An optimization method for discrete data transmission of motor

Abstract

The invention discloses an optimization method in a discrete data transmission process of a motor. A conjugate gradient method is used to process the encoded discrete data at the motor input end and the decoded discrete data at the motor output end during the transmission process. The discrete data are the angles measured by the inner ring of a motor magnetic encoder and the angles measured by the outer ring. The angles measured by the inner ring of the magnetic encoder are the encoded discrete data at the input end during the transmission process. The decoded discrete data at the output end calculated by a function model is compared with the original decoded discrete data at the output end, thereby solving the measurement deviation of the inner and outer ring positions of the magnetic encoder using two sensors. A function model is established, and the angle measured by the inner ring is substituted into the function model for calculation and converted into the outer ring angle. The data error precision is reduced, the bit error rate BER is reduced, the discrete data transmission time is reduced, the transmission precision is improved, the transmission anti-interference ability is improved, and the data transmission speed is improved and the response is accelerated during the transmission process.

Description

Technical Field

[0001] The invention belongs to the field of data transmission and data processing, and relates to an optimization method in a motor discrete data transmission process. Background Art

[0002] The transmission of discrete data is widely used and important in the field of modern communication technology. However, it is easily affected by its own factors such as excessive amount of transmitted data and external factors such as interference, which can easily lead to a decrease in transmission quality.

[0003] Especially for the transmission of discrete data, there is no particularly good mathematical model to handle its transmission process, which leads to large errors and precision in discrete data after the transmission is disturbed or due to the large amount of data itself. , the bit error rate BER is high, and the discrete data transmission time Longer.

[0004] Motor encoders are divided into magnetic encoders and photoelectric encoders. Although magnetic encoders have lower accuracy and resolution than photoelectric encoders, they are smaller in size and structure. However, due to size and structure limitations in the application field of robot joints, photoelectric encoders are not suitable.

[0005] However, the magnetic encoder uses two sensors to measure the positions of the inner and outer rings. If the data measured by the outer ring sensor is used directly, the accuracy will be low because the outer encoder is easily interfered with. The angle measured by the inner ring is transmitted to the outer ring encoder through the inner ring encoder and converted into the outer ring angle. The external factors such as interference during the transmission process may be caused by external equipment interference or defects in the equipment itself.

[0006] The conjugate gradient method has the advantages of high accuracy and fast iteration speed when solving equations. It can be used to solve related problems in the process of optimizing discrete data transmission, such as improving transmission accuracy, improving anti-interference ability, and improving data transmission speed and response during the transmission process.

[0007] There are few patents or literature reports on the optimization of the transmission process of motor discrete data. Due to its discrete nature, discrete data is directly transmitted without being processed by a suitable mathematical model. The transmission quality is not high, which can easily lead to adverse effects such as poor hardware quality control.

[0008] When the applicant applied the previous generation of data transmission, the transmission time t was relatively long, which means that the faster the data transmission speed, the slower the response, such as 200-500ns.

[0009] The shorter the discrete data transmission time t is, the faster the data transmission speed and the faster the response during the transmission process. Summary of the invention

[0010] The purpose of the present invention is to provide an optimization method in the process of discrete data transmission of a motor.

[0011] The problem to be solved by the present invention is to collect the angles measured by the inner ring and the outer ring of the motor magnetic encoder. The angles measured by the inner ring of the magnetic encoder are the encoded discrete data at the input end during the transmission process, and the angles measured by the outer ring of the magnetic encoder are the decoded discrete data at the output end during the transmission process. Based on the collected measured angles of the inner ring encoder and the outer ring encoder, a function model is established through a conjugate gradient algorithm to process the discrete data during the transmission process, improve the accuracy of the decoded discrete data at the output end, improve the anti-interference ability during the transmission process, and improve the data transmission speed and response during the transmission process.

[0012] The general steps of the iterative algorithm are as follows:

[0013] ①Given starting point , , , To stop the iteration accuracy;

[0014] ②Search directions generated according to certain criteria ,like Then stop, It is the iteration data accuracy;

[0015] ③Determine the iteration step length ,make ;

[0016] ④ Order , repeat step ②.

[0017] A technical solution adopted by an optimization method in the process of discrete data transmission of a motor is as follows:

[0018] S1: Collect the angles measured by the inner ring and outer ring of the motor magnetic encoder. The angle measured by the inner ring of the magnetic encoder is the encoded discrete data at the input end during the transmission process, and the angle measured by the outer ring of the magnetic encoder is the decoded discrete data at the output end during the transmission process.

