A torque control method for a permanent magnet synchronous motor for an aircraft electromechanical actuator
By using expanded state observer and discrete space vector modulation technology in aeronautical electromechanical actuators, the electromagnetic torque control is optimized, and the problems of parameter perturbation and dynamic response performance are solved, achieving high robustness and high dynamicity of the system , and shortens the algorithm calculation time.
Patent Information
- Application Number
- CN202411159456.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-22
- Publication Date
- 2024-11-22
- Estimated Expiration
- 2044-08-22
AI Technical Summary
Aviation electromechanical actuators have parameter perturbation problems when the temperature difference in the aircraft operating environment is large and affected by thunderstorms. The existing linear control strategies are difficult to meet the requirements of nonlinear dynamic response performance, and the control chip resources are limited, and the algorithm calculation time is long.
Using super-local prediction model based on expansion state observer and discrete space vector modulation technology, the cost function is constructed to optimize electromagnetic torque control, compensate parameter perturbation in real time, and improve dynamic performance, shorten the algorithm calculation time.
It improves the robustness and dynamic response performance of avionics and electromechanical actuators, reduces algorithm calculation time, and realizes high dynamicity and high reliability of the system.
Smart Images

Figure CN118694232B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of aviation motor control, and in particular to a torque control method of a permanent magnet synchronous motor for an aviation electromechanical actuator. Background Art
[0002] Aviation electromechanical actuators are an important part of modern aircraft and are used to control the movement and attitude of aircraft. Aviation electromechanical actuators have the advantages of high precision, strong reliability, fast response speed and simple maintenance. Compared with traditional hydraulic actuators, electromechanical actuators are driven by motors and do not require liquid media, so they are more energy-saving and environmentally friendly. At the same time, with the continuous development of aircraft, the technology of actuators is also constantly innovating and improving, and the requirements for electromechanical actuator control systems are also getting higher and higher.
[0003] In aircraft electromechanical actuators, the control system has a vital impact on the stability and safety of the aircraft. Aircraft electromechanical actuators require high-performance closed-loop control systems to meet actual dynamic and static application requirements, and need to seek control methods with fast response, high speed regulation accuracy, high reliability and strong robustness. When designing actual controllers, in order to improve real-time performance, it is necessary to minimize the computer's occupancy time and use control methods with simple algorithms and small calculations. In addition, it is also necessary to adopt control methods with strong adaptability and fast dynamic response.
[0004] Most of the airborne electric actuators use the traditional PID control method with simple algorithms and small computational complexity, but the robustness of this method when applied to parameter perturbation conditions is weak. In order to improve the robust control performance, modern control and intelligent control methods with strong adaptability have been introduced, such as adaptive control methods, fuzzy control methods, and neural network control methods. These control methods will adaptively change the control parameters according to the changes in system parameters, but the control system is relatively complex, and for fuzzy control and neural network control, better control performance needs to be based on perfect prior experimental data. In recent years, model predictive torque control has the advantages of simple structure, high dynamics, and multi-objective nonlinear optimal control, which is very suitable for airborne electric actuators. However, due to the limitation of model parameters, the robustness still needs to be improved.
[0005] In summary, considering the large temperature difference in the aircraft operating environment and the influence of thunderstorms, there is a problem of aviation electromechanical actuation motor parameter perturbation in aviation electromechanical actuation; secondly, the existing linear control strategy for controlling nonlinear aviation electromechanical actuators will limit the dynamic response performance; finally, due to the constraints of controller cost and space, the control chip resources of the electromechanical actuation system are small and the functions are complex to implement, and the calculation time of the control system algorithm needs to be shortened.
[0006] Therefore, it is necessary to provide a torque control method for a permanent magnet synchronous motor for an aircraft electromechanical actuator to solve the above problems. Summary of the invention
[0007] The present invention provides a torque control method for a permanent magnet synchronous motor for an aviation electromechanical actuator, so as to solve the problem that the aircraft operating environment has a large temperature difference and is affected by thunderstorms, and the aviation electromechanical actuation has the problem of perturbation of the parameters of the aviation electromechanical actuation motor; secondly, the existing linear control strategy controls the nonlinear aviation electromechanical actuator, which will limit the dynamic response performance; finally, due to the constraints of controller cost and space, the control chip resources of the electromechanical actuation system are small and the functions are complex to implement, and the calculation time of the control system algorithm needs to be shortened.
[0008] The present invention provides a method for controlling the torque of a permanent magnet synchronous motor for an aircraft electromechanical actuator, which adopts the following technical solution, including:
[0009] Obtaining the α-axis current, β-axis current, d-axis current and q-axis current of the permanent magnet synchronous motor according to the three-phase current of the permanent magnet synchronous motor; obtaining the electromagnetic torque, α-axis stator flux, β-axis stator flux, d-axis stator flux and q-axis stator flux according to the d-axis current and q-axis current of the permanent magnet synchronous motor;
[0010] According to the rotor angular velocity of the permanent magnet synchronous motor and a preset rotor angular velocity reference value, an electromagnetic torque reference value is obtained by using a proportional integral control algorithm;
[0011] The discrete space vector modulation technology is used to obtain the selected voltage vector; according to the electromagnetic torque, α-axis stator flux, β-axis stator flux, α-axis current, β-axis current and the selected voltage vector, a super-local prediction model based on the extended state observer is constructed, and the stator current prediction value, stator flux prediction value and electromagnetic torque prediction value of the permanent magnet synchronous motor at the next moment under each selected voltage vector are obtained by using the super-local prediction model;
[0012] Based on the bandwidth of the extended state observer, the electromagnetic torque reference value and the electromagnetic torque prediction value of the permanent magnet synchronous motor at the next moment under each candidate voltage vector, the dynamic bandwidth of the permanent magnet synchronous motor under each candidate voltage vector is obtained; according to the electromagnetic torque reference value, the preset stator flux reference value and the dynamic bandwidth, the cost function value of the permanent magnet synchronous motor under each candidate voltage vector is obtained;
[0013] The candidate voltage vector corresponding to the minimum cost function value is taken as the optimal voltage vector, and the switching state of the switch tube of the three-phase full-bridge inverter is changed according to the optimal voltage vector to control the torque of the permanent magnet synchronous motor.
