Motor rotor test method based on rotary transformer

Abstract

The invention belongs to the related technical field of motor control and test, and discloses a motor rotor test method based on a rotary transformer. A winding type rotary transformer and a rotor of a tested motor are coaxially nested and installed, and excitation sine wave signals for driving the rotary transformer are generated by an FPGA (Field Programmable Gate Array), a DAC (Digital-to-Analog Converter) device and an external signal conditioning circuit. When the motor rotor rotates, sine and cosine envelope signals output by the rotary transformer are acquired by using the double-path differential ADC signal acquisition circuit. And finally, performing algorithm processing on the acquired data by the FPGA to obtain the current rotor position of the tested motor, and further processing the data on the basis to obtain the angular velocity of rotor rotation. According to the invention, the position and the angular velocity of the motor rotor are detected, and the detection precision and the robustness are high, so that the detection device can normally detect the motor rotor in a severe environment.

Description

Technical Field

[0001] The present invention belongs to the technical field related to motor control and testing, and relates to a method for testing a motor rotor based on a resolver, so as to detect the angular position and angular velocity of the motor rotor. Background Art

[0002] Under the development trend of artificial intelligence robots and new energy vehicles, AC servo control systems have been widely applied in many high-tech fields, and reliable and accurate position detection and feedback devices have become important technologies in the servo control field.

[0003] With the continuous development of society, industrial automation motor control has continuously replaced manual production and manufacturing. For example, radars in the military field, numerically controlled machine tools with high control precision requirements, industrial robot control, textile, printing, and electric vehicle control in the field of traffic control, etc. In a motor servo system, without a reliable and accurate position detection and feedback device, the control system cannot ensure the normal operation of the motor.

[0004] Currently, the research on the angular position and angular velocity of the motor rotor is continuously deepening. The research on motor rotor detection integrates multiple fields such as machinery, computer, and control, and involves various software and hardware technologies such as sensor fusion and waveform decoding, meeting the test requirements of different types of motor rotors, enabling the angular position and angular velocity of the motor rotor to be detected in real time with high precision, and having practical application value for servo motor drive control. Summary of the Invention

[0005] In order to solve the problems existing in the background art, the present invention proposes a method for detecting a motor rotor based on a resolver to achieve real-time detection of the angular position and angular velocity of the motor rotor.

[0006] A method for detecting a motor rotor based on a resolver, characterized by comprising the following steps:

[0007] S1: The resolver mainly includes a wound resolver and a reluctance resolver. Here, the detection sensor uses a wound resolver;

[0008] S2: There are mainly two installation structures for the resolver: external type and embedded type. Here, the application scenario is to conduct general tests on motor rotors. Since not every motor to be tested has a built-in resolver, the motor rotor test sensor uses the external type;

[0009] S3: The resolver consists of a stator and a rotor. The rotor is horizontally nested and installed with the motor drive shaft, and the stator is horizontally coaxially fixedly installed with the rotor;

[0010] S4: Add a high-frequency sine wave signal to the stator winding. During the rotation of the motor, the two rotor windings of the secondary output waveform signals related to the stator position through the principle of electromagnetic induction;

[0011] S5: The output sine and cosine feedback signals are coupled by the excitation signal and the low-frequency signal generated by the rotation of the motor rotor;

[0012] S6: If rotor angle and speed information are to be obtained, it is necessary to demodulate the sine wave feedback signal and the cosine feedback signal to obtain the envelope signals of sine and cosine, and finally decode the envelope information through a certain decoding method;

[0013] S7: The generation of the excitation signal and the acquisition of the feedback signal of the resolver are both realized by discrete devices built by the system;

[0014] S8: The decoding scheme of the entire system adopts software decoding based on discrete devices;

[0015] S9: The amplitude of the output signal changes with the position of the motor rotor. The envelope signal is generated by the high-speed rotation of the motor, and then the peak value of the envelope signal is sampled by the peak sampling method;

