A method for eliminating abnormal motor speed values and compensating with model prediction

Abstract

The present invention discloses a method for eliminating abnormal values of motor speed and model predictive compensation. The method comprises the following steps: Step 1, calculate the maximum acceleration that the motor can reach under normal working conditions; Step 2, calculate the rotational speed ω(k) of the motor at the current moment; Step 3, if ω(k−1) is the correct speed, compare ω(k) with ω(k−1). If the absolute value of the difference between the two is greater than the acceleration critical value, it can be determined that the rotational speed calculated at this moment is an abnormal value and needs to be eliminated and compensated; Step 4, if ω(k) is determined to be an abnormal value and ω(k + 1) at the next moment is still an abnormal value, the counter continues to increment. When the value calculated at a certain moment during sampling is a normal value, the value of Cnt is reset to 1, and Steps 2 to 4 are looped. This method directly screens and eliminates abnormal values of the rotational speed, and compensates for the eliminated speed based on the kinematic mathematical model of the motor, which can ensure accurate calculation of the actual true speed of the motor when the speed signal is interfered with.

Description

Technical Field

[0001] The present invention relates to a method for suppressing speed interference of a permanent magnet synchronous motor, and more particularly to a method for eliminating abnormal speed values of the motor and model predictive compensation, which is applicable to the speed control of a permanent magnet synchronous motor based on encoder speed measurement. Background Art

[0002] In the speed control system of a permanent magnet synchronous motor, speed feedback is usually obtained by differential calculation of an encoder installed inside the motor. This result serves as the feedback value of the speed regulator and participates in controlling the speed of the motor. Therefore, reliable and accurate speed calculation is quite important for motor speed control. However, modern permanent magnet synchronous motor control systems usually use high-power switching tubes for control, and the large dv / dt and di / dt caused by their fast turn-on and turn-off chopping will bring serious electromagnetic interference. On the other hand, when the motor and the controller operate in a harsh electromagnetic environment, such as aerospace-grade products, the electromagnetic interference from the external environment cannot be ignored. These electromagnetic interferences will seriously affect the signal transmission of the encoder, cause position detection error codes, and thus introduce speed feedback spikes in speed calculation, leading to adverse consequences such as instantaneous loss of control of the motor or triggering of driver protection. Therefore, it is very necessary to propose a method for suppressing speed interference of a permanent magnet synchronous motor.

[0003] Existing technologies usually use low-pass filters to filter the calculated speed. However, the filter can only suppress the huge speed spikes generated by interference and cannot completely eliminate them. Such glitches will still have a great impact on the smooth control of the motor. Some scholars have proposed screening abnormal position values, but this method is not good at dealing with the phenomenon of position zero-crossing during normal operation. Therefore, it is necessary to discuss in multiple cases, which increases the complexity of the software and reduces the system reliability. Summary of the Invention

[0004] The purpose of the present invention is to provide a method for eliminating abnormal speed values of the motor and model predictive compensation. This method directly screens and eliminates abnormal speed values, and compensates for the eliminated speed based on the kinematic mathematical model of the motor, which can ensure accurate calculation of the actual true speed of the motor when the speed signal is interfered, and has great application value.

[0005] The purpose of the present invention is achieved through the following technical solutions:

[0006] A method for eliminating abnormal speed values of the motor and model predictive compensation includes the following steps:

[0007] Step 1: Determine the parameters of the motor, including the moment of inertia J, torque coefficient K t and the maximum current I max, thereby calculating the maximum acceleration a that the motor can reach under normal operating conditions according to the motion equation of the motor max = T·K t I max / J;

[0008] Step 2: Read the position information θ(k) of the encoder at t = k, and calculate the rotational speed ω(k) of the motor at the current moment through Euler forward difference:

[0009]

[0010] where T is the control period, and θ(k - 1) is the position information of the encoder at t = k - 1;

[0011] Step 3: If ω(k - 1) is the correct speed, compare the rotational speed ω(k) obtained in Step 2 with the rotational speed ω(k - 1) calculated at the previous moment according to the following expression:

