Permanent magnet synchronous motor sensorless control method based on prediction correction

Abstract

The invention discloses a permanent magnet synchronous motor sensorless control method based on prediction correction. The method comprises the following steps: step 1, constructing a motor mathematical model under a permanent magnet synchronous motor static coordinate system; 2, constructing a sliding-mode observer model and a current error equation; 3, adding a prediction current correction item in the sliding-mode observer, and adjusting the weight value of the prediction current correction item; 4, current estimation is carried out through a sliding mode observer, the current at the current moment is predicted through a discrete motor model, and the difference value between the actual current and the predicted current is calculated; 5, constructing a prediction equation and a correction equation of a Kalman filter, and performing back electromotive force correction estimation; and step 6, estimating the rotor speed and position according to the back electromotive force to realize motor control. According to the method, direct differential operation on measured values sensitive to noise is avoided, the rotor position and speed estimation precision is remarkably improved, particularly, the noise influence can be effectively suppressed in a high-noise environment, and the robustness and stability of the system are enhanced.

Description

TECHNICAL FIELD

[0001] The present application relates to permanent magnet synchronous motor control, and particularly relates to a permanent magnet synchronous motor position sensorless control method based on predictive correction. BACKGROUND

[0002] Permanent magnet synchronous motors have been widely used in industrial automation, electric vehicles and aerospace due to their high efficiency, high power density and excellent control performance.

[0003] In the field of permanent magnet synchronous motor control, traditional sliding mode observers are widely used for motor state estimation, especially in current observation. However, there is a significant deficiency in the practical application of traditional sliding mode observers, which is the high sensitivity to current noise. This problem mainly stems from the structure and working principle of traditional sliding mode observers. Traditional sliding mode observers drive state updates by measuring the deviation between the current and the estimated current, but in a noisy environment, current measurements are often disturbed by random noise. Since traditional sliding mode observers directly use these noisy measurements for state estimation, noise is directly introduced into the dynamics of the observer, resulting in large fluctuations and errors in the observed current and back-EMF estimates. This high sensitivity to current noise not only reduces the accuracy of state estimation, but also may cause system instability, especially in high-precision control applications, where this problem is particularly prominent. SUMMARY

[0004] To overcome the deficiencies in the prior art, the present application provides a permanent magnet synchronous motor position sensorless control method based on predictive correction to solve the technical problem of insufficient motor dynamic performance caused by current noise amplification in the prior art.

[0005] The present application provides a permanent magnet synchronous motor position sensorless control method based on predictive correction, comprising the following steps:

[0006] Step 1: Constructing a motor mathematical model in the stationary coordinate system of the permanent magnet synchronous motor;

[0007] Step 2: Constructing a sliding mode observer model and a current error equation;

[0008] Step 3: Adding a predictive current correction term to the sliding mode observer model, and adjusting the weight value of the predictive current correction term weight;

[0009] Step 4: Current estimation by sliding mode observer; according to the estimated current at the last moment and the current applied voltage, the current at the current moment is predicted by the discrete motor model, the difference between the actual current at the current moment and the predicted current at the current moment is calculated, and the sliding mode observer is corrected according to the difference;

[0010] Step 5: Constructing the prediction equation and the correction equation of the Kalman filter, and correcting and estimating the back electromotive force through the Kalman filter;

[0011] Step 6: Calculating the estimated rotor speed according to the corrected and estimated back electromotive force and the back electromotive force constant, obtaining the estimated rotor position through the estimated rotor speed, and controlling the permanent magnet synchronous motor through the estimated rotor speed and the estimated rotor position.

[0012] Further, in the step 1, the motor mathematical model is:

[0013]

[0014] In the formula, u α , u β is the voltage component in the stationary coordinate system; R s is the stator resistance; L s is the stator equivalent inductance; e α , e β is the back electromotive force component.

[0015] Further, in the step 1, the current error equation is:

[0016]

[0017] In the formula, is the current observation error; R s is the stator resistance; L s is the stator equivalent inductance; v α , v β is the control input of the sliding mode observer.

[0018] Further, in the step 3, the sliding mode observer model with the added predicted current correction term is:

[0019]

[0020] In the formula, L s is the stator equivalent inductance; R s is the stator resistance; u α , u β is the voltage component in the stationary coordinate system; e α , e β is the back electromotive force component; s α , s β is the designed sliding surface; is the current component in the stationary coordinate system; is the added predicted current correction term; and α is the weight value.

[0021] Further, in the step 3, the calculation formula of the weight value is:

[0022]

[0023] wherein, α min is the lower limit of the weight; α max is the upper limit of the weight; ||e|| is the amplitude of the current observation error in the stationary coordinate system; e max is the preset maximum allowable current error, when ||e|| > e max , then ||e|| = e max .

