Robust model predictive control method for permanent magnet synchronous motor based on error compensation

Abstract

The present invention discloses a robust permanent magnet synchronous motor model prediction control method based on error compensation, comprising the following steps: 1. calculating the dq axis current prediction value, stator flux prediction value and torque prediction value at time k; 2. obtaining the stator current actual value, stator flux actual value and torque actual value at time k; 3. calculating the stator current prediction error, torque prediction error and stator flux prediction error under the voltage space vector acting at time k; 4. calculating the stator current prediction error, stator flux prediction error and torque prediction error under the remaining unacted voltage space vectors at time k; 5. establishing a cost function after error compensation and selecting the optimal voltage space vector. The present invention compensates the torque prediction error and stator flux prediction error under all voltage space vectors at time k into the prediction model, so that the control of the permanent magnet synchronous motor is not affected by factors such as temperature rise and magnetic saturation, and has good robustness.

Description

Technical Field

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

[0002] In industrial manufacturing and driving, advanced industrial drive equipment can generate more kinetic energy output under the same consumption conditions, with high efficiency and environmental protection. Permanent magnet synchronous motors, as efficient drive equipment in the field of industrial manufacturing, are widely used in new energy vehicles, rail transportation, wind power and other fields due to their simple structure, small size and high power efficiency.

[0003] At present, as the use environment of permanent magnet synchronous motors becomes more diverse and complex, users have put forward higher requirements on the control performance of permanent magnet synchronous motors, mainly including transient response speed, operating efficiency, motor robustness and fault tolerance, etc. More and more advanced control methods have been introduced into permanent magnet synchronous motor control algorithms by domestic and foreign scholars, taking motors as research objects, aiming to achieve high-performance control strategies, such as vector control, fuzzy control, neural network control, model predictive control, etc.

[0004] Among them, model predictive control is a power controller switch sequence generation method that has attracted the attention of many scholars in recent years. Its goal is to select the most suitable switch state from all the switch state sets of the power controller. Different from the traditional PID algorithm, the advantage of model predictive control is that it has a simple structure, does not require a specific structural form, only emphasizes the prediction function, and draws on the idea of ​​optimization control, using rolling optimization to replace the unmodified global optimization. Model predictive control is a new type of optimization control algorithm used in areas that are not applicable in traditional industrial control processes, such as multi-objective constraints, nonlinear control, etc.

[0005] In practical applications, in some driving situations where both the steady-state performance and dynamic performance of the motor are required to be high, model predictive control has been introduced into the field of permanent magnet synchronous motor control, hoping to obtain excellent dynamic performance while having good steady-state performance. Whether it is model predictive current control or model predictive torque control, the basis for its implementation is the correct establishment of the mathematical model of the permanent magnet synchronous motor. However, affected by factors such as temperature rise and magnetic saturation, the parameters of the permanent magnet synchronous motor will change. The parameters of the permanent magnet synchronous motor will change, which will cause a large deviation between the mathematical model of the permanent magnet synchronous motor established by the motor parameters and its actual mathematical model, which will lead to deviations in the results predicted by the motor mathematical model, resulting in misjudgment in selecting the optimal vector using the evaluation function, resulting in a decrease in the control performance of the system, and even causing the system to crash.

[0006] Therefore, it is necessary to study how to improve the operating performance of the permanent magnet synchronous motor model predictive control system when the motor is affected by factors such as temperature rise and magnetic saturation. Summary of the invention

[0007] The technical problem to be solved by the present invention is to provide a robust permanent magnet synchronous motor model predictive control method based on error compensation in view of the deficiencies in the above-mentioned prior art, in which the torque prediction error and the stator flux prediction error under all voltage space vectors at time k are compensated into the prediction model, so that the control of the permanent magnet synchronous motor is not affected by factors such as temperature rise and magnetic saturation, and has good robustness.

