A resolver software decoding method and system considering measurement bias error

Abstract

The application relates to a resolver software decoding method and system considering measurement bias error, relates to the technical field of motor control, and solves the problem that the rotor angle frequency and position are fluctuated due to the bias of the resolver output signal in the existing resolver software demodulation technology. The method comprises the following steps: obtaining a rotor position estimation error signal considering the measurement bias error, multiplying the rotor position estimation error signal with an excitation signal, and passing through a low-pass filter to obtain a bias error signal caused by the measurement bias error; inputting the bias error signal into a second-order generalized integrator to obtain two orthogonal bias error signals; transforming the two orthogonal bias error signals into a dq axis coordinate system through coordinate transformation, and converting the two orthogonal bias error signals into a direct current bias error signal; estimating the direct current bias error signal through an integral controller to obtain an estimated measurement bias error signal, and compensating the estimated measurement bias error signal into the resolver output signal to complete decoding. The application is applied to the fields of equipment manufacturing, military industry and spaceflight.

Description

Technical Field

[0001] The present invention relates to the technical field of motor control, and in particular to a software decoding method for a rotary transformer taking measurement bias error into consideration. Background Art

[0002] Permanent magnet synchronous motors (PMSMs) are widely used in various fields of AC motor speed control systems due to their simple structure, high reliability, and high power density and efficiency. With the continuous development of high-performance industrial control applications such as equipment manufacturing, military and aerospace, the requirements for PMSM drive systems are becoming increasingly stringent. Accurate rotor position and speed are crucial for advanced motor control. Among the many position sensors currently available, resolvers are widely used in PMSM drives due to their inherent advantages, such as ease of installation, strong anti-interference capabilities, and wide temperature range. Rotor position is typically determined through a resolver-to-digital conversion process that demodulates the resolver's output signal. While using hardware chips to perform this process offers a simple approach, it increases the cost of industrial applications. Therefore, algorithm-based software decoding technology is currently a research focus. These technologies can be integrated with motor control algorithms in processors, enabling low-cost hardware implementation of rotor position and speed demodulation. However, resolver-to-digital conversion systems are subject to numerous non-idealities, such as excitation distortion, amplitude imbalance, sampling offset error, reference phase shift, and imperfect quadrature. These interferences can reduce the accuracy of the estimated rotor position, resulting in sinusoidal ripple in the estimated rotor position and speed.

[0003] High-performance control applications require resolver software demodulation to be simple, light on the processor computational burden, easy to implement, and to minimize the impact of system non-idealities on the estimated rotor speed and position. However, traditional resolver software demodulation algorithms have the following problems:

[0004] Due to the influence of the bias circuit and sampling circuit, the output signal of the resolver's secondary winding contains measurement bias error. The sinusoidal disturbance caused by the measurement bias error is inevitably reflected in the measured rotor angular frequency and position through the PI controller, causing the rotor angular frequency and position to fluctuate at the same frequency. When the measured rotor angular velocity and position are applied to a permanent magnet synchronous motor drive system, this will affect the motor's control performance, making it difficult to achieve high-precision control of the permanent magnet synchronous motor. Summary of the Invention

[0005] Aiming at the problem that the output signal of the rotary transformer is biased in the existing software-based demodulation technology of the rotary transformer, which causes fluctuations in the rotor angular frequency and position, the present invention proposes a rotary transformer software decoding method that takes measurement bias errors into consideration.

[0006] The present invention provides a software decoding method for a rotary transformer taking into account a measurement bias error, which uses a permanent magnet synchronous motor to obtain a rotor position signal. The method includes:

[0007] Step S1: According to the theoretical derivation process of the resolver software decoding strategy, the rotor position estimation error signal e is obtained taking into account the measurement bias error. θ ;

[0008] Step S2: Estimate the error signal e according to the rotor position θ Multiplying with the excitation signal and passing through a low-pass filter, the bias error signal caused by the measurement bias error is obtained;

[0009] Step S3: Inputting the bias error signal into a second-order generalized integrator to generate a 90° phase shift, thereby obtaining two orthogonal bias error signals;

[0010] Step S4: transforming the two orthogonal bias error signals into the dq axis coordinate system by using coordinate transformation, and converting the AC bias error signal into a DC bias error signal;

[0011] Step S5: using an integral controller to estimate the DC bias error signal, obtaining an estimated measurement bias error signal, and compensating the estimated measurement bias error signal to the resolver output signal to complete decoding.