[0024] Further, in the step 4, the specific formula for predicting the current at the current time is:

[0025]

[0026] wherein, is the estimated current at the previous time; u[k] is the current applied voltage.

[0027] Further, in the step 5, the prediction equation is:

[0028]

[0029] P[k|k-1] = AP[k-1|k-1]A T +Q

[0030] wherein, is the estimated back electromotive force at the previous time, A is the state transition matrix, is the state estimation after correction at time k-1, B is the control input matrix, u[k-1] is the voltage input applied at time k-1, P[k|k-1] is the covariance matrix of the estimated back electromotive force, P[k-1|k-1] is the covariance matrix after correction at the previous time, and Q is the covariance matrix of the process noise.

[0031] Further, in the step 5, the correction equation is:

[0032] K[k] = P[k|k-1]H T (HP[k|k-1]H T +R) -1

[0033]

[0034] P[k|k] = (I-K[k]H)P[k|k-1]

[0035] wherein, K[k] is the Kalman gain, H is the observation matrix, and R is the measurement noise covariance matrix, For the corrected estimated back electromotive force, y[k] is the actual measured value of the current, is the predicted measured value of the current, P[k|k] is the covariance matrix of the corrected estimated back electromotive force, and I is a unit matrix.

[0036] Further, in step 5, the specific method of correcting the estimated back electromotive force by the Kalman filter is: first, the back electromotive force is estimated by the prediction equation, and then the estimated back electromotive force is corrected by the correction equation.

[0037] The present application has the following beneficial effects:

[0038] The present application utilizes the error between the actual measured value and the predicted value of the current in the sliding mode controller to correct, avoids directly differentiating the measured value sensitive to noise, significantly improves the accuracy of rotor position and speed estimation, especially in a high noise environment, can effectively suppress the influence of noise on the estimation process, and improves the robustness and stability of the system.

[0039] The present application uses the Kalman filter to estimate the state, the Kalman filter combines the prediction equation and the correction equation, and dynamically adjusts the estimated value by using the Kalman gain, so as to realize optimal estimation in a noisy environment, significantly improve the accuracy of state estimation, especially in the face of measurement noise and model uncertainty, can effectively suppress noise interference, and provide more accurate back electromotive force, rotor speed and position estimation.

[0040] The present application uses a discrete motor model to predict the current, uses the discrete model of the motor, combines the estimated current at the last moment and the current applied voltage, predicts the current at the current moment, improves the accuracy of current prediction, provides a more reliable basis for subsequent error calculation and state correction, and further enhances the robustness of the system. BRIEF DESCRIPTION OF DRAWINGS

[0041] The features and advantages of the present application will be more clearly understood through reference to the accompanying drawings, which are schematic and should not be construed as any limitation on the present application, and in which:

[0042] Figure 1 is a flowchart of a specific embodiment of the present application;

[0043] Figure 2 is the actual angle and the estimated angle obtained by the traditional sliding mode observer;

[0044] Figure 3 is the actual angle and the estimated angle obtained by the specific embodiment of the present application;

[0045] Figure 4 is the back electromotive force waveform diagram obtained by the traditional sliding mode observer;

[0046] Figure 5 Figure of back electromotive force waveform obtained for the embodiment of the present application. DETAILED DESCRIPTION

[0047] In order to make the objects, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are some but not all of the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by a person of ordinary skill in the art without creative work fall within the scope of protection of the present application.

[0048] The present application will be further illustrated below with reference to the specific embodiments. Those skilled in the art should understand that the embodiments are only used to illustrate the present application and not used to limit the scope of the present application. Various modifications of the equivalent forms of the present application all fall within the scope of the appended claims of the present application.

[0049] As shown in Figure 1 , the present application provides a permanent magnet synchronous motor sensorless control method based on predictive correction, comprising the following steps:

[0050] Step 1: set the reference speed of the permanent magnet synchronous motor, obtain the actual voltage and current and construct the motor mathematical model in the stationary coordinate system of the permanent magnet synchronous motor through coordinate transformation, and the constructed motor mathematical model is:

[0051]

[0052] In the formula, u α , u β are voltage components in the stationary coordinate system; R s is the stator resistance; L s is the equivalent inductance of the stator; e α , e β are back electromotive force components.

[0053] Step 2: rewrite the motor mathematical model in the stationary coordinate system into a current state equation in the stationary coordinate system, and establish a sliding mode observer model and a current error equation according to the current state equation;

[0054] The current equation in the stationary coordinate system after rewriting the motor mathematical model is:

[0055]

[0056] In the formula, u α , u β are voltage components in the stationary coordinate system; i α , i β are current components in the stationary coordinate system; Rs For stator resistance; L s e is the stator equivalent inductance; α e β This is the back electromotive force component.

[0057] The sliding mode observer model established based on the current equation of equal order and corresponding format is as follows:

[0058]

[0059] In the formula, The current component in the stationary coordinate system; v α v β This is the control input for the sliding mode observer.