[0014] Preferably, the steps of obtaining the α-axis current, β-axis current, d-axis current and q-axis current of the permanent magnet synchronous motor according to the three-phase current of the permanent magnet synchronous motor are: the three-phase current of the permanent magnet synchronous motor is subjected to park transformation to obtain the α-axis current and β-axis current, and then the d-axis current and q-axis current of the permanent magnet synchronous motor are obtained after clark transformation.
[0015] Preferably, the expression of the electromagnetic torque reference value is:
[0016]
[0017] In the formula, For permanent magnet synchronous motor The electromagnetic torque reference value at the moment;
[0018] is the rotor angular velocity;
[0019] is the preset rotor angular velocity reference value;
[0020] is the proportional parameter of the proportional-integral control algorithm;
[0021] is the integral parameter of the proportional-integral control algorithm;
[0022] It is the integral operation of the proportional-integral control algorithm.
[0023] Preferably, the steps of obtaining the electromagnetic torque, the α-axis stator flux, the β-axis stator flux, the d-axis stator flux and the q-axis stator flux are:
[0024]
[0025]
[0026]
[0027] In the formula, is the d-axis stator flux;
[0028] is the q-axis stator flux;
[0029] is the electromagnetic torque;
[0030] is the number of permanent magnet pole pairs;
[0031] is the permanent magnet flux;
[0032] is the stator inductance;
[0033] is the d-axis current;
[0034] is the q-axis current;
[0035] Based on the d-axis stator flux and the q-axis stator flux, the α-axis stator flux is obtained by using the inverse Clark transformation and β-axis stator flux .
[0036] Preferably, the step of obtaining the selected voltage vector by using discrete space vector modulation technology is:
[0037] Based on the voltage amplitude of the basic voltage vector of the three-phase full-bridge inverter circuit of the permanent magnet synchronous motor, a two-equal-division virtual voltage vector and a three-equal-division virtual voltage vector are obtained;
[0038] Adding the bisected virtual voltage vector and the trisected virtual voltage vector to obtain a plurality of mixed virtual voltage vectors;
[0039] A basic voltage vector, a two-equally divided virtual voltage vector, a three-equally divided virtual voltage vector, and a mixed virtual voltage vector are used as voltage vectors to be selected.
[0040] Preferably, the steps of obtaining the virtual voltage vector divided into two equal parts and the virtual voltage vector divided into three equal parts are:
[0041] Dividing the voltage amplitude of each basic voltage vector by two to obtain a first target voltage vector;
[0042] Every two first target voltage vectors are added to obtain a 2-halved virtual voltage vector;
[0043] Dividing the voltage amplitude of each basic voltage vector by three to obtain a second target basic voltage vector;
[0044] Every two second target basic voltage vectors are added together to obtain a three-divided virtual voltage vector.
[0045] Preferably, the expression of the super local prediction model is:
[0046]
[0047]
[0048] In the formula, Stator current at time ; j is an imaginary unit;
[0049] The predicted value of stator current at time ; for The predicted value of stator current at the moment;
[0050] Stator flux at time ;
[0051] The predicted value of stator flux at time ; for The predicted value of stator flux at time; for α-axis stator flux at time ; for α-axis stator flux at time ;
[0052] for The predicted value of electromagnetic torque at the moment; for The predicted value of electromagnetic torque at the moment;
[0053] for The estimated value of the current disturbance at time for The estimated value of the current disturbance at time
[0054] for The estimated value of the flux disturbance at time; for The estimated value of the flux disturbance at time;
[0055] is the stator inductance;
[0056] is the number of permanent magnet pole pairs;
[0057] for The voltage vector to be selected input at any time;
[0058] Parameters for algorithm execution period;
[0059] is the first adjustable gain parameter of the extended state observer, is the second adjustable gain parameter of the extended state observer, is the third adjustable gain parameter of the extended state observer, is the fourth adjustable gain parameter of the extended state observer, where , , is the bandwidth of the extended state observer.
[0060] Preferably, the expression of the dynamic bandwidth of the permanent magnet synchronous motor under each selected voltage vector is:
[0061]
[0062] In the formula, is the dynamic bandwidth;
[0063] is the bandwidth of the extended state observer;
[0064] is the controller parameter, the range is 0-1;
[0065] is the hyperbolic tangent function;
[0066] For permanent magnet synchronous motor The electromagnetic torque reference value at the moment;
[0067] for The predicted value of electromagnetic torque at time.
[0068] Preferably, the expression of the cost function value of the permanent magnet synchronous motor under each candidate voltage vector is:
[0069]
[0070] In the formula, is the cost function value;
[0071] For permanent magnet synchronous motor The electromagnetic torque reference value at the moment;
[0072] For permanent magnet synchronous motor The predicted value of electromagnetic torque at the moment;
[0073] is the dynamic bandwidth;
[0074] is the preset stator flux reference value;
[0075] is the weight coefficient;
[0076] is the predicted value of stator flux.
[0077] The beneficial effects of the present invention are:
[0078] The torque control method of the surface-mounted permanent magnet synchronous motor for the aviation electromechanical actuator proposed in the present invention aims at the parameter perturbation problem, dynamic response problem and algorithm calculation time problem under the high and low temperature and strong vibration conditions of the aviation electromechanical actuator, and proposes a super-local prediction model based on the extended state observer, a discrete space vector modulation technology and a cost function. By using the super-local prediction model based on the extended state observer to compensate in real time for the error between the preset motor parameters and the actual motor parameters caused by the parameter perturbation, the robustness of the control system is improved. Aiming at the dynamic response problem of the aviation electromechanical actuator, a cost function is designed. By establishing the controller bandwidth and the candidate voltage vector The dynamic performance of the controller is added as a control target in the cost function, thereby enhancing the dynamic performance of the motor drive system; in view of the problem of algorithm calculation time of aviation electromechanical actuators, a discrete space voltage vector selection strategy is designed, and the candidate voltage vectors are evaluated through a predetermined evaluation process, and the initial optimal voltage vector is selected, and then the initial optimal voltage vector and the super local prediction model are used to predict the parameters, thereby reducing the amount of data and shortening the algorithm calculation time. That is, the present invention improves the dynamic and static performance of the drive system, shortens the algorithm calculation time, and makes the system have the advantages of strong robustness, high dynamics and high reliability. BRIEF DESCRIPTION OF THE DRAWINGS
[0079] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.
