exceed the rated speed. Therefore, the control system does not need to consider the inverter voltage limit. The excitation current is only used for magnetization control. The extended Lagrange multiplier method is used to find the minimum value of the objective function under the torque constraint, as follows:

[0050]

[0051] Where λ is the Lagrange multiplier; T eref is the given torque; p n is the pole pair number; f(i d ,i q ,i f ) is the objective function designed according to the control requirements of different operating conditions.

[0052] The operating conditions in the high-speed range include medium- and high-speed climbing and high-speed cruising. Because the motor speed in this range far exceeds the rated speed, the control system must employ field weakening to reduce the air gap flux, thereby widening the motor's speed range. The extended Lagrange multiplier method is used to solve for the minimum value of the objective function, which must satisfy both torque and voltage constraints, as follows:

[0053]

[0054] Where λ1 and λ2 are Lagrange multipliers; U dc is the DC bus voltage; f(i d ,i q ,i f ) is the objective function designed according to the control requirements of different operating conditions.

[0055] The above equations are solved by referring to the solution method of the extended Lagrange multiplier method, that is, the partial derivative of each variable in the equation is solved and set to 0. Then the Newton iteration method is used to solve the partial derivative equation to obtain the optimal solution for each variable.

[0056] According to the operating characteristics and control requirements of the four typical working conditions, the present invention designs the objective functions of the four typical working conditions respectively, as follows:

[0057] (1) Heavy load climbing condition

[0058] Under these operating conditions, the motor operates in the low-speed range, with high load torque and frequent changes. Therefore, by increasing the utilization of the excitation current in the current distribution, the motor's maximum output torque can be increased. More importantly, it is necessary to improve the current's rapid response capability to reduce the impact of sudden load torque on the motor's control performance. The objective function is designed as follows:

[0059]

[0060] Where, T emax is the maximum output torque; T e * is the given torque; n is the motor speed sampling feedback value; n bref is the field weakening base speed; Δn is the absolute value of the difference between the set speed and the feedback speed; and α is the acceleration of the motor speed. In this objective function, the copper loss ratio in the excitation winding decreases with increasing torque. Increasing the excitation current allocation in the current distribution improves the excitation current response speed, thereby improving the control system's torque responsiveness.

[0061] (2) Normal cruising conditions

[0062] This operating condition often involves variable speeds, such as acceleration and deceleration, starting and stopping. Therefore, it is necessary to increase the starting torque to improve the motor's response to sudden speed changes. The overall control environment for this condition is relatively stable, and copper losses must be minimized to improve the motor's steady-state performance. The objective function is designed as follows:

[0063]

[0064] Where n ref is a given speed. In this objective function, as the motor speed changes, the proportion of copper loss in the armature winding increases with the speed. Improving the utilization of the excitation current increases the motor's starting torque and the control system's responsiveness. Furthermore, when the speed stabilizes, the objective function distributes the current based on the proportion of minimum copper loss, maintaining the motor's lowest loss.

[0065] (3) Medium and high speed climbing conditions

[0066] This operating condition is in the medium-to-high speed range, and the control system needs to adopt weak magnetic control to widen the motor's speed range. However, this will also result in a decrease in starting torque and a slowdown in speed response. Therefore, under weak magnetic control, it is necessary to reduce the utilization rate of the excitation current and increase the utilization rate of the d-axis current. While ensuring the speed range, increasing the motor's starting torque can improve the utilization rate of the reluctance torque. The control system can respond quickly to sudden changes in the motor's speed to reduce the impact of sudden changes in speed on control performance. The objective function is designed as follows:

[0067]

[0068] In this objective function, the copper loss rate of the excitation winding increases with the increase of speed or torque. Field weakening control reduces the utilization rate of the excitation current, thereby improving the control system's responsiveness to speed and output torque.