[0019] S2: Clean the collected angle data, remove angle values ​​that are beyond a reasonable range, and process discontinuous jump values;

[0020] S3: converting the angle data after data cleaning into a digital signal using an analog-to-digital converter;

[0021] S4: according to the converted digital signal, draw the original transmission curve in a plane rectangular coordinate system, the abscissa of the curve is the encoded discrete data at the input end during the transmission process, and the ordinate of the curve is the decoded discrete data at the output end during the transmission process;

[0022] S5: Perform function fitting based on the drawn transmission curve and establish a function model , y The angle measured by the outer ring is converted into a digital signal. x It is the digital signal converted from the angle measured by the inner ring. A, B, C and D are all parameters. n = the number of revolutions of the inner ring of the magnetic encoder / the number of revolutions of the outer ring;

[0023] S6: According to the obtained function model , substitute the digital signal converted from the angle measured by the inner loop each time into x That is, the input end encodes discrete data during the transmission process, and the corresponding output end decodes discrete data during the transmission process. The decoded discrete data at the output end is converted into the outer loop angle through the analog-to-digital converter.

[0024] The conjugate gradient method is described as follows:

[0025] For a set of vectors , assuming the matrix is a symmetric positive definite matrix if any two vectors satisfy , then we call this set of vectors and the symmetric matrix Zhengding.

[0026] Assume that there is a quadratic function of the form ,in , , and In matrix form, and is the coefficient matrix, is a constant matrix.

[0027] right Derivative, and make the derivative result be , you will get , we can see that, let the derivative be of That is The solution.

[0028] Assuming the starting point The search direction is the steepest descent search direction ,so .

[0029] exist When , the known quantity is and , it is clear that these two quantities are linearly independent, so we construct a new quantity ,make .

[0030] Will and Substitution Available , and is the conjugate search direction, Search direction vector for the first iteration The coefficient of .

[0031] Using mathematical induction, we can derive the search step length: , in , is the conjugate search direction vector, For the k The conjugate search direction vector at iteration The coefficient of .

[0032] use Alternative In , we can get and use it to Taking the derivative and setting it to 0 gives the optimal search step length , ,in , is the iterative solution vector.

[0033] That is, the entire search iterative process of finding the optimal solution of the conjugate gradient method is as follows, using the residual Alternative ,from start, , In order to reduce the amount of calculation, some changes can be made to obtain the following steps:

[0034] ;

[0035] ;

[0036] ;

[0037] ;

[0038] ;

[0039] ;

[0040] in , All are residuals.

[0041] Furthermore, in the above step S2, the collected angle data is cleaned, including:

[0042] S21: Determine a reasonable range of angle acquisition according to the technical specifications of the motor sensor, and remove angle values ​​that exceed the reasonable range;

[0043] S22: For larger jump data, an interpolation method is used to fill in the values ​​near the jump point. For smaller jump data, a smoothing filter algorithm is used to smooth the data.

[0044] S23: Use visualization tools to check the cleaned data.

[0045] Furthermore, the function model in the above step S3 , the conjugate gradient algorithm is used to iteratively update the parameters A, B, C and D. The steps include:

[0046] S31: Function model Rewrite ;

[0047] S32: Calculation y The first-order partial derivatives with respect to parameters A, B, C, and D are: , , , ;

[0048] S33: Calculation y Second-order partial derivatives with respect to parameters A, B, C, and D ,in i and j represents one of the parameters A, B, C and D;

[0049] S34: Get , , , , , , , , , ;

[0050] Then the second-order partial derivative matrix H is ;

[0051] S35: For the function model , and the quadratic form The iterative steps are different. It is necessary to replace a in the iterative step with the desired quadratic partial derivative matrix H. , , , , , ;

[0052] S36: For step S35 The choice of the initial value of the iteration is to generate a length of m A random vector x , where each element has a mean of , standard deviation is Normal distribution , using a normally distributed random number generator to generate m Normal distribution Random numbers are put into the vector x in each element of .