[0016] S10: Perform the arctangent function operation on the sine and cosine signal data collected at the same moment to obtain the relative angular position of the rotor at that moment;

[0017] S11: If there is an error between the calculation result and the true value, it is necessary to compensate the system signal loop;

[0018] S12: If the motor rotor rotates at high speed, the method of calling the function calculation result cannot track the rotor angle position information in real time;

[0019] S13: If a high requirement is placed on the maximum tracking speed of the motor under test, it is necessary to use the improved CORDIC algorithm pipeline structure to process the data, which is easy to implement inside the FPGA, has a fast operation speed, and can output the operation result in each clock cycle;

[0020] Further, the stator and rotor of the resolver in S1 are assembled according to the correct drawing requirements.

[0021] Further, the mathematical expression of the sine wave excitation signal is:

[0022] E R1-R2 = Esinωt

[0023] where: E R1-R2 represents the excitation signal, E is the amplitude of the input excitation signal, ω is the angular frequency of the input excitation signal, and t is a certain moment.

[0024] Furthermore, the formulas for the sine and cosine feedback signals are as follows:

[0025]

[0026] Where: E R1-R2 represents the high-frequency sine excitation signal input to the stator-side winding on the primary side, E S1-S2 represents the sine feedback signal induced by the sine winding, E S2-S4 represents the cosine feedback signal induced by the cosine winding, K is the turns ratio between the primary and secondary sides, i.e., the voltage transformation ratio of the transformer, and θ is the rotor position angle.

[0027] Furthermore, the mathematical model of the CORDIC algorithm is as follows:

[0028]

[0029]

[0030] Where: x i and y i represent the coordinates of the i-th iteration in the two-dimensional plane, x i+1 and y i+1 represent the coordinates of the (i + 1)-th iteration in the two-dimensional plane, di refers to the iteration direction. If it is counterclockwise, it is -1, and if it is clockwise, it is +1. z i+1 refers to the sum of all vector angles at the (i + 1)-th moment, z i refers to the sum of all angles at the i-th moment, and θ i represents the angle value rotated at the i-th moment.

[0031] The present invention uses an improved CORDIC algorithm to implement the decoding of the output envelope signal of the resolver. Compared with a dedicated decoding chip, it has good environmental compatibility, fast data processing speed, and significantly reduced running time. This method can significantly improve the execution efficiency of software decoding. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] Figure 1 is the installation schematic diagram of the resolver and the motor of the present invention;

[0033] Figure 2 is the internal schematic diagram and wiring method of the resolver of the present invention;

[0034] Figure 3 is the system working block diagram of the present invention;

[0035] Figure 4 is the peak sampling schematic diagram;

[0036] Figure 5 is the CORDIC algorithm pipeline structure model; DETAILED DESCRIPTION OF THE INVENTION

[0037] To make the above objects, working modes, technical routes, and advantages of the relevant algorithms of the present invention clearer and more understandable, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0038] It should be noted that the following detailed descriptions are all exemplary and are intended to provide further explanations of the present invention.

[0039] Embodiment 1.

[0040] As Figure 1 shown, this embodiment provides a method for detecting the rotor of a motor based on a resolver, including the stator and rotor of the resolver, the permanent magnet synchronous motor body, and the drive harness.

[0041] The stator is used for:

[0042] Receiving an excitation signal, coupling the excitation signal into the stator through electromagnetic induction and voltage ratio transformation to generate a resolver, and then coupling it back to the sine-cosine feedback signal winding of the stator through reverse electromagnetic coupling, and leading it out through the harness.

[0043] The rotor is used for:

[0044] Coaxially and horizontally nested with the motor to be tested to generate a resolver.

[0045] The permanent magnet synchronous motor and the drive harness are used for:

[0046] Introducing the excitation signal and leading out the feedback signal, and driving the motor to rotate through the motor drive line.

[0047] As Figure 2 shown, it is a schematic diagram of the working principle and harness of the resolver.

[0048] Embodiment 2.