[0012] Furthermore, a preferred embodiment is proposed, wherein step S1 includes:

[0013]

[0014] Among them, LPF[] is a low-pass filter for the signal in the brackets, U exc The excitation signal generator provides the excitation signal to the primary winding, e is the error signal, K R is the ratio of the resolver, U E is the excitation signal amplitude, θ m is the actual position of the rotor, To estimate the rotor position signal, e O is the error signal caused by measurement bias error, ω e is the excitation signal angular frequency.

[0015] Furthermore, a preferred method is proposed, wherein the error signal e is calculated based on the actual output signal of the secondary winding and the estimated rotor position signal The sine and cosine values ​​of are multiplied and then the difference is obtained.

[0016] Furthermore, a preferred embodiment is proposed, wherein step S2 includes:

[0017] When the estimated rotor position is consistent with the actual position, Then the position error signal e θ for:

[0018] e θ ≈LPF[e O U E sin(ω e t)]

[0019] The position error signal and the excitation signal U exc Multiply and demodulate again through the filter to obtain the error signal e caused by the measurement offset error O :

[0020]

[0021] Among them, O sin and O cos Indicates the measured offset error of the two outputs of the resolver secondary winding.

[0022] Furthermore, a preferred embodiment is proposed, in which the second-order generalized integrator in step S3 is:

[0023]

[0024] Among them, G α (s) is the output signal e α and the input signal e O The transfer function between β (s) is the output signal e β and the input signal e O The transfer function between α and e β are the two output signals of the second-order generalized integrator, e O is the input signal of the second-order generalized integrator, K S is the control gain of the second-order generalized integrator, ω m is the mechanical angular velocity of the rotor, and s is a frequency domain variable.

[0025] Furthermore, a preferred embodiment is proposed, wherein step S4 includes:

[0026] The two orthogonal signals obtained by the second-order generalized integrator are transformed from the two-term stationary coordinate system to the two-phase rotating coordinate system using coordinate transformation to obtain the DC signal:

[0027]

[0028] Among them, e d and e q Represents the DC error signal obtained after coordinate transformation.

[0029] Further, it is proposed that the step S5 comprises:

[0030] The DC error signal is regulated by an integral controller, and the output of the regulating signal is the estimated measurement bias error signal compensation term:

[0031]

[0032] Wherein, And represents the estimated value of the measurement bias error of the two-way output of the secondary winding of the resolver, K si And K ci represent the gain coefficients of the two integral controllers.

[0033] Based on the same inventive concept, the present application also proposes a resolver software decoding system considering measurement bias error, which comprises:

[0034] A rotor position estimation error signal acquisition unit is configured to obtain a rotor position estimation error signal e θ considering measurement bias error according to a theoretical derivation process of a resolver software decoding strategy.

[0035] A bias error signal acquisition unit is configured to obtain a bias error signal caused by measurement bias error by multiplying the rotor position estimation error signal e θ with the excitation signal and passing through a low-pass filter.

[0036] A second-order generalized integrator processing unit is configured to input the bias error signal to a second-order generalized integrator to generate a 90° phase shift and obtain two orthogonal bias error signals.

[0037] A transformation unit is configured to transform the two orthogonal bias error signals into dq-axis coordinate system by coordinate transformation, and convert the alternating current bias error signal into a direct current bias error signal.

[0038] A compensation unit is configured to estimate the direct current bias error signal by an integral controller, obtain an estimated measurement bias error signal, and compensate the estimated measurement bias error signal into the resolver output signal to complete decoding.

[0039] Based on the same inventive concept, the present application also proposes a computer device comprising a memory and a processor, wherein the memory stores a computer program, and when the processor runs the computer program stored in the memory, the processor executes a resolver software decoding method considering measurement bias error according to any one of the above methods.

[0040] Based on the same inventive concept, the application further provides a computer readable storage medium, which stores a computer program, and the computer program, when executed by a processor, performs the steps of the resolver software decoding method considering measurement bias error according to any one of the above.

[0041] The application has the advantages that:

[0042] The resolver software decoding method considering measurement bias error can significantly improve the decoding accuracy of the rotor position and reduce the error caused by signal bias by considering the measurement bias error. Furthermore, the method has strong adaptability and can be applied to different types of resolvers, especially those working in a high dynamic range and susceptible to bias. The real-time bias error compensation technology can dynamically adjust the output signal, making the decoding process more stable. The low-pass filter and the second-order generalized integrator ensure noise suppression during signal processing, which helps to improve the anti-interference ability of the overall system. The optimization of the mathematical model and algorithm reduces the complexity of the entire decoding process, facilitating implementation and maintenance.