[0053] Furthermore, the function model obtained in the above step S4 , data error accuracy during transmission , bit error rate BER, discrete data transmission time The calculation is as follows:

[0054] Data error accuracy The calculation formula is: Data error accuracy = The difference between the decoded discrete data at the motor output and the encoded discrete data at the motor original input / the encoded discrete data at the motor original input, where the difference is the absolute value;

[0055] The bit error rate BER calculation formula is: bit error rate BER = number of error codes / total number of codes, where the total number of codes is substituted into the encoded discrete data at the original input end of the motor, and the error code number is substituted into the decoded discrete data at the output end of the motor after transmission, which has a large difference in error accuracy from the encoded discrete data at the original input end of the motor;

[0056] Discrete data transmission time The calculation formula is: Discrete data transmission time =The time when the decoded discrete data at the motor output end ends receiving the data - the time when the encoded discrete data at the motor input end starts uploading the data.

[0057] Data error accuracy Reflects the degree of deviation of discrete data after the conjugate gradient method is used, and the bit error rate BER reflects the anti-interference ability and transmission time of the model during transmission. Reflects the optimization capability of the function model for the discrete data transmission process of the motor;

[0058] Data error accuracy The lower the bit error rate (BER), the higher the transmission accuracy. The lower the bit error rate (BER), the stronger the anti-interference ability during transmission. The discrete data transmission time The shorter it is, the faster the data transmission speed and the faster the response during the transmission process.

[0059] Furthermore, the above step S6 includes:

[0060] S61: Each time the motor runs, the angle information is collected through the inner ring of the magnetic encoder and converted into a digital signal through the analog-to-digital converter;

[0061] S62: Combine the converted inner loop angle digital signal with the obtained function model , get the outer ring angle digital signal;

[0062] S63: converting the obtained outer ring angle digital signal into the angle information of the outer ring through an analog-to-digital converter, and the outer ring of the magnetic encoder directly controls the operation of the motor according to the angle information;

[0063] S64: After power is turned off and then on again, the motor uses the inner loop angle information before power off combined with the function model obtained After power is turned on, the outer ring angle digital signal is calculated. According to the outer ring angle digital signal, the motor outer ring is restored to the operating position before power failure, realizing the power-off memory function of the motor.

[0064] The optimization method in the process of discrete data transmission of a motor, wherein the discrete data acquisition is performed on the discrete transmission data of the motor, and the angle measured by the inner ring and the outer ring of the motor magnetic encoder are collected. The angle measured by the inner ring of the magnetic encoder is the input end encoded discrete data during the transmission process, and the angle measured by the outer ring of the magnetic encoder is the output end decoded discrete data during the transmission process. During the transmission process, the conjugate gradient algorithm is used to combine the original input end encoded discrete data of the motor with the function model After the transmission, the decoded discrete data of the motor output end is obtained, and the system can automatically process the encoded discrete data of the motor original input end according to the established function model. The function models established for different motors are different.

[0065] The conjugate gradient method has the advantages of high accuracy and fast iteration speed. When we collect multiple sets of corresponding input-end encoded discrete data and output-end decoded discrete data during the transmission process, we make a one-to-one correspondence and calculate and establish a function model based on the conjugate gradient method, and we get , , and .

[0066] The beneficial effects of the present invention are as follows: the angle measured by the inner ring and the outer ring of the motor magnetic encoder are collected, the angle measured by the inner ring of the magnetic encoder is the input end coded discrete data during the transmission process, and the angle measured by the outer ring of the magnetic encoder is the output end decoded discrete data during the transmission process. Multiple groups of corresponding input end coded discrete data and output end decoded discrete data are collected during the transmission process, and a one-to-one correspondence is performed and a function model is established based on the conjugate gradient method. The established function model , through the function model Compared with the original output-end decoded discrete data, the calculated output-end decoded discrete data solves the measurement deviation of the inner and outer ring positions of the magnetic encoder using two sensors. A function model is established, and the angle measured by the inner ring is substituted into the function model for calculation and converted into the outer ring angle. The data error accuracy The bit error rate BER is reduced and the discrete data transmission time is The transmission accuracy is improved, the anti-interference ability of the transmission is improved, and the data transmission speed is increased and the response is accelerated during the transmission process. In addition, the function models established for different motors are different, and the function model of the same motor does not need to be repeatedly established. BRIEF DESCRIPTION OF THE DRAWINGS

[0067] Figure 1 Shown is a flow chart of the present invention;

[0068] Figure 2 The figure shows the effect diagram of the function model established by the first group of conjugate gradient method;

[0069] Figure 3 The figure shows the effect diagram of the function model established by the second group of conjugate gradient method;

[0070] Figure 4 Shown is the effect diagram of the function model established by the third group of conjugate gradient method. DETAILED DESCRIPTION

[0071] The present invention is further clearly and completely described below, but the protection scope of the present invention is not limited thereto.

