WO2022134751A1 - 永磁同步电机最大功率及全速域效率最优控制电流轨迹搜索方法 - Google Patents

永磁同步电机最大功率及全速域效率最优控制电流轨迹搜索方法 Download PDF

Info

Publication number
WO2022134751A1
WO2022134751A1 PCT/CN2021/123463 CN2021123463W WO2022134751A1 WO 2022134751 A1 WO2022134751 A1 WO 2022134751A1 CN 2021123463 W CN2021123463 W CN 2021123463W WO 2022134751 A1 WO2022134751 A1 WO 2022134751A1
Authority
WO
WIPO (PCT)
Prior art keywords
current
iteration
amplitude
angle
current amplitude
Prior art date
Application number
PCT/CN2021/123463
Other languages
English (en)
French (fr)
Inventor
郑萍
乔光远
刘勇
佟诚德
隋义
白金刚
Original Assignee
哈尔滨工业大学
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Priority claimed from CN202011519801.0A external-priority patent/CN112468037B/zh
Priority claimed from CN202011519810.XA external-priority patent/CN112468038B/zh
Priority claimed from CN202011519799.7A external-priority patent/CN112468036B/zh
Priority claimed from CN202011519778.5A external-priority patent/CN112468033B/zh
Priority claimed from CN202011519783.6A external-priority patent/CN112468034B/zh
Application filed by 哈尔滨工业大学 filed Critical 哈尔滨工业大学
Publication of WO2022134751A1 publication Critical patent/WO2022134751A1/zh

Links

Images

Classifications

    • HELECTRICITY
    • H02GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
    • H02PCONTROL OR REGULATION OF ELECTRIC MOTORS, ELECTRIC GENERATORS OR DYNAMO-ELECTRIC CONVERTERS; CONTROLLING TRANSFORMERS, REACTORS OR CHOKE COILS
    • H02P21/00Arrangements or methods for the control of electric machines by vector control, e.g. by control of field orientation
    • H02P21/14Estimation or adaptation of machine parameters, e.g. flux, current or voltage
    • HELECTRICITY
    • H02GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
    • H02PCONTROL OR REGULATION OF ELECTRIC MOTORS, ELECTRIC GENERATORS OR DYNAMO-ELECTRIC CONVERTERS; CONTROLLING TRANSFORMERS, REACTORS OR CHOKE COILS
    • H02P21/00Arrangements or methods for the control of electric machines by vector control, e.g. by control of field orientation
    • H02P21/14Estimation or adaptation of machine parameters, e.g. flux, current or voltage
    • H02P21/20Estimation of torque
    • HELECTRICITY
    • H02GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
    • H02PCONTROL OR REGULATION OF ELECTRIC MOTORS, ELECTRIC GENERATORS OR DYNAMO-ELECTRIC CONVERTERS; CONTROLLING TRANSFORMERS, REACTORS OR CHOKE COILS
    • H02P21/00Arrangements or methods for the control of electric machines by vector control, e.g. by control of field orientation
    • H02P21/22Current control, e.g. using a current control loop
    • HELECTRICITY
    • H02GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
    • H02PCONTROL OR REGULATION OF ELECTRIC MOTORS, ELECTRIC GENERATORS OR DYNAMO-ELECTRIC CONVERTERS; CONTROLLING TRANSFORMERS, REACTORS OR CHOKE COILS
    • H02P6/00Arrangements for controlling synchronous motors or other dynamo-electric motors using electronic commutation dependent on the rotor position; Electronic commutators therefor
    • H02P6/34Modelling or simulation for control purposes

