A motor inverter system detection feedback analysis system and method

Abstract

The application discloses a motor inverter system detection feedback analysis system and method, relates to the technical field of motor control, and obtains sampling period data of a decoding chip and a torque current controller; obtains input and output signals of the decoding chip and rotor position signals received by the torque current controller, determines the transmission time of the rotor position signals output by the decoding chip to the torque current controller; compensates the rotor position signals output by the decoding chip based on the transmission time, the sampling period of the decoding chip and the sampling period of the torque current controller; determines the rotor position based on rotor position information of two kinds of heterogeneous redundant modes; calibrates the signal transmission time from the decoding chip to the torque current controller, compensates the rotor position information received by the torque current controller, so that the rotor position information is matched with the sampling node of the torque current controller; the two kinds of heterogeneous redundant modes are used to provide rotor position information, and the stability and safety of the system are provided.

Description

Technical Field

[0001] This invention relates to the field of motor control technology, specifically to a detection feedback analysis system and method for a motor inverter system. Background Technology

[0002] In recent years, with the rapid growth of new energy vehicles, torque control has become a key component of motor control systems within their motor drive inverter systems. It involves the precise adjustment of motor output torque to meet various application requirements. Torque control is closely related to torque safety; torque control provides the foundation for achieving precise motion, while torque safety ensures the safety of this process. Furthermore, there is a close relationship between the motor rotor angle and the torque control current. The torque control algorithm in the motor controller calculates the required current value based on the rotor angle and current feedback. To improve the performance and stability of the motor control system, obtaining the rotor position information of the controller's sampling nodes has become an urgent problem to solve. Summary of the Invention

[0003] The purpose of this invention is to provide a detection feedback analysis system and method for motor inverter systems to solve the problems mentioned in the background art.

[0004] In one aspect of the present invention, a detection feedback analysis method for a motor inverter system is provided, comprising:

[0005] S11, obtain the sampling period data of the decoding chip and the sampling period of the torque current controller;

[0006] S12, acquire the input and output signals of the decoding chip and the rotor position signal received by the torque current controller, and determine the transmission time of the rotor position signal output by the decoding chip to the torque current controller;

[0007] S13, the rotor position signal output by the decoding chip is compensated based on the transmission time, the sampling period of the decoding chip and the sampling period of the torque current controller;

[0008] S14: Obtain rotor position information using two heterogeneous redundancy modes respectively, and determine rotor position based on rotor position information from the two heterogeneous redundancy modes.

[0009] In step S12, determining the transmission time from the rotor position signal output by the decoding chip to the torque current controller further includes the following steps:

[0010] S21, acquire the input and output signals of the decoding chip in the time domain, with the horizontal axis representing time and the vertical axis representing input or output, match the output signal of the decoding chip with the input signal, and convert the time of the input signal in the time domain to the time of the corresponding input signal;

[0011] S22, acquire the rotor position signal received by the torque current controller in the time domain, shift the rotor position signal received by the torque current controller to the left, and calculate the root mean square error between the rotor position signal received by the torque current controller and the output signal of the decoding chip after matching in step S21. When the root mean square error reaches the minimum value, the time for shifting to the left is the transmission time Ts.

[0012] The output of the decoding chip is the rotor position information of the decoding chip sampling node, and the torque current controller also receives the rotor position information of the decoding chip sampling node. It takes a certain amount of time for the decoding chip to calculate and transmit the rotor position to the torque current controller based on the output of the resolver. Therefore, there is a certain time difference between the output of the decoding chip and the input of the torque current controller in the time domain. The output and input of the decoding chip have a one-to-one correspondence, so the time when the torque current controller obtains the input can be correlated with the time of the decoding chip sampling node.

[0013] In step S13, the compensation of the rotor position signal output by the decoding chip based on the transmission time, the sampling period of the decoding chip, and the sampling period of the torque current controller further includes the following steps:

[0014] S31, determine whether the sampling period of the decoding chip and the sampling period of the torque current controller are the same. If they are the same, proceed to step S32; otherwise, proceed to step S33. When the sampling period of the decoding chip and the sampling period of the torque current controller are the same, the decoding chip and the torque current controller can sample synchronously. Due to the transmission time, the rotor position signal received by the torque current controller may be data from previous sampling nodes. It is necessary to consider matching the rotor position signal received by the torque current controller with the current time.

[0015] S32, let the sampling period of the torque current controller be Tc, and the sampling period of the decoding chip be Tj. Divide the transmission time and the sampling period of the decoding chip to obtain the remainder. Let m1 be less than Tj, and k be an integer greater than or equal to zero. Let the current input of the torque current controller be x(n), then the rotor position signal received by the torque current controller is rad(nk). Compensate rad(nk) to obtain rad(n), and then the process ends.