[0072] An optimization method in the process of discrete data transmission of a motor adopts the following technical solution:

[0073] The present invention is further clearly and completely described below, but the protection scope of the present invention is not limited thereto.

[0074] An optimization method in the process of discrete data transmission of a motor adopts the following technical solution:

[0075] S1: Collect the angles measured by the inner ring and outer ring of the motor magnetic encoder. The angle measured by the inner ring of the magnetic encoder is the encoded discrete data at the input end during the transmission process, and the angle measured by the outer ring of the magnetic encoder is the decoded discrete data at the output end during the transmission process.

[0076] S2: Clean the collected angle data, remove angle values ​​that are beyond a reasonable range, and process discontinuous jump values;

[0077] S3: converting the angle data after data cleaning into a digital signal using an analog-to-digital converter;

[0078] S4: according to the converted digital signal, draw the original transmission curve in a plane rectangular coordinate system, the abscissa of the curve is the encoded discrete data at the input end during the transmission process, and the ordinate of the curve is the decoded discrete data at the output end during the transmission process;

[0079] S5: According to the drawn transmission curve, curve fitting is performed to establish a function model , y The angle measured by the outer ring is converted into a digital signal. x It is the digital signal converted from the angle measured by the inner ring. A, B, C and D are all parameters. n = the number of revolutions of the inner ring of the magnetic encoder / the number of revolutions of the outer ring;

[0080] S6: According to the obtained function model , substitute the digital signal converted from the angle measured by the inner loop each time into x That is, the input end encodes discrete data during the transmission process, and the corresponding output end decodes discrete data during the transmission process. The decoded discrete data at the output end is converted into the outer loop angle through the analog-to-digital converter.

[0081] The conjugate gradient method is described as follows:

[0082] For a set of vectors , assuming the matrix is a symmetric positive definite matrix if any two vectors satisfy , then we call this set of vectors and the symmetric matrix Zhengding.

[0083] Assume that there is a quadratic function of the form ,in , , and In matrix form, and is the coefficient matrix, is a constant matrix.

[0084] right Take the derivative and make the derivative result , you will get , we can see that, let the derivative be of That is The solution.

[0085] Assuming the starting point The search direction is the steepest descent search direction ,so .

[0086] exist When , the known quantity is and , it is clear that these two quantities are linearly independent, so we construct a new quantity ,make .

[0087] Will and Substitution Available , and is the conjugate search direction, Search direction vector for the first iteration The coefficient of .

[0088] Using mathematical induction, we can derive the search step length: , in , is the conjugate search direction vector, For the k The conjugate search direction vector at iteration The coefficient of .

[0089] use Alternative In , we can get and use it to Taking the derivative and setting it to 0 gives the optimal search step length , ,in , is the iterative solution vector.

[0090] That is, the entire search iterative process of finding the optimal solution of the conjugate gradient method is as follows, using the residual Alternative ,from start, , In order to reduce the amount of calculation, some changes can be made to obtain the following steps:

[0091] ;

[0092] ;

[0093] ;

[0094] ;

[0095] ;

[0096] ;

[0097] in , All are residuals.

[0098] Furthermore, in the above step S2, the collected angle data is cleaned, including:

[0099] S21: Determine a reasonable range of angle acquisition according to the technical specifications of the motor sensor, and remove angle values ​​that exceed the reasonable range;

[0100] S22: For larger jump data, an interpolation method is used to fill in the values ​​near the jump point, and for smaller jump data, a smoothing filter algorithm is used to smooth the data;

[0101] S23: Use visualization tools to check the cleaned data.