[0008] In order to solve the above technical problems, the technical solution adopted by the present invention is: a robust permanent magnet synchronous motor model predictive control method based on error compensation, characterized in that the method comprises the following steps:

[0009] Step 1: Obtain the actual value of the dq axis current at time k-1 in the synchronous rotating coordinate system, and calculate the predicted value of the dq axis current, the predicted value of the stator flux and the predicted value of the torque at time k in combination with the rated mathematical model of the permanent magnet synchronous motor;

[0010] Step 2: Obtain the actual value of the stator current, the actual value of the stator flux and the actual value of the torque at time k;

[0011] Step 3: Calculate the voltage space vector U acting at time k j (k) is the stator current prediction error, stator flux prediction error and torque prediction error under the condition of φ(k); j is the number of the voltage space vector acting at any time, j = 0, 1, ..., 7;

[0012] Step 4: Based on the calculation results of step 3, calculate the remaining unacted voltage space vector U at time k i (k) is the stator current prediction error, stator flux prediction error and torque prediction error under the condition of φ(k); i is the number of the voltage space vector that is not applied at any time, i=0, 1, ..., 7 and i≠j;

[0013] Step 5: Based on the torque prediction error and stator flux prediction error corresponding to all voltage space vectors at time k, establish a cost function after error compensation, select the voltage space vector that minimizes the cost function value as the optimal voltage space vector, and use the optimal voltage space vector for torque control.

[0014] The above-mentioned robust permanent magnet synchronous motor model predictive control method based on error compensation is characterized in that the specific steps of step 4 are:

[0015] Step 401: According to the formula Calculate the remaining unacted voltage space vector U at time k iStator current prediction error ΔI under (k) s i (k); where U j (k-1) is the voltage space vector acting at time k-1; ΔI s j (k) is the voltage space vector U acting at time k j The stator current prediction error under (k), ΔI s j (k-1) is the voltage space vector U acting at time k-1 j Stator current prediction error under (k-1), ΔI s j The calculation process of (k-1) and ΔI s j The calculation process of (k) is the same;

[0016] Step 402: According to the formula Calculate the remaining unacted voltage space vector U at time k i Stator flux prediction error Δψ under (k) s i (k); where Δψ s j (k) is the voltage space vector U acting at time k j The stator flux prediction error under (k), Δψ s j (k-1) is the voltage space vector U acting at time k-1 j Stator flux prediction error under (k-1), Δψ s j The calculation process of (k-1) and Δψ s j The calculation process of (k) is the same;

[0017] Step 403: According to the formula Calculate the remaining unacted voltage space vector U at time k i Torque prediction error ΔT under (k) e i (k); where ΔT e j (k) is the voltage space vector U acting at time k j (k) Torque prediction error, ΔT e j (k-1) is the voltage space vector U acting at time k-1 j Torque prediction error under (k-1), ΔT e j The calculation process of (k-1) and ΔT ej The calculation process of (k) is the same; U jq (k) is the voltage component of the voltage space vector acting at time k on the q axis, U iq (k) is the voltage component of the unacting voltage space vector on the q-axis at time k.

[0018] The above-mentioned robust permanent magnet synchronous motor model predictive control method based on error compensation is characterized in that the specific steps of step five are:

[0019] Step 501: Establish the cost function g(J) after error compensation = |T e ref -(T e JP (k+1)+ΔT e J (k))|+λ|ψ s ref -(ψ s JP (k+1)+Δψ s J (k))|;

[0020] Where, ΔT e J (k) is the space vector U of all voltages at time k J (k) Torque prediction error; Δψ s J (k) is the space vector U of all voltages at time k J Stator flux prediction error under (k), T e ref is the rated torque, ψ s ref is the rated stator flux, λ is the weight coefficient;

[0021] Step 502: According to the formula Calculate all voltage space vectors U at time k+1 J The predicted value of stator flux under (k) is ψ s JP (k+1) and torque prediction value T e JP (k+1); where L d and L q are the dq axis inductance, ψ f is the main magnetic flux of the permanent magnet, P n is the number of pole pairs of the permanent magnet synchronous motor, T s is the control period, R s is the stator winding resistance, ω e is the rotor electrical angular velocity, J is the serial number of all voltage space vectors, J = i∪j, Id JP (k+1) and I q JP (k+1) are all the voltage space vectors U at time k+1 J (k) The predicted value of dq axis current under s JP (k+1) is the total voltage space vector U at time k+1 J (k) is the predicted value of stator current; U Jd (k) and U Jq (k) are the voltage components of all voltage space vectors on the dq axis at time k; ψ Jd (k) and ψ Jq (k) are the actual values ​​of stator flux on the dq axis under the action of all voltage space vectors at time k;

[0022] Step 503: Select the voltage space vector that minimizes the value of the cost function g(J) as the optimal voltage space vector, and use the optimal voltage space vector to perform torque control.