[0043] The resolver software decoding method considering measurement bias error can significantly improve the decoding accuracy of the rotor position and reduce the error caused by signal bias by considering the measurement bias error. Furthermore, the method has strong adaptability and can be applied to different types of resolvers, especially those working in a high dynamic range and susceptible to bias. The real-time bias error compensation technology can dynamically adjust the output signal, making the decoding process more stable. The low-pass filter and the second-order generalized integrator ensure noise suppression during signal processing, which helps to improve the anti-interference ability of the overall system. The optimization of the mathematical model and algorithm reduces the complexity of the entire decoding process, facilitating implementation and maintenance.

[0044] Compared with the traditional resolver software decoding method, the method of the application can effectively compensate for bias error and suppress the sinusoidal fluctuations in the motor speed and rotor position error signal. For example, when the motor speed is the rated speed of 3000 r / min, the motor angular velocity fluctuation is reduced from 8 rad / s to 3 rad / s, and the rotor position error fluctuation is reduced from 3° to 1°. When the motor speed is 50% of the rated speed of 1500 r / min, the motor angular velocity fluctuation is reduced from 6 rad / s to 3 rad / s, and the rotor position error fluctuation is reduced from 3° to 1°.

[0045] The application is applied to industrial control fields such as equipment manufacturing, military aerospace, etc. BRIEF DESCRIPTION OF DRAWINGS

[0046] Figure 1This is a structural block diagram of a resolver software decoding method considering measurement bias error according to an eleventh embodiment;

[0047] Figure 2 This is a structural block diagram of the measurement bias error adaptive compensation method according to the eleventh embodiment;

[0048] Figure 3 This is a flow chart of the measurement bias error adaptive compensation method according to the eleventh embodiment;

[0049] Figure 4 1 is a structural block diagram of the second-order generalized integrator according to the eleventh embodiment;

[0050] Figure 5 1. The Bode plot of the closed-loop transfer function of the second-order generalized integrator according to the eleventh embodiment, wherein FIG (a) is a logarithmic amplitude-frequency characteristic curve of the closed-loop transfer function of the second-order generalized integrator, and FIG (b) is a logarithmic phase-frequency characteristic curve of the closed-loop transfer function of the second-order generalized integrator;

[0051] Figure 6 The experimental results of the resolver software decoding system based on the measurement bias error compensation method according to the eleventh embodiment at a rotation speed of 3000 r / min are shown;

[0052] Figure 7 Figure 11 shows the FFT analysis results of the motor speed waveform at a rated speed of 3000 r / min according to the eleventh embodiment. Figure (a) shows the FFT analysis result without the proposed method, and Figure (b) shows the FFT analysis result with the proposed method. The horizontal axis represents frequency, and the vertical axis represents amplitude.

[0053] Figure 8 The experimental results of the resolver software decoding system based on the measurement bias error compensation method described in the eleventh embodiment at a speed of 1500 r / min;

[0054] Figure 9 This is the FFT analysis result of the motor speed waveform at the rated speed of 1500r / min as described in the eleventh embodiment, wherein Figure (a) is the FFT analysis result without using the proposed method, and Figure (b) is the FFT analysis result using the proposed method. DETAILED DESCRIPTION

[0055] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in combination with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments.

[0056] Embodiment one, the method for decoding resolver considering measurement bias error, the method comprises:

[0057] Step S1: according to the theoretical derivation process of resolver software decoding strategy, the rotor position estimation error signal e θ considering measurement bias error is obtained

[0058] Step S2: according to the rotor position estimation error signal e θ , multiplied by the excitation signal, through low pass filter to obtain the bias error signal caused by measurement bias error;

[0059] Step S3: the bias error signal is input to the second order generalized integrator to produce 90° phase shift, and two orthogonal bias error signals are obtained;

[0060] Step S4: the two orthogonal bias error signals are transformed into dq axis coordinate system by coordinate transformation, and the alternating current bias error signal is converted into direct current bias error signal;

[0061] Step S5: the integral controller is used to estimate the direct current bias error signal, the estimated measurement bias error signal is obtained, and the estimated measurement bias error signal is compensated to the resolver output signal, and the decoding is completed.