[0016] S33, calculate the least common multiple P of the sampling period Tc of the torque current controller and the sampling period Tj of the decoding chip, calculate the cumulative sampling error DT, and when the cumulative sampling error DT reaches P, reset it to 0 and start accumulating again;

[0017] S34 sums the transmission time Ts and the sampling error DT, and calculates the remainder m2 with respect to the sampling period Tc of the torque current controller. When m2 is less than Tc, Less than or equal to First, the rotor position signal rad(nk) is compensated to rad(n), and then rad(n) is compensated to obtain the rotor position rad; when Greater than First, the rotor position signal rad(nk) is compensated to rad(n+1), and then rad(n+1) is compensated based on the remainder m2 to obtain the rotor position rad.

[0018] When the sampling period of the decoding chip and the sampling period of the torque current controller are different, the decoding chip may not have sampled at the sampling node of the torque current controller. In this case, further compensation is needed based on the sampling node data of the decoding chip. Since there is a certain error in the compensation, it is necessary to select the sampling node of the decoding chip that is closer to the sampling node of the torque current controller.

[0019] In step S32, the compensation of rad(nk) to obtain rad(n) further includes the following steps:

[0020] Historical rotor position data is obtained, and K-order prediction of rotor position is performed. Let rad(i) represent the historical data of the i-th rotor position. Then, rad(i), rad(i+1), ..., rad(i+pn1) are used as inputs, and rad(i+pn1+k) is used as the output to train a k-order prediction model. pn1+1 is the number of input features of the k-order prediction model. pn1 is adjusted according to the prediction effect. The current rotor position data rad(nk) and the previous pn1 rotor position data are input into the k-order prediction model to obtain rad(n).

[0021] In step S34, the step of first compensating the rotor position signal rad(nk) to rad(n+1) further includes the following steps:

[0022] The rotor position is predicted at order K+1, with rad(i), rad(i+1), ..., rad(i+pn2) as inputs and rad(i+pn2+k) as output. The k-order prediction model is trained, where pn2+1 is the number of input features of the k-order prediction model. pn2 is adjusted according to the prediction effect. The current rotor position data rad(nk) and the previous pn2 rotor position data are input into the k+1-order prediction model to obtain rad(n+1).

[0023] when Less than or equal to Let w(n) represent the rotor angular velocity, then the rotor position in rad is calculated using the following formula: ;

[0024] The torque current controller has a short sampling period, and the angular acceleration of the rotor remains basically constant within one period. Less than or equal to , It is closer to the sampling node of the torque current controller;

[0025] when Greater than hour, If the sampling node is closer to the torque current controller, the rotor position (rad) is calculated using the following formula: ;

[0026] The rotational speed is determined through the following steps:

[0027] The rotational speed is predicted by order K+1. Based on the rotor positions rad(i), rad(i+1), ..., rad(i+pn3+1), the rotational speeds w(i), w(i+1), ..., w(i+pn3) are obtained, where w(i) = [rad(i+1) - rad(i)] / Tj, w(i+1) = [rad(i+2) - rad(i+1)] / Tj, and w(i+pn3) = [rad(i+pn3+1) - rad(i+pn3)] / Tj. Using w(i), w(i+1), ..., w(i+pn3) as inputs and w(i+pn3+k+1) as output, train a k+1 order speed prediction model. pn3+1 is the number of input features of the k+1 order speed prediction model. Adjust pn3 according to the prediction effect. Obtain the current speed w(nk-1) based on the current rotor position data rad(nk). Input the current speed and the previous pn3 rotor position data into the k+1 order speed prediction model to obtain w(n).

[0028] In step S14, determining the rotor position based on the rotor position information of the two heterogeneous redundancy modes also includes the following steps: determining whether there is an anomaly in the heterogeneous redundancy mode by means of the following method:

[0029] When heterogeneous redundancy mode 1 Less than or equal to At that time, the rotational speed is calculated based on R1(n) and R1(n-1), and the calculation error with the output of the K+1 order rotational speed prediction model is compared. If the error is less than a threshold, the data of heterogeneous redundancy mode 1 is not abnormal; otherwise, the data of heterogeneous redundancy mode 1 is abnormal. R1(n) is the rotor position obtained by heterogeneous redundancy mode 1 after compensating for the current rotor position information. When the heterogeneous redundancy mode 1... Greater than At that time, the rotational speed is calculated based on R1(n+1) and R1(n), and the calculation error with the output of the K+1 order rotational speed prediction model is compared. If the error is less than the threshold, the data of heterogeneous redundancy mode 1 is not abnormal; otherwise, the data of heterogeneous redundancy mode 1 is abnormal. R1(n+1) is the rotor position obtained after heterogeneous redundancy mode 1 compensates for the current rotor position information. The judgment method of heterogeneous redundancy mode 2 is the same as that of heterogeneous redundancy mode 1.