[0012]

[0013] If the rotational speed ω(k) at the current moment satisfies the above expression, it is considered that the calculated rotational speed is correct, and the result can be used for motor control. In this case, reset the outlier counter Cnt to 1;

[0014] If the rotational speed ω(k) at the current moment does not satisfy the above expression, it is judged as an outlier. Then, the software - set outlier counter Cnt starts to accumulate from 1, and Cnt is incremented by 1 each time an outlier occurs. At this time, the true rotational speed at the current moment t = k will be model - predicted and compensated through the motor rotational speed ω(k - 1) at the previous moment and the motor kinematic equation. The model - prediction compensation expression for the true rotational speed at the current moment is shown as follows:

[0015]

[0016] where i q (k) is the sampled value of the current sensor, and this rotational speed value will replace the speed calculated from the position signal and be fed back to the speed regulator for control;

[0017] Step 4: If the rotational speed ω(k) at the k - th moment is judged as an outlier, the outlier judgment condition for the next moment k + 1 is shown as follows:

[0018]

[0019] And so on. If ω(k + 1) is still an outlier, the counter continues to accumulate. When the value calculated at a certain moment through sampling is a normal value, the value of the outlier counter Cnt is reset to 1, and then loop through Step 2, Step 3, and Step 4.

[0020] Compared with the prior art, the present invention has the following advantages:

[0021] 1. The present invention is applicable to the speed control system of a permanent magnet synchronous motor, and can significantly suppress the speed spikes introduced by interference, making the system have strong anti-interference ability.

[0022] 2. The method of the present invention can effectively eliminate the abnormal speed measurement values caused by electromagnetic interference in a complex electromagnetic environment, and perform precise model prediction compensation according to the mathematical model of the permanent magnet synchronous motor, greatly improving the anti-interference ability of the motor control system, ensuring the stable and reliable operation of the motor, and having extremely high application value. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] Figure 1 is a block diagram of an algorithm for eliminating abnormal rotation speed values and model prediction compensation proposed by the present invention;

[0024] Figure 2 is a flowchart of the algorithm. DETAILED DESCRIPTION OF THE INVENTION

[0025] The technical solutions of the present invention will be further described below in conjunction with the accompanying drawings, but are not limited thereto. Any modification or equivalent replacement of the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention shall be covered by the protection scope of the present invention.

[0026] The basic structure of the present invention is based on the traditional vector control algorithm of a three-phase surface-mounted permanent magnet synchronous motor, and adopts i * d = 0 control, where ACR represents the current regulator and ASR represents the speed regulator. Its working block diagram is as Figure 1 shown. Among them, the d-axis current reference i * d is set to 0. The a-phase and b-phase current feedbacks of the motor are sampled by high-precision current sensors configured in the driver, and are converted into dq-axis currents through clark and park transformations. The rotor position information of the motor is detected by an encoder. Generally speaking, in the driver, this position information is used to calculate the rotational speed by the forward difference method. Different from the traditional vector control structure, the present invention adds a speed abnormal value screening and model prediction compensation link after the above rotational speed calculation link, greatly improving the anti-interference ability of the system. As Figure 2 shown, the specific application steps are as follows:

[0027] Step 1: Determine the parameters of the motor, including the moment of inertia J, torque coefficient K t and the maximum current I max of the motor, and thus calculate the maximum acceleration a that the motor can reach under normal working conditions according to the motion equation of the motormax , The calculation formula is as follows:

[0028] a max = T·K t I max / J;

[0029] Among them, T is the control period.

[0030] Step 2: Read the position information θ(k) of the encoder at t = k, and calculate the current speed ω(k) of the motor through Euler forward difference:

[0031]

[0032] Among them, θ(k - 1) is the position information of the encoder at t = k - 1.