[0062] The method for decoding resolver considering measurement bias error provided by the embodiment can significantly improve the decoding accuracy of rotor position and reduce the error caused by signal bias by considering measurement bias error. Further, the method has strong adaptability and can be applied to different types of resolvers, especially those working in high dynamic range and susceptible to bias. The real-time bias error compensation technology can dynamically adjust the output signal, making the decoding process more stable. The low pass filter and the second order generalized integrator ensure noise suppression in the signal processing process, which helps to improve the anti-interference ability of the whole system. Through optimization of mathematical model and algorithm, the complexity of the whole decoding process is reduced, which is convenient for implementation and maintenance.

[0063] The principle of the method provided by the embodiment is as follows:

[0064] In step S1, the rotor position estimation error signal e θ obtained by theoretical derivation reflects the difference between the actual state of rotor position and the expected value, and further provides a basis for subsequent bias error processing.

[0065] In step S2, the error signal is multiplied by the excitation signal, and the bias error signal caused by measurement bias error is extracted by using low pass filter. This process helps to effectively distinguish the useful components and noise in the signal.

[0066] In step S3, the 90° phase shift is generated by the second-order generalized integrator, and two orthogonal bias error signals can be obtained. This is a key step in the decoding process, which makes the subsequent coordinate transformation more accurate.

[0067] In step S4, the orthogonal bias error signals are transformed into the dq axis coordinate system, which can convert the alternating current signal into direct current signal, simplifying the error processing process. The dq axis transformation is widely used in electrical engineering, effectively improving the intuitiveness of signal processing.

[0068] In step S5, the integral controller is used to estimate the direct current bias error signal, so that a more accurate measurement bias error signal can be obtained. By compensating this signal to the output signal of the resolver, the decoding process is completed, making the final output signal more accurate and reliable.

[0069] Embodiment two, this embodiment is a further limitation of the resolver software decoding method considering the measurement bias error according to embodiment one, and the step S1 comprises:

[0070]

[0071] Wherein, LPF[] is the low-pass filter of the signal in the bracket, U exc is the excitation signal generator providing the excitation signal to the primary winding, e is the error signal, K R is the resolver ratio, U E is the excitation signal amplitude, θ m is the actual position of the rotor, is the estimated rotor position signal, e O is the error signal caused by the measurement bias error, ω e is the angular frequency of the excitation signal.

[0072] The specific process of step S1 in this embodiment is:

[0073] The excitation signal generator provides the excitation signal U exc to the primary winding, and the expression is shown in formula (1):

[0074] U exc = U E sin(ω e t) (1)

[0075] In the formula: U E is the excitation signal amplitude, ω e is the angular frequency of the excitation signal.

[0076] The output signal of the resolver secondary winding can be represented by formula (2):

[0077]

[0078] Where: U sin and U cos Represents the output signal of the secondary winding, K R is the resolver ratio, θ m is the actual position of the rotor.

[0079] However, if the influence of the bias circuit and sampling circuit is taken into account, the actual secondary winding output signal will contain a measurement offset error, which can be expressed by formula (3):

[0080]

[0081] Where: O sin and O cos They represent the measurement offset errors of the two outputs of the resolver secondary winding.

[0082] The actual output signal of the secondary winding is compared with the estimated rotor position signal The error signal e can be obtained by multiplying the sine and cosine values ​​of and taking the difference, as shown in formula (4):

[0083]

[0084] Where: is the position measured by the rotor, e O represents the error signal caused by the measurement bias error, which can be expressed by formula (5):

[0085]

[0086] According to formula (4), the high-frequency excitation signal is superimposed on the error signal e. In order to filter out the high-frequency excitation signal, the error signal e in formula (4) is demodulated and the error signal e is combined with the excitation signal U exc After multiplication and low-pass filtering, the rotor position error signal e can be obtained. θ As shown in formula (6):

[0087]

[0088] Where: LPF[] is defined as low-pass filtering of the signal in the brackets.

[0089] Assumption: The cutoff frequency of the low-pass filter is greater than the rotation frequency of the resolver. This means that the low-pass filter does not cause a phase delay in the measured offset error signal.

[0090] If the bias error signal e is ignored O , rotor position error signal e θIt can be approximated by Taylor's formula as shown in formula (7):

[0091]

[0092] Finally, the PI controller is used to adjust the error signal in formula (7) so that it converges to 0, as shown in formula (8):

[0093]

[0094] Where: u m is the controller output, K p and K i is the control gain of the PI controller, and s is the frequency domain variable.