[0023] The above-mentioned robust permanent magnet synchronous motor model predictive control method based on error compensation is characterized in that: the specific steps of step one are:

[0024] Step 101: According to the formula Calculate the actual value of the dq axis current I at time k-1 d (k-1) and I q (k-1); where I a (k-1), I a (k-1), I a (k-1) are the actual values ​​of the three-phase stator current at time k-1, and θ(k-1) is the actual value of the rotor electrical angle at time k-1;

[0025] Step 102: According to the formula Calculate the predicted value of the dq axis current at time k I d P (k) and I q P (k); where U jd (k-1), U jq (k-1) are the voltage components of the voltage space vector acting at time k-1 on the dq axis; ψ d (k-1),ψ q (k-1) are the actual values ​​of stator flux on the dq axis at time k-1; T s is the control period, R s is the stator winding resistance, ω e is the rotor electrical angular velocity, L d , Lq are the dq axis inductance respectively;

[0026] Step 103: According to the formula Calculate the predicted value of stator flux at time k ψ s P (k) and torque prediction value T e P (k); where I s P (k) is the predicted value of stator current at time k, ψ f is the main magnetic flux of the permanent magnet, P n is the number of pole pairs of the permanent magnet synchronous motor.

[0027] The above-mentioned robust permanent magnet synchronous motor model predictive control method based on error compensation is characterized in that the specific steps of step three are:

[0028] Step 301: According to the formula ΔI s j (k) = I s P (k)-I s (k), calculate the voltage space vector U acting at time k j Stator current prediction error ΔI under (k) s j (k); where I s (k) is the actual value of the stator current at time k;

[0029] Step 302: According to the formula Δψ s j (k) = ψ s P (k)-ψ s (k), calculate the voltage space vector U acting at time k j Stator flux prediction error Δψ under (k) s j (k); where ψ s (k) is the actual value of stator flux at time k;

[0030] Step 303: According to the formula ΔT e j (k) = T e P (k)-T e (k), calculate the voltage space vector U acting at time k j Torque prediction error ΔT under (k) e j (k); where T e (k) is the actual torque value at time k.

[0031] Compared with the prior art, the beneficial effects of the present invention are as follows: the present invention calculates the torque prediction error and the stator flux prediction error under all voltage space vectors at time k, and compensates them into the prediction model as compensation factors at time k+1. The compensated torque and stator flux will not have torque fluctuations or stator flux deviations caused by parameter mismatch, thereby preventing the prediction control of the permanent magnet synchronous motor from being affected by factors such as temperature rise and magnetic saturation, thereby improving the parameter robustness of the model prediction error compensation control.

[0032] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] Figure 1 It is a schematic diagram of the control principle of the present invention.

[0034] Figure 2 The figure is a flowchart of the method of the present invention. DETAILED DESCRIPTION

[0035] like Figure 1 and Figure 2 As shown, the robust permanent magnet synchronous motor model predictive control method based on error compensation of the present invention comprises the following steps:

[0036] Step 1: Obtain the actual value of the dq axis current at time k-1 in the synchronous rotating coordinate system, and calculate the predicted value of the dq axis current, the predicted value of the stator flux and the predicted value of the torque at time k in combination with the rated mathematical model of the permanent magnet synchronous motor;

[0037] Step 2: Obtain the actual value of the stator current, the actual value of the stator flux and the actual value of the torque at time k;

[0038] Step 3: Calculate the voltage space vector U acting at time k j (k) is the stator current prediction error, stator flux prediction error and torque prediction error under the condition of φ(k); j is the number of the voltage space vector acting at any time, j = 0, 1, ..., 7;

[0039] Step 4: Based on the calculation results of step 3, calculate the remaining unacted voltage space vector U at time k i (k) is the stator current prediction error, stator flux prediction error and torque prediction error under the condition of φ(k); i is the number of the voltage space vector that is not applied at any time, i=0, 1, ..., 7 and i≠j;

[0040] Step 5: Based on the torque prediction error and stator flux prediction error corresponding to all voltage space vectors at time k, establish a cost function after error compensation, select the voltage space vector that minimizes the cost function value as the optimal voltage space vector, and use the optimal voltage space vector for torque control.

[0041] It should be noted that Figure 1 Where Sa, Sb and Sc are the switching states of the three-phase upper bridge arm of the inverter respectively.

[0042] It should be noted that, when j=1, i=0, 2, 3, 4, 5, 6, 7.

[0043] It should be noted that the k-1 time is the start time of the k-1th control cycle of the permanent magnet synchronous motor.