Definitions

  • the invention relates to a current trajectory search algorithm for the maximum power control of a permanent magnet synchronous motor and the optimal control of the efficiency in the full speed domain, a nonlinear flux linkage model of the permanent magnet synchronous motor and an online permanent magnet synchronous motor based on a neural network.
  • the invention discloses a maximum power control algorithm and a full-speed domain efficiency optimal online control algorithm, which belong to the field of electric motors.
  • Rare earth permanent magnet synchronous motors have the advantages of high power factor, high power density, high efficiency, and high reliability, and are widely used in electric vehicles, rail transit, household appliances, aerospace and defense industries. Rare earth permanent magnet motors can be divided into surface-mounted permanent magnet synchronous motors and built-in permanent magnet synchronous motors according to different rotor structures. The built-in permanent magnet synchronous motors have different AC and direct axis inductance. the reluctance torque, thereby improving the torque output capability of the motor.
  • the idea of maximum power control is usually used in the built-in permanent magnet synchronous motor.
  • the maximum power control method can maximize the use of the voltage capacity, current capacity and reluctance torque of the motor system, and improve the torque output capacity of the motor under the voltage limit and current limit. Under the limit, obtain the current operating point with the maximum output power under the current and voltage limit, and improve the maximum output power of the motor.
  • the traditional maximum power control algorithm is based on the mathematical model of the permanent magnet synchronous motor. According to the torque calculation formula and the voltage calculation formula, the current trajectory of the motor under the maximum power control is calculated.
  • the traditional maximum power control algorithm considers that the parameter values of the motor's AC and direct axis inductance, permanent magnet flux linkage and other parameters are fixed. This equivalent processing method is unreasonable.
  • the traditional maximum power control algorithm uses permanent magnet flux linkage and quadrature axis inductance. , direct axis inductance and other motor parameters, these motor parameters will change with the saturation degree of the motor iron core, and the higher the load saturation degree of the motor, the more obvious the motor inductance and other parameters change, the traditional algorithm uses fixed parameter values to calculate the maximum power control The current trajectory below is obviously unreasonable, the obtained current trajectory deviates from the actual maximum power control current trajectory, and accurate maximum power control cannot be achieved.
  • the traditional full-speed domain efficiency optimal control algorithm considers that the parameters such as the inductance of the motor's AC and direct axis, permanent magnet flux linkage and other parameters are fixed. This equivalent processing method is unreasonable.
  • the traditional full-speed domain efficiency optimal control algorithm uses permanent magnets. Motor parameters such as flux linkage, quadrature-axis inductance, and direct-axis inductance will change with the saturation degree of the motor core.
  • the parameter value calculation shows that the current trajectory under the full-speed domain efficiency optimal control is obviously unreasonable, and the obtained current trajectory deviates from the actual full-speed domain efficiency optimal control current trajectory, which cannot achieve accurate full-speed domain efficiency optimal control.
  • the purpose of the present invention is to solve the problem that the traditional algorithm uses a fixed parameter value to calculate the current trajectory under the maximum power control, there is a large deviation, and the accurate maximum power control cannot be realized, and to solve the traditional full-speed domain efficiency optimal control algorithm using the fixed parameter value to calculate , there is a problem that the deviation of the current trajectory is large and the accurate full-speed domain efficiency optimal control cannot be achieved.
  • a current trajectory search method and an online control method for the maximum power and full-speed domain efficiency optimal control of a permanent magnet synchronous motor are provided.
  • the method for searching the maximum power control current trajectory of the permanent magnet synchronous motor is as follows: when the motor runs below the base speed value, under the given torque command, speed command, voltage limit and current limit, MTPA control is adopted. The current operating point with the smallest current amplitude is obtained as the current trajectory; when the motor runs above the base speed value, under the given torque command, speed command, voltage limit and current limit, the maximum power control method in the field weakening area is used to obtain the output. The current operating point with the maximum power is used as the current trajectory.
  • the maximum power control method in the field weakening area includes two search methods: the current limit circle current trajectory search method is used when the current angle ⁇ is within the range of [ ⁇ a , ⁇ b ], and when ⁇ > ⁇ b
  • ⁇ a is the field weakening current angle of the permanent magnet motor when the current amplitude reaches the current limit value under MTPA control
  • ⁇ b is the field weakening current of the permanent magnet motor when the current amplitude reaches the current limit under MTPV control horn;
  • the process of using the MTPA control method to obtain the current operating point with the smallest current amplitude includes the current angle iteration cycle step and the current amplitude iteration cycle step.
  • the current angle iteration cycle step is performed, and the current angle iteration direction is the direction in which the current amplitude decreases;
  • the current amplitude iteration loop steps are nested to determine the current amplitude corresponding to each current angle.
  • the iteration direction of the current amplitude is the direction in which the error between the given torque and the actual torque decreases.
  • the process of using the MTPV control method to obtain the maximum current operating point of the output power includes the field-weakening current angle iterative cycle step and the field-weakening current amplitude iterative cycle step. direction; in the process of current angle iteration, the current amplitude iteration loop steps are nested to determine the current amplitude and maximum speed corresponding to each current angle.
  • the iteration direction of the current amplitude is the given torque and the actual rotation speed.
  • the process of using the MTPA control method to obtain the current operating point with the smallest current amplitude includes a current angle iterative cycle step and a current amplitude iterative cycle step;
  • the current angle iteration loop steps include:
  • the current amplitude objective function values I( ⁇ k ) and I( ⁇ k ) are obtained by calling the current amplitude iterative loop;
  • step A8 Determine whether the iteration is converged: if b k -ak ⁇ L 1 , execute step A9; otherwise, return to step A2;
  • L 1 is the current angle iteration accuracy
  • A9. Determine whether the current operating point meets the requirements of current limit and voltage limit at the same time: if I( ⁇ k ) ⁇ I lim & U( ⁇ k ) ⁇ U lim , I lim is the given current limit value, and U lim is the given voltage If the limit value is reached, output the MTPA current trajectory; otherwise, re-input the torque and speed commands, and then return to step A1;
  • the current amplitude iteration loop steps include:
  • the torque error objective function f(I) presses Get, where: is a given torque, T e (I, ⁇ ) is the torque corresponding to the current angle ⁇ , and T e (I, ⁇ ) is calculated and obtained according to the non-linear load AC-direction flux linkage model of the motor; the current angle ⁇ is the current angle iterative cycle Output current angle test points ⁇ k , ⁇ k ; I is the current amplitude;
  • the specific process of the current limit circle current trajectory search mode includes:
  • the initial value of the current angle iteration is ⁇ a , and ⁇ a is the field weakening current angle when the current amplitude of the permanent magnet motor reaches the current limit value I lim under MTPA control;
  • the current angle iteration termination value is ⁇ b , and ⁇ b is the field weakening current angle when the current amplitude of the permanent magnet motor reaches the current limit I lim under MTPV control;
  • the process of using the MTPV control method to obtain the maximum current operating point of the output power includes an iterative cycle step of field weakening current angle and an iterative cycle step of field weakening current amplitude;
  • the highest speed objective function values W( ⁇ k ) and W( ⁇ k ) are obtained by calling the current amplitude iterative loop;
  • L 1 is the current angle iteration accuracy
  • the torque error objective function f(I) presses Get, where: is a given torque, T e (I, ⁇ ) is the torque corresponding to the current angle ⁇ , and the current angle ⁇ is the current angle test points ⁇ k and ⁇ k output by the current angle iteration cycle; I is the current amplitude;
  • a series of current operating points are selected equidistantly or unequally within the current limit range of the motor, including equidistant or unequal distance current amplitude series values and equidistant or unequal distance current angle series values.
  • the point spacing is determined by the saturation degree of the motor. It is necessary to ensure that the magnetic permeability of the iron core between two adjacent current operating points remains unchanged, and the iron core is treated as a linear material;
  • the load alternating and direct-axis flux linkage model that is, the nonlinear flux linkage model of the permanent magnet synchronous motor:
  • ⁇ q (I, ⁇ ) ⁇ q ( id , i q ).
  • the torque T e (I, ⁇ ) is calculated and output by the non-linear load DC-axis flux linkage model of the motor, and is obtained according to the following formula:
  • T e (I, ⁇ ) p( ⁇ d (I, ⁇ )i q - ⁇ q (I, ⁇ )i d )
  • p is the number of pole pairs of the motor
  • id is the direct axis current of the motor
  • i q is the quadrature axis current of the motor
  • ⁇ d is the direct axis flux linkage of the motor
  • ⁇ q is the quadrature axis flux linkage of the motor.
  • the maximum speed W( ⁇ ) of the motor under a given voltage limit is
  • U lim is the limit value of the given voltage.
  • the voltage amplitude U( ⁇ ) is obtained as follows:
  • w is the electrical angular velocity of the motor
  • R 1 is the motor resistance
  • the present invention also provides another technical solution: an online control method for the maximum power control of the permanent magnet synchronous motor, and the current trajectory of the permanent magnet synchronous motor at multiple operating points is obtained by using the current trajectory search method for the maximum power control of the permanent magnet synchronous motor, Taking these current trajectories as sample data, training generates a maximum power control neural network model.
  • the input of the maximum power control neural network model is the speed, torque, current limit value and voltage limit value of the motor, and the output is the current amplitude and current angle;
  • Loading the maximum power control neural network model into the DSP or FPGA controller can realize the online control of the maximum power of the permanent magnet synchronous motor, and output the current amplitude and current angle in real time according to the speed and torque of the motor to control the online maximum power operation of the motor .
  • the search method includes two parts, the MTPA control current trajectory search method in the constant torque region and the maximum power control current trajectory search method in the field weakening region.
  • the method, wherein the maximum power control current trajectory search method in the field weakening region includes the current limit circle current trajectory search and the MTPV control current trajectory search.
  • An online maximum power control algorithm based on neural network model is provided.
  • the current trajectory obtained by the maximum power control search method based on the double golden section iterative method is used as sample data to train, test and verify the neural network model, establish a neural network model, and load the maximum power control neural network model to DSP or FPGA control In the controller, the online maximum power control of the permanent magnet synchronous motor can be realized.
  • the method for searching the current trajectory for the optimal control of the full-speed domain efficiency of the permanent magnet synchronous motor is as follows: when the motor runs below the base speed value, under the given torque command, rotational speed command, voltage limit, and current limit, The MTPA control method is used to obtain the current operating point with the smallest current amplitude as the current trajectory; when the motor runs above the base speed value, under the given torque command, speed command, voltage limit and current limit, the field weakening area is used for the best efficiency The control method obtains the current operating point with the smallest current amplitude as the current trajectory;
  • the process of using the MTPA control method to obtain the current operating point with the smallest current amplitude includes the current angle iteration cycle step and the current amplitude iteration cycle step.
  • the current angle iteration cycle step is performed, and the current angle iteration direction is the direction in which the current amplitude decreases;
  • the current amplitude iteration loop steps are nested to determine the current amplitude corresponding to each current angle.
  • the iteration direction of the current amplitude is the direction in which the error between the given torque and the actual torque decreases.
  • the process of obtaining the current operating point with the smallest current amplitude by adopting the optimal control method of the field weakening region efficiency includes the field weakening current angle iteration cycle step and the current amplitude iteration cycle step.
  • the current angle iteration cycle step is performed, and the current angle iteration direction is the voltage limit The direction in which the current amplitude decreases; in the process of current angle iteration, the current amplitude iteration loop steps are nested to determine the current amplitude corresponding to each current angle, and the iteration direction of the current amplitude is a given rotation.
  • the current amplitude is considered to have converged to the minimum value, and the optimal control current trajectory of the field weakening area is output.
  • the process of using the MTPA control method to obtain the current operating point with the smallest current amplitude includes the current angle iteration loop step and the current amplitude iteration loop step:
  • the current angle iteration loop steps include:
  • the current amplitude objective function values I( ⁇ k ) and I( ⁇ k ) are obtained by calling the current amplitude iterative loop;
  • step A8 Determine whether the iteration is converged: if b k -ak ⁇ L 1 , execute step A9; otherwise, return to step A2;
  • L 1 is the current angle iteration accuracy
  • A9. Determine whether the current operating point meets the requirements of current limit and voltage limit at the same time: if I( ⁇ k ) ⁇ I lim &U( ⁇ k ) ⁇ U lim , I lim is the given current limit value, and U lim is the given voltage If the limit value is reached, output the MTPA current trajectory; otherwise, re-input the torque and speed commands, and then return to step A1;
  • the current amplitude iteration loop steps include:
  • the torque error objective function f(I) presses Get, where: is a given torque, T e (I, ⁇ ) is the torque corresponding to the current angle ⁇ , and the current angle ⁇ is the current angle test points ⁇ k and ⁇ k output by the current angle iteration cycle; I is the current amplitude;
  • the process of obtaining the current operating point with the smallest current amplitude by adopting the optimal control mode of field weakening region efficiency includes the iterative looping step of field weakening current angle and the iterative looping step of current amplitude:
  • the current amplitude objective function values I( ⁇ k ) and I( ⁇ k ) are obtained by calling the current amplitude iterative loop;
  • L 1 is the current angle iteration accuracy
  • the current amplitude iteration loop steps include:
  • the torque error objective function f(I) presses Get, where: is a given torque, T e (I, ⁇ ) is the torque corresponding to the current angle ⁇ , and the current angle ⁇ is the current angle test points ⁇ k and ⁇ k output by the current angle iteration cycle; I is the current amplitude;
  • the torque T e (I, ⁇ ) is calculated and output by the non-linear load DC-axis flux linkage model of the motor, and is obtained according to the following formula:
  • T e (I, ⁇ ) p( ⁇ d (I, ⁇ )i q - ⁇ q (I, ⁇ )i d )
  • p is the number of pole pairs of the motor
  • id is the direct axis current of the motor
  • i q is the quadrature axis current of the motor
  • ⁇ d is the direct axis flux linkage of the motor
  • ⁇ q is the quadrature axis flux linkage of the motor.
  • a series of current operating points are selected equidistantly or unequally within the current limit range of the motor, including equidistant or unequal distance current amplitude series values and equidistant or unequal distance current angle series values.
  • the point spacing is determined by the saturation degree of the motor. It is necessary to ensure that the magnetic permeability of the iron core between two adjacent current operating points remains unchanged, and the iron core is treated as a linear material;
  • the load alternating and direct-axis flux linkage model that is, the nonlinear flux linkage model of the permanent magnet synchronous motor:
  • ⁇ q (I, ⁇ ) ⁇ q ( id , i q ).
  • the voltage amplitude U( ⁇ ) is obtained as follows:
  • w is the electrical angular velocity of the motor
  • R1 is the motor resistance
  • the present invention also provides another technical solution: an on-line control method for the optimal control of the efficiency of the permanent magnet synchronous motor in the full-speed domain, using the current trajectory search method for the optimal control of the efficiency in the field-weakening region of the permanent-magnet synchronous motor to obtain multiple current work in the full-speed domain. point, including the current operating point obtained by MTPA control below the base speed value, and the current operating point obtained by the optimal control method of field weakening region efficiency above the base speed value;
  • the training Taking these current operating points as sample data, the training generates the full-speed domain efficiency optimal control neural network model.
  • the input of the full-speed domain efficiency optimal control neural network model is the speed, torque, current limit and voltage limit of the motor, and the output is Current amplitude and current angle;
  • a search method based on the double golden section iteration method for the optimal control current trajectory of the full speed domain efficiency includes two parts, the efficiency optimal control current trajectory search method in the constant torque region and the efficiency in the field weakening region.
  • An optimal online control algorithm for full-speed domain efficiency based on neural network model is provided.
  • the current trajectory obtained by the full-speed domain efficiency optimal control search method based on the double golden section iterative method is used as sample data to train, test and verify the neural network model, establish a neural network model, and use the full-speed domain efficiency optimal control neural network model. Loaded into DSP or FPGA controller, it can realize the optimal online control of permanent magnet synchronous motor full speed domain efficiency.
  • the invention is not only aimed at the conventional permanent magnet synchronous motor, but also applies to the new type permanent magnet synchronous motor, such as the adjustable magnetic flux permanent magnet synchronous motor.
  • the magnetization state of the motor can be adjusted accordingly by applying charging and demagnetizing currents in the armature windings.
  • the motor can run in multiple magnetization states, but the motor operates in each magnetization state.
  • the principle is the same as that of the conventional permanent magnet synchronous motor, so the content of the present invention is also applicable to the new permanent magnet synchronous motor.
  • Fig. 1 is the load flux linkage model after saturation demagnetization of the series-parallel permanent magnet synchronous motor, in which Fig. 1(a) is the load direct-axis flux linkage model, and Fig. 1(b) is the load quadrature-axis flux linkage model;
  • Fig. 2 is the flow chart of adopting the MTPA control method to obtain the working point below the base speed value in the maximum power and full-speed domain efficiency optimal control current trajectory search method of the present invention
  • Fig. 3 is the flow chart of adopting the current limit circle current trajectory search method to obtain the operating point below the base speed value in the maximum power control current trajectory search method of the present invention
  • Fig. 4 is the flow chart of adopting the optimal control mode of field weakening area efficiency to obtain the working point below the base speed value in the maximum power control current trajectory search method of the present invention
  • Figure 5 is the torque-speed curve and power-speed curve calculated by the formula method when the maximum power of the motor is controlled and its finite element verification;
  • Fig. 6 is the torque-speed curve and the power-speed curve and the finite element verification of the torque-speed curve and the power-speed curve when the maximum power of the motor is calculated by the trajectory search method of the present invention
  • Fig. 7 is the schematic diagram of training, testing and verification errors of the maximum power control neural network model
  • Fig. 8 is the flow chart of obtaining the operating point by adopting the optimal control mode of field weakening region efficiency below the base speed value in the full-speed domain efficiency optimal control current trajectory search method of the present invention
  • FIG. 9 is a MAP diagram of the motor MTPA control efficiency calculated by using the traditional formula method.
  • Figure 11 is a schematic diagram of training, testing and validation errors of the full-speed domain efficiency optimal control neural network model.
  • the existing technical solutions have certain deficiencies in terms of accuracy, calculation amount, and implementation speed.
  • Motor parameters such as permanent magnet flux linkage, quadrature-axis inductance, and direct-axis inductance are used in the traditional maximum power algorithm. These motor parameters will change with the saturation degree of the motor core, and the higher the load saturation degree of the motor, the motor inductance, etc. The more obvious the parameter changes, the more unreasonable the traditional algorithm uses to calculate the current trajectory under the maximum power control with fixed parameter values, and the obtained current trajectory deviates from the actual maximum power control current trajectory.
  • the present invention does not calculate parameters such as AC and direct axis inductance, permanent magnet flux linkage, etc.
  • the search method of the present invention is based on the idea of the golden section, and can obtain the maximum output power under the current and voltage limits under the given speed range, current limit, and voltage limit. current operating point for maximum power control.
  • the present invention adopts different search methods in different stages of the motor, and its purpose is to realize the maximum power control in the full speed range, which is mainly divided into two sections: the constant speed region when the motor runs below the base speed value, and the field weakening region above the base speed value , in the constant speed area below the base speed value, the MTPA control method is used to obtain the current operating point with the smallest current amplitude as the current trajectory, and the field weakening area above the base speed value is subdivided into two stages: the initial use of the current limit circular current Trajectory search method, and MTPV control method is adopted in the later stage.
  • the condition for MTPA to end in the constant speed region is that the current reaches the limit value, and the MTPA cannot continue to be effectively controlled if the current is larger. Therefore, the MTPA control mode is ended.
  • the current limit circle current trajectory search method used in the initial field weakening region needs to determine the initial value of the iteration first.
  • the initial value ⁇ a of the current angle is obtained by the MTPA control method that has been operated in the constant speed area, and the current angle end value ⁇ b is obtained by running the MTPV control method in the field weakening area first.
  • the current angle reaches ⁇ b
  • the latter will be carried out in the MTPV control mode.
  • the MTPA control method includes two iterative processes: the current angle iterative cycle step and the current amplitude iterative cycle step.
  • the MTPV control method includes two iterative processes: the field weakening current angle iterative cycle step and the field weakening current amplitude iterative cycle step. Considering the nonlinearity of the inductance and the permanent magnet flux linkage, the current amplitude is difficult to obtain directly through the torque formula, so the current amplitude (field-weakening current amplitude) is nested in the iteration process of the current angle (field-weakening current angle).
  • the calculation of torque in the iterative process of current amplitude uses the AC-direction flux linkage model of the non-linear load of the motor, taking into account the nonlinear effects of inductance and permanent magnet flux linkage, and the calculation results are accurate.
  • the motor torque, load voltage, etc. can be accurately calculated, and parameters such as inductance and permanent magnet flux linkage are no longer required.
  • the variation law of the saturation degree of the iron core under different load conditions can realize the accurate modeling of the motor.
  • the above search method is used to obtain the current trajectories of the permanent magnet synchronous motor under different magnetization states and at multiple operating points, and these current trajectories are used as sample data to train, test and verify the neural network model.
  • the input of the maximum power control neural network model is the magnetization state, speed and torque of the motor, and the output is the current amplitude and current angle (or the direct axis current and the quadrature axis current).
  • the model can not only output the corresponding operating point in the sample data.
  • the current traces can also output the current traces of operating points other than the sample data, that is, the current traces of all operating points can be output.
  • the maximum power control neural network model (which can be expressed by the functional relationship of input and output) is loaded into the DSP or FPGA controller, and the online control of the maximum power control of the permanent magnet synchronous motor can be realized.
  • Embodiment 1 The present embodiment will be described below with reference to FIGS. 1 to 6 .
  • the method for searching the maximum power control current trajectory of a permanent magnet synchronous motor described in this embodiment is as follows: when the motor is running below the base speed value, at a given Under the torque command, speed command, voltage limit, and current limit, the MTPA control method is used to obtain the current operating point with the smallest current amplitude as the current trajectory; when the motor runs above the base speed value, at the given torque command, speed Under the command, voltage limit and current limit, the maximum power control method in the field weakening area is used to obtain the current operating point with the largest output power as the current trajectory.
  • the maximum power control method in the field weakening area includes two search methods: the current angle ⁇ is in [ ⁇ a , ⁇ b ] range, the current limit circle current trajectory search method is used, and the MTPV control method is used when ⁇ > ⁇ b .
  • the process of using the MTPA control method to obtain the current operating point with the smallest current amplitude includes the current angle iteration cycle step and the current amplitude iteration cycle step.
  • the current angle iteration cycle step is performed, and the current angle iteration direction is the direction in which the current amplitude decreases;
  • the current amplitude iteration loop steps are nested to determine the current amplitude corresponding to each current angle.
  • the iteration direction of the current amplitude is the direction in which the error between the given torque and the actual torque decreases.
  • the process of using the MTPV control method to obtain the maximum current operating point of the output power includes the field-weakening current angle iterative cycle step and the field-weakening current amplitude iterative cycle step. direction; in the process of current angle iteration, the current amplitude iteration loop steps are nested to determine the current amplitude and maximum speed corresponding to each current angle.
  • the iteration direction of the current amplitude is the given torque and the actual rotation speed.
  • the current amplitude selection range is (0, 2, 4, 7)
  • the current angle selection range is (0°, 5°, 10°,7)
  • the selected current operating point spacing is determined by the degree of saturation of the motor. It is necessary to ensure that the magnetic permeability of the iron core between two adjacent current operating points remains unchanged, and the iron core can be treated as a linear material.
  • calculate the motor load AC and direct-axis flux linkage data corresponding to the selected current operating point and interpolate the obtained load AC and direct-axis flux linkage data to obtain all current operating points within the current limit range.
  • the load alternating and direct-axis flux linkage model that is, the nonlinear flux linkage model of the permanent magnet synchronous motor:
  • the electromagnetic torque and load voltage of the motor can be accurately calculated.
  • the calculation formulas of electromagnetic torque and load voltage are as follows:
  • T e (I, ⁇ ) p( ⁇ d (I, ⁇ )i q - ⁇ q (I, ⁇ )i d )
  • T e (I, ⁇ ) is the electromagnetic torque
  • p is the number of pole pairs of the motor
  • id is the direct axis current of the motor
  • i q is the quadrature axis current of the motor
  • ⁇ d is the direct axis flux linkage of the motor
  • ⁇ q is the quadrature flux linkage of the motor.
  • w is the electrical angular velocity of the motor
  • R 1 is the motor resistance
  • U lim is the limit value of the given voltage.
  • This model combines the characteristics of the permanent magnet synchronous motor that can be treated as a piecewise linear model when considering the core saturation. It only needs to calculate the load flux linkage corresponding to a small part of the current operating point within the rated operating current range of the motor, and then use the feature of piecewise linearity. The load flux linkage of all current operating points is obtained by interpolation, and it is no longer necessary to calculate parameters such as inductance and permanent magnet flux linkage.
  • the model has a small amount of calculation and a fast calculation speed, and can accurately simulate the permanent magnet synchronous motor under different magnetization states and different loads. The variation law of the saturation degree of the iron core under different circumstances can realize the accurate modeling of the motor.
  • a model example is given below: take a series-parallel magnetic circuit type permanent magnet synchronous motor with 6 poles, 45 slots, a rated speed of 2100 rpm, and a rated torque of 12.2Nm after saturation demagnetization as an example , the nonlinear flux linkage model of the motor is obtained by means of finite element simulation.
  • the direct and quadrature flux linkages of the motor at the above-mentioned 49 current operating points under the saturated demagnetization state are obtained by simulation calculation, and the other current operating points between the two adjacent current operating points are calculated.
  • the corresponding flux linkage is interpolated to obtain the direct and quadrature load flux linkages corresponding to all current operating points of the series-parallel permanent magnet synchronous motor within the current limit value range, that is, the nonlinear flux linkage model of the motor, as shown in Figure 1.
  • the MTPA current control method based on the double golden section iteration method is used to obtain the current trajectory: under the given torque command, speed command, and motor magnetization state, the current operating point with the smallest current amplitude can be obtained.
  • MTPA control is realized, as shown in Figure 2 for details.
  • the process has two iteration loops: current angle iteration and current amplitude iteration.
  • the iteration of the current angle is performed. Under the given torque command, speed command, and motor magnetization state, the current angle iteration direction is the direction of the current amplitude reduction; while the current angle iteration is performed, the current amplitude is nested.
  • the iteration of is used to determine the current amplitude corresponding to each current angle, and the iteration direction of the current amplitude is the direction in which the error between the given torque and the actual torque decreases.
  • the iteration interval of the current angle is less than the given value, it is considered that the current amplitude has converged to the minimum value, that is, the MTPA operating point.
  • the values are I( ⁇ 1 ), I( ⁇ 1 ); I( ⁇ 2 ), I( ⁇ 2 ); I( ⁇ 3 ), I( ⁇ 3 )...
  • the current amplitude is difficult to obtain directly through the torque formula, so the amplitude iteration is nested in the current angle iteration process, and the torque calculation in the amplitude iteration process uses the non-linear method.
  • the linear load flux linkage model takes into account the nonlinear effects of inductance and permanent magnet flux linkage, and the current amplitude iteration results are accurate.
  • the following describes the implementation steps of the MTPA control based on the double golden section iterative method to obtain the current trajectory: including the current angle iteration loop step and the current amplitude iteration loop step.
  • the current angle iteration loop steps include:
  • [a 1 , b 1 ] is set to be [0°, 90°], and the iteration precision is set at the same time. As the iteration process continues, when the interval length is less than the given iteration precision, the iteration is considered to converge.
  • the input of the current amplitude objective function is the current angle
  • the output of the objective function is the current amplitude under a given torque
  • the objective function values I( ⁇ k ) and I( ⁇ k ) are obtained by calling the current amplitude iterative loop;
  • ⁇ k+1 ak+1 +0.618(b k+ 1 -ak +1 ),
  • ⁇ k+1 ak+1 +0.618(b k+ 1 -ak +1 )
  • ⁇ k+1 ak+1 +0.382(b k+ 1 -ak +1 ),
  • step A8 Determine whether the iteration is converged: if b k -ak ⁇ L 1 , execute step A9; otherwise, return to step A2;
  • L 1 is the current angle iteration accuracy
  • A9. Determine whether the current operating point meets the requirements of current limit and voltage limit at the same time: if I( ⁇ k ) ⁇ I lim &U( ⁇ k ) ⁇ U lim , I lim is the given current limit value, and U lim is the given voltage If the limit value is reached, output the MTPA current trajectory; otherwise, re-input the torque and speed commands, and then return to step A1;
  • the objective function value when k+1 is also called the current amplitude iteration loop is completed, according to step A8 to judge whether the iteration has converged, if not, continue Iterative loop; if it converges and meets the current limit and voltage limit requirements of step A9, output the MTPV trajectory; if it converges but does not meet the current limit and voltage limit requirements, it proves that the deviation of the parameters input by the system is large, then re-input the torque and speed commands, Re-execute both iteration loops from the beginning.
  • the current amplitude iteration loop steps include:
  • the initial value interval of the current value is set as [0A, 12A], and the iteration accuracy is set at the same time. As the iteration process continues, when the interval length is less than the given iteration accuracy, the iteration is considered to converge. .
  • the torque error objective function f(I) presses Get, where: is a given torque, T e (I, ⁇ ) is the torque corresponding to the current angle ⁇ , the current angle ⁇ does not change during the iteration of the current amplitude, and is a certain value, and the current angle ⁇ is the current angle iteration loop output
  • the torque T e (I, ⁇ ) is calculated and output by the non-linear load DC-axis flux linkage model of the motor, and can be obtained according to the following formula:
  • T e (I, ⁇ ) p( ⁇ d (I, ⁇ )i q - ⁇ q (I, ⁇ )i d )
  • p is the number of pole pairs of the motor
  • id is the direct axis current of the motor
  • i q is the quadrature axis current of the motor
  • ⁇ d is the direct axis flux linkage of the motor
  • ⁇ q is the quadrature axis flux linkage of the motor.
  • the maximum power control current trajectory in the field weakening area Due to the limitation of the current limit and voltage limit of the motor, with the increase of the motor speed, the current trajectory of the motor when it first enters the field weakening area is mainly limited by the current limit circle. At this time, the maximum power The control current trajectory coincides with the current limit circle of the motor. As the speed continues to increase, the current trajectory of the motor is mainly limited by the motor voltage limit circle. At this time, the maximum power control current trajectory is the MTPV control current trajectory.
  • the specific process of the current limit circle current trajectory search method includes:
  • the initial value of the current angle iteration is ⁇ a , and ⁇ a is the field weakening current angle when the current amplitude of the permanent magnet motor reaches the current limit value I lim under MTPA control;
  • the current angle iteration termination value is ⁇ b , and ⁇ b is the field weakening current angle when the current amplitude of the permanent magnet motor reaches the current limit I lim under MTPV control;
  • This step is used to determine the iterative range of the current limit circle search, and the current angles ⁇ a and ⁇ b are used as the initial value and the end value of the iteration.
  • the subsequent current operating point search in the field weakening region will use the MTPV control method.
  • the MTPV current trajectory search method based on the double golden section iterative method is shown in Fig. 4. This method can obtain the given torque, voltage limit and current under the given torque command, voltage limit command and current limit command The current operating point with the maximum output power under the limit realizes MTPV control.
  • the method has two iteration loops: current angle iteration and current amplitude iteration.
  • the iteration of the current angle on the left side of the flowchart is performed: under the given torque command, voltage limit command, and current limit command, the current angle iteration direction is the direction of the maximum speed increase; while the current angle iteration is performed, the nested
  • the iteration of the current amplitude is used to determine the current amplitude and the maximum speed corresponding to each current angle.
  • the iteration direction of the current amplitude is the direction in which the error between the given torque and the actual torque decreases.
  • the output of the current amplitude iteration process The results are used in the iterative process of the current angle. When the iterative interval of the current angle is less than the given value, the iteration is considered to converge, and the motor MTPV operating point is obtained.
  • the current amplitude is difficult to obtain directly through the torque formula, so the amplitude iteration is nested in the current angle iteration process, and the torque calculation in the amplitude iteration process uses the non-linear method.
  • the linear load flux linkage model takes into account the nonlinear effects of inductance and permanent magnet flux linkage, and the current amplitude iteration results are accurate.
  • the following describes the implementation steps of the MTPV current trajectory search method based on the double golden section iterative method: including the field weakening current angle iterative loop step and the field weakening current amplitude iterative loop step.
  • [a 1 , b 1 ] is set to be [0°, 90°], and the iteration precision is set at the same time. As the iteration process continues, when the interval length is less than the given iteration precision, the iteration is considered to converge.
  • the highest speed objective function values W( ⁇ k ) and W( ⁇ k ) are obtained by calling the current amplitude iterative loop;
  • ⁇ k+1 ak+1 +0.618(b k+ 1 -ak +1 ),
  • ⁇ k+1 ak+1 +0.382(b k+ 1 -ak +1 ),
  • L 1 is the current angle iteration accuracy
  • MTPV trajectory including given torque
  • a series of operating point data can be obtained by entering different torques.
  • I lim and U lim can also be adjusted according to specific conditions.
  • step C2 it is determined which test point to calculate when k+1 is calculated, and the objective function value at k+1 is also called the current amplitude to complete the iterative loop.
  • step C8 it is judged whether the iteration has converged, if not, continue.
  • the initial value interval of the current value is set as [0C, 14C], and the iteration accuracy is set at the same time.
  • the interval length is less than the given iteration accuracy, it is considered that Iterative convergence.
  • the torque error objective function f(I) presses Get, where: is a given torque, T e (I, ⁇ ) is the torque corresponding to the current angle ⁇ , the current angle ⁇ does not change during the iteration of the current amplitude, and is a certain value, and the current angle ⁇ is the current angle iteration loop output
  • the torque T e (I, ⁇ ) is calculated and output by the non-linear load DC-axis flux linkage model of the motor, and can be obtained according to the following formula:
  • T e (I, ⁇ ) p( ⁇ d (I, ⁇ )i q - ⁇ q (I, ⁇ )i d )
  • p is the number of pole pairs of the motor
  • id is the direct axis current of the motor
  • i q is the quadrature axis current of the motor
  • ⁇ d is the direct axis flux linkage of the motor
  • ⁇ q is the quadrature axis flux linkage of the motor.
  • the maximum power search method in this embodiment includes two parts, the MTPA control current trajectory search method in the constant torque region and the maximum power control current trajectory search method in the field weakening region, wherein the maximum power control current trajectory search method in the field weakening region includes a current limit circle Current Trajectory Search and MTPV Controlled Current Trajectory Search.
  • the flow chart of the current trajectory search method of MTPA control in the constant torque area of maximum power control based on the double golden section iteration method is shown in Figure 2. This method can be used under the given torque command, speed command, voltage limit, and current limit. Obtain the current operating point with the maximum output power of the motor under the voltage limit and current limit, and realize the maximum power control in the constant torque area.
  • the flow chart of the current limit circle current trajectory search method is shown in Figure 3, and the flow chart of the maximum power control field-weakening region MTPV control current trajectory search method based on the double golden section iteration method is shown in Figure 4.
  • This method can be used in a given Under the torque command, speed command, voltage limit and current limit, obtain the current operating point with the maximum output power of the motor under the voltage limit and current limit, and realize the maximum power control in the field weakening area.
  • Embodiment 2 The present embodiment will be described below with reference to FIG. 7 .
  • the online control method for the maximum power control of the permanent magnet synchronous motor described in this embodiment is obtained by using the current trajectory search method for the maximum power control of the permanent magnet synchronous motor described in Embodiment 1 to obtain the permanent magnet synchronous motor.
  • the current trajectories of the magnetic synchronous motor at multiple operating points are used as sample data to train and generate the maximum power control neural network model.
  • the input of the maximum power control neural network model is the motor speed, torque, current limit and Voltage limit value, the output is current amplitude and current angle;
  • Loading the maximum power control neural network model into the DSP or FPGA controller can realize the online control of the maximum power of the permanent magnet synchronous motor, and output the current amplitude and current angle in real time according to the speed and torque of the motor to control the online maximum power operation of the motor .
  • the neural network training process is: using the above search method to obtain the current trajectory of the permanent magnet synchronous motor at some operating points, and using these current trajectories as sample data to train, test and verify the neural network model, when the error is less than the set value.
  • the neural network structure and the weight and bias parameters of each neuron are determined.
  • the BP algorithm is used to calculate the weight of each node along the reverse direction of the neural network calculation according to the gradient of the error between the output value of the neural network and the sample value. Adjust with the bias.
  • the weight and bias of each node are adjusted according to the error.
  • the neural network structure and the weight and bias of each neuron are adjusted according to the error.
  • the setting parameters are determined, the establishment of the maximum power control neural network model is completed, and the training, testing and verification errors of the neural network model are shown in Figure 7.
  • the model can not only output the current trajectory of the corresponding operating point in the sample data, but also output other than the sample data.
  • the current trajectory of the operating point that is, the current trajectory of all operating points can be output.
  • the neural network model has four inputs, namely voltage limit, current limit, speed and torque, and two outputs, namely direct-axis current and quadrature-axis current.
  • the neural network model uses a hidden layer, and the hidden layer uses 9 neurons.
  • the present invention does not calculate parameters such as inductance of AC and direct axes, permanent magnet flux linkage, etc.
  • the search method of the present invention is based on the idea of the golden section, and can obtain the minimum current amplitude under the given torque command, rotational speed command, voltage limit, and current limit.
  • the current operating point can achieve the optimal control of the full-speed domain efficiency.
  • the motor runs below the base speed value, it is the constant speed region, and above the base speed value is the field weakening region.
  • the MTPA control method of the present invention obtains the current operating point with the smallest current amplitude when the base speed value is below the base speed value.
  • the optimal control method of the field weakening area as the current trajectory;
  • the MTPA control method includes the current angle iteration loop step and the current amplitude iteration loop step
  • the optimal control method of field weakening region efficiency includes the iterative loop step of field weakening current angle and the iterative loop step of current amplitude.
  • the current amplitude is difficult to obtain directly through the torque formula, so the current amplitude iteration is nested in the current angle (weakening current angle) iteration process, and the current amplitude iteration process is
  • the calculation of the medium torque uses the non-linear load DC-axis flux linkage model of the motor, taking into account the nonlinear effects of inductance and permanent magnet flux linkage, and the calculation results are accurate. Using this nonlinear load flux linkage model, the motor torque, load voltage, etc. can be accurately calculated, and parameters such as inductance and permanent magnet flux linkage are no longer required.
  • the variation law of the saturation degree of the iron core under different load conditions can realize the accurate modeling of the motor.
  • the above search method is used to obtain the current trajectories of the permanent magnet synchronous motor under different magnetization states and at multiple operating points, and these current trajectories are used as sample data to train, test and verify the neural network model.
  • the input of the neural network model for optimal online control of full-speed domain efficiency is the speed, torque, voltage limit and current limit of the motor, and the output is the current amplitude and current angle (or direct-axis current and quadrature-axis current).
  • the current trajectories of the corresponding operating points in the sample data can also be outputted for the current trajectories of operating points other than the sample data, that is, the current trajectories of all operating points can be output.
  • Loading the full-speed domain efficiency optimal online control neural network model (which can be expressed by the functional relationship of input and output) into the DSP or FPGA controller can realize the full-speed domain efficiency optimal online control of the permanent magnet synchronous motor.
  • Embodiment 1 The present embodiment will be described below with reference to Fig. 1, Fig. 2 and Fig. 8-Fig. 10.
  • the current trajectory search method for the optimal control of the efficiency of the permanent magnet synchronous motor in the full-speed domain described in this embodiment is that the motor runs below the base speed value.
  • the MTPA control method is used to obtain the current operating point with the smallest current amplitude as the current trajectory; when the motor runs above the base speed value, at the given Under the torque command, speed command, voltage limit and current limit, the current operating point with the smallest current amplitude is obtained by using the optimal control method of the field weakening area as the current trajectory;
  • the process of using the MTPA control method to obtain the current operating point with the smallest current amplitude includes the current angle iteration cycle step and the current amplitude iteration cycle step.
  • the current angle iteration cycle step is performed, and the current angle iteration direction is the direction in which the current amplitude decreases;
  • the current amplitude iteration loop steps are nested to determine the current amplitude corresponding to each current angle.
  • the iteration direction of the current amplitude is the direction in which the error between the given torque and the actual torque decreases.
  • the process of obtaining the current operating point with the smallest current amplitude by adopting the optimal control method of the field weakening region efficiency includes the field weakening current angle iteration cycle step and the current amplitude iteration cycle step.
  • the current angle iteration cycle step is performed, and the current angle iteration direction is the voltage limit The direction in which the current amplitude decreases; in the process of current angle iteration, the current amplitude iteration loop steps are nested to determine the current amplitude corresponding to each current angle, and the iteration direction of the current amplitude is a given rotation.
  • the current amplitude is considered to have converged to the minimum value, and the optimal control current trajectory of the field weakening area is output.
  • the detailed steps for establishing the AC-direction flux linkage model of the non-linear load of the motor are the same as the detailed steps for establishing the AC-direction flux linkage model of the non-linear load of the motor described in the above Embodiment 1, and are not repeated here.
  • the MTPA current control method based on the double golden section iteration method obtains the current trajectory: under the given torque command, speed command, and motor magnetization state, the current operating point with the smallest current amplitude can be obtained, so as to realize the MTPA control. For details, see shown in Figure 2.
  • the process has two iteration loops: current angle iteration and current amplitude iteration.
  • the iteration of the current angle is performed.
  • the current angle iteration direction is the direction in which the current amplitude decreases; while the current angle iteration is performed, the current amplitude is nested.
  • the iteration of the value is used to determine the current amplitude corresponding to each current angle.
  • the iteration direction of the current amplitude is the direction in which the error between the given torque and the actual torque decreases.
  • the values are I( ⁇ 1 ), I( ⁇ 1 ); I( ⁇ 2 ), I( ⁇ 2 ); I( ⁇ 3 ), I( ⁇ 3 )...
  • the current amplitude is difficult to obtain directly through the torque formula, so the amplitude iteration is nested in the current angle iteration process, and the torque calculation in the amplitude iteration process uses the non-linear method.
  • the linear load flux linkage model takes into account the nonlinear effects of inductance and permanent magnet flux linkage, and the current amplitude iteration results are accurate.
  • the following describes the implementation steps of the MTPA control based on the double golden section iterative method to obtain the current trajectory: including the current angle iteration loop step and the current amplitude iteration loop step.
  • the details of the current angle iteration cycle steps are the same as the current angle iteration cycle steps A1 to A9 described in the above-mentioned first embodiment, which will not be repeated here; and the details of the current amplitude iteration cycle steps are the same as the above-mentioned embodiments.
  • the current amplitude iterative loop steps B1 to B7 described in Section 1 will not be repeated here.
  • the process has two iteration loops: field weakening current angle iteration and current amplitude iteration.
  • the field weakening current angle is iterated.
  • the current angle iteration direction is the direction in which the current amplitude decreases under the voltage limit.
  • the iteration of the nested current amplitude is used to determine the current amplitude corresponding to each current angle, and the iteration direction of the current amplitude is the direction in which the error between the given torque and the actual torque decreases.
  • the iteration interval of the current angle is less than the given value, it is considered that the current amplitude has converged to the minimum value, that is, the optimal control operating point of the field weakening region efficiency.
  • the current amplitude is difficult to obtain directly through the torque formula, so the amplitude iteration is nested in the current angle iteration process, and the torque calculation in the amplitude iteration process uses the non-linear method.
  • the linear load flux linkage model takes into account the nonlinear effects of inductance and permanent magnet flux linkage, and the current amplitude iteration results are accurate.
  • the following describes the implementation steps for obtaining the current trajectory by the optimal control of the field weakening region efficiency based on the double golden section iterative method: including the field weakening current angle iterative loop step and the current amplitude iterative loop step.
  • [a 1 , b 1 ] is set to be [0°, 90°], and the iteration precision is set at the same time. As the iteration process continues, when the interval length is less than the given iteration precision, the iteration is considered to converge.
  • the current amplitude objective function values I( ⁇ k ) and I( ⁇ k ) are obtained by calling the current amplitude iterative loop;
  • the input of the current amplitude objective function is the current angle, and the output is the current amplitude at a given torque and speed.
  • ⁇ k+1 ak+1 +0.618(b k+ 1 -ak +1 ),
  • ⁇ k+1 ak+1 +0.618(b k+ 1 -ak +1 )
  • ⁇ k+1 ak+1 +0.382(b k+ 1 -ak +1 ),
  • L 1 is the current angle iteration accuracy
  • step C2 the objective function value at k+1 is also called the current amplitude iteration loop to complete, according to step C8 to judge whether the iteration is Convergence, if not, continue the iterative cycle; if it converges and meets the current limit requirements of step C10, output the optimal control current trajectory of the field weakening region efficiency; Torque, speed command, re-execute two iterative loops from the beginning.
  • any operating point can be obtained within the full-speed domain (base speed
  • base speed The current amplitude and phase that should be applied to achieve optimal efficiency control in the constant torque region below the base speed value
  • the field weakening region above the base speed value The influence of linear factors, the calculation results are accurate.
  • the optimal control current trajectory search method in the constant torque area is based on the idea of the golden section. , obtain the current operating point with the smallest current amplitude when the motor is running in the constant torque zone, and realize the optimal control of the efficiency in the constant torque zone, that is, MTPA control; when the motor is running in the field weakening zone, if the MTPA control is continued, the motor's The load terminal voltage will exceed the voltage limit value, and the direct-axis field weakening current must be increased to reduce the motor load terminal voltage.
  • the efficiency of the field weakening area is optimally controlled.
  • the current trajectory search method is based on the idea of the golden section. Under the command, voltage limit and current limit, obtain the current operating point with the smallest current amplitude when the motor is running in the field weakening area, and realize the optimal control of the efficiency in the field weakening area.
  • the search method has a small amount of calculation and a fast calculation speed.
  • Embodiment 2 The present embodiment will be described below with reference to FIGS. 1 and 2 and FIGS. 8-11 , and the online control method for the full-speed domain efficiency optimal control of a permanent magnet synchronous motor described in this embodiment.
  • the search method described in Embodiment 1 is used to obtain the current trajectories of the permanent magnet synchronous motor at a series of operating points under different magnetization states, and these current trajectories are used as sample data to train, test and verify the neural network model.
  • the input of the neural network model for optimal control of efficiency in the full speed domain is the speed, torque, voltage limit and current limit of the motor, and the output is the current amplitude and current angle (or direct-axis current and quadrature-axis current).
  • the gradient of the error between the output value and the sample value adjusts the weight and bias of each node along the reverse direction of the neural network calculation.
  • the weight and bias of each node are According to the adjustment of the error, when the error is less than the set value, the training is completed, and the neural network structure and the weight and bias parameters of each neuron are determined.
  • the training, testing and verification errors of the neural network model are shown in Figure 11.
  • the model can not only output
  • the current trajectories of the corresponding operating points in the sample data can also be outputted for the current trajectories of operating points other than the sample data, that is, the current trajectories of all operating points can be output.
  • the neural network model has four inputs, namely motor speed, torque, voltage limit and current limit, and two outputs, namely the direct-axis current and the quadrature-axis current.
  • the neural network model uses a hidden layer. 15 neurons were used.
  • the full-speed domain efficiency optimal control neural network model (which can be expressed by the functional relationship between input and output) is loaded into the DSP or FPGA controller, and the full-speed domain efficiency optimal online control of the permanent magnet synchronous motor can be realized.