[0069] (4) High-speed cruising conditions

[0070] This operating condition is in the high-speed range, so the control system requires deep field weakening control to achieve a wider speed range, which means improving the utilization of the excitation current. However, the most important control goal is to minimize copper losses during stable cruising, ensuring high stability and low losses in high-speed cruising conditions. The objective function is designed as follows:

[0071]

[0072] In this objective function, the proportion of copper loss in the armature winding increases with speed, reducing the proportion of armature current and improving the utilization of excitation current. This increases the depth of the field weakening strategy and broadens the speed range. When the speed stabilizes, the objective function distributes current based on the proportion of minimum copper loss to ensure that the motor operates with minimal losses.

[0073] The beneficial effects of the present invention are:

[0074] (1) This paper considers the control requirements of different operating conditions and analyzes the relationship between various operating conditions of electric vehicles and the current distribution of hybrid-excitation axial-field permanent magnet motors. A multi-mode current predictive control strategy for hybrid-excitation axial-field permanent magnet motors that meets diverse operating conditions is proposed. Control requirements and performance targets no longer focus on a single operating condition, but instead consider the diverse and complex operating conditions encountered during electric vehicle operation.

[0075] (2) The present invention uses the Lagrange multiplier method to design the optimization equation for the multi-mode current control strategy. The objective function of the optimization equation is designed based on the control requirements of different operating conditions. This avoids problems such as discontinuous distribution of excitation current and armature current during switching and sudden changes in motor current, thereby achieving more ideal control performance.

[0076] (3) The present invention adopts a model prediction current control strategy to replace the PI regulator and SVPWM of the current loop, which not only improves the responsiveness of the motor control system but also improves the control accuracy of the motor current.

[0077] (4) This invention proposes a multi-mode current prediction control strategy for a hybrid-excitation axial-field permanent magnet motor. This strategy not only meets the control requirements of various operating conditions, but also offers excellent control performance, rapid response, and is simple to implement and readily applicable. It also provides a new and feasible solution for the operation of other types of hybrid-excitation synchronous motors under various operating conditions. BRIEF DESCRIPTION OF THE DRAWINGS

[0078] Figure 1 This is a block diagram of the control system of the present invention;

[0079] Figure 2 This is the block diagram of multi-mode current control. DETAILED DESCRIPTION

[0080] The present invention will be further described below with reference to the accompanying drawings and specific embodiments.

[0081] The present invention proposes a multi-mode current prediction control strategy for a hybrid excitation axial magnetic field permanent magnet motor. The structural block diagram is as follows: Figure 1 The control system consists of a multi-mode current distribution module, a PI regulator, a duty cycle calculation module, a three-vector current prediction model module, a cost function module, a pulse width modulation module, a main power converter, an excitation power converter, a DC power supply, a photoelectric encoder, and a hybrid excitation axial magnetic field permanent magnet motor.

[0082] A DC power supply powers the main and excitation power converters. Hall-effect voltage sensors collect bus voltage, process it, and feed it into the controller. The outputs of the main and excitation power converters are connected to a hybrid-excitation axial-field permanent magnet motor. Hall-effect current sensors collect phase and excitation currents, process them, and feed them into the controller. An encoder collects rotor position signals, processes them, and feeds them into the controller to calculate the rotor position angle and angular velocity. The controller outputs six switching signals to drive the main power converter, which, through the excitation pulse-width modulation module, outputs four pulse-width signals to drive the excitation power converter.

[0083] The technical solution adopted by the present invention is a multi-mode current prediction control strategy for a hybrid excitation axial magnetic field permanent magnet motor, which is specifically implemented according to the following steps:

[0084] Step 1: At the current time k, the bus voltage U is collected from the main control circuit of the hybrid excitation axial magnetic field permanent magnet motor. dc (k), excitation voltage U f (k), phase current i a (k), i b (k), i c (k) and the excitation current i f (k) Collect signals from the motor encoder and send them to the controller for processing to obtain the motor rotor position angle θ and speed n;

[0085] Step 2: Substitute the phase current i obtained in step 1 a (k), i b (k), i c (k) After Park transformation, the d-axis current i in the synchronous rotating coordinate system at time k is obtained d (k) and q-axis current i q (k);