[0030] The final rotor position information R is obtained using the following formula. , and R1 and R2 are the rotor positions obtained by heterogeneous redundancy mode 1 and heterogeneous redundancy mode 2 after compensation based on the current rotor position information. The weights are adjusted according to whether there are any anomalies in the data of the heterogeneous redundancy mode. If there are anomalies in the heterogeneous redundancy mode, the weights are reduced.

[0031] The decoding chips in heterogeneous redundancy mode 1 and heterogeneous redundancy mode 2 can use different cycles. They perform rotor position compensation and anomaly detection independently without affecting each other. The rotor position information provided by both chips is only needed when determining the final rotor position information.

[0032] In another aspect of the invention, a detection feedback analysis system for a motor inverter system is provided, comprising a torque current controller, a speed measurement module, a data acquisition module, and a data analysis module; the output of the data acquisition module is connected to the input of the control module for acquiring torque current information of the motor; the output of the speed measurement module is connected to the input of the data analysis module for measuring the position of the motor rotor; the output of the data analysis module is connected to the input of the torque current controller for compensating the rotor position output by the speed measurement module and transmitting the compensated rotor position information to the torque current controller; the torque current controller generates a control signal to adjust the torque current based on the received rotor position information to maintain normal motor operation.

[0033] The rotational speed measurement module includes heterogeneous redundancy mode 1 and heterogeneous redundancy mode 2. In heterogeneous redundancy mode 1, the excitation and resolver feedback differential signals are sampled using an SDADC. In the input amplitude shaping / integrator, the resolver feedback sampled values ​​are shaped based on the synchronous excitation symbol output by the excitation phase synchronizer, and integrated at several sampling points within one excitation cycle to obtain the envelope Sin and Cos. After correction, the initial angle is obtained by using the arctangent Atan2 as a reference input to the observer / integrator, and the angle is output.

[0034] The heterogeneous redundancy mode 2 samples the excitation and resolver feedback differential signals through SARADC and then inputs them into the amplitude peak-valley selector. Based on the zero crossing of the synchronous excitation output by the excitation phase synchronizer, the resolver feedback ADC value at the peak and valley of the synchronous excitation wave is obtained. After the resolver feedback peak-valley ADC value is corrected, the angle is obtained by arctangent Atan2.

[0035] The data analysis module further includes a transmission time matching unit, a speed calculation unit, a rotor position compensation unit, a speed prediction unit, an anomaly detection unit, and a rotor position determination unit. The transmission time matching unit determines the time from when heterogeneous redundancy mode 1 and heterogeneous redundancy mode 2 receive signals from the rotary transformer to when the torque current controller receives rotor position information. The speed calculation unit calculates the speed based on the rotor position information received by the torque current controller. The rotor position compensation unit compensates for the rotor position information transmitted by heterogeneous redundancy mode 1 and heterogeneous redundancy mode 2 based on the sampling periods of heterogeneous redundancy mode 1 and heterogeneous redundancy mode 2, the sampling period of the torque current controller, and the transmission time. The speed prediction unit predicts the speed based on the speed calculated by the torque current controller. The anomaly detection unit determines whether there are anomalies in the data of heterogeneous redundancy mode 1 and heterogeneous redundancy mode 2, and adjusts the weights of heterogeneous redundancy mode 1 and heterogeneous redundancy mode 2 based on the determination result. The rotor position determination unit obtains the final rotor position information based on the rotor position information transmitted by heterogeneous redundancy mode 1 and heterogeneous redundancy mode 2.

[0036] The transmission time matching unit acquires the input and output signals of heterogeneous redundancy mode 1 in the time domain, with time as the horizontal axis and input or output as the vertical axis. It matches the output signal of heterogeneous redundancy mode 1 with the input signal, converting the time of the input signal in the time domain to the time of the corresponding input signal. It acquires the rotor position signal received by the torque current controller in the time domain, shifts the rotor position signal received by the torque current controller to the left, and calculates the root mean square error sum of the rotor position signal received by the torque current controller and the output signal of the matched heterogeneous redundancy mode 1. When the root mean square error sum reaches the minimum value, the time of shifting to the left is the transmission time of heterogeneous redundancy mode 1. The transmission time of heterogeneous redundancy mode 2 is determined in the same way.