[0080] Figure 1 It is a flow chart of a torque control method of a permanent magnet synchronous motor for an aircraft electromechanical actuator according to the present invention;
[0081] Figure 2 is a basic voltage vector diagram in an embodiment of the present invention;
[0082] Figure 3 A virtual voltage vector diagram divided into two equal parts in an embodiment of the present invention;
[0083] Figure 4 A virtual voltage vector diagram divided into three equal parts in an embodiment of the present invention;
[0084] Figure 5 A hybrid virtual voltage vector diagram in an embodiment of the present invention;
[0085] Figure 6 A voltage vector diagram to be selected in an embodiment of the present invention;
[0086] Figure 7is a candidate voltage vector diagram located in the first sector of the basic voltage vector diagram in the embodiment of the present invention;
[0087] Figure 8 It is a framework diagram of a robust model predictive torque control system of a surface-mounted permanent magnet synchronous motor for an aircraft electromechanical actuator in an embodiment of the present invention;
[0088] Fig. 9 In the embodiment of the present invention, the electrical parameters are changed to The simulation result diagram of the speed tracking performance of the traditional model predictive torque control method;
[0089] Fig.10 In the embodiment of the present invention, the electrical parameters are changed to The simulation result diagram of the stator flux tracking performance of the traditional model predictive torque control method;
[0090] Fig.11 In the embodiment of the present invention, the electrical parameters are changed to The simulation result diagram of the electromagnetic torque tracking performance of the traditional model predictive torque control method;
[0091] Fig.12 In the embodiment of the present invention, the electrical parameters are changed to Graph showing simulation results of the speed tracking performance of the torque control method of this embodiment;
[0092] Fig.13 In the embodiment of the present invention, the electrical parameters are changed to : is a simulation result diagram of the stator flux tracking performance of the torque control method of this embodiment;
[0093] Fig.14 In the embodiment of the present invention, the electrical parameters are changed to 3 is a diagram showing the simulation results of the electromagnetic torque tracking performance of the torque control method of this embodiment;
[0094] Fig.15 In the embodiment of the present invention, the electrical parameters are changed to The simulation result diagram of the speed tracking performance of the traditional model predictive torque control method;
[0095] Fig.16 In the embodiment of the present invention, the electrical parameters are changed to , , The simulation result diagram of the stator flux tracking performance of the traditional model predictive torque control method;
[0096] Fig.17 In the embodiment of the present invention, the electrical parameters are changed to , , The simulation result diagram of the electromagnetic torque tracking performance of the traditional model predictive torque control method;
[0097] Fig.18 In the embodiment of the present invention, the electrical parameters are changed to , , The simulation result diagram of the dynamic speed tracking performance of the torque control method of this embodiment;
[0098] Fig.19 In the embodiment of the present invention, the electrical parameters are changed to , , : is a simulation result diagram of the stator flux tracking performance of the torque control method of this embodiment;
[0099] Fig. 20 In the embodiment of the present invention, the electrical parameters are changed to , , 3 is a diagram showing the simulation results of the electromagnetic torque tracking performance of the torque control method of this embodiment;
[0100] Fig.21 In the embodiment of the present invention, the electrical parameters are changed to , , The simulation result diagram of phase current THD of the traditional model prediction torque control method;
[0101] Fig. 22 In the embodiment of the present invention, the electrical parameters are changed to , , : is a simulation result diagram of phase current THD of the torque control method of this embodiment;
[0102] Fig.23 In the embodiment of the present invention, the electrical parameters are changed to , , The simulation result diagram of phase current THD of the traditional model prediction torque control method;
[0103] Fig.24 In the embodiment of the present invention, the electrical parameters are changed to , , FIG. 5 is a diagram showing simulation results of phase current THD of the torque control method of this embodiment. DETAILED DESCRIPTION
[0104] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0105] An embodiment of a torque control method for a permanent magnet synchronous motor for an aircraft electromechanical actuator of the present invention is as follows: Figure 1 As shown, including:
[0106] S1, obtaining the α-axis current, β-axis current, d-axis current and q-axis current of the permanent magnet synchronous motor, as well as the electromagnetic torque, α-axis stator flux, β-axis stator flux, d-axis stator flux and q-axis stator flux;
[0107] Specifically, obtain the three-phase current of the permanent magnet synchronous motor, the three-phase current is the a-phase current , b-phase current , c-phase current ; Obtain the d-axis current and q-axis current of the permanent magnet synchronous motor according to the three-phase current of the permanent magnet synchronous motor; Obtain the electromagnetic torque, d-axis stator flux and q-axis stator flux according to the d-axis current and q-axis current of the permanent magnet synchronous motor.
[0108] Step 11, the steps of obtaining the α-axis current, β-axis current, d-axis current and q-axis current of the permanent magnet synchronous motor are as follows: the three-phase current of the permanent magnet synchronous motor is subjected to park transformation to obtain the α-axis current and β-axis current, and then the d-axis current and q-axis current of the permanent magnet synchronous motor are obtained after clark transformation.
[0109] Among them, the expression of park transformation is as follows:
[0110] (1)
[0111] In formula (1), is the α-axis current, is the β-axis current.
[0112] The expression of Clark transformation is as follows:
[0113] (2)
[0114] In formula (2), is the α-axis current, is the β-axis current; is the rotor electrical angle, , is the number of permanent magnet pole pairs.
[0115] Step 12: The steps of obtaining the electromagnetic torque, the α-axis stator flux, the β-axis stator flux, the d-axis stator flux, and the q-axis stator flux according to the d-axis current and the q-axis current of the permanent magnet synchronous motor are as follows:
[0116] In this embodiment, the d-axis current is used according to the permanent magnet synchronous motor current and flux linkage model. and q-axis current Calculate the electromagnetic torque , d-axis stator flux , q-axis stator flux , that is, the expression of the permanent magnet synchronous motor current and flux linkage model is:
[0117] (3)
[0118] In formula (3), is the d-axis stator flux; is the q-axis stator flux; is the electromagnetic torque; is the number of permanent magnet pole pairs; is the permanent magnet flux; is the stator inductance; is the d-axis current; is the q-axis current.