[0086] Step 3: Compare the given speed with the speed obtained in step 1. The speed deviation is passed through the PI regulator to obtain the given torque T e * .like Figure 2 As shown in the figure, a multi-mode current control strategy is used to allocate a constant current, that is, according to the given values ​​of the motor speed n and acceleration α, the actual values ​​are compared, the objective functions corresponding to the four typical operating conditions are selected, and the corresponding Lagrangian constraint equations are substituted to calculate the d-axis current reference value i dref , q-axis current reference value i qref , excitation current reference value i fref Specifically:

[0087] The low-speed range includes operating conditions such as heavy-load climbing and normal cruising. The output torque requirement in this range is high, and the motor speed does not exceed the rated speed. Therefore, the control system does not need to consider the inverter voltage limit. The excitation current is only used for magnetization control. The extended Lagrange multiplier method is used to find the minimum value of the objective function under the torque constraint, as follows:

[0088]

[0089] Where λ is the Lagrange multiplier; T eref is the given torque; i d 、i q 、i f are d-axis current, q-axis current, and excitation current respectively; p n is the pole pair number; ψ f is the permanent magnet flux; L d , L q are d-axis inductance and q-axis inductance respectively; M f is the mutual inductance between the armature winding and the excitation winding; f(i d ,i q ,i f ) is the objective function designed according to the control requirements of different operating conditions.

[0090] The operating conditions in the high-speed range include medium- and high-speed climbing and high-speed cruising. Because the motor speed in this range far exceeds the rated speed, the control system must employ field weakening to reduce the air gap flux, thereby widening the motor's speed range. The extended Lagrange multiplier method is used to solve for the minimum value of the objective function, which must satisfy both torque and voltage constraints, as follows:

[0091]

[0092] Where λ1 and λ2 are Lagrange multipliers; U dc is the DC bus voltage; ω e is the electrical angular velocity; f(i d ,i q ,i f ) is the objective function designed according to the control requirements of different operating conditions.

[0093] The above equations are solved by referring to the solution method of the extended Lagrange multiplier method, that is, the partial derivative of each variable in the equation is solved and set to 0. Then the Newton iteration method is used to solve the partial derivative equation to obtain the optimal solution for each variable.

[0094] According to the operating characteristics and control requirements of the four typical working conditions, the present invention designs the objective functions of the four typical working conditions respectively, as follows:

[0095] (1) Heavy load climbing condition

[0096] Under these operating conditions, the motor operates in the low-speed range, with high load torque and frequent changes. Therefore, by increasing the utilization of the excitation current in the current distribution, the motor's maximum output torque can be increased. More importantly, it is necessary to improve the current's rapid response capability to reduce the impact of sudden load torque on the motor's control performance. The objective function is designed as follows:

[0097]

[0098] Where R s is the armature winding resistance; R f is the resistance value of the excitation winding; T emax is the maximum output torque; T e * is the given torque; n is the motor speed sampling feedback value; n bref is the field weakening base speed; Δn is the absolute value of the difference between the set speed and the feedback speed; and α is the acceleration of the motor speed. In this objective function, the copper loss ratio in the excitation winding decreases with increasing torque. Increasing the excitation current allocation in the current distribution improves the excitation current response speed, thereby improving the control system's torque responsiveness.

[0099] (2) Normal cruising conditions

[0100] This operating condition often involves variable speeds, such as acceleration and deceleration, starting and stopping. Therefore, it is necessary to increase the starting torque to improve the motor's response to sudden speed changes. The overall control environment for this condition is relatively stable, and copper losses must be minimized to improve the motor's steady-state performance. The objective function is designed as follows:

[0101]

[0102] Where n ref is a given speed. In this objective function, as the motor speed changes, the proportion of copper loss in the armature winding increases with the speed. Improving the utilization of the excitation current increases the motor's starting torque and the control system's responsiveness. Furthermore, when the speed stabilizes, the objective function distributes the current based on the proportion of minimum copper loss, maintaining the motor's lowest loss.