[0037] Compared with the prior art, the beneficial effects achieved by the present invention are: calibrating the signal transmission time from the decoding chip to the torque current controller, compensating the rotor position information received by the torque current controller, so that the rotor position information matches the sampling node of the torque current controller; and providing rotor position information through two heterogeneous redundancy modes, thereby providing system stability and security. Attached Figure Description

[0038] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used in conjunction with embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings:

[0039] Figure 1 This is a schematic diagram of the structure of a motor inverter system detection feedback analysis system according to an embodiment of the present invention. Detailed Implementation

[0040] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0041] In an embodiment of the present invention, please refer to Figure 1 A detection feedback analysis system for a motor inverter system is provided, comprising: a torque current controller, a speed measurement module, a data acquisition module, and a data analysis module; the output of the data acquisition module is connected to the input of the control module for acquiring the torque current information of the motor; the output of the speed measurement module is connected to the input of the data analysis module for measuring the position of the motor rotor; the output of the data analysis module is connected to the input of the torque current controller for compensating the rotor position output by the speed measurement module and transmitting the compensated rotor position information to the torque current controller; the torque current controller generates a control signal to adjust the torque current based on the received rotor position information to maintain normal motor operation.

[0042] The speed measurement module has heterogeneous redundancy mode 1: angle conversion mode 0 and heterogeneous redundancy mode 2: angle conversion mode 1.

[0043] Angular Conversion Mode 0: The excitation and resolver feedback differential signals are rapidly sampled using an SDADC (Sigma DeltaAnalog-to-Digital Converter) and then input into an Input Amplitude Recifier / Integrator. Based on the synchronous excitation symbol output by the Excitation Phase Synchronizer, the resolver feedback sampled values ​​are shaped and integrated across multiple sampling points within one excitation cycle to obtain the envelope Sin and Cos. After Sin and Cos correction (InputCorrection), the arctangent Atan2 is used to obtain a preliminary angle, which is then input to the PID / Integrator (PID / Integrator) as a reference. The output angle is then compensated by Angle Compensator 0 to obtain a compensated angle aligned with the torque control current sampling time.

[0044] Angular Conversion Mode 1: The excitation and resolver feedback differential signals are rapidly sampled using a SARADC (Successive Approximation Register Analog-to-Digital Converter) and then input into an Input Amplitude Peak-Trough Selector. Based on the zero-crossing of the synchronous excitation output from the Excitation Phase Synchronizer, the resolver feedback ADC values ​​at the peaks and troughs of the synchronous excitation wave are obtained; based on the zero-crossing of the original excitation wave, the excitation ADC values ​​at the peaks and troughs of the original excitation wave are obtained. After correction (Input Correction), the resolver feedback peak-trough ADC values ​​are used to obtain the angle using the arctangent (Atan2), and the rotational speed is calculated using a speed calculator. The two angles are then compensated by Angle Compensator 1 to obtain a compensated angle aligned with the torque safety current sampling time.

[0045] The data analysis module further includes a transmission time matching unit, a speed calculation unit, a rotor position compensation unit, a speed prediction unit, an anomaly detection unit, and a rotor position determination unit. The transmission time matching unit determines the time from when heterogeneous redundancy mode 1 and heterogeneous redundancy mode 2 receive signals from the rotary transformer to when the torque current controller receives rotor position information. The speed calculation unit calculates the speed based on the rotor position information received by the torque current controller. The rotor position compensation unit compensates for the rotor position information transmitted by heterogeneous redundancy mode 1 and heterogeneous redundancy mode 2 based on the sampling periods of heterogeneous redundancy mode 1 and heterogeneous redundancy mode 2, the sampling period of the torque current controller, and the transmission time. The speed prediction unit predicts the speed based on the speed calculated by the torque current controller. The anomaly detection unit determines whether there are anomalies in the data of heterogeneous redundancy mode 1 and heterogeneous redundancy mode 2, and adjusts the weights of heterogeneous redundancy mode 1 and heterogeneous redundancy mode 2 based on the determination result. The rotor position determination unit obtains the final rotor position information based on the rotor position information transmitted by heterogeneous redundancy mode 1 and heterogeneous redundancy mode 2.

[0046] The transmission time matching unit acquires the input and output signals of heterogeneous redundancy mode 1 in the time domain, with time as the horizontal axis and input or output as the vertical axis. It matches the output signal of heterogeneous redundancy mode 1 with the input signal, converting the time of the input signal in the time domain to the time of the corresponding input signal. It acquires the rotor position signal received by the torque current controller in the time domain, shifts the rotor position signal received by the torque current controller to the left, and calculates the root mean square error sum of the rotor position signal received by the torque current controller and the output signal of the matched heterogeneous redundancy mode 1. When the root mean square error sum reaches the minimum value, the time of shifting to the left is the transmission time of heterogeneous redundancy mode 1. The transmission time of heterogeneous redundancy mode 2 is determined in the same way.