[0119] Based on the d-axis stator flux and the q-axis stator flux, the α-axis stator flux is obtained by using the inverse Clark transformation and β-axis stator flux , the expression is:
[0120] (4)
[0121] In formula (4), is the α-axis stator flux, is the β-axis stator flux; is the rotor electrical angle, is the d-axis stator flux, is the q-axis stator flux.
[0122] S2, obtaining an electromagnetic torque reference value;
[0123] Specifically, the electromagnetic torque reference value is obtained according to the rotor angular velocity of the permanent magnet synchronous motor, a preset rotor angular velocity reference value and a rotor angle, and by using a proportional-integral control algorithm.
[0124] In this embodiment, the rotor angular velocity is and the preset rotor angular velocity reference value is , the rotor angular velocity and the preset rotor angular velocity reference The input speed loop PI controller calculates the electromagnetic torque reference value , that is, the electromagnetic torque reference value The expression is:
[0125] (5)
[0126] In the formula, For permanent magnet synchronous motor The electromagnetic torque reference value at the moment; is the rotor angular velocity; is the preset rotor angular velocity reference value; is the proportional parameter of the proportional-integral control algorithm; is the integral parameter of the proportional-integral control algorithm; It is the integral operation of the proportional-integral control algorithm.
[0127] S3, obtaining a candidate voltage vector, constructing a super-local prediction model based on an extended state observer, and obtaining a stator current prediction value, a stator flux prediction value, and an electromagnetic torque prediction value of the permanent magnet synchronous motor at the next moment under each candidate voltage vector;
[0128] Specifically, discrete space vector modulation technology is used to obtain a candidate voltage vector; according to the electromagnetic torque, d-axis stator flux, q-axis stator flux and the candidate voltage vector, a hyperlocal prediction model based on an extended state observer is constructed, and the hyperlocal prediction model is used to obtain the stator current prediction value, stator flux prediction value and electromagnetic torque prediction value of the permanent magnet synchronous motor at the next moment under each candidate voltage vector.
[0129] Step 31: Get the voltage vector to be selected:
[0130] In this embodiment, discrete space vector modulation technology is used to obtain a voltage vector to be selected, and the specific steps are: based on the voltage amplitude of the basic voltage vector of the three-phase full-bridge inverter circuit of the permanent magnet synchronous motor, a two-equal virtual voltage vector and a three-equal virtual voltage vector are obtained; the two-equal virtual voltage vector and the three-equal virtual voltage vector are added to obtain a variety of mixed virtual voltage vectors; the basic voltage vector, the two-equal virtual voltage vector, the three-equal virtual voltage vector and the mixed virtual voltage vector are used as the voltage vector to be selected. In this embodiment, the basic voltage vector is represented as follows Figure 2 As shown, the voltage amplitude of each basic voltage vector is divided by two to obtain a first target voltage vector; every two first target voltage vectors are added to obtain a halved virtual voltage vector, and 18 halved virtual voltage vectors can be obtained; the halved virtual voltage vectors are as follows Figure 3As shown; the voltage amplitude of each basic voltage vector is divided by three to obtain a second target basic voltage vector; every two second target basic voltage vectors are added to obtain a three-divided virtual voltage vector, and 30 three-divided virtual voltage vectors can be obtained. The three-divided virtual voltage vectors are as follows Figure 4 By sequentially adding all the two-part virtual voltage vectors to the three-part virtual voltage vectors, 18 mixed virtual voltage vectors can be obtained. The mixed virtual voltage vectors are as follows: Figure 5 shown.
[0131] Among them, the expression of the voltage vector to be selected is:
[0132] (6)
[0133] In the formula, represents the basic voltage vector; Represents the bisection of the virtual voltage vector, expressed as ; It represents the virtual voltage vector divided into three equal parts, and the expression is ; Represents the hybrid virtual voltage vector, expressed as + ; represents the calculation cycle, , ; represents a first basic voltage vector; represents a second fundamental voltage vector; represents the third fundamental voltage vector; represents a fourth fundamental voltage vector; represents the fifth fundamental voltage vector; represents the sixth fundamental voltage vector; represents the seventh fundamental voltage vector; represents the eighth fundamental voltage vector.
[0134] Among them, the expression of 12 equally divided virtual voltage vectors is:
[0135] (7)
[0136] In the formula, represents the first bisected virtual voltage vector; represents the second bisected virtual voltage vector; represents the third bisected virtual voltage vector; represents the fourth bisected virtual voltage vector; represents the fifth bisected virtual voltage vector; represents the sixth bisected virtual voltage vector; represents the seventh bisected virtual voltage vector; represents the eighth bisected virtual voltage vector; represents the ninth bisected virtual voltage vector; represents the tenth bisected virtual voltage vector; represents the eleventh bisected virtual voltage vector; represents the twelfth bisected virtual voltage vector. Since the seventh basic voltage vector and the eighth fundamental voltage vector are all zero voltage vectors, so the eighth basic voltage vector is used in the formula Using the seventh fundamental voltage vector express.