[0102] Furthermore, the function model in the above step S3 , the conjugate gradient algorithm is used to iteratively update the parameters A, B, C and D. The steps include:

[0103] S31: Function model Rewrite ;

[0104] S32: Calculation y The first-order partial derivatives with respect to parameters A, B, C, and D are: , , , ;

[0105] S33: Calculation y Second-order partial derivatives with respect to parameters A, B, C, and D ,in i and j represents one of the parameters A, B, C and D;

[0106] S34: Get , , , , , , , , , ;

[0107] Then the second-order partial derivative matrix H is ;

[0108] S35: For the function model , and the quadratic form The iterative steps are different. It is necessary to replace a in the iterative step with the desired quadratic partial derivative matrix H. , , , , , ;

[0109] S36: For step S35 The choice of the initial value of the iteration is to generate a length of m A random vector x , where each element has a mean of , standard deviation is Normal distribution , using a normally distributed random number generator to generate m Normal distribution Random numbers are put into the vector x in each element of .

[0110] Furthermore, the function model obtained in the above step S4 , data error accuracy during transmission , bit error rate BER, discrete data transmission time The calculation is as follows:

[0111] Data error accuracy The calculation formula is: Data error accuracy = The difference between the decoded discrete data at the motor output and the encoded discrete data at the motor original input / the encoded discrete data at the motor original input, where the difference is the absolute value;

[0112] The bit error rate BER calculation formula is: bit error rate BER = number of error codes / total number of codes, where the total number of codes is substituted into the encoded discrete data at the original input end of the motor, and the error code number is substituted into the decoded discrete data at the output end of the motor after transmission, which has a large difference in error accuracy from the encoded discrete data at the original input end of the motor;

[0113] Discrete data transmission time The calculation formula is: Discrete data transmission time =The time when the decoded discrete data at the motor output end ends receiving the data - the time when the encoded discrete data at the motor input end starts uploading the data.

[0114] Furthermore, the above step S6 includes:

[0115] S61: Each time the motor runs, the angle information is collected through the inner ring of the magnetic encoder and converted into a digital signal through the analog-to-digital converter;

[0116] S62: Combine the converted inner loop angle digital signal with the obtained function model , get the outer ring angle digital signal;

[0117] S63: converting the obtained outer ring angle digital signal into the angle information of the outer ring through an analog-to-digital converter, and the outer ring of the magnetic encoder directly controls the operation of the motor according to the angle information;

[0118] S64: After power is turned off and then on again, the motor uses the inner loop angle information before power off combined with the function model obtained After power is turned on, the outer ring angle digital signal is calculated. According to the outer ring angle digital signal, the motor outer ring is restored to the operating position before power failure, realizing the power-off memory function of the motor.

[0119] Embodiment 1:

[0120] The first group of motor magnetic encoder inner ring rotation number / outer ring rotation number is 9, that is, n = 9, generate a random vector of length 111 x , where each element follows a normal distribution with a mean of 1503.45 and a standard deviation of 519.81 , use the normal distribution random number generator to generate 111 normally distributed random numbers Random numbers are put into the vector x In each element of , the function model of the first group of motor original output terminal decoded discrete data and motor original input terminal encoded discrete data using the conjugate gradient method is: , Figure 2 To compare the function model obtained by the conjugate gradient method with the original transmission data, the data error accuracy is calculated one by one. , error accuracy The maximum is 0.4316%, according to the error accuracy To calculate the bit error rate BER, the bit error rates BER are , discrete data transmission time is 50.21 ns This shows that the conjugate gradient method is used to optimize the motor discrete data transmission process, and the error accuracy Low, high transmission accuracy, low bit error rate BER, indicating high anti-interference ability, discrete data transmission time The smaller the data transmission speed, the faster the response.

[0121] Embodiment 2:

[0122] The second group of motor magnetic encoder inner ring rotation number / outer ring rotation number is 10, that is, n = 10, generating a random vector with a length of 142 x , where each element follows a normal distribution with a mean of 1576.08 and a standard deviation of 351.77 , use the normal distribution random number generator to generate 142 normally distributed random numbers Random numbers are put into the vector x In each element of , the function model of the second group of motor original output terminal decoded discrete data and motor original input terminal encoded discrete data using the conjugate gradient method is: , Figure 3 To compare the function model obtained by the conjugate gradient method with the original transmission data, the data error accuracy is calculated one by one. , error accuracy The maximum is 0.5256%, according to the error accuracy To calculate the bit error rate BER, the bit error rates BER are , discrete data transmission time is 50.26 ns This shows that the conjugate gradient method is used to optimize the motor discrete data transmission process, and the error accuracy Low, high transmission accuracy, low bit error rate BER, indicating high anti-interference ability, discrete data transmission time The smaller the data transmission speed, the faster the response.