[0047] In an embodiment of the present invention, a detection feedback analysis method for a motor inverter system is provided, comprising:

[0048] S11, obtain the sampling period data of the decoding chip and the sampling period of the torque current controller;

[0049] S12, acquire the input / output signals of the decoding chip and the rotor position signal received by the torque current controller, and determine the transmission time of the rotor position signal output by the decoding chip to the torque current controller:

[0050] S21, acquire the input and output signals of the decoding chip in the time domain, with the horizontal axis representing time and the vertical axis representing input or output, match the output signal of the decoding chip with the input signal, and convert the time of the input signal in the time domain to the time of the corresponding input signal;

[0051] S22, acquire the rotor position signal received by the torque current controller in the time domain, shift the rotor position signal received by the torque current controller to the left, and calculate the root mean square error between the rotor position signal received by the torque current controller and the output signal of the decoding chip after matching in step S21. When the root mean square error reaches the minimum value, the time for shifting to the left is the transmission time Ts.

[0052] S13, the rotor position signal output by the decoding chip is compensated based on the transmission time, the sampling period of the decoding chip, and the sampling period of the torque current controller:

[0053] S31, determine whether the sampling period of the decoding chip and the sampling period of the torque current controller are the same. If they are the same, proceed to step S32; otherwise, proceed to step S33.

[0054] S32, let the sampling period of the torque current controller be Tc, and the sampling period of the decoding chip be Tj. Divide the transmission time and the sampling period of the decoding chip to obtain the remainder. Let m1 be less than Tj, k be an integer greater than or equal to zero, and m1 be the remainder. Let the current input of the torque current controller be x(n), then the rotor position signal received by the torque current controller is rad(nk). Compensate rad(nk) to obtain rad(n). Obtain historical data of rotor position and perform K-order prediction on rotor position. Let rad(i) represent the historical data of the i-th rotor position. Then take rad(i), rad(i+1), ..., rad(i+pn1) as input and rad(i+pn1+k) as output to train the k-order prediction model. pn1+1 is the number of input features of the k-order prediction model. Adjust pn1 according to the prediction effect. Input the current rotor position data rad(nk) and the previous pn1 rotor position data into the k-order prediction model to obtain rad(n), and the process ends.

[0055] The prediction model can be selected based on the performance of the torque current controller. First, the model is trained using a computer. Models such as long short-term memory neural networks, recurrent neural networks, convolutional neural network units, and linear regression can be used for prediction. Then, the trained model is stored in the torque current controller. The torque current controller can call the model to perform compensation based on the received rotor position information.

[0056] S33, calculate the least common multiple P of the sampling period Tc of the torque current controller and the sampling period Tj of the decoding chip, calculate the cumulative sampling error DT, and when the cumulative sampling error DT reaches P, reset it to 0 and start accumulating again; the cumulative sampling error DT increases in the form of |Tc-Tj|, 2|Tc-Tj|, ..., P;

[0057] S34 sums the transmission time Ts and the sampling error DT, and calculates the remainder m2 with respect to the sampling period Tc of the torque current controller. m2 is less than Tc;

[0058] when Less than or equal to First, the rotor position signal rad(nk) is compensated to rad(n), and then rad(n) is compensated to obtain the rotor position rad; the method of obtaining rad(n) is the same as in step S32; let w(n) represent the rotor angular velocity, then the rotor position rad is calculated using the following formula: ;

[0059] when Greater than First, the rotor position signal rad(nk) is compensated to rad(n+1), and then rad(n+1) is compensated based on the remainder m2 to obtain the rotor position rad. A K+1 order prediction is performed on the rotor position, with rad(i), rad(i+1), ..., rad(i+pn2) as inputs and rad(i+pn2+k) as output. A k-order prediction model is trained, where pn2+1 is the number of input features for the k-order prediction model. pn2 is adjusted based on the prediction performance. The current rotor position data rad(nk) and the previous pn2 rotor position data are input into the k+1 order prediction model to obtain rad(n+1). The rotor position rad is calculated using the following formula: ;

[0060] The rotational speed is determined through the following steps:

[0061] The rotational speed is predicted by order K+1. Based on the rotor positions rad(i), rad(i+1), ..., rad(i+pn3+1), the rotational speeds w(i), w(i+1), ..., w(i+pn3) are obtained, where w(i) = [rad(i+1) - rad(i)] / Tj, w(i+1) = [rad(i+2) - rad(i+1)] / Tj, and w(i+pn3) = [rad(i+pn3+1) - rad(i+pn3)] / Tj. Using w(i), w(i+1), ..., w(i+pn3) as inputs and w(i+pn3+k+1) as output, train a k+1 order speed prediction model. pn3+1 is the number of input features of the k+1 order speed prediction model. Adjust pn3 according to the prediction effect. Obtain the current speed w(nk-1) based on the current rotor position data rad(nk). Input the current speed and the previous pn3 rotor position data into the k+1 order speed prediction model to obtain w(n).