[0103] (3) Medium and high speed climbing conditions

[0104] This operating condition is in the medium-to-high speed range, and the control system needs to adopt weak magnetic control to widen the motor's speed range. However, this will also result in a decrease in starting torque and a slowdown in speed response. Therefore, under weak magnetic control, it is necessary to reduce the utilization rate of the excitation current and increase the utilization rate of the d-axis current. While ensuring the speed range, increasing the motor's starting torque can improve the utilization rate of the reluctance torque. The control system can respond quickly to sudden changes in the motor's speed to reduce the impact of sudden changes in speed on control performance. The objective function is designed as follows:

[0105]

[0106] In this objective function, the copper loss rate of the excitation winding increases with the increase of speed or torque. Field weakening control reduces the utilization rate of the excitation current, thereby improving the control system's responsiveness to speed and output torque.

[0107] (4) High-speed cruising conditions

[0108] This operating condition is in the high-speed range, so the control system requires deep field weakening control to achieve a wider speed range, which means improving the utilization of the excitation current. However, the most important control goal is to minimize copper losses during stable cruising, ensuring high stability and low losses in high-speed cruising conditions. The objective function is designed as follows:

[0109]

[0110] In this objective function, the proportion of copper loss in the armature winding increases with speed, reducing the proportion of armature current and improving the utilization of excitation current. This increases the depth of the field weakening strategy and broadens the speed range. When the speed stabilizes, the objective function distributes current based on the proportion of minimum copper loss to ensure that the motor operates with minimal losses.

[0111] Step 4: According to the d-axis current reference value i obtained in step 3 dref , q-axis current reference value i qref , excitation current reference value i fref , the d-axis current i obtained in step 2 d (k), q-axis current i q (k), the excitation current i obtained in step 1 f (k) The duty cycle t of the applied voltage vector is calculated using the prediction model of the multi-vector model predictive current control ijz , t ab Among them, the state space equation of the multi-vector model predictive current control discrete prediction model is expressed as:

[0112]

[0113] Where x(k) is the state vector, u(k) is the input vector, y(k) is the output vector, C is the identity matrix, and h(k) is the disturbance term.

[0114] The matrices and vectors in the above formula are as follows:

[0115]

[0116] Where, L f is the self-inductance of the excitation winding; T s is the sampling time; u d 、u q 、u f They are d-axis voltage, q-axis voltage, and excitation voltage respectively.

[0117] Step 5: Duty cycle t obtained from step 4 ijz , t ab , synthesize multiple sets of virtual action voltage vectors. Use the cost function to select the optimal virtual action voltage vector and output the corresponding duty cycle to the main power converter and the excitation power converter. The cost function is as follows:

[0118] min{J}=|i d (k+2)-i dref |+|i q (k+2)-i qref |+|i f (k+2)-i fref |

[0119] Where i d (k+2), i q (k+2), i f (k+2) are the predicted values ​​of d-axis current, q-axis current and excitation current at time k+2 respectively; i dref 、i qref 、i fref They are the d-axis current given value, q-axis current given value, and excitation current given value respectively.

[0120] There are two basic control strategies for existing hybrid excitation axial magnetic field permanent magnet motors, namely i d=0 control and minimum copper loss control, although they can meet the requirements of certain single working conditions, it is difficult to meet the requirements of various working conditions with complex requirements. The present invention provides a multi-mode current prediction control strategy for a hybrid excitation axial magnetic field permanent magnet motor, which analyzes in detail the control requirements of different operating conditions of electric vehicles. Taking into account the different emphasis of each working condition on the control requirements, the current distribution ratio can be adjusted online to achieve an effective current distribution control strategy for the hybrid excitation axial magnetic field permanent magnet motor under variable working conditions, thereby adapting to the control requirements of variable operating conditions. The motor drive system in the entire operating area has stronger robustness, faster dynamic response, higher operating efficiency, and better steady-state performance.