Landscapes

  • Engineering & Computer Science (AREA)
  • Power Engineering (AREA)
  • Control Of Ac Motors In General (AREA)

Abstract

本申请属于电机领域,提供了一种永磁同步电机最大功率及全速域效率最优控制电流轨迹搜索方法。本发明解决了传统算法使用固定参数值计算最大功率及全速域效率控制下的电流轨迹存在偏差大,无法实现准确的最大功率及全速域效率控制的问题,实现永磁同步电机最大功率最优在线控制和永磁同步电机全速域效率最优在线控制。

Description

永磁同步电机最大功率及全速域效率最优控制电流轨迹搜索方法 技术领域
本发明涉及一种永磁同步电机最大功率控制时以及全速域效率最优控制时的电流轨迹搜索算法,一种永磁同步电机非线性磁链模型和一种基于神经网络的永磁同步电机在线最大功率控制算法以及全速域效率最优在线控制算法,属于电机领域。
背景技术
近年来传统汽车保有量激增,造成的环境污染问题日益严重,逐步成为加剧全球变暖和温室效应的重要因素之一。同时,传统汽车使用内燃机,其能量转化率较低,且十分依赖石油等不可再生资源,环境污染和能源危机的双重压力促使传统汽车产业逐步向新能源汽车方向发展。稀土永磁同步电机具有高功率因数、高功率密度、高效率、高可靠性等优点,被广泛应用于电动汽车,轨道交通,家用电器,航空航天和国防工业等领域。稀土永磁电机按转子结构不同可以分为表贴式永磁同步电机和内置式永磁同步电机,其中内置式永磁同步电机的交、直轴电感不同,利用电感的不对称性可以产生额外的磁阻转矩,进而提高电机的转矩输出能力。
为最大程度地利用磁阻转矩,提高电机的输出转矩,实现电机在全速域的高功率运行,最大功率控制的思想通常被用于内置式永磁同步电机。采用最大功率控制方法能够最大限度地利用电机系统的电压容量、电流容量和磁阻转矩,提高电机在电压限制和电流限制下的转矩输出能力,在给定的转速范围、电流极限、电压极限下,获取电流、电压限制下输出功率最大的电流工作点,提高电机的最大输出功率。传统的最大功率控制算法基于永磁同步电机的数学模型,根据转矩计算公式和电压计算公式,计算出电机在最大功率控制下的电流轨迹。
但传统的最大功率控制算法认为电机的交直轴电感、永磁磁链等参数值固定,这种等效处理方式是不合理的,传统最大功率控制算法中用到永磁磁链、交轴电感、直轴电感等电机参数,这些电机参数会随着电机铁心饱和程度的变化而变化,且电机的负载饱和程度越高,电机电感等参数变化越明显,传统算法使用固定参数值计算最大功率控制下的电流轨迹明显不合理,得到的电流轨迹与实际最大功率控制电流轨迹有偏差,无法实现准确的最大功率控制。
且传统的全速域效率最优控制算法认为电机的交直轴电感、永磁磁链等参数值固定,这种等效处理方式是不合理的,传统全速域效率最优控制算法中用到永磁磁链、交轴电感、直轴电感等电机参数,这些电机参数会随着电机铁心饱和程度的变化而变化,且电机的负载饱和程度越高,电机电感等参数变化越明显,传统算法使用固定参数值计算全速域效率最优控制下的电流轨迹明显不合理,得到的电流轨迹与实际全速域效率最优控制电流轨迹有偏差,无法实现准确的全速域效率最优控制。
发明内容
本发明目的是为了解决传统算法使用固定参数值计算最大功率控制下的电流轨迹存在偏差大,无法实现准确的最大功率控制的问题,以及解决传统的全速域效率最优控制算法使用固定参数值计算,存在电流轨迹偏差大,无法实现准确的全速域效率最优控制的问题,提供了一种永磁同步电机最大功率及全速域效率最优控制电流轨迹搜索方法和在线控制方法。
本发明所述永磁同步电机最大功率控制电流轨迹搜索方法,该方法为:电机运行在基速值以下时,在给定的转矩指令、转速指令、电压极限、电流极限下,采用MTPA控制方式获取电流幅值最小的电流工作点作为电流轨迹;电机运行在基速值以上时,在给定的转矩指令、转速指令、电压极限、电流极限下,采用弱磁区最大功率控制方式获取输出功率最大的电流工作点作为电流轨迹,弱磁区最大功率控制方式包括两种搜索方式:电流角θ在[θ ab]范围内采用电流极限圆电流轨迹搜索方式,在θ>θ b时采用MTPV控制方式,θ a为永磁电机在MTPA控制下电流幅值达到电流极限值时的弱磁电流角,θ b为永磁电机在MTPV控制下电流幅值达到电流极限时的弱磁电流角;
采用MTPA控制方式获取电流幅值最小的电流工作点的过程包括电流角迭代循环步骤和电流幅值迭代循环步骤,首先进行电流角迭代循环步骤,电流角迭代方向为电流幅值减小的方向;在进行电流角迭代过程中,嵌套电流幅值迭代循环步骤,用以确定每个电流角对应的电流幅值,电流幅值的迭代方向为给定转矩与实际转矩误差减小的方向,当电流角的迭代区间小于给定电流角迭代精度,认为电流幅值已经收敛至最小值,输出MTPA电流轨迹作为最大功率控制电流轨迹;
采用MTPV控制方式获取输出功率最大电流工作点的过程包括弱磁电流角迭代循环步骤和弱磁电流幅值迭代循环步骤,首先进行弱磁电流角迭代循环步骤,电流角迭代方向为最高转速增加的方向;在进行电流角迭代的过程中,嵌套电流幅值迭代循环步骤,用以确定每个电流角对应的电流幅值及最高转速,电流幅值的迭代方向为给定转矩与实际转矩误差减小的方向,当电流角的迭代区间小于给定电流角迭代精度,认为电机转速已经收敛至最大值,电机在电压限制下的输出功率收敛至最大值,输出MTPV电流轨迹作为最大功率控制电流轨迹。
优选地,采用MTPA控制方式获取电流幅值最小的电流工作点的过程包括电流角迭代循环步骤和电流幅值迭代循环步骤;
电流角迭代循环步骤包括:
A1、初始化电流角初值区间[a 1,b 1],并计算电流角试探点初值λ 1、β 1
λ 1=a 1+0.382(b 1-a 1)、β 1=a 1+0.618(b 1-a 1);
A2、判断两电流角试探点处电流幅值目标函数值I(λ k)和I(β k)是否存在关系I(λ k)>I(β k),电流角迭代次数k=1,2,3...判断结果为是,执行步骤A3;判断结果为否执行步骤A5;
电流幅值目标函数值I(λ k)和I(β k)通过调用电流幅值迭代循环获取;
A3、令a k+1=λ k,b k+1=b k,λ k+1=β k,β k+1=a k+1+0.618(b k+1-a k+1),
A4、调用电流幅值迭代循环获取电流幅值目标函数值I(β k+1),然后执行步骤A7;
A5、令a k+1=a k,b k+1=β k,β k+1=λ k,λ k+1=a k+1+0.382(b k+1-a k+1),
A6、调用电流幅值迭代循环获取电流幅值目标函数值I(λ k+1),然后执行步骤A7;
A7、令k=k+1;
A8、判断迭代是否收敛:若b k-a k<L 1,执行步骤A9;否则,返回步骤A2;
其中L 1为电流角迭代精度;
A9、判断电流工作点是否同时满足电流极限与电压极限的要求:若I(λ k)≤I lim&U(λ k)≤U lim,I lim为给定电流极限值,U lim为给定电压极限值,输出MTPA电流轨迹;否则,重新输入转矩、转速指令,再返回执行步骤A1;
MTPA电流轨迹为:电流幅值I=I(λ k)、电流角θ=λ k
电流幅值迭代循环步骤包括:
B1、初始化电流幅值的初值区间:[c 1,d 1],并计算电流幅值试探点初值μ 1、v 1
μ 1=c 1+0.382(d 1-c 1)、v 1=c 1+0.618(d 1-c 1);
B2、计算两电流幅值试探点处的转矩误差目标函数值:f(μ 1)、f(v 1),
转矩误差目标函数f(I)按
Figure PCTCN2021123463-appb-000001
获取,其中:
Figure PCTCN2021123463-appb-000002
为给定转矩,T e(I,θ)为电流角θ对应的转矩,T e(I,θ)根据电机非线性负载交直轴磁链模型计算获取;电流角θ为电流角迭代循环输出的电流角试探点λ k、β k;I为电流幅值;
B3、判断两电流幅值试探点处转矩误差目标函数值f(μ h)和f(ν h)是否存在关系f(μ h)>f(ν h),电流幅值迭代次数h=1,2,3…判断结果为是,执行步骤B4;判断结果为否执行步骤B5;
B4、令c h+1=μ h,d h+1=d h,μ h+1=v h,v h+1=c h+1+0.618(d h+1-c h+1),
计算目标函数值f(v h+1),然后步骤B6;
B5、令c h+1=c h,d h+1=v h,v h+1=μ h,μ h+1=c h+1+0.382(d h+1-c h+1),
计算目标函数值f(μ h+1),然后步骤B6;
B6、令h=h+1,
B7、判断迭代是否收敛:若d h-c h<L 2,输出给定电流角对应的电流幅值I(θ)、电压幅值U(θ),输出结果用于电流角的迭代搜索过程;否则,返回步骤B3;其中L 2为电流幅值迭代精度。
优选地,电流极限圆电流轨迹搜索方式的具体过程包括:
E1、电流极限圆电流轨迹搜索的初始化:
电流角迭代初始值为θ a,θ a为永磁电机在MTPA控制下电流幅值达到电流极限值I lim时的弱磁电流角;
电流角迭代终止值为θ b,θ b为永磁电机在MTPV控制下电流幅值达到电流极限I lim时的弱磁电流角;
E2、根据电机非线性负载交直轴磁链模型计算转矩T e(I,θ)和最高转速W(θ),并输出沿电流极限圆工作点轨迹I,θ,T(θ),W(θ):
I=I lim
θ=θ s,迭代次数s=1,2,3…,θ 1=θ a
T(θ)=T(I,θ s),
W(θ)=W(I,θ s,U lim),
E3、令θ s+1=θ s+Δθ,Δθ为迭代步进角度增幅;
E4、令s=s+1;
E5、判断迭代是否收敛:若θ s<θ b,返回执行步骤E2;否则结束迭代循环。
优选地,采用MTPV控制方式获取输出功率最大电流工作点的过程包括弱磁电流角迭代循环步骤和弱磁电流幅值迭代循环步骤;
弱磁电流角迭代循环步骤包括:
C1、初始化电流角初值区间[a 1,b 1],并计算电流角试探点初值λ 1、β 1
λ 1=a 1+0.382(b 1-a 1)、β 1=a 1+0.618(b 1-a 1);
C2、判断两电流角试探点处最高转速目标函数值W(λ k)和W(β k)是否存在关系W(λ k)<W(β k),电流角迭代次数k=1,2,3…判断结果为是,执行步骤C3;判断结果为否执行步骤C5;
最高转速目标函数值W(λ k)和W(β k)通过调用电流幅值迭代循环获取;
C3、令a k+1=λ k,b k+1=b k,λ k+1=β k,β k+1=a k+1+0.618(b k+1-a k+1),
C4、调用电流幅值迭代循环获取最高转速目标函数值W(β k+1),然后执行步骤C7;
C5、令a k+1=a k,b k+1=β k,β k+1=λ k,λ k+1=a k+1+0.382(b k+1-a k+1),
C6、调用电流幅值迭代循环获取最高转速目标函数值W(λ k+1),然后执行步骤C7;
C7、令k=k+1;
C8、判断迭代是否收敛:若b k-a k<L 1,执行步骤C9;否则,返回步骤C2;
其中L 1为电流角迭代精度;
C9、判断电流工作点是否满足电流极限的要求:若I(λ k)≤I lim,I lim为给定电流极限值,输出MTPV轨迹;否则,重新输入转矩指令,再返回执行步骤C1;
MTPV轨迹包括给定转矩
Figure PCTCN2021123463-appb-000003
给定电压极限和电流极限下的电机最高转速w=W(θ),电流幅值I=I(λ k)和电流角θ=λ k
弱磁电流幅值迭代循环步骤包括:
D1、初始化电流幅值的初值区间:[c 1,d 1],并计算电流幅值试探点初值μ 1、v 1
μ 1=c 1+0.382(d 1-c 1)、v 1=c 1+0.618(d 1-c 1);
D2、计算两电流幅值试探点处的转矩误差目标函数值:f(μ 1)、f(v 1),
转矩误差目标函数f(I)按
Figure PCTCN2021123463-appb-000004
获取,其中:
Figure PCTCN2021123463-appb-000005
为给定转矩,T e(I,θ)为电流角θ对应的转矩,电流角θ为电流角迭代循环输出的电流角试探点λ k、β k;I为电流幅值;
D3、判断两电流幅值试探点处转矩误差目标函数值f(μ h)和f(ν h)是否存在关系f(μ h)>f(ν h),电流幅值迭代次数h=1,2,3…判断结果为是,执行步骤D4;判断结果为否执行步骤D5;
D4、令c h+1=μ h,d h+1=d h,μ h+1=v h,v h+1=c h+1+0.618(d h+1-c h+1),
计算目标函数值f(v h+1),然后步骤D6;
D5、令c h+1=c h,d h+1=v h,v h+1=μ h,μ h+1=c h+1+0.382(d h+1-c h+1),计算目标函数值f(μ h+1),然后步骤D6;
D6、令h=h+1,
D7、判断迭代是否收敛:若d h-c h<L 2,输出给定电流角对应的电流幅值I(θ)、计算并输出给定转矩和给定电压极限下的电机最高转速W(θ),输出结果用于电流角的迭代搜索过程;否则,返回步骤D3;其中L 2为电流幅值迭代精度。
优选地,电机非线性负载交直轴磁链模型的建立过程:
在电机的电流极限范围内等距或不等距的选取一系列电流工作点,包括等距或不等距电流幅值系列值及等距或不等距电流角系列值,所选取的电流工作点间距由电机的饱和程度决定,需要保证相邻两电流工作点之间的铁心磁导率保持不变,铁心按线性材料处理;
采用仿真或实验的方式,计算所选取的电流工作点对应的电机负载交、直轴磁链数据,并将得到的负载交、直轴磁链数据进行插值,得到电流极限范围内所有电流工作点的负载交、直轴磁链模型,即永磁同步电机的非线性磁链模型:
ψ d(I,θ)=ψ d(i d,i q)
ψ q(I,θ)=ψ q(i d,i q)。
优选地,转矩T e(I,θ)由电机非线性负载交直轴磁链模型计算输出,按如下公式获取:
T e(I,θ)=p(ψ d(I,θ)i qq(I,θ)i d)
其中,p为电机极对数,i d为电机的直轴电流,i q为电机的交轴电流,ψ d为电机的直轴磁链,ψ q为电机的交轴磁链。
优选地,给定电压极限下的电机最高转速W(θ)按
Figure PCTCN2021123463-appb-000006
获取,
式中:U lim为给定电压极限值。
优选地,电压幅值U(θ)按下式获取:
Figure PCTCN2021123463-appb-000007
其中直轴电压
Figure PCTCN2021123463-appb-000008
交轴电压
Figure PCTCN2021123463-appb-000009
w为电机的电角速度,R 1为电机电阻。
本发明还提供另一个技术方案:永磁同步电机最大功率控制在线控制方法,采用所述的永磁同步电机最大功率控制电流轨迹搜索方法得到永磁同步电机在多个工作点下的电流轨迹,将这些电流轨迹作为样本数据,训练生成最大功率控制神经网络模型,最大功率控制神经网络模型的输入为电机的转速、转矩、电流极限值和电压极限值,输出为电流幅值与电流角;
将最大功率控制神经网络模型加载至DSP或FPGA控制器中,可以实现永磁同步电机最大功率在线控制,根据电机的转速和转矩实时输出电流幅值与电流角用于控制电机在线最大功率运行。
永磁同步电机最大功率控制电流轨迹搜索方法的有益效果为:
(1)提供了一种充分考虑电机非线性的负载磁链模型,充分考虑了不同磁化状态下、不同负载情况下铁心饱和等非线性因素对电机模型的影响规律,可以准确模拟电机在不同磁化状态下、不同负载情况下的非线性特性,不需要计算电感、永磁磁链等参数,可以准确计算电机转矩、负载电压等。
(2)提供了一种基于双黄金分割迭代法的最大功率控制电流轨迹搜索方法,该搜索方法包括两部分,恒转矩区的MTPA控制电流轨迹搜索方法和弱磁区的最大功率控制电流轨迹搜索方法,其中弱磁区最大功率控制电流轨迹搜索方法包括电流极限圆电流轨迹搜索和MTPV控制电流轨迹搜索。利用电机的负载磁链模型,搜索过程迭代收敛速度快,计算量小,可以快速、准确地实现永磁同步电机最大功率控制,提高电机运行性能。
(3)提供了一种基于神经网络模型的在线最大功率控制算法。将基于双黄金分割迭代法的最大功率控制搜索方法得到的电流轨迹作为样本数据,对神经网络模型进行训练、测试与验证,建立神经网络模型,将最大功率控制神经网络模型加载至DSP或FPGA控制器中,可以实现永磁同步电机在线最大功率控制。
本发明所述永磁同步电机全速域效率最优控制电流轨迹搜索方法,该方法为:电机运行在基速值以下时,在给定的转矩指令、转速指令、电压极限、电流极限下,采用MTPA控制方式获取电流幅值最小的电流工作点作为电流轨迹;电机运行在基速值以上时,在给定的转矩指令、转速指令、电压极限、电流极限下,采用弱磁区效率最优控制方式获取电流幅值最小的电流工作点作为电流轨迹;
采用MTPA控制方式获取电流幅值最小的电流工作点的过程包括电流角迭代循环步骤和电流幅值迭代循环步骤,首先进行电流角迭代循环步骤,电流角迭代方向为电流幅值减小的方向;在进行电流角迭代过程中,嵌套电流幅值迭代循环步骤,用以确定每个电流角对应的电流幅值,电流幅值的迭代方向为给定转矩与实际转矩误差减小的方向,当电流角的迭代区间小于给定电流角迭代精度,认为电流幅值已经收敛至最小值,输出MTPA电流轨迹;
采用弱磁区效率最优控制方式获取电流幅值最小的电流工作点的过程包括弱磁电流角迭代循环步骤和电流幅值迭代循环步骤,首先进行电流角迭代循环步骤,电流角迭代方向为电压极限下电流幅值减小的方向;在进行电流角迭代的过程中,嵌套电流幅值迭代循环步骤,用以确定每个电流角对应的电流幅值,电流幅值的迭代方向为给定转矩与实际转矩误差减小的方向,当电流角的迭代区间小于给定电流角迭代精度,认为电流幅值已经收敛至最小值,输出弱磁区效率最优控制电流轨迹。
优选地,采用MTPA控制方式获取电流幅值最小的电流工作点的过程包括电流角迭代循环步骤和电流幅值迭代循环步骤:
电流角迭代循环步骤包括:
A1、初始化电流角初值区间[a 1,b 1],并计算电流角试探点初值λ 1、β 1
λ 1=a 1+0.382(b 1-a 1)、β 1=a 1+0.618(b 1-a 1);
A2、判断两电流角试探点处电流幅值目标函数值I(λ k)和I(β k)是否存在关系I(λ k)>I(β k),电流角迭代次数k=1,2,3...判断结果为是,执行步骤A3;判断结果为否执行步骤A5;
电流幅值目标函数值I(λ k)和I(β k)通过调用电流幅值迭代循环获取;
A3、令a k+1=λ k,b k+1=b k,λ k+1=β k,β k+1=a k+1+0.618(b k+1-a k+1),
A4、调用电流幅值迭代循环获取电流幅值目标函数值I(β k+1),然后执行步骤A7;
A5、令a k+1=a k,b k+1=β k,β k+1=λ k,λ k+1=a k+1+0.382(b k+1-a k+1),
A6、调用电流幅值迭代循环获取电流幅值目标函数值I(λ k+1),然后执行步骤A7;
A7、令k=k+1;
A8、判断迭代是否收敛:若b k-a k<L 1,执行步骤A9;否则,返回步骤A2;
其中L 1为电流角迭代精度;
A9、判断电流工作点是否同时满足电流极限与电压极限的要求:若I(λ k)≤I lim&U(λ k)≤U lim,I lim为给定电流极限值,U lim为给定电压极限值,输出MTPA电流轨迹;否则,重新输入转矩、转速指令,再返回执行步骤A1;
电流幅值迭代循环步骤包括:
B1、初始化电流幅值的初值区间:[c 1,d 1],并计算电流幅值试探点初值μ 1、v 1
μ 1=c 1+0.382(d 1-c 1)、v 1=c 1+0.618(d 1-c 1);
B2、计算两电流幅值试探点处的转矩误差目标函数值:f(μ 1)、f(v 1),
转矩误差目标函数f(I)按
Figure PCTCN2021123463-appb-000010
获取,其中:
Figure PCTCN2021123463-appb-000011
为给定转矩,T e(I,θ)为电流角θ对应的转矩,电流角θ为电流角迭代循环输出的电流角试探点λ k、β k;I为电流幅值;
B3、判断两电流幅值试探点处转矩误差目标函数值f(μ h)和f(ν h)是否存在关系f(μ h)>f(ν h),电流幅值迭代次数h=1,2,3…判断结果为是,执行步骤B4;判断结果为否执行步骤B5;
B4、令c h+1=μ h,d h+1=d h,μ h+1=v h,v h+1=c h+1+0.618(d h+1-c h+1),
计算目标函数值f(v h+1),然后步骤B6;
B5、令c h+1=c h,d h+1=v h,v h+1=μ h,μ h+1=c h+1+0.382(d h+1-c h+1),
计算目标函数值f(μ h+1),然后步骤B6;
B6、令h=h+1,
B7、判断迭代是否收敛:若d h-c h<L 2,输出给定电流角对应的电流幅值I(θ)、电压幅值U(θ),输出结果用于电流角的迭代搜索过程;否则,返回步骤B3;其中L 2为电流幅值迭代精度。
优选地,采用弱磁区效率最优控制方式获取电流幅值最小的电流工作点的过程包括弱磁电流角迭代循环步骤和电流幅值迭代循环步骤:
弱磁电流角迭代循环步骤包括:
C1、初始化电流角初值区间[a 1,b 1],并计算电流角试探点初值λ 1、β 1
λ 1=a 1+0.382(b 1-a 1)、β 1=a 1+0.618(b 1-a 1);
C2、判断负载电压目标函数值U(β k)和电压极限值U lim的大小关系,若U(β k)>U lim,执行步骤C6;否则,执行步骤C3;
负载电压目标函数值U(β k)通过调用电流幅值迭代循环获取,电流角迭代次数k=1,2,3…;
C3、判断两电流角试探点处电流幅值目标函数值I(λ k)和I(β k)是否存在关系I(λ k)>I(β k),
判断结果为是,执行步骤C4;判断结果为否执行步骤C6;
电流幅值目标函数值I(λ k)和I(β k)通过调用电流幅值迭代循环获取;
C4、令a k+1=λ k,b k+1=b k,λ k+1=β k,β k+1=a k+1+0.618(b k+1-a k+1),
C5、调用电流幅值迭代循环获取电流幅值目标函数值I(β k+1),然后执行步骤C8;
C6、令a k+1=a k,b k+1=β k,β k+1=λ k,λ k+1=a k+1+0.382(b k+1-a k+1),
C7、调用电流幅值迭代循环获取电流幅值目标函数值I(λ k+1),然后执行步骤C8;
C8、令k=k+1;
C9、判断迭代是否收敛:若b k-a k<L 1,执行步骤C10;否则,返回步骤C2;
其中L 1为电流角迭代精度;
C10、判断电流工作点是否同时满足电流极限的要求:若I(λ k)≤I lim,U lim为给定电流极限值,输出弱磁区效率最优控制电流轨迹;否则,重新输入转矩、转速指令,再返回执行步骤C1;
电流幅值迭代循环步骤包括:
B1、初始化电流幅值的初值区间:[c 1,d 1],并计算电流幅值试探点初值μ 1、v 1
μ 1=c 1+0.382(d 1-c 1)、v 1=c 1+0.618(d 1-c 1);
B2、计算两电流幅值试探点处的转矩误差目标函数值:f(μ 1)、f(v 1),
转矩误差目标函数f(I)按
Figure PCTCN2021123463-appb-000012
获取,其中:
Figure PCTCN2021123463-appb-000013
为给定转矩,T e(I,θ)为电流角θ对应的转矩,电流角θ为电流角迭代循环输出的电流角试探点λ k、β k;I为电流幅值;