[0137] Among them, the calculation expressions of 30 three-divided virtual voltage vectors are as follows:
[0138] (8)
[0139] In the formula, represents the first three-divided virtual voltage vector; represents the second three-divided virtual voltage vector; represents the third trisection virtual voltage vector; represents the fourth three-divided virtual voltage vector; represents the fifth three-divided virtual voltage vector; represents the sixth trisection virtual voltage vector; represents the seventh trisection virtual voltage vector; represents the eighth three-divided virtual voltage vector; represents the ninth trisection virtual voltage vector; represents the tenth trisection virtual voltage vector; represents the eleventh trisection virtual voltage vector; represents the twelfth trisection virtual voltage vector; represents the thirteenth trisection virtual voltage vector; represents the fourteenth trisection virtual voltage vector; represents the fifteenth three-divided virtual voltage vector; represents the sixteenth trisection virtual voltage vector; represents the seventeenth trisection virtual voltage vector; represents the eighteenth trisection virtual voltage vector; represents the nineteenth trisection virtual voltage vector; represents the twentieth trisection virtual voltage vector; represents the twenty-first trisection virtual voltage vector; represents the twenty-second trisection virtual voltage vector; represents the twenty-third trisection virtual voltage vector; represents the twenty-fourth trisection virtual voltage vector; represents the twenty-fifth trisection virtual voltage vector; represents the twenty-sixth trisection virtual voltage vector; represents the twenty-seventh trisection virtual voltage vector; represents the twenty-eighth trisection virtual voltage vector; represents the twenty-ninth trisection virtual voltage vector; represents the 30th three-divided virtual voltage vector. Since the seventh basic voltage vector and the eighth fundamental voltage vector are all zero voltage vectors, so the eighth basic voltage vector is used in the formula Using the seventh fundamental voltage vector express.
[0140] Among them, the calculation expressions of 18 hybrid virtual voltage vectors are as follows:
[0141] (9)
[0142] In the formula, represents the first virtual voltage vector; represents the second virtual voltage vector; represents the third virtual voltage vector; represents the fourth virtual voltage vector; represents the fifth virtual voltage vector; represents the sixth virtual voltage vector; represents the seventh virtual voltage vector; represents the eighth virtual voltage vector; represents the ninth virtual voltage vector; represents the tenth virtual voltage vector; represents an eleventh virtual voltage vector; represents the twelfth virtual voltage vector; represents a thirteenth virtual voltage vector; represents a fourteenth virtual voltage vector; represents the fifteenth virtual voltage vector; represents the sixteenth virtual voltage vector; represents a seventeenth virtual voltage vector; represents the eighteenth virtual voltage vector, since the seventh basic voltage vector and the eighth fundamental voltage vector are all zero voltage vectors, so the eighth basic voltage vector is used in the formula Using the seventh fundamental voltage vector express.
[0143] At this point, all the voltage vectors to be selected can be obtained.
[0144] Step 32: The steps of constructing a hyperlocal prediction model based on an extended state observer are:
[0145] The model expression of the surface mounted permanent magnet synchronous motor is as follows:
[0146] (10)
[0147] In the formula, represents the voltage vector; Indicates the stator resistance value; is the stator current value; is the stator flux value; is the stator inductance; is the rotor electrical angular velocity; is the permanent magnet flux value; is the rotor electrical angle; j is the imaginary unit; is the electromagnetic torque; is the number of permanent magnet pole pairs, represents the derivative of a variable, is the index.
[0148] The state space lumped model is established according to formula (10). The expression of the state space lumped model is as follows:
[0149] (11)
[0150] In the formula, the state variable matrix ; Input variables ; Parameter matrix ; Among them, the lumped perturbation matrix ;
[0151] In the formula, is the stator inductance perturbation value; is the stator resistance perturbation value; is the flux perturbation value of the permanent magnet; represents the voltage vector; Indicates the stator resistance value; is the stator current value; is the stator inductance; is the rotor electrical angular velocity; is the permanent magnet flux value; is the rotor electrical angle; j is the imaginary unit.
[0152] The second-order lumped model is established according to formula (11). The expression of the second-order lumped model is as follows:
[0153] (12)
[0154] In the formula, the variable , , , ; Maximum value of aggregate disturbance .
[0155] According to formula (12), the second-order extended state observer model is established. The expression of the second-order extended state observer model is as follows:
[0156] (13)
[0157] In the formula, is the first estimated variable; is the second estimated variable; is the first gain parameter of the extended state observer; is the second gain parameter of the extended state observer.
[0158] Taking the stator current and stator flux as state variables, the double extended state observer expression is as follows:
[0159] (14)
[0160] The expression of the hyperlocal prediction model is:
[0161] (15)
[0162] In the formula, in the formula, Stator current at time ; j is an imaginary unit; The predicted value of stator current at time ; for The predicted value of stator current at the moment; Stator flux at time ; The predicted value of stator flux at time ; for The predicted value of stator flux at time; for α-axis stator flux at time ; for α-axis stator flux at time ; for The predicted value of electromagnetic torque at the moment; for The predicted value of electromagnetic torque at the moment; for The estimated value of the current disturbance at time for The estimated value of the current disturbance at time for The estimated value of the flux disturbance at time; for The estimated value of the flux disturbance at time; is the stator inductance; is the number of permanent magnet pole pairs; for The voltage vector to be selected input at any time; Parameters for algorithm execution period; is the first adjustable gain parameter of the extended state observer, is the second adjustable gain parameter of the extended state observer, is the third adjustable gain parameter of the extended state observer, is the fourth adjustable gain parameter of the extended state observer, where according to the Hurwitz stability criterion, , , is the bandwidth of the extended state observer.
[0163] Since the parameters of the extended state observer , , , It will affect the control performance. Therefore, it is necessary to analyze the basis for parameter selection. The basic principles are as follows:
[0164] According to formula (13), the prediction error matrix of the double extended state observer is constructed. The expression of the prediction error matrix is:
[0165] (16)
[0166] In the formula, the error term , , , ; Lumped disturbance term and Satisfy and ; For disturbance The extreme value of for The extreme value of .
[0167] The characteristic equation is constructed according to formula (16), and the expression of the characteristic equation is:
[0168] (17)
[0169] If the extended state observer parameters satisfy and , then the system satisfies the stability theorem.
[0170] When the stability condition is met, according to Jury's theorem, the system bandwidth range is:
[0171] (18)
[0172] In the formula, is a discrete calculation cycle; is the bandwidth of the extended state observer.
[0173] Step 33, the step of obtaining the d-axis stator current prediction value, the q-axis stator current prediction value, the d-axis stator flux prediction value, the q-axis stator flux prediction value and the electromagnetic torque prediction value of the permanent magnet synchronous motor at the next moment under each candidate voltage vector:
[0174] By inputting the candidate voltage vector into equation (14), the predicted value of the d-axis stator current, the predicted value of the q-axis stator current, the predicted value of the d-axis stator flux, the predicted value of the q-axis stator flux and the predicted value of the electromagnetic torque of the permanent magnet synchronous motor at the next moment can be obtained.