[0062] S14, Obtain rotor position information using two heterogeneous redundancy modes respectively, and determine rotor position based on rotor position information from the two heterogeneous redundancy modes.

[0063] Determine whether there are anomalies in heterogeneous redundancy modes using the following methods:

[0064] When heterogeneous redundancy mode 1 Less than or equal to At that time, the rotational speed is calculated based on R1(n) and R1(n-1), and the calculation error with the output of the K+1 order rotational speed prediction model is compared. If the error is less than a threshold, the data of heterogeneous redundancy mode 1 is not abnormal; otherwise, the data of heterogeneous redundancy mode 1 is abnormal. R1(n) is the rotor position obtained by heterogeneous redundancy mode 1 after compensating for the current rotor position information. When the heterogeneous redundancy mode 1... Greater than At that time, the rotational speed is calculated based on R1(n+1) and R1(n), and the calculation error with the output of the K+1 order rotational speed prediction model is compared. If the error is less than the threshold, the data of heterogeneous redundancy mode 1 is not abnormal; otherwise, the data of heterogeneous redundancy mode 1 is abnormal. R1(n+1) is the rotor position obtained after heterogeneous redundancy mode 1 compensates for the current rotor position information. The judgment method of heterogeneous redundancy mode 2 is the same as that of heterogeneous redundancy mode 1.

[0065] The final rotor position information R is obtained using the following formula. , and As weights, R1 and R2 are the rotor positions obtained after compensation based on the current rotor position information for heterogeneous redundancy mode 1 and heterogeneous redundancy mode 2, respectively. Adjustments are made based on whether there are anomalies in the heterogeneous redundancy mode data; if anomalies exist in the heterogeneous redundancy mode, the weights are reduced. Optionally, and All are set to 0.5. When a heterogeneous redundancy mode malfunctions, the weights are reset to 0, and only the output of the other heterogeneous redundancy mode is used as the rotor position.

[0066] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0067] Finally, it should be noted that the above descriptions are merely preferred embodiments of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A detection feedback analysis method for an electric motor inverter system, characterized by, The method comprises the following steps: S11, acquiring sampling period data of a decoding chip and sampling period of a torque current controller; S12, acquiring input and output signals of the decoding chip and rotor position signals received by the torque current controller, and determining transmission time of the rotor position signals output by the decoding chip to the torque current controller; S13, compensating the rotor position signals output by the decoding chip based on the transmission time, the sampling period of the decoding chip and the sampling period of the torque current controller; In step S13, the compensating the rotor position signals output by the decoding chip based on the transmission time, the sampling period of the decoding chip and the sampling period of the torque current controller further comprises the following steps: S31, judging whether the sampling period of the decoding chip and the sampling period of the torque current controller are the same, if yes, entering step S32, otherwise, performing step S33; S32, let the sampling period of the torque current controller be Tc, the sampling period of the decoding chip be Tj, let the transmission time and the sampling period of the decoding chip be divided to obtain a remainder : , m1 is less than Tj, k is an integer greater than or equal to zero, let the current input of the torque current controller at present be x(n), the rotor position signal received by the torque current controller is rad(n-k), rad(n-k) is compensated to obtain rad(n), and the process ends. S33, calculating the least common multiple P of the sampling period Tc of the torque current controller and the sampling period Tj of the decoding chip, and calculating cumulative sampling error DT, when the cumulative sampling error DT reaches P, resetting to 0 to start accumulating again; S34 sums the delivery time Ts and the sampling error DT and calculates the remainder m2 of the torque current controller sampling period Tc, m2 is less than Tc, when is less than or equal to Tc, the rotor position signal rad(n-k) is first compensated to rad(n), and rad(n) is then compensated to obtain the rotor position rad; when is greater than Tc, the rotor position signal rad(n-k) is first compensated to rad(n+1), and rad(n+1) is then compensated based on the remainder m2 to obtain the rotor position rad; S14, acquiring rotor position information in two heterogeneous redundant modes respectively, and determining the rotor position based on the rotor position information in the two heterogeneous redundant modes, wherein the two heterogeneous redundant modes comprise a heterogeneous redundant mode 1 and a heterogeneous redundant mode 2; The heterogeneous redundant mode 1 samples the excitation and the resolver feedback differential signal through an SDADC, shapes the resolver feedback sampling value based on the synchronous excitation symbol output by an excitation phase synchronizer in an input amplitude shaper / integrator, integrates the resolver feedback sampling value at a plurality of sampling points in an excitation period to obtain envelope lines Sin and Cos, obtains a preliminary angle as a reference through the Atan2 after the Sin and Cos are corrected, and inputs the preliminary angle into an observer / integrator to output an angle; The heterogeneous redundant mode 2 samples the excitation and the resolver feedback differential signal through an SARADC, inputs the resolver feedback differential signal into an amplitude peak-valley selector, acquires resolver feedback ADC values at synchronous excitation peaks and valleys based on synchronous excitation zero-crossing output by an excitation phase synchronizer, and obtains an angle through the Atan2 after the resolver feedback peak-valley ADC values are corrected.