B3、判断两电流幅值试探点处转矩误差目标函数值f(μ h)和f(ν h)是否存在关系f(μ h)>f(ν h),电流幅值迭代次数h=1,2,3…判断结果为是,执行步骤B4;判断结果为否执行步骤B5;
B4、令c h+1=μ h,d h+1=d h,μ h+1=v h,v h+1=c h+1+0.618(d h+1-c h+1),
计算目标函数值f(v h+1),然后步骤B6;
B5、令c h+1=c h,d h+1=v h,v h+1=μ h,μ h+1=c h+1+0.382(d h+1-c h+1),
计算目标函数值f(μ h+1),然后步骤B6;
B6、令h=h+1,
B7、判断迭代是否收敛:若d h-c h<L 2,输出给定电流角对应的电流幅值I(θ)、电压幅值U(θ),输出结果用于电流角的迭代搜索过程;否则,返回步骤B3;其中L 2为电流幅值迭代精度。
优选地,电流轨迹为:电流幅值I=I(λ k)、电流角θ=λ k
优选地,转矩T e(I,θ)由电机非线性负载交直轴磁链模型计算输出,按如下公式获取:
T e(I,θ)=p(ψ d(I,θ)i qq(I,θ)i d)
其中,p为电机极对数,i d为电机的直轴电流,i q为电机的交轴电流,ψ d为电机的直轴磁链,ψ q为电机的交轴磁链。
优选地,电机非线性负载交直轴磁链模型的建立过程:
在电机的电流极限范围内等距或不等距的选取一系列电流工作点,包括等距或不等距电流幅值系列值及等距或不等距电流角系列值,所选取的电流工作点间距由电机的饱和程度决定,需要保证相邻两电流工作点之间的铁心磁导率保持不变,铁心按线性材料处理;
采用仿真或实验的方式,计算所选取的电流工作点对应的电机负载交、直轴磁链数据,并将得到的负载交、直轴磁链数据进行插值,得到电流极限范围内所有电流工作点的负载交、直轴磁链模型,即永磁同步电机的非线性磁链模型:
ψ d(I,θ)=ψ d(i d,i q)
ψ q(I,θ)=ψ q(i d,i q)。
优选地,电压幅值U(θ)按下式获取:
Figure PCTCN2021123463-appb-000014
其中,直轴电压
Figure PCTCN2021123463-appb-000015
交轴电压
Figure PCTCN2021123463-appb-000016
w为电机的电角速度,R 1为电机电阻。
本发明还提供另一个技术方案:永磁同步电机全速域效率最优控制在线控制方法,采用所述永磁同步电机弱磁区效率最优控制电流轨迹搜索方法获取全速域范围内的多个电流工作点,包括基速值以下采用MTPA控制方式获取的电流工作点,和基速值以上采用弱磁区效率最优控制方式获取的电流工作点;
将这些电流工作点作为样本数据,训练生成全速域效率最优控制神经网络模型,全速域效率最优控制神经网络模型的输入为电机的转速、转矩、电流极限值和电压极限值,输出为电流幅值与电流角;
将全速域效率最优控制神经网络模型加载至DSP或FPGA控制器中,可以实现永磁同步电机在全速域范围内效率最优在线 控制,根据电机的转速和转矩实时输出电流幅值与电流角用于控制电机运行。
本发明的有益效果:
(1)提供了一种充分考虑电机非线性的负载磁链模型,充分考虑了不同磁化状态下、不同负载情况下铁心饱和等非线性因素对电机模型的影响规律,可以准确模拟电机在不同磁化状态下、不同负载情况下的非线性特性,不需要计算电感、永磁磁链等参数,可以准确计算电机转矩、负载电压等。
(2)提供了一种基于双黄金分割迭代法的全速域效率最优控制电流轨迹搜索方法,该搜索方法包括两部分,恒转矩区的效率最优控制电流轨迹搜索方法和弱磁区的效率最优控制电流轨迹搜索方法,每个搜索方法具有两个迭代循环:弱磁电流角迭代和电流幅值迭代。利用电机的负载磁链模型,搜索过程迭代收敛速度快,计算量小,可以快速、准确地实现永磁同步电机全速域效率最优控制,提高电机运行性能。
(3)提供了一种基于神经网络模型的全速域效率最优在线控制算法。将基于双黄金分割迭代法的全速域效率最优控制搜索方法得到的电流轨迹作为样本数据,对神经网络模型进行训练、测试与验证,建立神经网络模型,将全速域效率最优控制神经网络模型加载至DSP或FPGA控制器中,可以实现永磁同步电机全速域效率最优在线控制。
本发明不仅针对常规永磁同步电机,对新型永磁同步电机,如可调磁通永磁同步电机等同样适用,可调磁通永磁同步电机结构与常规永磁同步电机结构相似,由于采用了低矫顽力永磁体,电机的磁化状态可以通过在电枢绕组中施加充、去磁电流进行相应地调整,电机可以运行在多个磁化状态下,但电机在每个磁化状态下的运行原理与常规永磁同步电机一致,所以本发明的内容同样适用于新型永磁同步电机。
附图说明
图1是串并联永磁同步电机饱和去磁后的负载磁链模型,其中图1(a)是负载直轴磁链模型,图1(b)是负载交轴磁链模型;
图2是本发明最大功率及全速域效率最优控制电流轨迹搜索方法中基速值以下采用MTPA控制方式获取工作点的流程图;
图3是本发明最大功率控制电流轨迹搜索方法中基速值以下采用电流极限圆电流轨迹搜索方式获取工作点的流程图;
图4是本发明最大功率控制电流轨迹搜索方法中基速值以下采用弱磁区效率最优控制方式获取工作点的流程图;
图5是公式法计算得到的电机最大功率控制时的转矩-转速曲线和功率-转速曲线及其有限元验证;
图6是采用本发明轨迹搜索方法计算得到的电机最大功率控制时的转矩-转速曲线和功率-转速曲线及其有限元验证;
图7是最大功率控制神经网络模型的训练、测试与验证误差的示意图;
图8是本发明全速域效率最优控制电流轨迹搜索方法中基速值以下采用弱磁区效率最优控制方式获取工作点的流程图;
图9是使用传统公式法计算得到的电机MTPA控制效率MAP图;
图10是采用本发明轨迹搜索方法计算得到的电机全速域效率最优控制效率MAP图;
图11是全速域效率最优控制神经网络模型的训练、测试与验证误差的示意图。
具体实施方式
现有的技术方案,如公式法、查表法等,在准确性、计算量、实施速度等方面具有一定的不足。传统最大功率算法中用到永磁磁链、交轴电感、直轴电感等电机参数,这些电机参数会随着电机铁心饱和程度的变化而变化,且电机的负载饱和程度越高,电机电感等参数变化越明显,传统算法使用固定参数值计算最大功率控制下的电流轨迹明显不合理,得到的电流轨迹与实际最大功率控制电流轨迹有偏差。
实施例一
本发明不计算交直轴电感、永磁磁链等参数,本发明搜索方法基于黄金分割的思想,可以在给定的转速范围、电流极限、电压极限下,获取电流、电压限制下输出功率最大的电流工作点,实现最大功率控制。本发明在电机不同阶段采用不同的搜索方式,其目的是在全速域范围内实现最大功率控制,主要分成两段:机运行在基速值以下为恒转速区域,基速值以上为弱磁区域,在基速值以下的恒转速区域采用MTPA控制方式获取电流幅值最小的电流工作点作为电流轨迹,在基速值以上的弱磁区域又细分为两个阶段:初期采用电流极限圆电流轨迹搜索方式,后期采用MTPV控制方式。在恒转速区域MTPA结束的条件是电流达到极限值,电流再大MTPA将不能再继续有效控制,因此结束MTPA控制方式,弱磁区域初期采用的电流极限圆电流轨迹搜索方式需要先确定迭代初始值和终止值两个参数,其中电流角初始值θ a通过恒转速区域运行过的MTPA控制方式获取,电流角终止值θ b则通过在弱磁区先运行MTPV控制方式获取,当电流角达到θ b时结束弱磁区域初期阶段的电流极限圆搜索,后面的将采用MTPV控制方式进行。
MTPA控制方式包括两个迭代过程:电流角迭代循环步骤和电流幅值迭代循环步骤,MTPV控制方式包括两个迭代过程:弱磁电流角迭代循环步骤和弱磁电流幅值迭代循环步骤。考虑到电感和永磁磁链的非线性,电流幅值难以通过转矩公式直接求得,所以在电流角(弱磁电流角)迭代过程中嵌套了电流幅值(弱磁电流幅值)迭代,电流幅值迭代过程中转矩的计算使用了电机非线性负载交直轴磁链模型,考虑了电感和永磁磁链非线性的影响,计算结果准确。使用该非线性负载磁链模型可以准确的计算电机转矩、负载电压等,不再需要计算电感,永磁磁链等参数,计算量小,计算速度快,能够准确模拟永磁同步电机不同磁化状 态下、不同负载情况下铁心饱和程度的变化规律,实现电机的准确建模。利用上述搜索方法得到永磁同步电机在不同充磁状态下,多个工作点下的电流轨迹,将这些电流轨迹作为样本数据,对神经网络模型进行训练、测试与验证。最大功率控制神经网络模型的输入为电机的磁化状态、转速和转矩,输出为电流幅值与电流角(或直轴电流与交轴电流),该模型不仅可以输出样本数据中相应工作点的电流轨迹,还可以输出样本数据以外的工作点的电流轨迹,即可以输出所有工作点的电流轨迹。将最大功率控制神经网络模型(可以用输入输出的函数关系来表达)加载至DSP或FPGA控制器中,可以实现永磁同步电机最大功率控制在线控制。
具体实施方式一:下面结合图1~图6说明本实施方式,本实施方式所述永磁同步电机最大功率控制电流轨迹搜索方法,该方法为:电机运行在基速值以下时,在给定的转矩指令、转速指令、电压极限、电流极限下,采用MTPA控制方式获取电流幅值最小的电流工作点作为电流轨迹;电机运行在基速值以上时,在给定的转矩指令、转速指令、电压极限、电流极限下,采用弱磁区最大功率控制方式获取输出功率最大的电流工作点作为电流轨迹,弱磁区最大功率控制方式包括两种搜索方式:电流角θ在[θ ab]范围内采用电流极限圆电流轨迹搜索方式,在θ>θ b时采用MTPV控制方式,θ a为永磁电机在MTPA控制下电流幅值达到电流极限值时的弱磁电流角,θ b为永磁电机在MTPV控制下电流幅值达到电流极限时的弱磁电流角;
采用MTPA控制方式获取电流幅值最小的电流工作点的过程包括电流角迭代循环步骤和电流幅值迭代循环步骤,首先进行电流角迭代循环步骤,电流角迭代方向为电流幅值减小的方向;在进行电流角迭代过程中,嵌套电流幅值迭代循环步骤,用以确定每个电流角对应的电流幅值,电流幅值的迭代方向为给定转矩与实际转矩误差减小的方向,当电流角的迭代区间小于给定电流角迭代精度,认为电流幅值已经收敛至最小值,输出MTPA电流轨迹作为最大功率控制电流轨迹;
采用MTPV控制方式获取输出功率最大电流工作点的过程包括弱磁电流角迭代循环步骤和弱磁电流幅值迭代循环步骤,首先进行弱磁电流角迭代循环步骤,电流角迭代方向为最高转速增加的方向;在进行电流角迭代的过程中,嵌套电流幅值迭代循环步骤,用以确定每个电流角对应的电流幅值及最高转速,电流幅值的迭代方向为给定转矩与实际转矩误差减小的方向,当电流角的迭代区间小于给定电流角迭代精度,认为电机转速已经收敛至最大值,电机在电压限制下的输出功率收敛至最大值,输出MTPV电流轨迹作为最大功率控制电流轨迹。
首先建立电机非线性负载交直轴磁链模型;详细的,建立电机非线性负载交直轴磁链模型的详细步骤如下:
针对永磁同步电机不同磁化状态下、不同负载情况下铁心饱和程度变化明显,电机参数变化明显的特点,首先提出并建立一种非线性磁链模型,来模拟电机在不同磁化状态下、不同负载情况下的非线性特性。
在电机的电流极限范围内等距或不等距的选取一系列电流工作点,如电流幅值选取范围为(0,2,4,…),电流角选取范围为(0°,5°,10°,…),所选取的电流工作点间距由电机的饱和程度决定,需要保证相邻两电流工作点之间的铁心磁导率保持不变,铁心可以作为线性材料处理。采用仿真或实验的方式,计算所选取的电流工作点对应的电机负载交、直轴磁链数据,并将得到的负载交、直轴磁链数据进行插值,得到电流极限范围内所有电流工作点的负载交、直轴磁链模型,即永磁同步电机的非线性磁链模型:
ψ d(I,θ)=ψ d(i d,i q)
ψ q(I,θ)=ψ q(i d,i q)
直轴磁链模型:ψ d(I,θ)=ψ d(i d,i q),根据电机的交直轴电流就可以对应计算出电机的直轴磁链ψ d
交轴磁链模型:ψ q(I,θ)=ψ q(i d,i q),根据电机的交直轴电流就可以对应计算出电机的交轴磁链ψ q
根据得到的非线性磁链模型,可以准确地计算电机的电磁转矩、负载电压等,电磁转矩和负载电压的计算公式如下所示:
转矩计算公式:
T e(I,θ)=p(ψ d(I,θ)i qq(I,θ)i d)
其中,T e(I,θ)为电磁转矩,p为电机极对数,i d为电机的直轴电流,i q为电机的交轴电流,ψ d为电机的直轴磁链,ψ q为电机的交轴磁链。
电压幅值
Figure PCTCN2021123463-appb-000017
其中直轴电压
Figure PCTCN2021123463-appb-000018
交轴电压
Figure PCTCN2021123463-appb-000019
w为电机的电角速度,R 1为电机电阻。
给定电压极限下的电机最高转速W(θ)按
Figure PCTCN2021123463-appb-000020
获取,
式中:U lim为给定电压极限值。
该模型结合永磁同步电机考虑铁心饱和时可以处理为分段线性模型的特点,只需要计算电机额定运行电流范围内的一小部分电流工作点对应的负载磁链,再利用分段线性的特点插值得到所有电流工作点的负载磁链,同时不再需要计算电感,永磁磁链等参数,该模型计算量小,计算速度快,且能够准确模拟永磁同步电机不同磁化状态下、不同负载情况下铁心饱和程度的变化规律,实现电机的准确建模。
下面给出一个模型实施例:以一个极数为6,槽数为45,额定转速为2100转/分,饱和去磁后额定转矩为12.2Nm的串并联磁路型永磁同步电机为例,通过有限元仿真的手段获得电机的非线性磁链模型。此时电机磁化状态为饱和去磁,电机的电流给定为:直轴电流i d取值为(0,-2,-4,-6,-8,-10,-12)(A),共7个离散的电流点;交轴电流i q取值为(0,2,4,6,8,10,12)(A),共7个离散的电流点;共有7×7=49个离散的电流工作点。通过有限元仿真软件,仿真计算得到电机在饱和去磁状态下在上述的49个电流工作点处的电机直、交轴磁链,并对相邻两个电流工作点之间的其他电流工作点对应的磁链进行插值,得到串并联永磁同步电机在电流极限值范围内所有电流工作点对应的直、交轴负载磁链,即电机的非线性磁链模型,如附图1所示。
在恒转矩区采用基于双黄金分割迭代法的MTPA电流控制方式获取电流轨迹:可以在给定的转矩指令、转速指令、电机充磁状态下,获取电流幅值最小的电流工作点,从而实现MTPA控制,具体参见图2所示。
该过程具有两个迭代循环:电流角迭代和电流幅值迭代。首先进行电流角的迭代,在给定的转矩指令、转速指令、电机充磁状态下,电流角迭代方向为电流幅值减小的方向;在进行电流角迭代的同时,嵌套电流幅值的迭代,用以确定每个电流角对应的电流幅值,电流幅值的迭代方向为给定转矩与实际转矩误差减小的方向。当电流角的迭代区间小于给定值,认为电流幅值已经收敛至最小值,即MTPA工作点。
电流角迭代循环步骤中的目标函数值I(λ k)和I(β k)通过调用电流幅值迭代循环获取,k=1,2,3…即需要调用电流幅值迭代循环获取的目标函数值有I(λ 1)、I(β 1);I(λ 2)、I(β 2);I(λ 3)、I(β 3)…,输出至电流幅值迭代循环的参数为电流角试探点λ k、β k,k=1时,θ=λ 1和β 1两个值,需要进行两次电流幅值迭代循环,k=2,3…时,θ=λ k或β k,进行一次电流幅值迭代循环即可,经电流幅值迭代输出I(θ),即相当于输出I(λ k)或I(β k)作为目标函数值返回电流角迭代循环中。
考虑到电感和永磁磁链的非线性,电流幅值难以通过转矩公式直接求得,所以在电流角迭代过程中嵌套了幅值迭代,幅值迭代过程中转矩的计算使用了非线性负载磁链模型,考虑了电感和永磁磁链非线性的影响,电流幅值迭代结果准确。
下面介绍基于双黄金分割迭代法的MTPA控制获取电流轨迹的实施步骤:包括电流角迭代循环步骤和电流幅值迭代循环步骤。
电流角迭代循环步骤包括:
A1、初始化电流角初值区间[a 1,b 1],并计算电流角试探点初值λ 1、β 1
λ 1=a 1+0.382(b 1-a 1)、β 1=a 1+0.618(b 1-a 1);
比如[a 1,b 1]取值为[0°,90°],同时设定迭代精度,随着迭代过程的不断进行,当区间长度小于给定的迭代精度时,认为迭代收敛。
A2、判断两电流角试探点处电流幅值目标函数值I(λ k)和I(β k)是否存在关系I(λ k)>I(β k),电流角迭代次数k=1,2,3...
判断结果为是,执行步骤A3;判断结果为否执行步骤A5;
电流幅值目标函数的输入为电流角,目标函数的输出为给定转矩下的电流幅值,目标函数值I(λ k)和I(β k)通过调用电流幅值迭代循环获取;
A3、令a k+1=λ k,b k+1=b k,则
λ k+1=a k+1+0.382(b k+1-a k+1)=a k+0.382(b k-a k)+0.382(b k-a k-0.382(b k-a k))=a k+0.618(b k-a k)=β k
β k+1=a k+1+0.618(b k+1-a k+1),
A4、调用电流幅值迭代循环获取电流幅值目标函数值I(β k+1),然后执行步骤A7;
本步骤中不用执行计算λ k+1的调用步骤,因为I(λ k+1)=I(β k),即利用上次迭代的结果即可。由于使用黄金分割系数确定下一次迭代时的试探点,在进行下一次试探点选取的时候,其中一个试探点直接取自上一次迭代时的试探点,只需重新计算另一个试探 点,节省了计算资源,计算量小,计算速度快。
A5、令a k+1=a k,b k+1=β k,则
β k+1=a k+1+0.618(b k+1-a k+1)
=a k+0.618(a k+0.618(b k-a k)-a k)
=a k+0.382(b k-a k)=λ k
λ k+1=a k+1+0.382(b k+1-a k+1),
A6、调用电流幅值迭代循环获取电流幅值目标函数值I(λ k+1),然后执行步骤A7;
本步骤中不用执行计算I(β k+1)的调用步骤,因为I(β k+1)=I(λ k),即利用上次迭代的结果即可。由于使用黄金分割系数确定下一次迭代时的试探点,在进行下一次试探点选取的时候,其中一个试探点直接取自上一次迭代时的试探点,只需重新计算另一个试探点,节省了计算资源,计算量小,计算速度快。
A7、令k=k+1;
A8、判断迭代是否收敛:若b k-a k<L 1,执行步骤A9;否则,返回步骤A2;
其中L 1为电流角迭代精度;
A9、判断电流工作点是否同时满足电流极限与电压极限的要求:若I(λ k)≤I lim&U(λ k)≤U lim,I lim为给定电流极限值,U lim为给定电压极限值,输出MTPA电流轨迹;否则,重新输入转矩、转速指令,再返回执行步骤A1;
输出MTPA电流轨迹为:电流幅值I=I(λ k)、电流角θ=λ k的工作点,输入不同的转速、转矩可获取一系列工作点数据。
k=1时,将试探点初值λ 1、β 1输入至电流幅值迭代中,通过调用电流幅值迭代循环计算出目标函数值I(λ 1)、I(β 1)并返回电流角迭代循环中,根据步骤A2的判断结果决定计算k+1时计算哪个试探点,k+1时的目标函数值也是调用电流幅值迭代循环完成,根据步骤A8判断迭代是否收敛,若不收敛继续迭代循环;若收敛且满足步骤A9的电流极限、电压极限要求,输出MTPV轨迹,若收敛但不满足电流极限、电压极限要求,证明系统输入的参数偏差大,则重新输入转矩、转速指令,从头重新执行两个迭代循环。
电流幅值迭代循环步骤包括:
B1、初始化电流幅值的初值区间:[c 1,d 1],并计算电流幅值试探点初值μ 1、v 1
μ 1=c 1+0.382(d 1-c 1)、v 1=c 1+0.618(d 1-c 1);
比如当电流极限值为12A,电流值的初值区间定为[0A,12A],同时设定迭代精度,随着迭代过程的不断进行,当区间长度小于给定的迭代精度时,认为迭代收敛。
B2、计算两电流幅值试探点处的转矩误差目标函数值:f(μ 1)、f(v 1),
转矩误差目标函数f(I)按
Figure PCTCN2021123463-appb-000021
获取,其中:
Figure PCTCN2021123463-appb-000022
为给定转矩,T e(I,θ)为电流角θ对应的转矩,电流角θ在电流幅值迭代的过程中不变,为一确定值,电流角θ为电流角迭代循环输出的电流角试探点λ k、β k;I为电流幅值,i d=I sinθ,i q=I cosθ;
转矩T e(I,θ)由电机非线性负载交直轴磁链模型计算输出,按如下公式获取:
T e(I,θ)=p(ψ d(I,θ)i qq(I,θ)i d)
其中,p为电机极对数,i d为电机的直轴电流,i q为电机的交轴电流,ψ d为电机的直轴磁链,ψ q为电机的交轴磁链。
B3、判断两电流幅值试探点处转矩误差目标函数值f(μ h)和f(ν h)是否存在关系f(μ h)>f(ν h),电流幅值迭代次数h=1,2,3…判断结果为是,执行步骤B4;判断结果为否执行步骤B5;
B4、令c h+1=μ h,d h+1=d h,则
μ h+1=c h+1+0.382(d h+1-c h+1)=c h+0.382(d h-c h)+0.382(d h-c h-0.382(d h-c h))=c h+0.618(d h-c h)=v hv h+1=c h+1+0.618(d h+1-c h+1),
计算目标函数值f(v h+1),然后步骤B6;
B5、令c h+1=c h,d h+1=v h,则
v h+1=c h+1+0.618(d h+1-c h+1)=c h+0.618(c h+0.618(d h-c h)-c h)=c h+0.382(d h-c h)=μ h
μ h+1=c h+1+0.382(d h+1-c h+1),
计算目标函数值f(μ h+1),然后步骤B6;
B6、令h=h+1,
B7、判断迭代是否收敛:若d h-c h<L 2,输出给定电流角对应的电流幅值I(θ)、电压幅值U(θ),输出结果用于电流角的 迭代搜索过程;否则,返回步骤B3;其中L 2为电流幅值迭代精度。
在弱磁区的最大功率控制电流轨迹搜索:由于电机的电流极限和电压极限的限制,随着电机转速的升高,刚进入弱磁区时电机的电流轨迹主要受电流极限圆限制,此时最大功率控制电流轨迹与电机的电流极限圆重合,随着转速持续升高,电机的电流轨迹主要受电机电压极限圆的限制,此时最大功率控制电流轨迹为MTPV控制电流轨迹。
电流极限圆电流轨迹搜索方式的具体过程包括:
E1、电流极限圆电流轨迹搜索的初始化:
电流角迭代初始值为θ a,θ a为永磁电机在MTPA控制下电流幅值达到电流极限值I lim时的弱磁电流角;
电流角迭代终止值为θ b,θ b为永磁电机在MTPV控制下电流幅值达到电流极限I lim时的弱磁电流角;
本步骤用于确定电流极限圆搜索的迭代范围,以电流角θ a和θ b作为迭代的初始值和终止值。
E2、根据电机非线性负载交直轴磁链模型计算转矩T e(I,θ)和最高转速W(θ),并输出沿电流极限圆工作点轨迹I,θ,T(θ),W(θ):
I=I lim
θ=θ s,迭代次数s=1,2,3…,θ 1=θ a
T(θ)=T(I,θ s),
W(θ)=W(I,θ s,U lim),
E3、令θ s+1=θ s+Δθ,Δθ为迭代步进角度增幅;
E4、令s=s+1;
E5、判断迭代是否收敛:若θ s<θ b,返回执行步骤E2;否则结束迭代循环。
当电流角超过θ b之后,在弱磁区域后续的电流工作点搜索将采用MTPV控制方式。
基于双黄金分割迭代法的MTPV电流轨迹搜索方法,具体参见图4所示,该方法可以在给定的转矩指令、电压极限指令、电流极限指令下,获取给定转矩、电压极限和电流极限下输出功率最大的电流工作点,实现MTPV控制。