[0175] S4, obtaining the dynamic bandwidth and cost function value of the permanent magnet synchronous motor under each candidate voltage vector;
[0176] Specifically, based on the bandwidth of the extended state observer, the electromagnetic torque reference value and the predicted value of the electromagnetic torque of the permanent magnet synchronous motor at the next moment under each candidate voltage vector, the dynamic bandwidth of the permanent magnet synchronous motor under each candidate voltage vector is obtained; according to the electromagnetic torque reference value, the preset stator flux reference value and the dynamic bandwidth, the cost function value of the permanent magnet synchronous motor under each candidate voltage vector is obtained.
[0177] Among them, the expression of the dynamic bandwidth of the permanent magnet synchronous motor under each candidate voltage vector is:
[0178] (19)
[0179] In the formula, is the dynamic bandwidth; is the bandwidth of the extended state observer; is the controller parameter, usually ranging from 0 to 1; is the hyperbolic tangent function; For permanent magnet synchronous motor The electromagnetic torque reference value at the moment; for The predicted value of electromagnetic torque at time.
[0180] Among them, the expression of the cost function value is:
[0181] (20)
[0182] In the formula, is the cost function value; For permanent magnet synchronous motor The electromagnetic torque reference value at the moment; For permanent magnet synchronous motor The predicted value of electromagnetic torque at the moment; is the dynamic bandwidth; is the preset stator flux reference value; is the weight coefficient; for The predicted value of stator flux at time .
[0183] S5, obtaining the optimal voltage vector and controlling the torque of the permanent magnet synchronous motor;
[0184] The candidate voltage vector corresponding to the minimum cost function value is taken as the optimal voltage vector, and the switching state of the switch tube of the three-phase full-bridge inverter is changed according to the optimal voltage vector to control the torque of the permanent magnet synchronous motor.
[0185] In the step of obtaining the optimal voltage vector: if the following is adopted Figure 6 The 68 candidate voltage vectors shown in the figure evaluate the cost function values of the candidate voltage vectors in turn, which will greatly increase the algorithm calculation time. In order to reduce the amount of calculation, this embodiment proposes a discrete space voltage vector selection strategy. The steps of the discrete space voltage vector selection strategy are:
[0186] Step 51: First, use 6 amplitudes as the effective voltage vector amplitude times the virtual voltage vector ( , , , , , ) and zero voltage vector ( , ) respectively calculate the corresponding cost function values; step 52, taking the candidate voltage vector corresponding to the minimum cost function value among the cost function values calculated in step 51 as the local optimal voltage vector; step 53, if the local optimal voltage vector is a zero voltage vector, then evaluating the candidate voltage vector closest to the zero voltage vector, that is, evaluating and calculating the 6 amplitudes in turn as the effective voltage vector amplitudes times the virtual voltage vector ( , , , , , ), the virtual voltage vector corresponding to the minimum cost function value in step 53 is the optimal voltage vector; if the local optimal voltage vector obtained in step 52 is a non-zero voltage vector, the voltage vectors near the non-zero voltage vector are evaluated until the optimal voltage vector is obtained, and the evaluation process ends. Specifically, Figure 7 Taking the first sector in as an example, if the local optimal voltage vector evaluated in the first step is the candidate voltage vector V1, the first cost function value calculation is performed on the hexagonal vertex candidate voltage vectors V2, V3, V6, V7, V10, and V11 with V1 as the center; if the local optimal voltage vector in the first cost function value calculation result is the candidate voltage vector V2, then the candidate voltage vector V12 and the candidate voltage vector V4 closest to the candidate voltage vector V2 are evaluated, and finally the candidate voltage vector corresponding to the minimum cost function value in the evaluation results of the candidate voltage vector V12 and the candidate voltage vector V4 is the optimal voltage vector; if the local optimal voltage vector is still V1 when the first cost function value calculation is performed on the hexagonal vertex candidate voltage vectors V2, V3, V6, V7, V10, and V11 with V1 as the center, the candidate voltage vector V5 and the candidate voltage vector V8 closest to V1 are evaluated, and the candidate voltage vector corresponding to the minimum cost function value finally obtained is the optimal voltage vector of the sampling period.
[0187] The present embodiment is described below with reference to the accompanying drawings:
[0188] The specific parameters of the surface-mounted permanent magnet synchronous motor for the aircraft electromechanical actuator in this embodiment are shown in Table 1. The framework of the robust model predictive torque control system in this embodiment is shown in Table 1. Figure 8 As shown; In order to verify the effectiveness and robustness of the control method of this embodiment, Matlab / Simulink software is used to implement the traditional model prediction torque control method and the torque control method of this embodiment respectively, and the simulation results are compared. The simulation cycle is set to 1e-5, and the simulation adopts a fixed discrete step solver. The simulation conditions are: the speed reference value is set to 1000 rpm, a 1Nm step load torque is suddenly added at 0.015 seconds, and the reference value of the stator flux amplitude is set to the permanent magnet flux amplitude.