[0060] By comparing the current equation with the sliding mode observer model, the current error equation is obtained as follows:

[0061]

[0062] In the formula, the current observation error is:

[0063] Step 3: Add a predicted current correction term to the sliding mode observer model, and adjust the weight values ​​of the predicted current correction term.

[0064] The predicted current correction term is equivalent to the difference between the actual current and the predicted current at the current moment. The sliding mode observer model with the predicted current correction term is as follows:

[0065]

[0066] In the formula, L s R is the stator equivalent inductance; s For stator resistance; u α u β The voltage component in the stationary coordinate system; e α e β s is the back electromotive force component; α s β For the designed sliding surface; The current component is in the stationary coordinate system; The added predicted current correction term; α is the weight value used to adjust the weight of the predicted current correction term. When there is an error between the actual current and the predicted current, the error is used as a correction signal. After being adjusted by the gain parameter α, the observer's output is adjusted in real time. This allows the observer to provide high-accuracy current estimation even in the presence of noise and model uncertainty.

[0067] Based on the error between the actual current and the predicted current, the weight value α of the predicted current correction term can be obtained as follows:

[0068]

[0069] In the formula, α min Let α be the lower limit of the weight. max Here, ||e|| represents the upper limit of the weight, and ||e|| represents the magnitude of the current observation error in the stationary coordinate system. max The preset maximum allowable current error is defined when ||e|| > e. max When ||e| = e max .

[0070] Step 4: Estimate the current using a sliding mode observer; based on the estimated current at the previous moment and the currently applied voltage, predict the current at the current moment using a discrete motor model, calculate the difference between the actual current at the current moment and the predicted current at the current moment, and correct the sliding mode observer based on the difference.

[0071] The specific formula for predicting the current at the current moment is:

[0072]

[0073] In the formula, The current is the estimated current at the previous moment; u[k] is the voltage currently applied.

[0074] The error between the actual current sample value and the estimated current at the previous moment is:

[0075]

[0076] In the formula, i[k] is the actual current sampling value; This is the estimated current at the previous moment.

[0077] Step 5: Construct the prediction equation and correction equation of the Kalman filter. First, the back EMF is estimated using the prediction equation, and then the estimated back EMF is corrected using the correction equation.

[0078] The prediction equation is:

[0079]

[0080] P[k|k-1]=AP[k-1|k-1]A T +Q

[0081] In the formula, Let A be the estimated back electromotive force from the previous time step, and let A be the state transition matrix. is the state estimation corrected at time k-1, B is a control input matrix, u[k-1] is a voltage input applied at time k-1, P[k|k-1] is a covariance matrix of the estimated back electromotive force, P[k-1|k-1] is a covariance matrix corrected at the last time, and Q is a covariance matrix of process noise;

[0082] The correction equation is:

[0083] K[k] = P[k|k-1]H T (HP[k|k-1]H T +R) -1

[0084]

[0085] P[k|k] = (I-K[k]H)P[k|k-1]

[0086] wherein K[k] is a Kalman gain, H is an observation matrix, R is a covariance matrix of measurement noise, is the estimated back electromotive force after correction, y[k] is an actual current measurement value, is a predicted current measurement value, and P[k|k] is a covariance matrix of the estimated back electromotive force after correction, and I is an identity matrix. The Kalman filter uses a prediction equation to preliminarily estimate the back electromotive force at the current time, which is based on the system model and the estimation result at the last time; subsequently, a correction equation adjusts the preliminary estimation by introducing actual measurement data to improve the estimation accuracy; the calculation of the Kalman gain K[k] enables the measurement residual to be incorporated into the state estimation in an optimal manner, thereby achieving accurate estimation of the back electromotive force.

[0087] Step 6: calculating the estimated rotor speed according to the corrected estimated back electromotive force and the back electromotive force constant, obtaining the estimated rotor position through the estimated rotor speed, and controlling the permanent magnet synchronous motor through the estimated rotor speed and the estimated rotor position.

[0088] The specific formula for calculating the estimated rotor speed according to the corrected estimated back electromotive force and the back electromotive force constant is:

[0089]

[0090] The specific formula for obtaining the estimated rotor position through the estimated rotor speed is:

[0091]

[0092] The process of the application is simulated and described by the following examples:

[0093] The simulation environment is MATLAB / Simulink, and the stator resistance R is sets = 2.875 Ω, d-q axis inductance L s = 0.85 mH, pole pair number Pn = 4, rotor permanent magnet flux linkage ψ f = 0.175 wb.

[0094] As shown in Figure 2 , 3 , Figure 2 It can be seen that the traditional method will amplify the current noise in the current integral link, so that the observation angle and the actual angle have a certain deviation; the method of the present application corrects the error between the actual measurement value and the predicted value by predicting the future state of the motor, avoids the difference operation of the measurement value sensitive to the noise directly, and therefore can greatly reduce the error between the observation angle and the actual angle, greatly improves the system stability.