[0123] Embodiment three:

[0124] The third group of motor magnetic encoder inner ring rotation number / outer ring rotation number is 11, that is, n = 11, generating a random vector of length 146 x , where each element follows a normal distribution with a mean of 701.58 and a standard deviation of 279.44 , use the normal distribution random number generator to generate 111 normally distributed random numbers Random numbers are put into the vector x In each element of , the function model of the third group of motor original output terminal decoded discrete data and motor original input terminal encoded discrete data using the conjugate gradient method is: , Figure 4To compare the function model obtained by the conjugate gradient method with the original transmission data, the data error accuracy is calculated one by one. , error accuracy The maximum is 0.3256%, according to the error accuracy To calculate the bit error rate BER, the bit error rates BER are , discrete data transmission time is 49.23 ns This shows that the conjugate gradient method is used to optimize the motor discrete data transmission process, and the error accuracy Low, high transmission accuracy, low bit error rate BER, indicating high anti-interference ability, discrete data transmission time The smaller the data transmission speed, the faster the response.

[0125] The optimization method for the transmission of discrete data of a motor, wherein the discrete data is collected by the angle measured by the inner ring and the outer ring of the motor magnetic encoder, wherein the angle measured by the inner ring of the magnetic encoder is the input end encoded discrete data during the transmission process, and the angle measured by the outer ring of the magnetic encoder is the output end decoded discrete data during the transmission process, and the conjugate gradient algorithm is used to combine the original input end encoded discrete data of the motor with the function model during the transmission process The decoded discrete data of the motor output end after transmission is obtained, which solves the measurement deviation of the inner and outer ring positions of the magnetic encoder using two sensors. A function model is established, and the angle measured by the inner ring is substituted into the function model for calculation and converted into the outer ring angle. The system can automatically process the encoded discrete data of the original input end of the motor according to the established function model. The function models established for different motors are different.

Claims

1. An optimization method in the process of discrete data transmission of a motor, characterized in that: The following steps are involved: S1: Collect the angles measured by the inner ring and outer ring of the motor magnetic encoder. The angle measured by the inner ring of the magnetic encoder is the encoded discrete data at the input end during the transmission process, and the angle measured by the outer ring of the magnetic encoder is the decoded discrete data at the output end during the transmission process. S2: Clean the collected angle data, remove angle values ​​that are beyond a reasonable range, and process discontinuous jump values; S3: converting the angle data after data cleaning into a digital signal using an analog-to-digital converter; S4: according to the converted digital signal, draw the original transmission curve in a plane rectangular coordinate system, the abscissa of the curve is the encoded discrete data at the input end during the transmission process, and the ordinate of the curve is the decoded discrete data at the output end during the transmission process; S5: According to the drawn transmission curve, perform function fitting to establish a function model , y The angle measured by the outer ring is converted into a digital signal. x It is the digital signal converted from the angle measured by the inner ring. A, B, C and D are all parameters. n = the number of revolutions of the inner ring of the magnetic encoder / the number of revolutions of the outer ring; S6: According to the obtained function model , substitute the digital signal converted from the angle measured by the inner loop each time into x That is, the input end encodes discrete data during the transmission process, and the corresponding output end decodes discrete data during the transmission process. The decoded discrete data at the output end is converted into the outer loop angle through the analog-to-digital converter.

2. The optimization method in the process of discrete data transmission of a motor according to claim 1, characterized in that: The step S6 comprises: S61: Each time the motor runs, the angle information is collected through the inner ring of the magnetic encoder and converted into a digital signal through the analog-to-digital converter; S62: Combine the converted inner loop angle digital signal with the obtained function model , get the outer ring angle digital signal; S63: converting the obtained outer ring angle digital signal into the angle information of the outer ring through an analog-to-digital converter, and the outer ring of the magnetic encoder directly controls the operation of the motor according to the angle information; S64: After power is turned off and then on again, the motor uses the inner loop angle information before power off combined with the function model obtained After power is turned on, the outer ring angle digital signal is calculated. According to the outer ring angle digital signal, the motor outer ring is restored to the operating position before power failure, realizing the power-off memory function of the motor.

3. The optimization method in the process of discrete data transmission of a motor according to claim 1, characterized in that: The S6 function model Specifically: .

Images