[0033] Step 3: If ω(k - 1) is the correct speed, compare the speed ω(k) obtained in Step 2 with the speed ω(k - 1) calculated at the previous moment. If the absolute value of the difference between the two is greater than the acceleration critical value T·K t I max / J, it can be determined that the speed calculated at this moment is an abnormal value and needs to be eliminated and compensated. Among them: The effective condition for judging the abnormal value is calculated according to the maximum acceleration that the motor can actually achieve and the control period of the driver, and the expression is as shown in the following formula:

[0034]

[0035] If the current speed ω(k) satisfies the above expression, it is considered that the calculated speed is correct, and the result can be used for motor control. In this case, reset the software-set abnormal value counter Cnt to 1;

[0036] If the current speed ω(k) does not satisfy the above expression, it is judged as an abnormal value. Then, the software-set abnormal value counter Cnt starts to accumulate and count from 1, and each time an abnormality occurs, the Cnt counter is incremented by 1. At this time, the true speed at the current moment t = k will be compensated by model prediction through the motor speed ω(k - 1) at the previous moment and the motor kinematic equation. The model prediction compensation expression for the true speed at the current moment is as shown in the following formula:

[0037]

[0038] Among them, i q (k) is the sampling value of the current sensor, and this speed value will replace the speed calculated from the position signal and be fed back to the speed regulator for control.

[0039] Step 4: If the rotational speed ω(k) at time k is determined to be an outlier, the outlier determination condition for the next time k+1 is shown in the following formula:

[0040]

[0041] where Cnt is an outlier counter set by the software; additionally, the reference value for comparison is always the correct speed obtained from the nearest sample to the current time.

[0042] And so on. If ω(k+1) is still an outlier, the counter continues to increment. When the value calculated from a certain sample is a normal value, the value of the outlier counter Cnt is reset to 1. Then, Steps 2, 3, and 4 are looped to implement the complete process of the anti-interference algorithm proposed in the present invention.

Claims

1. A method for eliminating abnormal values of motor speed and compensating by model prediction, characterized in that the method comprises the following steps: Step 1: Determine the parameters of the motor, including the moment of inertia J, torque coefficient K t and the maximum current I max , and thus calculate the maximum acceleration a that the motor can reach under normal working conditions according to the motion equation of the motor max ; Step 2: Read the position information θ(k) of the encoder at t = k, and calculate the current speed ω(k) of the motor by Euler forward difference; Step 3: If ω(k - 1) is the correct speed, compare the speed ω(k) obtained in Step 2 with the speed ω(k - 1) calculated at the previous moment according to the following expression: wherein, T is the control period; If the rotational speed ω(k) at the current moment satisfies the above expression, it is considered that the calculated rotational speed is correct, and the result can be used for motor control. In this case, the outlier counter Cnt is reset to 1; If the rotational speed ω(k) at the current moment does not satisfy the above expression, it is judged as an outlier. Then, the outlier counter Cnt set by the software starts to accumulate from 1, and Cnt is incremented by 1 each time an outlier occurs. At this time, the true rotational speed at the current moment t = k will be model-predicted and compensated through the motor rotational speed ω(k - 1) at the previous moment and the motor kinematic equation, replacing the speed calculated from the position signal, and fed back to the speed regulator for control; Step 4: If the rotational speed ω(k) at time k is judged as an outlier, the outlier judgment condition for the next moment k + 1 is shown in the following formula: And so on. If ω(k + 1) is still an outlier, the counter continues to accumulate. When the value calculated by sampling at a certain moment is a normal value, the value of the outlier counter Cnt is reset to 1, and then steps 2, 3, and 4 are looped.

2. The method for eliminating abnormal values of motor speed and compensating by model prediction according to claim 1, characterized in that the calculation formula of the maximum acceleration a max is as follows: a max = T·K t I max / J.

3. The method for eliminating abnormal values of motor speed and compensating by model prediction according to claim 1, characterized in that the calculation formula of the current speed ω(k) of the motor is as follows: wherein, θ(k - 1) is the position information of the encoder at t = k - 1.

4. The method for eliminating abnormal values of motor speed and compensating by model prediction according to claim 1, characterized in that the model prediction compensation expression of the true speed at the current moment is as shown in the following formula: wherein, i q (k) is the sampled value of the current sensor.

Images