该方法具有两个迭代循环:电流角迭代和电流幅值迭代。首先进行流程图左侧的电流角的迭代:在给定的转矩指令、电压极限指令、电流极限指令下,电流角迭代方向为最高转速增加的方向;在进行电流角迭代的同时,嵌套电流幅值的迭代,用以确定每个电流角对应的电流幅值及最高转速,电流幅值的迭代方向为给定转矩与实际转矩误差减小的方向,电流幅值迭代过程的输出结果用于电流角的迭代过程,当电流角的迭代区间小于给定值,认为迭代收敛,得到电机MTPV工作点。
考虑到电感和永磁磁链的非线性,电流幅值难以通过转矩公式直接求得,所以在电流角迭代过程中嵌套了幅值迭代,幅值迭代过程中转矩的计算使用了非线性负载磁链模型,考虑了电感和永磁磁链非线性的影响,电流幅值迭代结果准确。
下面介绍基于双黄金分割迭代法的MTPV电流轨迹搜索方法的实施步骤:包括弱磁电流角迭代循环步骤和弱磁电流幅值迭代循环步骤。
弱磁电流角迭代循环步骤包括:
C1、初始化电流角初值区间[a 1,b 1],并计算电流角试探点初值λ 1、β 1
比如[a 1,b 1]取值为[0°,90°],同时设定迭代精度,随着迭代过程的不断进行,当区间长度小于给定的迭代精度时,认为迭代收敛。
λ 1=a 1+0.382(b 1-a 1)、β 1=a 1+0.618(b 1-a 1);
C2、判断两电流角试探点处最高转速目标函数值W(λ k)和W(β k)是否存在关系W(λ k)<W(β k),电流角迭代次数k=1,2,3…判断结果为是,执行步骤C3;判断结果为否执行步骤C5;
最高转速目标函数值W(λ k)和W(β k)通过调用电流幅值迭代循环获取;
C3、令a k+1=λ k,b k+1=b k,则
λ k+1=a k+1+0.382(b k+1-a k+1)=a k+0.382(b k-a k)+0.382(b k-a k-0.382(b k-a k))=a k+0.618(b k-a k)=β k
β k+1=a k+1+0.618(b k+1-a k+1),
C4、调用电流幅值迭代循环获取最高转速目标函数值W(β k+1),然后执行步骤C7;
本步骤中不用执行计算W(λ k+1)的调用步骤,因为W(λ k+1)=W(β k),即利用上次迭代的结果即可。由于使用黄金分割系数确定下一次迭代时的试探点,在进行下一次试探点选取的时候,其中一个试探点直接取自上一次迭代时的试探点,只需重新计算另一个试探点,节省了计算资源,计算量小,计算速度快。
C5、令a k+1=a k,b k+1=β k,则
β k+1=a k+1+0.618(b k+1-a k+1)=a k+0.618(a k+0.618(b k-a k)-a k)=a k+0.382(b k-a k)=λ k
λ k+1=a k+1+0.382(b k+1-a k+1),
C6、调用电流幅值迭代循环获取目标函数值W(λ k+1),然后执行步骤C7;
本步骤中不用执行计算W(β k+1)的调用步骤,因为W(β k+1)=W(λ k),即利用上次迭代的结果即可。由于使用黄金分割系数确定下一次迭代时的试探点,在进行下一次试探点选取的时候,其中一个试探点直接取自上一次迭代时的试探点,只需重新计算另一个试探点,节省了计算资源,计算量小,计算速度快。
C7、令k=k+1;
C8、判断迭代是否收敛:若b k-a k<L 1,执行步骤C9;否则,返回步骤C2;
其中L 1为电流角迭代精度;
C9、判断电流工作点是否满足电流极限的要求:若I(λ k)≤I lim,I lim为给定电流极限值,输出MTPV轨迹;否则,重新输入转矩指令,再返回执行步骤C1;
MTPV轨迹包括给定转矩
Figure PCTCN2021123463-appb-000023
给定电压极限和电流极限下的电机最高转速w=W(θ),电流幅值I=I(λ k)和电流角θ=λ k。输入不同的转转矩可获取一系列工作点数据。当然也可根据具体情况调整I lim、U lim
k=1时,将试探点初值λ 1、β 1输入至电流幅值迭代中,通过调用电流幅值迭代循环计算出目标函数值W(λ 1)、W(β 1)并返回电流角迭代循环中,根据步骤C2的判断结果决定计算k+1时计算哪个试探点,k+1时的目标函数值也是调用电流幅值迭代循环完成,根据步骤C8判断迭代是否收敛,若不收敛继续迭代循环;若收敛且满足步骤C9的电流极限要求,输出MTPV轨迹,若收敛但不满足电流极限要求,证明系统输入的参数偏差大,则重新输入转矩指令,从头重新执行两个迭代循环。
弱磁电流幅值迭代循环步骤包括:
D1、初始化电流幅值的初值区间:[c 1,d 1],并计算电流幅值试探点初值μ 1、v 1
μ 1=c 1+0.382(d 1-c 1)、v 1=c 1+0.618(d 1-c 1);
比如当电流极限值I lim=14C,电流值的初值区间定为[0C,14C],同时设定迭代精度,随着迭代过程的不断进行,当区间长度小于给定的迭代精度时,认为迭代收敛。
D2、计算两电流幅值试探点处的转矩误差目标函数值:f(μ 1)、f(ν 1),
转矩误差目标函数f(I)按
Figure PCTCN2021123463-appb-000024
获取,其中:
Figure PCTCN2021123463-appb-000025
为给定转矩,T e(I,θ)为电流角θ对应的转矩,电流角θ在电流幅值迭代的过程中不变,为一确定值,电流角θ为电流角迭代循环输出的电流角试探点λ k、β k;I为电流幅值,i d=I sinθ,i q=I cosθ;
转矩T e(I,θ)由电机非线性负载交直轴磁链模型计算输出,按如下公式获取:
T e(I,θ)=p(ψ d(I,θ)i qq(I,θ)i d)
其中,p为电机极对数,i d为电机的直轴电流,i q为电机的交轴电流,ψ d为电机的直轴磁链,ψ q为电机的交轴磁链。
D3、判断两电流幅值试探点处转矩误差目标函数值f(μ h)和f(ν h)是否存在关系f(μ h)>f(v h),电流幅值迭代次数h=1,2,3…判断结果为是,执行步骤D4;判断结果为否执行步骤D5;
D4、令c h+1=μ h,d h+1=d h,则
μ h+1=c h+1+0.382(d h+1-c h+1)=c h+0.382(d h-c h)+0.382(d h-c h-0.382(d h-c h))=c h+0.618(d h-c h)=v h
v h+1=c h+1+0.618(d h+1-c h+1),
计算目标函数值f(v h+1),然后步骤D6;
D5、令c h+1=c h,c h+1=v h,则
v h+1=c h+1+0.618(d h+1-c h+1)=c h+0.618(c h+0.618(d h-c h)-c h)=c h+0.382(d h-c h)=μ h
μ h+1=c h+1+0.382(d h+1-c h+1),
计算目标函数值f(μ h+1),然后步骤D6;
D6、令h=h+1,
D7、判断迭代是否收敛:若d h-c h<L 2,输出给定电流角对应的电流幅值I(θ)、计算并输出给定转矩和给定电压极限下的电机最高转速W(θ),输出结果用于电流角的迭代搜索过程;否则,返回步骤D3;其中L 2为电流幅值迭代精度。
本实施方式所述最大功率搜索方法包括两部分,恒转矩区的MTPA控制电流轨迹搜索方法和弱磁区的最大功率控制电流轨迹搜索方法,其中弱磁区最大功率控制电流轨迹搜索方法包括电流极限圆电流轨迹搜索和MTPV控制电流轨迹搜索。基于双黄金分割迭代法的最大功率控制恒转矩区MTPA控制电流轨迹搜索方法的流程图如图2所示,该方法可以在给定的转矩指令、转速指令、电压极限、电流极限下,获取电压极限和电流极限下电机输出功率最大的电流工作点,实现恒转矩区最大功率控制。电流极限圆电流轨迹搜索方法的流程图如图3所示,基于双黄金分割迭代法的最大功率控制弱磁区MTPV控制电流轨迹搜索方法的流程图如图4所示,该方法可以在给定的转矩指令、转速指令、电压极限、电流极限下,获取电压极限和电流极限下电机输出功率最大的电流工作点,实现弱磁区最大功率控制。
使用该搜索方法计算串并联永磁同步电机最大功率控制时的电流轨迹并计算施加相应电流轨迹后的电机转矩-转速曲线和功率-转速曲线,如附图6所示,并同时使用公式法计算电机最大功率控制时的转矩-转速曲线和功率-转速曲线,如附图5所示。通过两图的对比可以看出,在相同的电压、电流极限下,该迭代搜索方法计算出的电机最大输出功率时的电流工作点准确性较高,且相同转速下电机的输出功率更高,可以实现永磁电机的最大功率控制。同时根据计算过程可以看出该搜索方法的计算量小,计算速度快。
具体实施方式二:下面结合图7说明本实施方式,本实施方式所述永磁同步电机最大功率控制在线控制方法,采用实施方式一所述的永磁同步电机最大功率控制电流轨迹搜索方法得到永磁同步电机在多个工作点下的电流轨迹,将这些电流轨迹作为样本数据,训练生成最大功率控制神经网络模型,最大功率控制神经网络模型的输入为电机的转速、转矩、电流极限值和电压极限值,输出为电流幅值与电流角;
将最大功率控制神经网络模型加载至DSP或FPGA控制器中,可以实现永磁同步电机最大功率在线控制,根据电机的转速和转矩实时输出电流幅值与电流角用于控制电机在线最大功率运行。
神经网络训练过程为:利用上述搜索方法得到永磁同步电机在部分工作点下的电流轨迹,将这些电流轨迹作为样本数据,对神经网络模型进行训练、测试与验证,当误差小于设定值后训练完成,神经网络结构以及各个神经元的权重和偏置参数确定,利用BP算法根据神经网络输出值与样本值之间的误差的梯度,沿着神经网络计算的逆向方向对各节点的权值与偏置进行调节,在每个样本的训练过程中,各节点的权值和偏置都依据误差得到调节,当误差小于设定值后训练完成,神经网络结构以及各个神经元的权重和偏置参数确定,最大功率控制神经网络模型建立完成,神经网络模型训练、测试与验证误差如图7所示,该模型不仅可以输出样本数据中相应工作点的电流轨迹,还可以输出样本数据以外的工作点的电流轨迹,即可以输出所有工作点的电流轨迹。该神经网络模型有四个输入,分别为电压极限、电流极限、转速和转矩,有两个输出,分别为直轴电流和交轴电流,神经网络模型采用一层隐藏层,隐藏层中采用9个神经元。
实施例二
本发明不计算交直轴电感、永磁磁链等参数,本发明搜索方法基于黄金分割的思想,可以在给定的转矩指令、转速指令、、电压极限、电流极限下,获取电流幅值最小的电流工作点,实现全速域效率最优控制控制。电机运行在基速值以下为恒转速区域,基速值以上为弱磁区域,为了实现全速域效率最优,本发明在基速值以下时MTPA控制方式获取电流幅值最小的电流工作点作为电流轨迹;电机运行在基速值以上时,采用弱磁区效率最优控制方式获取电流幅值最小的电流工作点作为电流轨迹;采用MTPA控制方式包括电流角迭代循环步骤和电流幅值迭代循环步骤,采用弱磁区效率最优控制方式包括弱磁电流角迭代循环步骤和电流幅值迭代循环步骤。
考虑到电感和永磁磁链的非线性,电流幅值难以通过转矩公式直接求得,所以在电流角(弱磁电流角)迭代过程中嵌套了电流幅值迭代,电流幅值迭代过程中转矩的计算使用了电机非线性负载交直轴磁链模型,考虑了电感和永磁磁链非线性的影响,计算结果准确。使用该非线性负载磁链模型可以准确的计算电机转矩、负载电压等,不再需要计算电感,永磁磁链等参数,计算量小,计算速度快,能够准确模拟永磁同步电机不同磁化状态下、不同负载情况下铁心饱和程度的变化规律,实现电机的准确建模。利用上述搜索方法得到永磁同步电机在不同充磁状态下,多个工作点下的电流轨迹,将这些电流轨迹作为样本数据,对神经网络模型进行训练、测试与验证。全速域效率最优在线控制神经网络模型的输入为电机的转速、转矩、电压极限、电流极限,输出为电流幅值与电流角(或直轴电流与交轴电流),该模型不仅可以输出样本数据中相应工作点的电流轨迹,还可以输出样本数据以外的工作点的电流轨迹,即可以输出所有工作点的电流轨迹。将全速域效率最优在线控制神经网络模型(可以用输入输出的函数关系来表达)加载至DSP或FPGA控制器中,可以实现永磁同步电机全速域效率最优在线控制。
具体实施方式一:下面结合图1、图2以及图8-图10说明本实施方式,本实施方式所述永磁同步电机全速域效率最优控制电流轨迹搜索方法,电机运行在基速值以下时,在给定的转矩指令、转速指令、电压极限、电流极限下,采用MTPA控制方式获取电流幅值最小的电流工作点作为电流轨迹;电机运行在基速值以上时,在给定的转矩指令、转速指令、电压极限、电流极限下,采用弱磁区效率最优控制方式获取电流幅值最小的电流工作点作为电流轨迹;
采用MTPA控制方式获取电流幅值最小的电流工作点的过程包括电流角迭代循环步骤和电流幅值迭代循环步骤,首先进行电流角迭代循环步骤,电流角迭代方向为电流幅值减小的方向;在进行电流角迭代过程中,嵌套电流幅值迭代循环步骤,用以确定每个电流角对应的电流幅值,电流幅值的迭代方向为给定转矩与实际转矩误差减小的方向,当电流角的迭代区间小于给定电流角迭代精度,认为电流幅值已经收敛至最小值,输出MTPA电流轨迹;
采用弱磁区效率最优控制方式获取电流幅值最小的电流工作点的过程包括弱磁电流角迭代循环步骤和电流幅值迭代循环 步骤,首先进行电流角迭代循环步骤,电流角迭代方向为电压极限下电流幅值减小的方向;在进行电流角迭代的过程中,嵌套电流幅值迭代循环步骤,用以确定每个电流角对应的电流幅值,电流幅值的迭代方向为给定转矩与实际转矩误差减小的方向,当电流角的迭代区间小于给定电流角迭代精度,认为电流幅值已经收敛至最小值,输出弱磁区效率最优控制电流轨迹。
首先建立电机非线性负载交直轴磁链模型:
具体的,建立电机非线性负载交直轴磁链模型的详细步骤同上述实施例一中记载的建立电机非线性负载交直轴磁链模型的详细步骤,这里不再累述。
基于双黄金分割迭代法的MTPA电流控制方式获取电流轨迹:可以在给定的转矩指令、转速指令、电机充磁状态下,获取电流幅值最小的电流工作点,从而实现MTPA控制,具体参见图2所示。
该过程具有两个迭代循环:电流角迭代和电流幅值迭代。首先进行电流角的迭代,在给定的转矩指令、转速指令、电压极限、电流极限下,电流角迭代方向为电流幅值减小的方向;在进行电流角迭代的同时,嵌套电流幅值的迭代,用以确定每个电流角对应的电流幅值,电流幅值的迭代方向为给定转矩与实际转矩误差减小的方向。当电流角的迭代区间小于给定值,认为电流幅值已经收敛至最小值,即MTPA工作点。
电流角迭代循环步骤中的目标函数值I(λ k)和I(β k)通过调用电流幅值迭代循环获取,k=1,2,3…即需要调用电流幅值迭代循环获取的目标函数值有I(λ 1)、I(β 1);I(λ 2)、I(β 2);I(λ 3)、I(β 3)…,输出至电流幅值迭代循环的参数为电流角试探点λ k、β k,k=1时,θ=λ 1和β 1两个值,需要进行两次电流幅值迭代循环,k=2,3…时,θ=λ k或β k,进行一次电流幅值迭代循环即可,经电流幅值迭代输出I(θ),即相当于输出I(λ k)或I(β k)作为目标函数值返回电流角迭代循环中。
考虑到电感和永磁磁链的非线性,电流幅值难以通过转矩公式直接求得,所以在电流角迭代过程中嵌套了幅值迭代,幅值迭代过程中转矩的计算使用了非线性负载磁链模型,考虑了电感和永磁磁链非线性的影响,电流幅值迭代结果准确。
下面介绍基于双黄金分割迭代法的MTPA控制获取电流轨迹的实施步骤:包括电流角迭代循环步骤和电流幅值迭代循环步骤。
具体的,该电流角迭代循环步骤的详细内容同上述实施例一中记载的电流角迭代循环步骤A1至A9,这里不再累述;并且该电流幅值迭代循环步骤的详细内容同上述实施例一中记载的电流幅值迭代循环步骤B1至B7,这里不再累述。
基于双黄金分割迭代法的弱磁区效率最优控制获取电流轨迹:可以在给定的转矩指令、转速指令、电压极限、电流极限下,获取电流幅值最小的电流工作点,实现弱磁区效率最优控制,具体参见图2所示。
该过程具有两个迭代循环:弱磁电流角迭代和电流幅值迭代。首先进行弱磁电流角的迭代,在给定的转矩指令、转速指令、电压极限和电流极限下,电流角迭代方向为电压极限下,电流幅值减小的方向;在进行电流角迭代的同时,嵌套电流幅值的迭代,用以确定每个电流角对应的电流幅值,电流幅值的迭代方向为给定转矩与实际转矩误差减小的方向。当电流角的迭代区间小于给定值,认为电流幅值已经收敛至最小值,即弱磁区效率最优控制工作点。
考虑到电感和永磁磁链的非线性,电流幅值难以通过转矩公式直接求得,所以在电流角迭代过程中嵌套了幅值迭代,幅值迭代过程中转矩的计算使用了非线性负载磁链模型,考虑了电感和永磁磁链非线性的影响,电流幅值迭代结果准确。
下面介绍基于双黄金分割迭代法的弱磁区效率最优控制获取电流轨迹的实施步骤:包括弱磁电流角迭代循环步骤和电流幅值迭代循环步骤。
具体的,该电流幅值迭代循环步骤的详细内容同上述实施例一中记载的电流幅值迭代循环步骤B1至B7,这里不再累述。
弱磁电流角迭代循环步骤包括:
C1、初始化电流角初值区间[a 1,b 1],并计算电流角试探点初值λ 1、β 1
λ 1=a 1+0.382(b 1-a 1)、β 1=a 1+0.618(b 1-a 1);
比如[a 1,b 1]取值为[0°,90°],同时设定迭代精度,随着迭代过程的不断进行,当区间长度小于给定的迭代精度时,认为迭代收敛。
C2、判断负载电压目标函数值U(β k)和电压极限值U lim的大小关系,若U(β k)>U lim,执行步骤C6;否则,执行步骤C3;
负载电压目标函数值U(β k)通过调用电流幅值迭代循环获取,电流角迭代次数k=1,2,3…;
调用电流幅值迭代循环输出U(θ)=U(β k)或U(λ k),本实施方式只用到U(β k),负载电压目标函数的输入为电流角,输出为给定转矩、转速下的负载电压。
C3、判断两电流角试探点处电流幅值目标函数值I(λ k)和I(β k)是否存在关系I(λ k)>I(β k),
判断结果为是,执行步骤C4;判断结果为否执行步骤C6;
电流幅值目标函数值I(λ k)和I(β k)通过调用电流幅值迭代循环获取;
电流幅值目标函数的输入为电流角,输出为给定转矩、转速下的电流幅值。
C4、令a k+1=λ k,b k+1=b k,则
λ k+1=a k+1+0.382(b k+1-a k+1)=a k+0.382(b k-a k)+0.382(b k-a k-0.382(b k-a k))=a k+0.618(b k-a k)=β k
β k+1=a k+1+0.618(b k+1-a k+1),
C5、调用电流幅值迭代循环获取电流幅值目标函数值I(β k+1),然后执行步骤C8;
本步骤中不用执行计算λ k+1的调用步骤,因为I(λ k+1)=I(β k),即利用上次迭代的结果即可。由于使用黄金分割系数确定下一次迭代时的试探点,在进行下一次试探点选取的时候,其中一个试探点直接取自上一次迭代时的试探点,只需重新计算另一个试探点,节省了计算资源,计算量小,计算速度快。
C6、令a k+1=a k,b k+1=β k,则
β k+1=a k+1+0.618(b k+1-a k+1)
=a k+0.618(a k+0.618(b k-a k)-a k)
=a k+0.382(b k-a k)=λ k
λ k+1=a k+1+0.382(b k+1-a k+1),
C7、调用电流幅值迭代循环获取电流幅值目标函数值I(λ k+1),然后执行步骤C8;
本步骤中不用执行计算I(β k+1)的调用步骤,因为I(β k+1)=I(λ k),即利用上次迭代的结果即可。由于使用黄金分割系数确定下一次迭代时的试探点,在进行下一次试探点选取的时候,其中一个试探点直接取自上一次迭代时的试探点,只需重新计算另一个试探点,节省了计算资源,计算量小,计算速度快。
C8、令k=k+1;
C9、判断迭代是否收敛:若b k-a k<L 1,执行步骤C10;否则,返回步骤C2;
其中L 1为电流角迭代精度;
C10、判断电流工作点是否同时满足电流极限的要求:若I(λ k)≤I lim,I lim为给定电流极限值,输出弱磁区效率最优控制电流轨迹;否则,重新输入转矩、转速指令,再返回执行步骤C1;
弱磁区效率最优控制电流轨迹为:电流幅值I=I(λ k)、电流角θ=λ k,输入不同的转速、转矩可获取一系列工作点数据。
k=1时,将试探点初值λ 1、β 1输入至电流幅值迭代中,通过调用电流幅值迭代循环计算出目标函数值I(λ 1)、I(β 1)、U(β 1)并返回电流角迭代循环中,根据步骤C2的判断结果决定计算k+1时计算哪个试探点,k+1时的目标函数值也是调用电流幅值迭代循环完成,根据步骤C8判断迭代是否收敛,若不收敛继续迭代循环;若收敛且满足步骤C10的电流极限要求,输出弱磁区效率最优控制电流轨迹,若收敛但不满足电流极限要求,证明系统输入的参数偏差大,则重新输入转矩、转速指令,从头重新执行两个迭代循环。
通过上述的基于双黄金分割迭代法的全速域效率最优控制电流轨迹搜索方法可以获得任一工作点(给定转矩指令、转速指令、电压极限、电流极限)在全速域范围内(基速值以下的恒转矩区)和基速值以上的弱磁区实现效率最优控制时应该施加的电流幅值及相位,该搜索方法迭代收敛速度快,计算量小,且考虑了铁心饱和等非线性因素的影响,计算结果准确。
电机在恒转矩区运行时,电机的负载端电压未达到电机极限值,恒转矩区的效率最优控制电流轨迹搜索方法基于黄金分割的思想,可以在给定的转矩指令、转速指令下,获取电机在恒转矩区运行时电流幅值最小的电流工作点,实现恒转矩区的效率最优控制、即MTPA控制;电机在弱磁区运行时,若继续采用MTPA控制,电机的负载端电压会超过电压极限值,必须增加直轴弱磁电流以降低电机负载端电压,弱磁区的效率最优控制电流轨迹搜索方法基于黄金分割的思想,可以在给定的转矩指令、转速指令、电压极限、电流极限下,获取电机在弱磁区运行时电流幅值最小的电流工作点,实现弱磁区的效率最优控制。
使用该搜索方法计算串并联永磁同步电机弱磁区效率最优控制时的电流轨迹并计算施加相应电流轨迹后的电机效率MAP图,如附图10所示,并同时使用公式法计算电机全速域效率最优控制时的效率MAP图,如附图9所示。通过两图的对比可以看出,在相同的电压、电流极限下,该迭代搜索方法计算出的电机弱磁运行范围更大,且转折速度后的每个转速点对应的最大转矩也更高,同时可以看出迭代搜索方法算出的MAP图的高效区占比更大,所以该迭代搜索方法在计算全速域工作点时准确性较高。同时根据计算过程可以看出该搜索方法的计算量小,计算速度快。
具体实施方式二:下面结合图1和图2以及图8-图11说明本实施方式,本实施方式所述永磁同步电机全速域效率最优控制在线控制方法。
利用实施方式一所述搜索方法得到永磁同步电机在不同充磁状态下,一系列工作点下的电流轨迹,将这些电流轨迹作为样本数据,对神经网络模型进行训练、测试与验证。全速域效率最优控制神经网络模型的输入为电机的转速、转矩、电压极限、电流极限,输出为电流幅值与电流角(或直轴电流与交轴电流),利用BP算法根据神经网络输出值与样本值之间的误差的梯度,沿着神经网络计算的逆向方向对各节点的权值与偏置进行调节,在每个样本的训练过程中,各节点的权值和偏置都依据误差得到调节,当误差小于设定值后训练完成,神经网络结构以及各个神经元的权重和偏置参数确定,神经网络模型训练、测试与验证误差如图11所示,该模型不仅可以输出样本数据中相应工作点的电流轨迹,还可以输出样本数据以外的工作点的电流轨迹,即可以输出所有工作点的电流轨迹。该神经网络模型有四个输入,分别为电机转速、转矩、电压极限和电流极限,有两个输出,分别为直轴电流和交轴电流,神经网络模型采用一层隐藏层,隐藏层中采用15个神经元。将该全速域效率最优控制神经网络模型(可以用输入输出的函数关系来表达)加载至DSP或FPGA控制器中,可以实现永磁同步电机全速域效率最优在线控制。