[0095] According to formula (8), the angular velocity and position of the rotor can be measured as shown in formula (9):

[0096]

[0097] Where: ω m is the rotor angular velocity, is the rotor position.

[0098] In this implementation, by introducing parameters such as the excitation signal, actual position, and estimated position in step S1, and processing the signal with a low-pass filter, the error caused by measurement bias can be more accurately estimated. This enables the decoder to better cope with the nonlinear effects caused by bias error. The consideration of the amplitude and angular frequency of the excitation signal in step S1 means that the decoding method can dynamically adjust based on the characteristics of the excitation signal. This feature improves the system's responsiveness to environmental changes and signal fluctuations.

[0099] Furthermore, the signal filtering process using a low-pass filter (LPF) in this embodiment effectively suppresses high-frequency noise and reduces interference with the error estimate, thereby improving the reliability and accuracy of decoding. Because this method can dynamically adjust the error estimate based on different excitation signals and actual rotor position, it can maintain good decoding performance in different application scenarios (such as different operating frequencies and load conditions). By combining the excitation signal with the error signal and processing it through a low-pass filter, a more optimized signal processing flow is formed. This method not only improves the accuracy of the error estimate but also simplifies the complexity of subsequent steps.

[0100] Implementation method 3: This implementation method is a further limitation of the software decoding method of the rotary transformer considering the measurement bias error described in implementation method 2. The error signal e is calculated based on the actual output signal of the secondary winding and the estimated rotor position signal. The sine and cosine values ​​of are multiplied and then the difference is obtained.

[0101] Implementation 4: This implementation further limits the resolver software decoding method considering measurement bias error described in Implementation 2. Step S2 includes:

[0102] When the estimated rotor position is consistent with the actual position, Then the position error signal e θ for:

[0103] e θ ≈LPF[e O U E sin(ω e t)]

[0104] The position error signal and the excitation signal U exc Multiply and demodulate again through the filter to obtain the error signal e caused by the measurement offset error O :

[0105]

[0106] Among them, O sin and O cos Indicates the measured offset error of the two outputs of the resolver secondary winding.

[0107] In this embodiment, the measurement accuracy of the rotary transformer can be significantly improved by estimating the rotor position and compensating for the measurement bias error. Embodiment 2 effectively eliminates or reduces the error caused by the measurement bias by calculating the position error signal and multiplying it with the excitation signal. This method can quickly respond to dynamically changing input signals by estimating the rotor position and calculating the error signal in real time. Through low-pass filter processing, high-frequency noise can be effectively filtered out, maintaining the rotary transformer's sensitivity to low-frequency signal changes and improving its dynamic response capability. During implementation, the signal processed by the low-pass filter can reduce the impact of external noise and interference, making the decoding process more stable and reliable. Reducing the impact of external factors on measurement results is particularly important in complex industrial environments. This decoding method can dynamically adjust the measurement bias compensation strategy based on changes in the actual working environment. This adaptability enables the rotary transformer to maintain efficient operation under different working conditions and meet various application requirements.

[0108] Implementation 5: This implementation further limits the resolver software decoding method considering the measurement bias error described in Implementation 1. The second-order generalized integrator in step S3 is:

[0109]

[0110] Among them, G α(s) is the output signal e α and the input signal e O The transfer function between β (s) is the output signal e β and the input signal e O The transfer function between α and e β are the two output signals of the second-order generalized integrator, e O is the input signal of the second-order generalized integrator, K S is the control gain of the second-order generalized integrator, ω m is the mechanical angular velocity of the rotor, and s is a frequency domain variable.

[0111] The second-order generalized integrator in this embodiment can effectively process and eliminate bias errors in the signal, which is crucial in measuring resolvers. By precisely controlling the control gain, the output signal can be adjusted and optimized in real time, thereby improving overall measurement accuracy.

[0112] Implementation 6: This implementation further limits the resolver software decoding method considering measurement bias error described in Implementation 5. Step S4 includes:

[0113] The two orthogonal signals obtained by the second-order generalized integrator are transformed from the two-term stationary coordinate system to the two-phase rotating coordinate system using coordinate transformation to obtain the DC signal:

[0114]

[0115] Among them, e d and e q Represents the DC error signal obtained after coordinate transformation.