[0044] It should be noted that, without adding additional hardware equipment and computational complexity, the torque prediction error at moment k is compensated in the torque prediction at the next moment; at the same time, in order to ensure the accuracy of the torque prediction error compensation, the prediction errors corresponding to all voltage space vectors will be obtained within one moment; this method can achieve reliable torque prediction through online real-time error compensation when the parameters of the permanent magnet synchronous motor change, thereby improving the parameter robustness of the model predictive torque control.

[0045] In this embodiment, the changes in the inductance parameters and the stator flux parameters in the motor are taken into account, thereby taking into account the torque prediction error and the stator flux prediction error under the action of different voltage vectors. Compared with the two types of compensation methods, namely, single vector non-fixed compensation or multi-vector non-real-time update compensation, this method has a better compensation effect and is more robust.

[0046] In this embodiment, the specific steps of step 4 are:

[0047] Step 401: According to the formula Calculate the remaining unacted voltage space vector U at time k i Stator current prediction error ΔI under (k) s i (k); where U j (k-1) is the voltage space vector acting at time k-1; ΔI s j (k) is the voltage space vector U acting at time k j The stator current prediction error under (k), ΔI s j (k-1) is the voltage space vector U acting at time k-1 j Stator current prediction error under (k-1), ΔI s j The calculation process of (k-1) and ΔI s j The calculation process of (k) is the same;

[0048] Step 402: According to the formula Calculate the remaining unacted voltage space vector U at time k i Stator flux prediction error Δψ under (k) s i (k); where Δψ s j (k) is the voltage space vector U acting at time k j Stator flux prediction error under (k), Δψ s j (k-1) is the voltage space vector U acting at time k-1 j Stator flux prediction error under (k-1), Δψ s j The calculation process of (k-1) and Δψ s j The calculation process of (k) is the same;

[0049] Step 403: According to the formula Calculate the remaining unacted voltage space vector U at time k i Torque prediction error ΔT under (k) e i (k); where ΔT e j (k) is the voltage space vector U acting at time k j (k) Torque prediction error, ΔT e j (k-1) is the voltage space vector U acting at time k-1 j Torque prediction error under (k-1), ΔT e j The calculation process of (k-1) and ΔT e j The calculation process of (k) is the same; U jq (k) is the voltage component of the voltage space vector acting at time k on the q axis, U iq (k) is the voltage component of the unacting voltage space vector on the q-axis at time k.

[0050] It should be noted that by substituting the stator flux prediction error and the torque prediction error at time k into the cost function at time k+1, the optimal voltage space vector at time k+1 is obtained, thereby improving the parameter robustness of the permanent magnet synchronous motor model predictive torque control;

[0051] In this embodiment, the specific steps of step five are:

[0052] Step 501: Establish the cost function g(J) after error compensation = |T e ref -(T e JP(k+1)+ΔT e J (k))|+λ|ψ s ref -(ψ s JP (k+1)+Δψ s J (k))|;

[0053] Where, ΔT e J (k) is the space vector U of all voltages at time k J (k) Torque prediction error; Δψ s J (k) is the space vector U of all voltages at time k J Stator flux prediction error under (k), T e ref is the rated torque, ψ s ref is the rated stator flux, λ is the weight coefficient;

[0054] Step 502: According to the formula Calculate all voltage space vectors U at time k+1 J The predicted value of stator flux under (k) is ψ s JP (k+1) and torque prediction value T e JP (k+1); where L d and L q are the dq axis inductance, ψ f is the main magnetic flux of the permanent magnet, P n is the number of pole pairs of the permanent magnet synchronous motor, T s is the control period, R s is the stator winding resistance, ω e is the rotor electrical angular velocity, J is the serial number of all voltage space vectors, J = i∪j, I d JP (k+1) and I q JP (k+1) are all the voltage space vectors U at time k+1 J (k) The predicted value of dq axis current under s JP (k+1) is the total voltage space vector U at time k+1 J (k) is the predicted value of stator current; U Jd (k) and U Jq (k) are the voltage components of all voltage space vectors on the dq axis at time k; ψ Jd (k) and ψ Jq(k) are the actual values ​​of stator flux on the dq axis under the action of all voltage space vectors at time k;

[0055] Step 503: Select the voltage space vector that minimizes the value of the cost function g(J) as the optimal voltage space vector, and use the optimal voltage space vector to perform torque control.

[0056] It should be noted that the weight coefficient λ is designed based on the principle of equal priority, that is, it is determined by the ratio of rated torque to rated stator flux. In actual use, it needs to be adjusted according to actual experiments.

[0057] In this embodiment, the value of λ is 100.