[0189] Table 1
[0190]
[0191] In order to highlight the control performance of the torque control method of this embodiment under parameter perturbation conditions, when the electrical parameters become 0.5 times the stator inductance, 0.1 times the stator resistance, and 2 times the permanent magnet flux ( , , The simulation results of the dynamic and static control performance of the traditional model prediction torque control method are as follows: Figures 9 to 11 As shown; when the electrical parameters become 0.5 times the stator inductance, 0.1 times the stator resistance, 2 times the permanent magnet flux ( , , ) The simulation results of the dynamic and static control performance of the torque control method of this embodiment are as follows: Figure 12 to Figure 14 As shown; Fig. 9 The traditional model predictive torque control method is presented when the electrical parameters are changed to 0.5 times the stator inductance, 0.1 times the stator resistance, and 2 times the permanent magnet flux ( , , ) when the speed tracking performance, Fig.10 The traditional model predictive torque control method is presented when the electrical parameters are changed to 0.5 times the stator inductance, 0.1 times the stator resistance, and 2 times the permanent magnet flux ( , , ) when the stator flux tracking performance; Fig.11 The traditional model predictive torque control method is presented when the electrical parameters are changed to 0.5 times the stator inductance, 0.1 times the stator resistance, and 2 times the permanent magnet flux ( , , ) when electromagnetic torque tracking performance; Fig.12 The torque control method of this embodiment is presented when the electrical parameters are changed to 0.5 times the stator inductance, 0.1 times the stator resistance, and 2 times the permanent magnet flux ( , , ) when the speed tracking performance, Fig.13 The torque control method of this embodiment is presented when the electrical parameters are changed to 0.5 times the stator inductance, 0.1 times the stator resistance, and 2 times the permanent magnet flux ( , , ) when the stator flux tracking performance; Fig.14 The torque control method of this embodiment is presented when the electrical parameters are changed to 0.5 times the stator inductance, 0.1 times the stator resistance, and 2 times the permanent magnet flux ( , , ) when the electromagnetic torque tracking performance; that is, Fig. 9 It can be seen that when the motor parameters change, the motor speed of the traditional model predictive torque control method has a large speed drop amplitude under loading conditions; Fig.12 It can be seen that the loading condition after the motor parameters change in the torque control method of this embodiment has a smaller speed drop amplitude. By comparison, it can be seen that the proposed method can maintain high dynamic tracking performance of the speed after the parameters change. Fig.10It can be seen that in the traditional model predictive torque control method, when the motor parameters change, there is a tracking error between the α-axis stator flux feedback value and the reference value, and between the β-axis stator flux feedback value and the reference value; Fig.13 It can be seen that in the torque control method of this embodiment, when the motor parameters change, the α-axis stator flux feedback value and the β-axis stator flux feedback value can both follow the reference value in real time. By comparison, it can be seen that the torque control method of this embodiment can reduce the stator flux tracking error generated when the motor parameters change. Fig.11 It can be seen that when the motor parameters change, there is a tracking error between the electromagnetic torque feedback value and the electromagnetic torque reference value; Fig.14 It can be seen that when the motor parameters change, the electromagnetic torque feedback value can follow the reference value in real time. By comparison, it can be seen that the torque control method of this embodiment can reduce the torque tracking error generated when the motor parameters change.
[0192] The electrical parameters become The simulation results of the dynamic and static control performance of the traditional model predictive torque control method are shown in the figure below. Figures 15 to 17 As shown, the electrical parameters become The simulation results of the dynamic and static control performance of the torque control method of this embodiment are shown in FIG. Figure 18 to Figure 20 As shown, Fig.15 The traditional model predictive torque control method is presented when the electrical parameters change Speed tracking performance, Fig.16 The traditional model predictive torque control method is presented when the electrical parameters change Stator flux tracking performance when Fig.17 The traditional model predictive torque control method is presented when the electrical parameters change Electromagnetic torque tracking performance when Fig.18 The torque control method of this embodiment is presented when the electrical parameters become Speed tracking performance, Fig.19 The torque control method of this embodiment is presented when the electrical parameters become Stator flux tracking performance when Fig. 20 The torque control method of this embodiment is presented when the electrical parameters become Electromagnetic torque tracking performance.
[0193] in, Fig.11 , Fig.14 , Fig.17 and Fig. 20 In is the electromagnetic torque reference value, is the electromagnetic torque feedback value, Fig.10 , Fig.13 , Fig.16 and Fig.19 In is the α-axis stator flux reference value, is the β-axis stator flux reference value, is the α-axis stator flux feedback value, is the stator flux feedback value of the β axis. Fig.15 and Fig.18 , Fig.16 and Fig.19 , Fig.17 and Fig. 20 By comparison, it can be seen that the stator flux and electromagnetic torque of the traditional model prediction torque control method have large tracking errors, especially under speed regulation and loading conditions. The torque tracking error is positive, which will reduce the dynamic performance of the control method and increase the speed regulation time under dynamic conditions. Compared with the traditional model prediction torque control method, the torque control method of this embodiment can eliminate the tracking bias problem caused by the change of motor parameters. That is, Fig.12 , Fig.14 Fig.14 as well as Fig.18 , Fig.19 , Fig. 20 It can be seen that the electromagnetic torque and stator flux amplitude of the torque control method of this embodiment both reach the reference value, thereby improving the dynamic response performance of the motor during starting conditions and sudden load conditions. Figures 9 to 20 It can be seen from the simulation comparison results that the torque control method of this embodiment can reduce the tracking error caused by the motor parameter perturbation and has high dynamic performance and high robustness performance.
[0194] In order to compare the steady-state performance of the traditional model prediction torque control method and the torque control method of this embodiment, the parameters of the torque control method using the traditional model prediction torque control method and the torque control method of this embodiment are respectively , , and , , The phase current THD is analyzed and the electrical parameters become , , The simulation results of the phase current THD of the traditional model predictive torque control method are as follows: Fig.21 As shown, the electrical parameters become , , The results of analyzing the phase current THD of the torque control method of this embodiment are as follows: Fig. 22 As shown. When the electrical parameters become , , The simulation results of the phase current THD of the traditional model predictive torque control method are as follows: Fig.23 As shown, the electrical parameters become , , The results of analyzing the phase current THD of the torque control method of this embodiment are as follows: Fig.24 As shown in the figure, by collecting the phase current waveform of one cycle, the base frequency is set to 83.3Hz. Fig.21 and Fig. 22 , Fig.23 and Fig.24 It can be seen that the parameters are , , When the conventional model predicts the phase current THD of the torque control method, the phase current THD is 1.27%, while the phase current THD of the torque control method of this embodiment is 0.66%. , , The traditional model predicts that the phase current THD of the torque control method is 4.24%, and the phase current THD of the torque control method of this embodiment is 1.69%. Therefore, the phase current waveform of the torque control method of this embodiment has fewer harmonic components and smoother control performance, that is, the torque control method of the present invention can reduce the harmonic components of the phase current and improve reliability.