[0116] In this embodiment, by using coordinate transformation to convert the AC bias error signal into a DC bias error signal, the error caused by signal changes can be effectively eliminated. After converting to the dq axis coordinate system, the system can better identify and compensate for the bias error, thereby improving the accuracy of the measurement. In the dq axis coordinate system, the signal usually appears in the form of DC, which makes the subsequent signal processing and decoding algorithms simpler. Processing DC signals is much easier than processing AC signals because DC signals have no frequency changes, which can reduce computational complexity. This method converts the orthogonal signal into a DC signal, so that the error compensation algorithm can be applied more effectively.

[0117] Implementation 7: This implementation further limits the resolver software decoding method considering measurement bias error described in Implementation 6. Step S5 includes:

[0118] The DC error signal is regulated by an integral controller, and the output regulated signal is the estimated measurement bias error signal compensation term:

[0119]

[0120] in, and They represent the estimated values ​​of the measurement offset errors of the two outputs of the resolver secondary winding, K si and K ci Represent the gain coefficients of the two integral controllers respectively.

[0121] In this embodiment, by using an integral controller to estimate the DC bias error, the decoding error caused by the measurement bias can be effectively reduced, thereby improving the accuracy of the output signal. This compensation mechanism can correct the deviation caused by factors such as system characteristics, environmental changes or device aging in real time, making the decoding result more reliable. The integral controller can adjust the continuous bias error, and its output signal can respond quickly to the error change. Compared with other compensation methods, the integral controller shows better stability and dynamic response capabilities when dealing with slow changes or steady-state errors. Integrating the compensation logic of the bias error directly into the software decoding process can simplify hardware design and implementation and reduce costs.

[0122] Embodiment 8: A resolver software decoding system considering measurement bias error according to this embodiment includes:

[0123] The rotor position estimation error signal acquisition unit is used to obtain the rotor position estimation error signal e considering the measurement bias error according to the theoretical derivation process of the resolver software decoding strategy. θ ;

[0124] Bias error signal acquisition unit, used to estimate the error signal e according to the rotor position θ Multiplying with the excitation signal and passing through a low-pass filter, the bias error signal caused by the measurement bias error is obtained;

[0125] a second-order generalized integrator processing unit, configured to input the bias error signal into the second-order generalized integrator to generate a 90° phase shift, thereby obtaining two orthogonal bias error signals;

[0126] A transformation unit, configured to transform the two orthogonal bias error signals into a dq-axis coordinate system by using coordinate transformation, and to transform the AC bias error signal into a DC bias error signal;

[0127] The compensation unit is used to estimate the DC bias error signal using an integral controller, obtain an estimated measurement bias error signal, and compensate the estimated measurement bias error signal to the resolver output signal to complete decoding.

[0128] Implementation method 9. A computer device described in this implementation method includes a memory and a processor, wherein a computer program is stored in the memory. When the processor runs the computer program stored in the memory, the processor executes a rotary transformer software decoding method considering measurement bias error according to any one of implementation methods 1 to 7.

[0129] Implementation method 10. A computer-readable storage medium according to this implementation method stores a computer program. When the computer program is executed by a processor, the steps of the software decoding method for a rotary transformer considering measurement bias error as described in any one of implementation methods 1 to 7 are executed.

[0130] Implementation Method 11: Combination Figures 1 to 7 This embodiment provides a specific example of the software decoding method for a rotary transformer taking into account the measurement bias error described in the first embodiment, and is also used to explain the second to seventh embodiments. Specifically:

[0131] This embodiment describes a resolver software decoding strategy that takes into account measurement bias error. The method includes the following steps:

[0132] Step 1: According to Figure 1 The resolver software decoding system block diagram shown in Figure 1 obtains the rotor position estimation error signal e considering the measurement bias error. θ .

[0133] Step 2: Use the rotor position estimation error signal e obtained in step 1 taking into account the measurement bias error θ , multiply it with the excitation signal and pass it through a low-pass filter to obtain the bias error signal e caused by the measurement bias error O .

[0134] For the rotor position estimation error signal e considering the measurement bias error θ , which is of the form:

[0135]

[0136] When the estimated rotor position is consistent with the actual position, it can be assumed that Then we can get the error signal e θ In the following form:

[0137] e θ ≈LPF[e O U E sin(ω e t)]

[0138] Multiplying it with the excitation signal and demodulating it again through the filter, we can get the error signal e caused by the measurement offset error. O :

[0139]

[0140] Step 3: Convert the error signal e O The input to the second-order generalized integrator produces a 90° phase shift, and two orthogonal bias error signals e are obtained. α 、e β Input signal e O To two output signals e α and e β The transfer functions are:

[0141]

[0142] The following combination Figure 4 Explain and analyze the second-order generalized integrator:

[0143] The second-order generalized integrator can track and phase-shift AC signals of a specific frequency and can be used to separate and filter voltage signals. Figure 3 As shown, it starts from the input signal e O To two output signals e α and e β The transfer functions are:

[0144]

[0145] Where: K S is the control gain of the second-order generalized integrator.