[0095] As shown in Figure 4 , 5 It can be seen that the back electromotive force waveform curve of the permanent magnet synchronous motor position control based on the prediction correction of the embodiment of the present application is smoother, and the chattering is smaller.

[0096] Although the embodiments of the present application are described in combination with the drawings, various modifications and variations can be made by those skilled in the art without departing from the spirit and scope of the present application, and such modifications and variations all fall within the scope defined by the appended claims.

Claims

1. A sensorless control method for a permanent magnet synchronous motor based on predictive correction, characterized in that, Includes the following steps: Step 1: Construct a mathematical model of the permanent magnet synchronous motor in the stationary coordinate system; Step 2: Construct the sliding mode observer model and the current error equation; Step 3: Add a predicted current correction term to the sliding mode observer model, and adjust the weight values ​​of the predicted current correction term. Step 4: Estimate the current using a sliding mode observer; based on the estimated current at the previous moment and the currently applied voltage, predict the current at the current moment using a discrete motor model, calculate the difference between the actual current at the current moment and the predicted current at the current moment, and correct the sliding mode observer based on the difference. Step 5: Construct the prediction equation and correction equation of the Kalman filter, and use the Kalman filter to correct and predict the back electromotive force. Step 6: Calculate the estimated rotor speed based on the corrected back EMF and back EMF constant, obtain the estimated rotor position through the estimated rotor speed, and control the permanent magnet synchronous motor through the estimated rotor speed and estimated rotor position.

2. The sensorless control method for permanent magnet synchronous motors based on predictive correction as described in claim 1, characterized in that, In step 1, the mathematical model of the motor is: In the formula, u α u β The voltage component is in the stationary coordinate system; R s For stator resistance; L s e is the stator equivalent inductance; α e β This is the back electromotive force component.

3. The sensorless control method for permanent magnet synchronous motors based on predictive correction as described in claim 1, characterized in that, In step 1, the current error equation is: In the formula, For current observation error; R s For stator resistance; L s The stator equivalent inductance; v α v β This is the control input for the sliding mode observer.

4. The sensorless control method for permanent magnet synchronous motors based on predictive correction as described in claim 1 or 2, characterized in that, In step 3, the sliding mode observer model with the added predicted current correction term is: In the formula, L s R is the stator equivalent inductance; s For stator resistance; u α u β The voltage component in the stationary coordinate system; e α e β s is the back electromotive force component; α s β For the designed sliding surface; The current component is in the stationary coordinate system; This is the added predicted current correction term; α is the weight value.

5. The sensorless control method for permanent magnet synchronous motors based on predictive correction as described in claim 4, characterized in that... In step 3, the formula for calculating the weight value is: In the formula, α min α is the lower limit of the weight; max ||e|| represents the upper limit of the weight; ||e|| represents the magnitude of the current observation error in the stationary coordinate system; e max The preset maximum allowable current error is defined when ||e|| > e. max When ||e|| = e max .

6. The sensorless control method for permanent magnet synchronous motors based on predictive correction as described in claim 1, characterized in that, In step 4, the specific formula for predicting the current at the current moment is: In the formula, The current is the estimated current at the previous moment; u[k] is the voltage currently applied.

7. The sensorless control method for permanent magnet synchronous motors based on predictive correction as described in claim 1, characterized in that, In step 5, the prediction equation is: In the formula, Let A be the estimated back electromotive force from the previous time step, and let A be the state transition matrix. For the state estimate after correction at time k-1, B is the control input matrix, u[k-1] is the voltage input applied at time k-1, P[k|k-1] is the covariance matrix of the predicted back electromotive force, P[k-1|k-1] is the covariance matrix after correction at the previous time, and Q is the covariance matrix of the process noise.

8. The sensorless control method for permanent magnet synchronous motors based on predictive correction as described in claim 1 or 7, characterized in that, In step 5, the correction equation is: K[k]=P[k|k-1]H T (HP[k|k-1]H T +R) -1 P[k|k]=(IK[k]H)P[k|k-1] In the formula, K[k] is the Kalman gain, H is the observation matrix, and R is the measurement noise covariance matrix. The corrected estimated back electromotive force is given by y[k], where y[k] is the actual measured value of the current. P[k|k] is the predicted current measurement, P[k|k] is the covariance matrix of the corrected predicted back electromotive force, and I is the identity matrix.

9. The sensorless control method for permanent magnet synchronous motors based on predictive correction as described in claim 7 or 8, characterized in that, In step 5, the specific method for correcting and estimating the back electromotive force using a Kalman filter is as follows: first, the back electromotive force is estimated using a prediction equation, and then the estimated back electromotive force is corrected using a correction equation.

Images