Claims (17)

  1. 永磁同步电机最大功率控制电流轨迹搜索方法,其特征在于,该方法为:电机运行在基速值以下时,在给定的转矩指令、转速指令、电压极限、电流极限下,采用MTPA控制方式获取电流幅值最小的电流工作点作为电流轨迹;电机运行在基速值以上时,在给定的转矩指令、转速指令、电压极限、电流极限下,采用弱磁区最大功率控制方式获取输出功率最大的电流工作点作为电流轨迹,弱磁区最大功率控制方式包括两种搜索方式:电流角θ在[θ ab]范围内采用电流极限圆电流轨迹搜索方式,在θ>θ b时采用MTPV控制方式,θ a为永磁电机在MTPA控制下电流幅值达到电流极限值时的弱磁电流角,θ b为永磁电机在MTPV控制下电流幅值达到电流极限时的弱磁电流角;
    采用MTPA控制方式获取电流幅值最小的电流工作点的过程包括电流角迭代循环步骤和电流幅值迭代循环步骤,首先进行电流角迭代循环步骤,电流角迭代方向为电流幅值减小的方向;在进行电流角迭代过程中,嵌套电流幅值迭代循环步骤,用以确定每个电流角对应的电流幅值,电流幅值的迭代方向为给定转矩与实际转矩误差减小的方向,当电流角的迭代区间小于给定电流角迭代精度,认为电流幅值已经收敛至最小值,输出MTPA电流轨迹作为最大功率控制电流轨迹;
    采用MTPV控制方式获取输出功率最大电流工作点的过程包括弱磁电流角迭代循环步骤和弱磁电流幅值迭代循环步骤,首先进行弱磁电流角迭代循环步骤,电流角迭代方向为最高转速增加的方向;在进行电流角迭代的过程中,嵌套电流幅值迭代循环步骤,用以确定每个电流角对应的电流幅值及最高转速,电流幅值的迭代方向为给定转矩与实际转矩误差减小的方向,当电流角的迭代区间小于给定电流角迭代精度,认为电机转速已经收敛至最大值,电机在电压限制下的输出功率收敛至最大值,输出MTPV电流轨迹作为最大功率控制电流轨迹。
  2. 根据权利要求1所述永磁同步电机最大功率控制电流轨迹搜索方法,其特征在于,采用MTPA控制方式获取电流幅值最小的电流工作点的过程包括电流角迭代循环步骤和电流幅值迭代循环步骤;
    电流角迭代循环步骤包括:
    A1、初始化电流角初值区间[a 1,b 1],并计算电流角试探点初值λ 1、β 1
    λ 1=a 1+0.382(b 1-a 1)、β 1=a 1+0.618(b 1-a 1);
    A2、判断两电流角试探点处电流幅值目标函数值I(λ k)和I(β k)是否存在关系I(λ k)>I(β k),电流角迭代次数k=1,2,3...
    判断结果为是,执行步骤A3;判断结果为否执行步骤A5;
    电流幅值目标函数值I(λ k)和I(β k)通过调用电流幅值迭代循环获取;
    A3、令a k+1=λ k,b k+1=b k,λ k+1=β k,β k+1=a k+1+0.618(b k+1-a k+1),
    A4、调用电流幅值迭代循环获取电流幅值目标函数值I(β k+1),然后执行步骤A7;
    A5、令a k+1=a k,b k+1=β k,β k+1=λ k,λ k+1=a k+1+0.382(b k+1-a k+1),
    A6、调用电流幅值迭代循环获取电流幅值目标函数值I(λ k+1),然后执行步骤A7;
    A7、令k=k+1;
    A8、判断迭代是否收敛:若b k-a k<L 1,执行步骤A9;否则,返回步骤A2;
    其中L 1为电流角迭代精度;
    A9、判断电流工作点是否同时满足电流极限与电压极限的要求:若I(λ k)≤I lim&U(λ k)≤U lim,I lim为给定电流极限值,U lim为给定电压极限值,输出MTPA电流轨迹;否则,重新输入转矩、转速指令,再返回执行步骤A1;
    MTPA电流轨迹为:电流幅值I=I(λ k)、电流角θ=λ k
    电流幅值迭代循环步骤包括:
    B1、初始化电流幅值的初值区间:[c 1,d 1],并计算电流幅值试探点初值μ 1、v 1
    μ 1=c 1+0.382(d 1-c 1)、v 1=c 1+0.618(d 1-c 1);
    B2、计算两电流幅值试探点处的转矩误差目标函数值:f(μ 1)、f(v 1),
    转矩误差目标函数f(I)按
    Figure PCTCN2021123463-appb-100001
    获取,其中:
    Figure PCTCN2021123463-appb-100002
    为给定转矩,T e(I,θ)为电流角θ对应的转矩,T e(I,θ)根据电机非线性负载交直轴磁链模型计算获取;电流角θ为电流角迭代循环输出的电流角试探点λ k、β k;I为电流幅值;
    B3、判断两电流幅值试探点处转矩误差目标函数值f(μ h)和f(ν h)是否存在关系f(μ h)>f(ν h),电流幅值迭代次数h=1,2,3…
    判断结果为是,执行步骤B4;判断结果为否执行步骤B5;
    B4、令c h+1=μ h,d h+1=d h,μ h+1=v h,v h+1=c h+1+0.618(d h+1-c h+1),
    计算目标函数值f(v h+1),然后步骤B6;
    B5、令c h+1=c h,d h+1=v h,v h+1h,μ h+1=c h+1+0.382(d h+1-c h+1),
    计算目标函数值f(μ h+1),然后步骤B6;
    B6、令h=h+1,
    B7、判断迭代是否收敛:若d h-c h<L 2,输出给定电流角对应的电流幅值I(θ)、电压幅值U(θ),输出结果用于电流角的迭代搜索过程;否则,返回步骤B3;其中L 2为电流幅值迭代精度。
  3. 根据权利要求1所述永磁同步电机最大功率控制电流轨迹搜索方法,其特征在于,电流极限圆电流轨迹搜索方式的具体过程包括:
    E1、电流极限圆电流轨迹搜索的初始化:
    电流角迭代初始值为θ a,θ a为永磁电机在MTPA控制下电流幅值达到电流极限值I lim时的弱磁电流角;
    电流角迭代终止值为θ b,θ b为永磁电机在MTPV控制下电流幅值达到电流极限I lim时的弱磁电流角;
    E2、根据电机非线性负载交直轴磁链模型计算转矩T e(I,θ)和最高转速W(θ),并输出沿电流极限圆工作点轨迹I,θ,T(θ),W(θ):
    I=I lim
    θ=θ s,迭代次数s=1,2,3…,θ 1=θ a
    T(θ)=T(I,θ s),
    W(θ)=W(I,θ s,U lim),
    E3、令θ s+1=θ s+Δθ,Δθ为迭代步进角度增幅;
    E4、令s=s+1;
    E5、判断迭代是否收敛:若θ s<θ b,返回执行步骤E2;否则结束迭代循环。
  4. 根据权利要求1所述永磁同步电机最大功率控制电流轨迹搜索方法,其特征在于,采用MTPV控制方式获取输出功率最大电流工作点的过程包括弱磁电流角迭代循环步骤和弱磁电流幅值迭代循环步骤;
    弱磁电流角迭代循环步骤包括:
    C1、初始化电流角初值区间[a 1,b 1],并计算电流角试探点初值λ 1、β 1
    λ 1=a 1+0.382(b 1-a 1)、β 1=a 1+0.618(b 1-a 1);
    C2、判断两电流角试探点处最高转速目标函数值W(λ k)和W(β k)是否存在关系W(λ k)<W(β k),电流角迭代次数k=1,2,3…判断结果为是,执行步骤C3;判断结果为否执行步骤C5;
    最高转速目标函数值W(λ k)和W(β k)通过调用电流幅值迭代循环获取;
    C3、令a k+1=λ k,b k+1=b k,λ k+1=β k,β k+1=a k+1+0.618(b k+1-a k+1),
    C4、调用电流幅值迭代循环获取最高转速目标函数值W(β k+1),然后执行步骤C7;
    C5、令a k+1=a k,b k+1=β k,β k+1=λ k,λ k+1=a k+1+0.382(b k+1-a k+1),
    C6、调用电流幅值迭代循环获取最高转速目标函数值W(λ k+1),然后执行步骤C7;
    C7、令k=k+1;
    C8、判断迭代是否收敛:若b k-a k<L 1,执行步骤C9;否则,返回步骤C2;
    其中L 1为电流角迭代精度;
    C9、判断电流工作点是否满足电流极限的要求:若I(λ k)≤I lim,I lim为给定电流极限值,输出MTPV轨迹;否则,重新输入转矩指令,再返回执行步骤C1;
    MTPV轨迹包括给定转矩
    Figure PCTCN2021123463-appb-100003
    给定电压极限和电流极限下的电机最高转速w=W(θ),电流幅值I=I(λ k)和电流角θ=λ k
    弱磁电流幅值迭代循环步骤包括:
    D1、初始化电流幅值的初值区间:[c 1,d 1],并计算电流幅值试探点初值μ 1、v 1
    μ 1=c 1+0.382(d 1-c 1)、v 1=c 1+0.618(d 1-c 1);
    D2、计算两电流幅值试探点处的转矩误差目标函数值:f(μ 1)、f(v 1),
    转矩误差目标函数f(I)按
    Figure PCTCN2021123463-appb-100004
    获取,其中:
    Figure PCTCN2021123463-appb-100005
    为给定转矩,T e(I,θ)为电流角θ对应的转矩,电流角θ为电流角迭代循环输出的电流角试探点λ k、β k;I为电流幅值;
    D3、判断两电流幅值试探点处转矩误差目标函数值f(μ h)和f(ν h)是否存在关系f(μ h)>f(ν h),电流幅值迭代次数h=1,2,3…
    判断结果为是,执行步骤D4;判断结果为否执行步骤D5;
    D4、令c h+1=μ h,d h+1=d h,μ h+1=v h,v h+1=c h+1+0.618(d h+1-c h+1),
    计算目标函数值f(v h+1),然后步骤D6;
    D5、令c h+1=c h,d h+1=v h,v h+1=μ h,μ h+1=c h+1+0.382(d h+1-c h+1),
    计算目标函数值f(μ h+1),然后步骤D6;
    D6、令h=h+1,
    D7、判断迭代是否收敛:若d h-c h<L 2,输出给定电流角对应的电流幅值I(θ)、计算并输出给定转矩和给定电压极限下的电机最高转速W(θ),输出结果用于电流角的迭代搜索过程;否则,返回步骤D3;其中L 2为电流幅值迭代精度。
  5. 根据权利要求2、3或4所述永磁同步电机最大功率控制电流轨迹搜索方法,其特征在于,电机非线性负载交直轴磁链模型的建立过程:
    在电机的电流极限范围内等距或不等距的选取一系列电流工作点,包括等距或不等距电流幅值系列值及等距或不等距电流角系列值,所选取的电流工作点间距由电机的饱和程度决定,需要保证相邻两电流工作点之间的铁心磁导率保持不变,铁心按线性材料处理;
    采用仿真或实验的方式,计算所选取的电流工作点对应的电机负载交、直轴磁链数据,并将得到的负载交、直轴磁链数据进行插值,得到电流极限范围内所有电流工作点的负载交、直轴磁链模型,即永磁同步电机的非线性磁链模型:
    ψ d(I,θ)=ψ d(i d,i q)
    ψ q(I,θ)=ψ q(i d,i q)。
  6. 根据权利要求1所述永磁同步电机最大功率控制电流轨迹搜索方法,其特征在于,转矩T e(I,θ)由电机非线性负载交直轴磁链模型计算输出,按如下公式获取:
    T e(I,θ)=p(ψ d(I,θ)i qq(I,θ)i d)
    其中,p为电机极对数,i d为电机的直轴电流,i q为电机的交轴电流,ψ d为电机的直轴磁链,ψ q为电机的交轴磁链。
  7. 根据权利要求4所述永磁同步电机最大功率控制电流轨迹搜索方法,其特征在于,给定电压极限下的电机最高转速W(θ)按
    Figure PCTCN2021123463-appb-100006
    获取,
    式中:U lim为给定电压极限值。
  8. 根据权利要求2所述永磁同步电机最大功率控制电流轨迹搜索方法,其特征在于,电压幅值U(θ)按下式获取:
    Figure PCTCN2021123463-appb-100007
    其中直轴电压
    Figure PCTCN2021123463-appb-100008
    交轴电压
    Figure PCTCN2021123463-appb-100009
    w为电机的电角速度,R 1为电机电阻。
  9. 永磁同步电机最大功率控制在线控制方法,其特征在于,采用权利要求1~8任一权利要求所述的永磁同步电机最大功率控制电流轨迹搜索方法得到永磁同步电机在多个工作点下的电流轨迹,将这些电流轨迹作为样本数据,训练生成最大功率控制神经网络模型,最大功率控制神经网络模型的输入为电机的转速、转矩、电流极限值和电压极限值,输出为电流幅值与电流角;
    将最大功率控制神经网络模型加载至DSP或FPGA控制器中,可以实现永磁同步电机最大功率在线控制,根据电机的转速和转矩实时输出电流幅值与电流角用于控制电机在线最大功率运行。
  10. 永磁同步电机全速域效率最优控制电流轨迹搜索方法,其特征在于,该方法为:电机运行在基速值以下时,在给定的 转矩指令、转速指令、电压极限、电流极限下,采用MTPA控制方式获取电流幅值最小的电流工作点作为电流轨迹;电机运行在基速值以上时,在给定的转矩指令、转速指令、电压极限、电流极限下,采用弱磁区效率最优控制方式获取电流幅值最小的电流工作点作为电流轨迹;
    采用MTPA控制方式获取电流幅值最小的电流工作点的过程包括电流角迭代循环步骤和电流幅值迭代循环步骤,首先进行电流角迭代循环步骤,电流角迭代方向为电流幅值减小的方向;在进行电流角迭代过程中,嵌套电流幅值迭代循环步骤,用以确定每个电流角对应的电流幅值,电流幅值的迭代方向为给定转矩与实际转矩误差减小的方向,当电流角的迭代区间小于给定电流角迭代精度,认为电流幅值已经收敛至最小值,输出MTPA电流轨迹;
    采用弱磁区效率最优控制方式获取电流幅值最小的电流工作点的过程包括弱磁电流角迭代循环步骤和电流幅值迭代循环步骤,首先进行电流角迭代循环步骤,电流角迭代方向为电压极限下电流幅值减小的方向;在进行电流角迭代的过程中,嵌套电流幅值迭代循环步骤,用以确定每个电流角对应的电流幅值,电流幅值的迭代方向为给定转矩与实际转矩误差减小的方向,当电流角的迭代区间小于给定电流角迭代精度,认为电流幅值已经收敛至最小值,输出弱磁区效率最优控制电流轨迹。
  11. 根据权利要求10所述的永磁同步电机全速域效率最优控制电流轨迹搜索方法,其特征在于,所述采用MTPA控制方式获取电流幅值最小的电流工作点的过程包括电流角迭代循环步骤和电流幅值迭代循环步骤:
    电流角迭代循环步骤包括:
    A1、初始化电流角初值区间[a 1,b 1],并计算电流角试探点初值λ 1、β 1
    λ 1=a 1+0.382(b 1-a 1)、β 1=a 1+0.618(b 1-a 1);
    A2、判断两电流角试探点处电流幅值目标函数值I(λ k)和I(β k)是否存在关系I(λ k)>I(β k),电流角迭代次数k=1,2,3...
    判断结果为是,执行步骤A3;判断结果为否执行步骤A5;
    电流幅值目标函数值I(λ k)和I(β k)通过调用电流幅值迭代循环获取;
    A3、令a k+1=λ k,b k+1=b k,λ k+1=β k,β k+1=a k+1+0.618(b k+1-a k+1),
    A4、调用电流幅值迭代循环获取电流幅值目标函数值I(β k+1),然后执行步骤A7;
    A5、令a k+1=a k,b k+1=β k,β k+1=λ k,λ k+1=a k+1+0.382(b k+1-a k+1),
    A6、调用电流幅值迭代循环获取电流幅值目标函数值I(λ k+1),然后执行步骤A7;
    A7、令k=k+1;
    A8、判断迭代是否收敛:若b k-a k<L 1,执行步骤A9;否则,返回步骤A2;
    其中L 1为电流角迭代精度;
    A9、判断电流工作点是否同时满足电流极限与电压极限的要求:若I(λ k)≤I lim&U(λ k)≤U lim,I lim为给定电流极限值,U lim为给定电压极限值,输出MTPA电流轨迹;否则,重新输入转矩、转速指令,再返回执行步骤A1;
    电流幅值迭代循环步骤包括:
    B1、初始化电流幅值的初值区间:[c 1,d 1],并计算电流幅值试探点初值μ 1、v 1
    μ 1=c 1+0.382(d 1-c 1)、v 1=c 1+0.618(d 1-c 1);
    B2、计算两电流幅值试探点处的转矩误差目标函数值:f(μ 1)、f(v 1),
    转矩误差目标函数f(I)按
    Figure PCTCN2021123463-appb-100010
    获取,其中:
    Figure PCTCN2021123463-appb-100011
    为给定转矩,T e(I,θ)为电流角θ对应的转矩,电流角θ为电流角迭代循环输出的电流角试探点λ k、β k;I为电流幅值;
    B3、判断两电流幅值试探点处转矩误差目标函数值f(μ h)和f(ν h)是否存在关系f(μ h)>f(ν h),电流幅值迭代次数h=1,2,3…
    判断结果为是,执行步骤B4;判断结果为否执行步骤B5;
    B4、令c h+1=μ h,d h+1=d h,μ h+1=v h,v h+1=c h+1+0.618(d h+1-c h+1),
    计算目标函数值f(v h+1),然后步骤B6;
    B5、令c h+1=c h,d h+1=v h,v h+1=μ h,μ h+1=c h+1+0.382(d h+1-c h+1),
    计算目标函数值f(μ h+1),然后步骤B6;
    B6、令h=h+1,
    B7、判断迭代是否收敛:若d h-c h<L 2,输出给定电流角对应的电流幅值I(θ)、电压幅值U(θ),输出结果用于电流角的迭代搜索过程;否则,返回步骤B3;其中L 2为电流幅值迭代精度。
  12. 根据权利要求10所述的永磁同步电机全速域效率最优控制电流轨迹搜索方法,其特征在于,采用弱磁区效率最优控制方式获取电流幅值最小的电流工作点的过程包括弱磁电流角迭代循环步骤和电流幅值迭代循环步骤:
    所述弱磁电流角迭代循环步骤包括:
    C1、初始化电流角初值区间[a 1,b 1],并计算电流角试探点初值λ 1、β 1
    λ 1=a 1+0.382(b 1-a 1)、β 1=a 1+0.618(b 1-a 1);
    C2、判断负载电压目标函数值U(β k)和电压极限值U lim的大小关系,若U(β k)>U lim,执行步骤C6;否则,执行步骤C3;
    负载电压目标函数值U(β k)通过调用电流幅值迭代循环获取,电流角迭代次数k=1,2,3…;
    C3、判断两电流角试探点处电流幅值目标函数值I(λ k)和I(β k)是否存在关系I(λ k)>I(β k),
    判断结果为是,执行步骤C4;判断结果为否执行步骤C6;
    电流幅值目标函数值I(λ k)和I(β k)通过调用电流幅值迭代循环获取;
    C4、令a k+1=λ k,b k+1=b k,λ k+1=β k,β k+1=a k+1+0.618(b k+1-a k+1),
    C5、调用电流幅值迭代循环获取电流幅值目标函数值I(β k+1),然后执行步骤C8;
    C6、令a k+1=a k,b k+1=β k,β k+1=λ k,λ k+1=a k+1+0.382(b k+1-a k+1),
    C7、调用电流幅值迭代循环获取电流幅值目标函数值I(λ k+1),然后执行步骤C8;
    C8、令k=k+1;
    C9、判断迭代是否收敛:若b k-a k<L 1,执行步骤C10;否则,返回步骤C2;
    其中L 1为电流角迭代精度;
    C10、判断电流工作点是否同时满足电流极限的要求:若I(λ k)≤I lim,I lim为给定电流极限值,输出弱磁区效率最优控制电流轨迹;否则,重新输入转矩、转速指令,再返回执行步骤C1;
    电流幅值迭代循环步骤包括:
    B1、初始化电流幅值的初值区间:[c 1,d 1],并计算电流幅值试探点初值μ 1、v 1
    μ 1=c 1+0.382(d 1-c 1)、v 1=c 1+0.618(d 1-c 1);
    B2、计算两电流幅值试探点处的转矩误差目标函数值:f(μ 1)、f(v 1),
    转矩误差目标函数f(I)按
    Figure PCTCN2021123463-appb-100012
    获取,其中:
    Figure PCTCN2021123463-appb-100013
    为给定转矩,T e(I,θ)为电流角θ对应的转矩,电流角θ为电流角迭代循环输出的电流角试探点λ k、β k;I为电流幅值;
    B3、判断两电流幅值试探点处转矩误差目标函数值f(μ h)和f(ν h)是否存在关系f(μ h)>f(ν h),电流幅值迭代次数h=1,2,3…
    判断结果为是,执行步骤B4;判断结果为否执行步骤B5;
    B4、令c h+1=μ h,d h+1=d h,μ h+1=v h,v h+1=c h+1+0.618(d h+1-c h+1),
    计算目标函数值f(v h+1),然后步骤B6;
    B5、令c h+1=c h,d h+1=v h,v h+1=μ h,μ h+1=c h+1+0.382(d h+1-c h+1),
    计算目标函数值f(μ h+1),然后步骤B6;
    B6、令h=h+1,
    B7、判断迭代是否收敛:若d h-c h<L 2,输出给定电流角对应的电流幅值I(θ)、电压幅值U(θ),输出结果用于电流角的迭代搜索过程;否则,返回步骤B3;其中L 2为电流幅值迭代精度。
  13. 根据权利要求10-12任一权利要求所述永磁同步电机全速域效率最优控制电流轨迹搜索方法,其特征在于,电流轨迹为:电流幅值I=I(λ k)、电流角θ=λ k
  14. 根据权利要求12所述永磁同步电机全速域效率最优控制电流轨迹搜索方法,其特征在于,转矩T e(I,θ)由电机非线性负载交直轴磁链模型计算输出,按如下公式获取:
    T e(I,θ)=p(ψ d(I,θ)i qq(I,θ)i d)
    其中,p为电机极对数,i d为电机的直轴电流,i q为电机的交轴电流,ψ d为电机的直轴磁链,ψ q为电机的交轴磁链。
  15. 根据权利要求14所述永磁同步电机全速域效率最优控制电流轨迹搜索方法,其特征在于,电机非线性负载交直轴磁链模型的建立过程:
    在电机的电流极限范围内等距或不等距的选取一系列电流工作点,包括等距或不等距电流幅值系列值及等距或不等距电流角系列值,所选取的电流工作点间距由电机的饱和程度决定,需要保证相邻两电流工作点之间的铁心磁导率保持不变,铁心按线性材料处理;
    采用仿真或实验的方式,计算所选取的电流工作点对应的电机负载交、直轴磁链数据,并将得到的负载交、直轴磁链数据进行插值,得到电流极限范围内所有电流工作点的负载交、直轴磁链模型,即永磁同步电机的非线性磁链模型:
    ψ d(I,θ)=ψ d(i d,i q)
    ψ q(I,θ)=ψ q(i d,i q)。
  16. 根据权利要求15所述永磁同步电机全速域效率最优控制电流轨迹搜索方法,其特征在于,电压幅值U(θ)按下式获取:
    Figure PCTCN2021123463-appb-100014
    其中,直轴电压
    Figure PCTCN2021123463-appb-100015
    交轴电压
    Figure PCTCN2021123463-appb-100016
    w为电机的电角速度,R 1为电机电阻。
  17. 永磁同步电机全速域效率最优控制在线控制方法,其特征在于,采用权利要求10~16任一权利要求所述的永磁同步电机弱磁区效率最优控制电流轨迹搜索方法获取全速域范围内的多个电流工作点,包括基速值以下采用MTPA控制方式获取的电流工作点,和基速值以上采用弱磁区效率最优控制方式获取的电流工作点;
    将这些电流工作点作为样本数据,训练生成全速域效率最优控制神经网络模型,全速域效率最优控制神经网络模型的输入为电机的转速、转矩、电流极限值和电压极限值,输出为电流幅值与电流角;
    将全速域效率最优控制神经网络模型加载至DSP或FPGA控制器中,可以实现永磁同步电机在全速域范围内效率最优在线控制,根据电机的转速和转矩实时输出电流幅值与电流角用于控制电机运行。
PCT/CN2021/123463 2020-12-21 2021-10-13 永磁同步电机最大功率及全速域效率最优控制电流轨迹搜索方法 WO2022134751A1 (zh)