[0146] According to formula (21), at frequency ω m At , the amplitude and phase angle of the closed-loop frequency characteristics of the second-order generalized integrator are:

[0147]

[0148] According to formula (22), when the frequency of the input signal is the same as the set frequency, the output signal e of the second-order generalized integrator is α The output signal will be the same as the input signal, e β Compared to the input signal, the amplitude remains unchanged, but there is a 90° phase delay. Therefore, when the input signal is of frequency ω m The sinusoidal signal sin(ω m t), the output signal of the second-order generalized integrator is:

[0149]

[0150] Step 4: Measuring the Sinusoidal Fluctuation Caused by Bias Error Even after passing through a second-order generalized integrator, it is still difficult to directly estimate the bias error. Because its output is a pair of orthogonal signals with a 90° phase difference, this pair of signals can be considered to be located in a two-phase stationary coordinate system. Based on this, coordinate transformation can be used to transform this pair of orthogonal signals into a two-phase rotating coordinate system, resulting in a pair of DC error signals. This is specifically implemented as follows:

[0151]

[0152] According to step 3, the DC error signal e d and e q for:

[0153]

[0154] Step 5: According to the DC error signal e in step 4 d 、e q It can be seen that the measurement bias error will cause a DC deviation on the dq axis of the two-phase rotating coordinate system. The deviation on the d axis is caused by the output U sin The deviation on the q axis is caused by the bias error on the channel. cos Based on this, we can dq The DC deviation on the axis is equivalent to the measurement bias error on the two output channels of the resolver. Based on this, the DC error signal is adjusted using an integral controller to obtain the estimated measurement bias error compensation term as follows:

[0155]

[0156] Where: and They represent the estimated values ​​of the measurement bias errors of the two outputs of the resolver, K si and K ci Represent the gain coefficients of the two integral controllers respectively.

[0157] Finally, the estimated measurement deviation value is compensated to the output signal of the secondary winding of the resolver, which is:

[0158]

[0159] Where: They are the compensated resolver secondary winding output signals respectively.

[0160] By compensating for the measurement bias error, the rotor estimated position error e can be effectively reduced. θ The sinusoidal disturbance in the rotor position and speed signals can be reduced, thereby improving the control performance of the motor.

[0161] Experimental results:

[0162] according to Figure 6 When the motor speed is 3000r / min, the proposed method can reduce the angular velocity pulsation from 8rad / s to 3rad / s, and the rotor position error pulsation from 3° to 1°. Figure 7 , the amplitude of the first harmonic of the motor speed is reduced from 2.2% to 0.35%. Figure 8 When the motor speed is 50% of the rated speed (1500r / min), the proposed method can reduce the angular velocity pulsation from 6rad / s to 3rad / s, and the rotor position error pulsation from 3° to 1°. Figure 9 , the amplitude of the first harmonic of the motor speed is reduced from 1.31% to 0.35%. Experimental results show that the proposed scheme can effectively compensate for the bias error and suppress the sinusoidal fluctuations in the motor speed and rotor position error signals, thereby enhancing the driving performance of the motor system.

[0163] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Thus, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0164] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems) and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowcharts and / or block diagrams, as well as combinations of processes and / or blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowcharts and / or block diagrams. Figure 1 a process or multiple processes and / or boxes Figure 1 These computer program instructions can also be stored in a computer-readable memory that can guide a computer or other programmable data processing device to work in a specific way, so that the instructions stored in the computer-readable memory produce a product including the instruction device, which implements the function specified in the process. Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0165] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0166] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit its scope of protection. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in the field should understand that after reading the present invention, those skilled in the art may still make various changes, modifications or equivalent substitutions to the specific implementation methods of the invention, but these changes, modifications or equivalent substitutions are all within the scope of protection of the pending claims of the invention.