[0058] In this embodiment, the specific steps of step one are:

[0059] Step 101: According to the formula Calculate the actual value of the dq axis current I at time k-1 d (k-1) and I q (k-1); where I a (k-1), I a (k-1), I a (k-1) are the actual values ​​of the three-phase stator current at time k-1, and θ(k-1) is the actual value of the rotor electrical angle at time k-1;

[0060] Step 102: According to the formula Calculate the predicted value of the dq axis current at time k I d P (k) and I q P (k); where U jd (k-1), U jq (k-1) are the voltage components of the voltage space vector acting at time k-1 on the dq axis; ψ d (k-1),ψ q (k-1) are the actual values ​​of stator flux on the dq axis at time k-1; T s is the control period, R s is the stator winding resistance, ω e is the rotor electrical angular velocity, L d , L q are the dq axis inductance respectively;

[0061] In this embodiment, a voltage-type flux observer with a compensation low-pass filter is used to obtain the stator flux at time k-1, thereby obtaining the actual value of the stator flux on the dq axis at time k-1 ψ d (k-1) and ψ q (k-1);

[0062] Step 103: According to the formula Calculate the predicted value of stator flux at time k ψ s P (k) and torque prediction value T e P (k); where I s P (k) is the predicted value of stator current at time k, ψ f is the main magnetic flux of the permanent magnet, P n is the number of pole pairs of the permanent magnet synchronous motor.

[0063] In this embodiment, the specific steps of step three are:

[0064] Step 301: According to the formula ΔI s j (k) = I s P (k)-I s (k), calculate the voltage space vector U acting at time k j Stator current prediction error ΔI under (k) s j (k); where I s (k) is the actual value of the stator current at time k;

[0065] Step 302: According to the formula Δψ s j (k) = ψ s P (k)-ψ s (k), calculate the voltage space vector U acting at time k j Stator flux prediction error Δψ under (k) s j (k); where ψ s (k) is the actual value of stator flux at time k;

[0066] Step 303: According to the formula ΔT e j (k) = T e P (k)-T e (k), calculate the voltage space vector U acting at time k j Torque prediction error ΔT under (k) e j (k); where T e (k) is the actual torque value at time k.

[0067] The above description is only a preferred embodiment of the present invention and does not limit the present invention in any way. Any simple modification, change and equivalent structural change made to the above embodiment based on the technical essence of the present invention still falls within the protection scope of the technical solution of the present invention.

Claims

1. A robust permanent magnet synchronous motor model predictive control method based on error compensation, characterized in that: The method comprises the following steps: Step 1: Obtain the actual value of the dq axis current at time k-1 in the synchronous rotating coordinate system, and calculate the predicted value of the dq axis current, the predicted value of the stator flux and the predicted value of the torque at time k in combination with the rated mathematical model of the permanent magnet synchronous motor; Step 2: Obtain the actual value of the stator current, the actual value of the stator flux and the actual value of the torque at time k; Step 3: Calculate the voltage space vector U acting at time k j (k) is the stator current prediction error, stator flux prediction error and torque prediction error under the condition of φ(k); j is the number of the voltage space vector acting at any time, j = 0, 1, ..., 7; Step 4: Based on the calculation results of step 3, calculate the remaining unacted voltage space vector U at time k i (k) is the stator current prediction error, stator flux prediction error and torque prediction error under the condition of φ(k); i is the number of the voltage space vector that is not applied at any time, i=0, 1, ..., 7 and i≠j; Step 5: Based on the torque prediction error and stator flux prediction error corresponding to all voltage space vectors at time k, a cost function after error compensation is established, and the voltage space vector that minimizes the cost function value is selected as the optimal voltage space vector, and the optimal voltage space vector is used for torque control; The specific steps of step 4 are: Step 401: According to the formula Calculate the remaining unacted voltage space vector U at time k i Stator current prediction error ΔI under (k) s i (k); where U j (k-1) is the voltage space vector acting at time k-1; ΔI s j (k) is the voltage space vector U acting at time k j The stator current prediction error under (k), ΔI s j (k-1) is the voltage space vector U acting at time k-1 j Stator current prediction error under (k-1), ΔI s j The calculation process of (k-1) and ΔI s j The calculation process of (k) is the same; Step 402: According to the formula Calculate the remaining unacted voltage space vector U at time k i Stator flux prediction error Δψ under (k) s i (k); where Δψ s j (k) is the voltage space vector U acting at time k j Stator flux prediction error under (k), Δψ s j (k-1) is the voltage space vector U acting at time k-1 j Stator flux prediction error under (k-1), Δψ s j The calculation process of (k-1) and Δψ s j The calculation process of (k) is the same; Step 403: According to the formula Calculate the remaining unacted voltage space vector U at time k i Torque prediction error ΔT under (k) e i (k); where ΔT e j (k) is the voltage space vector U acting at time k j (k) Torque prediction error, ΔT e j (k-1) is the voltage space vector U acting at time k-1 j Torque prediction error under (k-1), ΔT e j The calculation process of (k-1) and ΔT e j The calculation process of (k) is the same; U jq (k) is the voltage component of the voltage space vector acting at time k on the q axis, U iq (k) is the voltage component of the unacting voltage space vector on the q-axis at time k.