2. The detection feedback analysis method of a motor inverter system according to claim 1, wherein, In step S12, the determining transmission time of the rotor position signals output by the decoding chip to the torque current controller further comprises the following steps: S21, acquiring input and output signals of the decoding chip in a time domain, taking time as the horizontal coordinate and taking input or output as the vertical coordinate, matching the output signal with the input signal of the decoding chip, and converting the time of the output signal in the time domain into the time of the corresponding input signal; S22, acquiring rotor position signals received by the torque current controller in the time domain, performing left translation on the rotor position signals received by the torque current controller, calculating the root mean square error sum of the rotor position signals received by the torque current controller and the output signal of the decoding chip matched in step S21, and when the root mean square error sum reaches a minimum value, the time of the left translation is the transmission time Ts.

3. The method of claim 2, wherein the feedback analysis is performed by a feedback analysis unit of the motor inverter system. In step S32, the compensation of rad(n-k) to obtain rad(n) further includes the following steps: Obtaining historical data of the rotor position, performing K-order prediction on the rotor position, taking rad(i) as the historical data of the i-th rotor position, rad(i), rad(i+1), … rad(i+pn1) as the input, and rad(i+pn1+k) as the output, training the k-order prediction model, and adjusting pn1 according to the prediction effect; inputting the current rotor position data rad(n-k) and the previous pn1 rotor position data into the k-order prediction model to obtain rad(n); In step S34, the compensation of the rotor position signal rad(n-k) to rad(n+1) further includes the following steps: Performing K+1-order prediction on the rotor position, taking rad(i), rad(i+1), … rad(i+pn2) as the input, and rad(i+pn2+k) as the output, training the k-order prediction model, and adjusting pn2 according to the prediction effect; inputting the current rotor position data rad(n-k) and the previous pn2 rotor position data into the k+1-order prediction model to obtain rad(n+1).

4. The method of claim 3, wherein the feedback analysis is performed by a feedback analysis unit of the motor inverter system. When Less than or equal to When Tc, let w(n) represent the rotor angular velocity, then the rotor position rad is calculated by the following formula: ; When greater than Tc, then the rotor position rad is calculated by the following equation: ; The speed is determined by the following steps: Performing K+1-order prediction on the speed, obtaining the speed w(i), w(i+1), … w(i+pn3) according to the rotor position rad(i), rad(i+1), … rad(i+pn3+1), w(i)=[rad(i+1)-rad(i)] / Tj, w(i+1)=[rad(i+2)-rad(i+1)] / Tj … w(i+pn3)=[rad(i+pn3+1)-rad(i+pn3)] / Tj, taking w(i), w(i+1), … w(i+pn3) as the input, and w(i+pn3+k+1) as the output, training the k+1-order speed prediction model, and adjusting pn3 according to the prediction effect; obtaining the current speed w(n-k-1) according to the current rotor position data rad(n-k), inputting the current speed and the previous pn3 rotor position data into the k+1-order speed prediction model to obtain w(n).

5. The method of claim 4, wherein the feedback analysis is performed by a feedback analysis unit of the motor inverter system. In step S14, determining the rotor position based on the rotor position information of the two heterogeneous redundant modes further includes the following steps: Judging whether the heterogeneous redundant mode is abnormal by the following way: When m2 of heterogeneous redundancy mode 1 is less than or equal to At time Tc, the rotational speed is calculated based on R1(n) and R1(n-1), and the calculation error is compared with the output of the K+1 order rotational speed prediction model. If the error is less than a threshold, the data of heterogeneous redundancy mode 1 is not abnormal; otherwise, the data of heterogeneous redundancy mode 1 is abnormal. R1(n) is the rotor position obtained by heterogeneous redundancy mode 1 after compensating the current rotor position information. When m2 of heterogeneous redundancy mode 1 is greater than... At time Tc, the rotational speed is calculated based on R1(n+1) and R1(n). The calculation error between this error and the output of the K+1 order rotational speed prediction model is considered. If the error is less than a threshold, the data of heterogeneous redundancy mode 1 is not abnormal; otherwise, the data of heterogeneous redundancy mode 1 is abnormal. R1(n+1) is the rotor position obtained by heterogeneous redundancy mode 1 after compensating for the rotor position information at the next sampling time. The judgment method for heterogeneous redundancy mode 2 is the same as that for heterogeneous redundancy mode 1. The position information R of the final rotor is obtained by the following formula, , and is the weight, R1 and R2 are the rotor positions obtained after compensation according to the current rotor position information of the isomerous redundant mode 1 and the isomerous redundant mode 2, and the weight is adjusted according to whether there is an abnormality in the data of the isomerous redundant mode. If the isomerous redundant mode has an abnormality, the weight is reduced.