[0049] As Figure 3 shown, this embodiment provides a method for testing the rotor of a motor based on a resolver, which includes: the FPGA core part, the DA conversion module, the AD conversion module, and the communication module.

[0050] S1: The FPGA core part acts as the processing brain of the entire system, responsible for generating the excitation signal, decoding the sine-cosine envelope signal, processing and calculating internal data, and communicating with the Ethernet port of the computer, etc.

[0051] S2: The excitation signal of the resolver is first queried from the FPGA sine wave ROM lookup table, then passed through the digital-to-analog converter, low-pass filter, and finally output the required sine wave signal after amplitude adjustment and power amplification.

[0052] S3: Since the FPGA chip can only process digital signals, it is necessary to perform digital-to-analog conversion operations on the two-channel orthogonal analog quantities output by the resolver.

[0053] S4: The angle information is saved in the form of internal data of the FPGA through the acquisition circuit and waits for arithmetic processing.

[0054] S5: Since the classical CORDIC algorithm has a large domain range in the two-dimensional plane coordinates and is not convenient for calculation, it is not applicable to the software decoding of the resolver.

[0055] The improved CORDIC decoding algorithm measures the full angle, and the angle of each iteration is a specific value θ i+1 = d i arctan2 -i , d i represents that the iteration direction is 1 or -1.

[0056] The improved angle superposition formula:

[0057]

[0058] That is, the maximum range of the measured angle is (-π / 2, π / 2).

[0059] S6: If you want to calculate the angle within (-π, π), you need to preprocess the initial angle and postprocess the output angle. The angle preprocessing module is responsible for transforming the points in the entire two-dimensional plane into the range of (-π / 4, π / 4), and at the same time avoiding the measurement of angles near ±π / 2.

[0060] S7: The angle postprocessing module is responsible for restoring the angle processed by the angle preprocessing module and the decoding module to the correct angle and outputting it.

[0061] S8: As Figure 4 shown, an exciting sine wave signal is applied to the input side of the resolver, and the function expression is:

[0062] E R1-R2 = Esinωt

[0063] Where: E R1-R2 represents the excitation signal, E is the amplitude of the input excitation signal, ω is the angular frequency of the input excitation signal, and t is a certain moment.

[0064] S9: As Figure 4 shown, sine and cosine envelope signals will be output at the feedback end of the resolver, and their function expressions are:

[0065]

[0066] Where: E R1-R2The high-frequency sine excitation signal input to the stator-side winding representing the primary side, E S1-S2 The sine feedback signal induced by the sine winding is represented by E S2-S4 The cosine feedback signal induced by the cosine winding is represented by K, which is the turn ratio of the primary and secondary sides, i.e., the voltage transformation ratio of the transformer, and θ is the rotor position angle.

[0067] S10: As Figure 4 shown, the amplitude of the resolver output signal changes with the rotor position. Under high-speed resolvers, the sine and cosine signals output by the stator will form an envelope signal.

[0068] S11: The peak sampling method needs to trigger sampling according to the peak signal of the modulation wave, and the envelope extraction of the resolver output signal is carried out based on the peak sampling method.

[0069] S12: As Figure 5 shown, based on the improved CORDIC algorithm function model:

[0070]

[0071]

[0072] Where: x i and y i represent the coordinates of the i-th iteration in the two-dimensional plane, X i+1 and y i+1 represent the coordinates of the (i + 1)-th iteration in the two-dimensional plane, d i refers to the iteration direction. If it is counterclockwise, it is -1, then if it is clockwise, it is +1, and z i+1 refers to the sum of all vector angles at the (i + 1)-th moment, and z i refers to the sum of all angles at the i-th moment, and θ i represents the angle value rotated at the i-th moment.

[0073] S13: The resolver waveform data and calculation results collected by the lower computer are uploaded to the upper computer for display through Ethernet and high-speed serial ports respectively, and the waveform of the resolver of the motor under test and the rotor angle position and angular velocity information during operation are recorded.