[0195] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims
1. A method for controlling the torque of a permanent magnet synchronous motor for an aircraft electromechanical actuator, characterized in that: include: Obtaining an α-axis current, a β-axis current, a d-axis current, and a q-axis current of the permanent magnet synchronous motor according to the three-phase current of the permanent magnet synchronous motor; Obtaining electromagnetic torque, α-axis stator flux, β-axis stator flux, d-axis stator flux and q-axis stator flux according to the d-axis current and q-axis current of the permanent magnet synchronous motor; According to the rotor angular velocity of the permanent magnet synchronous motor and the preset rotor angular velocity reference value, the electromagnetic torque reference value is obtained by using the proportional integral control algorithm; the expression of the electromagnetic torque reference value is: In the formula, For permanent magnet synchronous motor The electromagnetic torque reference value at the moment; is the rotor angular velocity; is the preset rotor angular velocity reference value; is the proportional parameter of the proportional-integral control algorithm; is the integral parameter of the proportional-integral control algorithm; It is the integral operation of proportional integral control algorithm; The discrete space vector modulation technology is used to obtain the selected voltage vector; according to the electromagnetic torque, α-axis stator flux, β-axis stator flux, α-axis current, β-axis current and the selected voltage vector, a super-local prediction model based on the extended state observer is constructed, and the stator current prediction value, stator flux prediction value and electromagnetic torque prediction value of the permanent magnet synchronous motor at the next moment under each selected voltage vector are obtained by using the super-local prediction model; Among them, the steps of obtaining the voltage vector to be selected by using the discrete space vector modulation technology are: based on the voltage amplitude of the basic voltage vector of the three-phase full-bridge inverter circuit of the permanent magnet synchronous motor, obtain a two-element virtual voltage vector and a three-element virtual voltage vector; add the two-element virtual voltage vector and the three-element virtual voltage vector to obtain a variety of mixed virtual voltage vectors; use the basic voltage vector, the two-element virtual voltage vector, the three-element virtual voltage vector and the mixed virtual voltage vector as the voltage vector to be selected; the steps of obtaining the two-element virtual voltage vector and the three-element virtual voltage vector are: divide the voltage amplitude of each basic voltage vector by two to obtain a first target voltage vector; add every two first target voltage vectors to obtain a two-element virtual voltage vector; divide the voltage amplitude of each basic voltage vector by three to obtain a second target basic voltage vector; add every two second target basic voltage vectors to obtain a three-element virtual voltage vector; Among them, the expression of the super local prediction model is: In the formula, Stator current at time ; j is an imaginary unit; The predicted value of stator current at time ; for The predicted value of stator current at the moment; Stator flux at time ; The predicted value of stator flux at time ; for The predicted value of stator flux at time; for α-axis stator flux at time ; for α-axis stator flux at time ; for The predicted value of electromagnetic torque at the moment; for The predicted value of electromagnetic torque at the moment; for The estimated value of the current disturbance at time instant; for The estimated value of the current disturbance at time for The estimated value of the flux disturbance at time; for The estimated value of the flux disturbance at time; is the stator inductance; is the number of permanent magnet pole pairs; for The voltage vector to be selected input at any time; Parameters for algorithm execution period; is the first adjustable gain parameter of the extended state observer, is the second adjustable gain parameter of the extended state observer, is the third adjustable gain parameter of the extended state observer, is the fourth adjustable gain parameter of the extended state observer, where , , is the bandwidth of the extended state observer; Based on the bandwidth of the extended state observer, the electromagnetic torque reference value and the electromagnetic torque prediction value of the permanent magnet synchronous motor at the next moment under each candidate voltage vector, the dynamic bandwidth of the permanent magnet synchronous motor under each candidate voltage vector is obtained; according to the electromagnetic torque reference value, the preset stator flux reference value and the dynamic bandwidth, the cost function value of the permanent magnet synchronous motor under each candidate voltage vector is obtained; wherein, the expression of the dynamic bandwidth of the permanent magnet synchronous motor under each candidate voltage vector is: In the formula, is the dynamic bandwidth; is the bandwidth of the extended state observer; is the controller parameter; is the hyperbolic tangent function; For permanent magnet synchronous motor The electromagnetic torque reference value at the moment; for The predicted value of electromagnetic torque at the moment; The candidate voltage vector corresponding to the minimum cost function value is taken as the optimal voltage vector, and the switching state of the switch tube of the three-phase full-bridge inverter is changed according to the optimal voltage vector to control the torque of the permanent magnet synchronous motor.
2. The torque control method of a permanent magnet synchronous motor for an aircraft electromechanical actuator according to claim 1, characterized in that: The steps of obtaining the α-axis current, β-axis current, d-axis current and q-axis current of the permanent magnet synchronous motor according to the three-phase current of the permanent magnet synchronous motor are as follows: the three-phase current of the permanent magnet synchronous motor is subjected to park transformation to obtain the α-axis current and β-axis current, and then the d-axis current and q-axis current of the permanent magnet synchronous motor are obtained after clark transformation.
3. The torque control method of a permanent magnet synchronous motor for an aircraft electromechanical actuator according to claim 1, characterized in that: The steps of obtaining the electromagnetic torque, the α-axis stator flux, the β-axis stator flux, the d-axis stator flux and the q-axis stator flux are as follows: In the formula, is the d-axis stator flux; is the q-axis stator flux; is the electromagnetic torque; is the number of permanent magnet pole pairs; is the permanent magnet flux; is the stator inductance; is the d-axis current; is the q-axis current; Based on the d-axis stator flux and the q-axis stator flux, the α-axis stator flux is obtained by using the inverse Clark transformation and β-axis stator flux .
4. The torque control method of a permanent magnet synchronous motor for an aircraft electromechanical actuator according to claim 1, characterized in that: The expression of the cost function value of the permanent magnet synchronous motor under each candidate voltage vector is: In the formula, is the cost function value; For permanent magnet synchronous motor The electromagnetic torque reference value at the moment; For permanent magnet synchronous motor The predicted value of electromagnetic torque at the moment; is the dynamic bandwidth; is the preset stator flux reference value; is the weight coefficient; is the predicted value of stator flux.