Claims

1. A software decoding method for a resolver taking into account measurement bias error, characterized in that: The method comprises: Step S1: According to the theoretical derivation process of the resolver software decoding strategy, the rotor position estimation error signal considering the measurement bias error is obtained. e θ ; Step S2: Estimate the error signal based on the rotor position e θ Multiplying with the excitation signal and passing through a low-pass filter, the bias error signal caused by the measurement bias error is obtained; Step S3: Inputting the bias error signal into a second-order generalized integrator to generate a 90° phase shift, thereby obtaining two orthogonal bias error signals; Step S4: transforming the two orthogonal bias error signals into the dq axis coordinate system by using coordinate transformation, and converting the AC bias error signal into a DC bias error signal; Step S5: using an integral controller to estimate the DC bias error signal, obtaining an estimated measurement bias error signal, and compensating the estimated measurement bias error signal to the resolver output signal to complete decoding.

2. The method for software decoding a resolver considering a measurement bias error according to claim 1, wherein: The step S1 comprises: Among them, LPF[] is a low-pass filter for the signal in the brackets. U exc The excitation signal generator provides an excitation signal to the primary winding. e is the error signal, is the resolver ratio, is the excitation signal amplitude, is the actual position of the rotor, To estimate the rotor position signal, e O is the error signal due to measurement offset error, ω e is the excitation signal angular frequency.

3. The method for software decoding a resolver considering a measurement bias error according to claim 2, wherein: The error signal e According to the actual output signal of the secondary winding and the estimated rotor position signal φ m The sine and cosine values ​​of are multiplied and then the difference is obtained.

4. The method for software decoding a resolver considering a measurement bias error according to claim 2, wherein: The step S2 comprises: When the estimated rotor position is consistent with the actual position, φ m ≈ θ m , then the position error signal e θ for: The position error signal and the excitation signal U exc Multiply and demodulate again through the filter to get the error signal caused by the measurement offset error e O : in, O sin and O cos Indicates the measured offset error of the two outputs of the resolver secondary winding.

5. The method for software decoding a resolver considering a measurement bias error according to claim 1, wherein: The second-order generalized integrator in step S3 is: in, Output signal e α and input signal e O The transfer function between Output signal e β and input signal e O The transfer function between e α and e β are the two output signals of the second-order generalized integrator, e O is the input signal of the second-order generalized integrator, K S is the control gain of the second-order generalized integrator, ω m is the mechanical angular velocity of the rotor, s is a frequency domain variable.

6. The method for software decoding a resolver considering measurement bias error according to claim 5, wherein: The step S4 comprises: The two orthogonal signals obtained by the second-order generalized integrator are transformed from the two-phase stationary coordinate system to the two-phase rotating coordinate system using coordinate transformation to obtain the DC signal: in, e d and e q represents the DC error signal obtained after coordinate transformation, To estimate the rotor position signal.

7. The method for software decoding a resolver considering a measurement bias error according to claim 6, wherein: The step S5 comprises: The DC error signal is regulated by an integral controller, and the output regulated signal is the estimated measurement bias error signal compensation term: in, and represents the estimated value of the measured offset error of the two outputs of the resolver secondary winding, K si and K ci Represents the gain coefficients of the two integral controllers.

8. A resolver software decoding system taking into account measurement bias error, characterized in that: The system comprises: The rotor position estimation error signal acquisition unit is used to obtain the rotor position estimation error signal taking into account the measurement bias error according to the theoretical derivation process of the resolver software decoding strategy. e θ ; Bias error signal acquisition unit, used to estimate the error signal according to the rotor position e θ Multiplying with the excitation signal and passing through a low-pass filter, the bias error signal caused by the measurement bias error is obtained; a second-order generalized integrator processing unit, configured to input the bias error signal into the second-order generalized integrator to generate a 90° phase shift, thereby obtaining two orthogonal bias error signals; A transformation unit, configured to transform the two orthogonal bias error signals into a dq-axis coordinate system by using coordinate transformation, and to transform the AC bias error signal into a DC bias error signal; The compensation unit is used to estimate the DC bias error signal using an integral controller, obtain an estimated measurement bias error signal, and compensate the estimated measurement bias error signal to the resolver output signal to complete decoding.

9. A computer device, characterized in that: The invention comprises a memory and a processor, wherein a computer program is stored in the memory, and when the processor runs the computer program stored in the memory, the processor executes the software decoding method for a rotary transformer considering a measurement bias error according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that The computer-readable storage medium stores a computer program, which, when executed by a processor, executes the steps of the method for software decoding a resolver considering a measurement bias error according to any one of claims 1 to 7.

Images