[0074] The above is only the preferred embodiment of the present invention. In addition, for those of ordinary skill in the art, various modifications and variations can be made to the invention while maintaining the principle of the present invention. If the modifications and variations fall within the scope of the claims of the present invention and their equivalent technologies, these modifications and variations should also be regarded as within the protection scope of the present invention.

Claims

1. A method for testing a motor rotor based on a resolver, characterized in that, Including: S1: Resolvers mainly include wound-type resolvers and reluctance resolvers. Here, the detection sensor uses a wound-type resolver; S2: There are mainly two installation structures for resolvers: external type and embedded type. The application scenario here is to conduct general tests on the motor rotor. Since not every motor under test has a built-in resolver, the motor rotor test sensor uses the external type; S3: A resolver consists of a stator and a rotor. The rotor is horizontally nested and installed on the motor drive shaft, and the stator and the rotor are horizontally coaxially fixedly installed; S4: A high-frequency sine wave signal is added to the stator winding. During the rotation of the motor, two rotor windings in the secondary output waveform signals related to the stator position through the principle of electromagnetic induction; S5: The output sine and cosine feedback signals are coupled by the excitation signal and the low-frequency signal generated by the rotation of the motor rotor; S6: If rotor angle and speed information are to be obtained, it is necessary to demodulate the sine wave feedback signal and the cosine feedback signal to obtain the envelope signals of sine and cosine, and finally decode the envelope information through a certain decoding method; S7: The generation of the excitation signal and the acquisition of the feedback signal of the resolver are both realized by discrete devices built by the system; S8: The decoding scheme of the entire system uses software decoding based on discrete devices; S9: The amplitude of the output signal changes with the position of the motor rotor. An envelope signal is generated by the high-speed rotation of the motor, and then the peak value of the envelope signal is sampled using the peak sampling method; S10: Performing an arctangent function operation on the sine and cosine signal data collected at the same moment can obtain the relative angular position of the rotor at that moment; S11: If there is an error between the calculation result and the true value, it is necessary to compensate the system signal loop; S12: If the motor rotor rotates at a high speed, the method of calling the function calculation result cannot track the rotor angle position information in real time; S13: If a high requirement is placed on the maximum tracking speed of the motor under test, it is necessary to use an improved CORDIC algorithm pipeline structure to process the data, which is easy to implement inside the FPGA, has a fast operation speed, and can output the operation result in each clock cycle.

2. The method according to claim 1, wherein The built-in motor in S2 refers to a motor that comes with a resolver sensor internally during production.

3. The method according to claim 1, wherein The mathematical expression of the sine wave excitation signal in S4: E R1-R2 = E sin ωt Where: E R1-R2 represents the excitation signal, E is the amplitude of the input excitation signal, ω is the angular frequency of the input excitation signal, and t is a certain moment.

4. The method according to claim 1, characterized in that, The formula for the output sine and cosine feedback signals is: Where: E R1-R2 represents the high-frequency sinusoidal excitation signal input to the stator-side winding of the primary side, E S1-S2 represents the sinusoidal feedback signal induced by the sinusoidal winding, E S2-S4 represents the cosine feedback signal induced by the cosine winding, K is the turn ratio of the primary and secondary sides, i.e., the voltage transformation ratio of the transformer, and θ is the rotor position angle.

5. The method according to claim 1, wherein The improved CORDIC algorithm is written in hardware description language to obtain the result, while the traditional decoding algorithm directly calls the arctangent function.

6. The method according to claim 5, wherein The decoding scheme is based on the improved CORDIC algorithm. Through multiple iterations and the principle of successive approximation, the current angular position and angular velocity information of the motor rotor are obtained.

7. The method according to claim 6, wherein For the improved CORDIC algorithm, coordinate transformation needs to be performed on the points falling in the second and third quadrants in the two-dimensional rectangular coordinate system to improve the convergence domain and accuracy of the iteration.

Images