Claims

1. A multi-mode current prediction control method for a hybrid excitation axial magnetic field permanent magnet motor, characterized in that: Please follow the steps below to implement: Step 1: At the current time k, the bus voltage U is collected from the main control circuit of the hybrid excitation axial magnetic field permanent magnet motor. dc (k), excitation voltage U f (k), a, b, c phase current i a (k), i b (k), i c (k) and the excitation current i f (k) Calculate the motor rotor position angle θ and speed n based on the signal collected by the motor encoder; Step 2: Substitute the phase current i obtained in step 1 a (k), i b (k), i c (k) After Park transformation, the d-axis current i in the synchronous rotating coordinate system at time k is obtained d (k) and q-axis current i q (k); Step 3: Compare the given speed with the speed obtained in step 1. The speed deviation is passed through the PI regulator to obtain the given torque T e * , a multi-mode current control strategy is used to allocate a constant current, that is, according to the given values ​​of motor speed n and acceleration α, the actual values ​​are compared, the objective functions corresponding to four typical operating conditions are selected, and the corresponding Lagrangian constraint equations are substituted to calculate the d-axis current reference value i dref , q-axis current reference value i qref , excitation current reference value i fref ; Step 4: According to the d-axis current reference value i obtained in step 3 dref , q-axis current reference value i qref , excitation current reference value i fref , the d-axis current i obtained in step 2 d (k), q-axis current i q (k), the excitation current i obtained in step 1 f (k) The duty cycle t of the applied voltage vector is calculated using the discrete prediction model of the multi-vector model predictive current control ijz , t ab ; The state space equation of the discrete prediction model of the multi-vector model predictive current control is expressed as: Where x(k) is the state vector, u(k) is the input vector, y(k) is the output vector, C is the identity matrix, and h(k) is the disturbance term. The matrices and vectors in the above formula are as follows: Where, L f is the self-inductance of the excitation winding; T s is the sampling time; u d 、u q 、u f They are d-axis voltage, q-axis voltage, and excitation voltage respectively; The cost function J is constructed as follows: min{J}=|i d (k+2)-i dref |+|i q (k+2)-i qref |+|i f (k+2)-i fref | Where i d (k+2), i q (k+2), i f (k+2) are the predicted values ​​of d-axis current, q-axis current and excitation current at time k+2 respectively; i dref 、i qref 、i fref They are d-axis current given value, q-axis current given value, and excitation current given value respectively; Step 5: Duty cycle t obtained from step 4 ijz , t ab , synthesize multiple groups of virtual action voltage vectors, select the optimal virtual action voltage vector using the cost function, and output the corresponding duty cycle to the main power converter and the excitation power converter.

2. The multi-mode current prediction control method for a hybrid excitation axial magnetic field permanent magnet motor according to claim 1, characterized in that: The specific steps of allocating a constant current using the multi-mode current control strategy in step 3 are as follows: The low-speed range includes operating conditions such as heavy-load climbing and normal cruising. The output torque requirement in this range is high, and the motor speed does not exceed the rated speed. Therefore, the control system does not need to consider the inverter voltage limit. The excitation current is only used for magnetization control, and an extended Lagrange multiplier method is used to find the minimum value of the objective function under the torque constraint. The operating conditions in the high-speed zone include medium- and high-speed climbing conditions and high-speed cruising conditions. Since the motor speed in this zone far exceeds the rated speed, the control system needs to use field weakening control to reduce the air gap flux, thereby widening the motor's speed regulation range. The extended Lagrange multiplier method is used to solve the minimum value of the objective function, which must simultaneously meet torque and voltage constraints. The above equations are solved by referring to the solution method of the extended Lagrange multiplier method, that is, the partial derivative of each variable in the equation is solved and set to 0, and then the Newton iteration method is used to solve the partial derivative equation to obtain the optimal solution of each variable.

3. The multi-mode current prediction control method for a hybrid excitation axial magnetic field permanent magnet motor according to claim 2, characterized in that: The extended Lagrange multiplier method is used to find the minimum value of the objective function under the torque constraint: Where λ is the Lagrange multiplier; T eref is the given torque; i d 、i q 、i f are d-axis current, q-axis current, and excitation current respectively; p n is the pole pair number; ψ f is the permanent magnet flux; L d 、L q are d-axis inductance and q-axis inductance respectively; M f is the mutual inductance between the armature winding and the excitation winding; f(i d ,i q ,i f ) is the objective function designed according to the control requirements of different operating conditions.