6. A detection feedback analysis system for an inverter system of an electric machine, for executing the detection feedback analysis method of any one of claims 1 to 5, characterized by Including: A torque current controller, a speed measurement module, a data acquisition module, and a data analysis module; The output end of the data acquisition module is connected with the input end of the control module, for obtaining the torque current information of the motor; the output end of the rotating speed measurement module is connected with the input end of the data analysis module, for measuring the position of the motor rotor; the output end of the data analysis module is connected with the input end of the torque current controller, for transmitting the rotor position information after compensation to the torque current controller based on the rotor position output by the rotating speed measurement module; The torque current controller generates a control signal to adjust the torque current according to the received rotor position information, to maintain the normal operation of the motor.

7. The detection feedback analysis system of an inverter system of an electric machine according to claim 6, characterized by, The rotating speed measurement module comprises a heterogeneous redundancy mode 1 and a heterogeneous redundancy mode 2; the heterogeneous redundancy mode 1 samples the excitation and the resolver feedback differential signal by an SDADC, shapes the resolver feedback sampling value based on the synchronous excitation symbol output by an excitation phase synchronizer in an input amplitude shaper / integrator, integrates the resolver feedback sampling value at a plurality of sampling points in an excitation period to obtain envelope lines Sin and Cos, obtains a preliminary angle by Atan2 after correction of Sin and Cos, and inputs the preliminary angle as a reference into an observer / integrator to output an angle; The heterogeneous redundancy mode 2 samples the excitation and the resolver feedback differential signal by an SARADC, inputs the resolver feedback differential signal into an amplitude peak-valley selector, obtains the resolver feedback ADC value at the synchronous excitation peak and valley based on the synchronous excitation zero-crossing output by the excitation phase synchronizer, and obtains an angle by Atan2 after correction of the resolver feedback peak-valley ADC value.

8. The detection feedback analysis system of an inverter system of an electric machine according to claim 6, characterized by, The data analysis module further comprises a transmission time matching unit, a rotating speed calculation unit, a rotor position compensation unit, a rotating speed prediction unit, an abnormality detection unit and a rotor position determination unit; the transmission time matching unit is used to determine the time for the heterogeneous redundancy mode 1 and the heterogeneous redundancy mode 2 to receive the resolver signal to the time for the torque current controller to receive the rotor position information; The rotating speed calculation unit calculates the rotating speed according to the rotor position information received by the torque current controller; The rotor position compensation unit compensates the rotor position information transmitted by the heterogeneous redundancy mode 1 and the heterogeneous redundancy mode 2 according to the sampling period of the heterogeneous redundancy mode 1 and the heterogeneous redundancy mode 2, the sampling period of the torque current controller and the transmission time; the rotating speed prediction unit predicts the rotating speed according to the rotating speed calculated by the torque current controller; The abnormality detection unit is used to judge whether the data of the heterogeneous redundancy mode 1 and the heterogeneous redundancy mode 2 is abnormal, and adjusts the weight of the heterogeneous redundancy mode 1 and the heterogeneous redundancy mode 2 according to the judgment result; The rotor position determination unit obtains the final rotor position information according to the rotor position information transmitted by the heterogeneous redundancy mode 1 and the heterogeneous redundancy mode 2.

9. The detection feedback analysis system of an inverter system of an electric machine according to claim 6, characterized by, The transfer time matching unit acquires the input signal and the output signal of the heterogeneous redundant mode 1 in the time domain, the horizontal coordinate is time, and the vertical coordinate is input or output, matches the output signal with the input signal of the heterogeneous redundant mode 1, and converts the time of the input signal in the time domain into the time of the corresponding input signal; The rotor position signal received by the torque current controller in the time domain is acquired, the rotor position signal received by the torque current controller is translated to the left, the root mean square error sum of the rotor position signal received by the torque current controller and the matched output signal of the heterogeneous redundant mode 1 is calculated, and when the root mean square error sum reaches a minimum value, the time of the left translation is the transfer time of the heterogeneous redundant mode 1; the transfer time of the heterogeneous redundant mode 2 is determined in the same way.

Images