2. The robust permanent magnet synchronous motor model predictive control method based on error compensation according to claim 1 is characterized in that: The specific steps of step five are: Step 501: Establish the cost function g(J) after error compensation = |T e ref -(T e JP (k+1)+ΔT e J (k))|+λ|ψ s ref -(ψ s JP (k+1)+Δψ s J (k))|; Where, ΔT e J (k) is the space vector U of all voltages at time k J (k) Torque prediction error; Δψ s J (k) is the space vector U of all voltages at time k J Stator flux prediction error under (k), T e ref is the rated torque, ψ s ref is the rated stator flux, λ is the weight coefficient; Step 502: According to the formula Calculate all voltage space vectors U at time k+1 J The predicted value of stator flux under (k) is ψ s JP (k+1) and torque prediction value T e JP (k+1); where L d and L q are the dq axis inductance, ψ f is the main magnetic flux of the permanent magnet, P n is the number of pole pairs of the permanent magnet synchronous motor, T s is the control period, R s is the stator winding resistance, ω e is the rotor electrical angular velocity, J is the serial number of all voltage space vectors, J = i∪j, I d JP (k+1) and I q JP (k+1) are all the voltage space vectors U at time k+1 J (k) The predicted value of dq axis current under s JP (k+1) is the space vector U of all voltages at time k+1 J (k) is the predicted value of stator current; U Jd (k) and U Jq (k) are the voltage components of all voltage space vectors on the dq axis at time k; ψ Jd (k) and ψ Jq (k) are the actual values ​​of stator flux on the dq axis under the action of all voltage space vectors at time k; Step 503: Select the voltage space vector that minimizes the value of the cost function g(J) as the optimal voltage space vector, and use the optimal voltage space vector to perform torque control.

3. The robust permanent magnet synchronous motor model predictive control method based on error compensation according to claim 1 is characterized in that: The specific steps of step one are: Step 101: According to the formula Calculate the actual value of the dq axis current I at time k-1 d (k-1) and I q (k-1); where I a (k-1), I a (k-1), I a (k-1) are the actual values ​​of the three-phase stator current at time k-1, and θ(k-1) is the actual value of the rotor electrical angle at time k-1; Step 102: According to the formula Calculate the predicted value of the dq axis current at time k I d P (k) and I q P (k); where U jd (k-1), U jq (k-1) are the voltage components of the voltage space vector acting at time k-1 on the dq axis; ψ d (k-1),ψ q (k-1) are the actual values ​​of stator flux on the dq axis at time k-1; T s is the control period, R s is the stator winding resistance, ω e is the rotor electrical angular velocity, L d , L q are the dq axis inductance respectively; Step 103: According to the formula Calculate the predicted value of stator flux at time k ψ s P (k) and torque prediction value T e P (k); where I s P (k) is the predicted value of stator current at time k, ψ f is the main magnetic flux of the permanent magnet, P n is the number of pole pairs of the permanent magnet synchronous motor.

4. The robust permanent magnet synchronous motor model predictive control method based on error compensation according to claim 1 is characterized in that: The specific steps of step three are: Step 301: According to the formula ΔI s j (k) = I s P (k)-I s (k), calculate the voltage space vector U acting at time k j Stator current prediction error ΔI under (k) s j (k); where I s (k) is the actual value of the stator current at time k; Step 302: According to the formula Δψ s j (k) = ψ s P (k)-ψ s (k), calculate the voltage space vector U acting at time k j Stator flux prediction error Δψ under (k) s j (k); where ψ s (k) is the actual value of stator flux at time k; Step 303: According to the formula ΔT e j (k) = T e P (k)-T e (k), calculate the voltage space vector U acting at time k j Torque prediction error ΔT under (k) e j (k); where T e (k) is the actual torque value at time k.

Images