4. The multi-mode current prediction control method for a hybrid excitation axial magnetic field permanent magnet motor according to claim 2, characterized in that: The objective function needs to satisfy the constraints of torque and voltage at the same time: Where λ1 and λ2 are Lagrange multipliers; U dc is the DC bus voltage; ω e is the electrical angular velocity; f(i d ,i q ,i f ) is the objective function designed according to the control requirements of different operating conditions.

5. The multi-mode current prediction control method for a hybrid excitation axial magnetic field permanent magnet motor according to claim 2, characterized in that: The objective function is designed as follows based on the control requirements of four typical working conditions: (1) Heavy load climbing condition Under this operating condition, the motor runs in the low-speed range, with large load torque and frequent changes. Therefore, the maximum output torque of the motor can be increased by improving the utilization rate of the excitation current in the current distribution. More importantly, it is necessary to improve the rapid response capability of the current to reduce the impact of sudden load torque on the motor control performance. The objective function is designed as follows: Where R s is the armature winding resistance; R f is the resistance value of the excitation winding; T emax is the maximum output torque; T e * is the given torque; n is the motor speed sampling feedback value; n bref is the field weakening base speed; Δn is the absolute value of the difference between the given speed and the feedback speed; α is the acceleration of the motor speed. In this objective function, the copper loss ratio in the excitation winding decreases with the increase of torque. Increasing the distribution ratio of the excitation current in the current distribution increases the response speed of the excitation current, thereby improving the control system's ability to respond to torque. (2) Normal cruising conditions This operating condition often involves acceleration and deceleration, starting and stopping speed changes; Therefore, it is necessary to increase the starting torque to improve the response speed of the motor to sudden speed changes. The overall control environment of this working condition is relatively stable, and it is necessary to maintain the minimum copper loss to improve the steady-state performance of the motor. The objective function is designed as follows: Where n ref is a given speed. In this objective function, when the motor speed changes, the proportion of copper loss in the armature winding increases with the increase of speed, improving the utilization rate of the excitation current to increase the starting torque of the motor and the responsiveness of the control system. At the same time, when the speed tends to be stable, the objective function will distribute the current according to the proportion of minimum copper loss to keep the motor running with the lowest loss; (3) Medium and high speed climbing conditions This operating condition is in the medium and high speed range. The control system needs to adopt weak magnetic control to widen the speed range of the motor. However, it will also lead to a decrease in starting torque and a slowdown in speed response. Therefore, under weak magnetic control, it is necessary to reduce the utilization rate of the excitation current and increase the utilization rate of the d-axis current. Under the premise of ensuring the speed range, the starting torque of the motor is increased and the utilization rate of the reluctance torque is increased. The control system can respond quickly to the sudden change in speed of the motor to reduce the impact of the sudden change in speed on the control performance. The objective function is designed as follows: In this objective function, the copper loss rate of the excitation winding increases with the increase of speed or torque; the field weakening control reduces the utilization rate of the excitation current, thereby improving the control system's responsiveness to speed and output torque; (4) High-speed cruising conditions This operating condition is in the high-speed range, and the control system needs to perform deep field weakening control to obtain a wider speed range, that is, to improve the utilization rate of the excitation current. However, the most important control goal is to keep the copper loss of the motor to a minimum during stable cruising, to ensure high stability and low loss of the motor under high-speed cruising conditions. The objective function is designed as follows: In this objective function, the proportion of copper loss in the armature winding increases with the increase of speed, reducing the proportion of armature current and improving the utilization rate of excitation current, thereby increasing the depth of the weak magnetic strategy to widen the speed range. When the speed tends to be stable, the objective function distributes the current according to the proportion of minimum copper loss to ensure that the motor runs with minimum loss.

Images