Applications Claiming Priority (10)

Application Number Priority Date Filing Date Title
CN202011519801.0A CN112468037B (zh) 2020-12-21 2020-12-21 永磁同步电机mtpv控制电流轨迹搜索方法及在线控制方法
CN202011519801.0 2020-12-21
CN202011519810.X 2020-12-21
CN202011519810.XA CN112468038B (zh) 2020-12-21 2020-12-21 永磁同步电机mtpa控制电流轨迹搜索方法及在线控制方法
CN202011519799.7A CN112468036B (zh) 2020-12-21 2020-12-21 永磁同步电机全速域效率最优控制电流轨迹搜索方法及在线控制方法
CN202011519778.5A CN112468033B (zh) 2020-12-21 2020-12-21 永磁同步电机最大功率控制电流轨迹搜索方法及在线控制方法
CN202011519783.6A CN112468034B (zh) 2020-12-21 2020-12-21 永磁同步电机弱磁区效率最优控制电流轨迹搜索方法及在线控制方法
CN202011519778.5 2020-12-21
CN202011519783.6 2020-12-21
CN202011519799.7 2020-12-21

Publications (1)

Publication Number Publication Date
WO2022134751A1 true WO2022134751A1 (zh) 2022-06-30

Family

ID=82158764

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2021/123463 WO2022134751A1 (zh) 2020-12-21 2021-10-13 永磁同步电机最大功率及全速域效率最优控制电流轨迹搜索方法

Country Status (1)

Country Link
WO (1) WO2022134751A1 (zh)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117335701A (zh) * 2023-10-08 2024-01-02 浙江大学 永磁同步电机拓宽极限边界的显式模型预测弱磁控制方法

Citations (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103872959A (zh) * 2014-03-21 2014-06-18 合肥工业大学 一种增强型永磁同步电机的弱磁控制方法
KR20160141942A (ko) * 2015-06-01 2016-12-12 엘에스산전 주식회사 전동기 제어 방법
CN109873590A (zh) * 2019-04-09 2019-06-11 湘潭大学 一种电动汽车用内嵌式永磁同步电机的弱磁扩速方法
CN110311611A (zh) * 2019-06-27 2019-10-08 国电南瑞科技股份有限公司 一种永磁同步电机弱磁控制方法和系统
CN111756291A (zh) * 2020-07-08 2020-10-09 东风电子科技股份有限公司 新能源汽车永磁同步电机控制器实现抑制电压饱和失控控制的方法
CN111884552A (zh) * 2020-07-02 2020-11-03 华中科技大学 基于电压反馈的永磁同步电机弱磁优化控制方法及系统
CN112468036A (zh) * 2020-12-21 2021-03-09 哈尔滨工业大学 永磁同步电机全速域效率最优控制电流轨迹搜索方法及在线控制方法
CN112468037A (zh) * 2020-12-21 2021-03-09 哈尔滨工业大学 永磁同步电机mtpv控制电流轨迹搜索方法及在线控制方法
CN112468033A (zh) * 2020-12-21 2021-03-09 哈尔滨工业大学 永磁同步电机最大功率控制电流轨迹搜索方法及在线控制方法
CN112468034A (zh) * 2020-12-21 2021-03-09 哈尔滨工业大学 永磁同步电机弱磁区效率最优控制电流轨迹搜索方法及在线控制方法
CN112468038A (zh) * 2020-12-21 2021-03-09 哈尔滨工业大学 永磁同步电机mtpa控制电流轨迹搜索方法及在线控制方法

Patent Citations (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103872959A (zh) * 2014-03-21 2014-06-18 合肥工业大学 一种增强型永磁同步电机的弱磁控制方法
KR20160141942A (ko) * 2015-06-01 2016-12-12 엘에스산전 주식회사 전동기 제어 방법
CN109873590A (zh) * 2019-04-09 2019-06-11 湘潭大学 一种电动汽车用内嵌式永磁同步电机的弱磁扩速方法
CN110311611A (zh) * 2019-06-27 2019-10-08 国电南瑞科技股份有限公司 一种永磁同步电机弱磁控制方法和系统
CN111884552A (zh) * 2020-07-02 2020-11-03 华中科技大学 基于电压反馈的永磁同步电机弱磁优化控制方法及系统
CN111756291A (zh) * 2020-07-08 2020-10-09 东风电子科技股份有限公司 新能源汽车永磁同步电机控制器实现抑制电压饱和失控控制的方法
CN112468036A (zh) * 2020-12-21 2021-03-09 哈尔滨工业大学 永磁同步电机全速域效率最优控制电流轨迹搜索方法及在线控制方法
CN112468037A (zh) * 2020-12-21 2021-03-09 哈尔滨工业大学 永磁同步电机mtpv控制电流轨迹搜索方法及在线控制方法
CN112468033A (zh) * 2020-12-21 2021-03-09 哈尔滨工业大学 永磁同步电机最大功率控制电流轨迹搜索方法及在线控制方法
CN112468034A (zh) * 2020-12-21 2021-03-09 哈尔滨工业大学 永磁同步电机弱磁区效率最优控制电流轨迹搜索方法及在线控制方法
CN112468038A (zh) * 2020-12-21 2021-03-09 哈尔滨工业大学 永磁同步电机mtpa控制电流轨迹搜索方法及在线控制方法

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117335701A (zh) * 2023-10-08 2024-01-02 浙江大学 永磁同步电机拓宽极限边界的显式模型预测弱磁控制方法

Similar Documents

Publication Publication Date Title
WO2022134661A1 (zh) 可调磁通永磁同步电机全速域效率最优控制磁化状态选择方法及在线控制方法
CN112468038B (zh) 永磁同步电机mtpa控制电流轨迹搜索方法及在线控制方法
Miao et al. MTPA fitting and torque estimation technique based on a new flux-linkage model for interior-permanent-magnet synchronous machines
Ren et al. Direct torque control of permanent-magnet synchronous machine drives with a simple duty ratio regulator
CN107896080B (zh) 内嵌式永磁同步电机mtpa曲线拟合方法及控制系统
US20220294375A1 (en) Spmsm sensorless composite control method with dual sliding-mode observers
CN109347385B (zh) 模型参考自适应算法自适应率参数整定方法
CN107659231B (zh) 一种超高速永磁同步电机转速控制方法
CN111953250A (zh) 一种注入谐波电流的永磁同步电机转矩脉动抑制方法
CN112468034B (zh) 永磁同步电机弱磁区效率最优控制电流轨迹搜索方法及在线控制方法
CN112865639B (zh) 含路况复现功能的电动汽车永磁同步电机控制系统
CN112468033B (zh) 永磁同步电机最大功率控制电流轨迹搜索方法及在线控制方法
WO2022134751A1 (zh) 永磁同步电机最大功率及全速域效率最优控制电流轨迹搜索方法
CN111478633A (zh) 一种永磁同步电机无差拍预测控制方法
Chen et al. Multiobjective optimization design of unequal halbach array permanent magnet vernier motor based on optimization algorithm
Hu et al. Improved loss model and loss minimization control strategy for linear induction machine
CN114499307A (zh) 一种永磁同步电机电流环解耦控制方法
CN112468036B (zh) 永磁同步电机全速域效率最优控制电流轨迹搜索方法及在线控制方法
CN112468037B (zh) 永磁同步电机mtpv控制电流轨迹搜索方法及在线控制方法
CN104579092B (zh) 电机的控制方法、控制系统及电机电感的计算方法、装置
CN112468032B (zh) 永磁同步电机的全速域效率map图生成方法
Hu et al. Dynamic loss minimization control of linear induction machine
CN117277878A (zh) 一种基于相角补偿的电机带载启动控制方法
CN114244216A (zh) 一种永磁同步电机参数辨识方法、装置及系统
CN103701388A (zh) 一种用于永磁直驱风电系统速度辨识策略的选择方法

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 21908752

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 21908752

Country of ref document: EP

Kind code of ref document: A1

122 Ep: pct application non-entry in european phase

Ref document number: 21908752

Country of ref document: EP

Kind code of ref document: A1