Permanent magnet synchronous motor contact terminal high resistance fault real-time detection and non-inductive performance compensation method and system

Abstract

The application discloses a permanent magnet synchronous motor contact terminal high-resistance fault real-time detection and non-inductive performance compensation method and system, and the method comprises the following steps: establishing a motor model with asymmetric three-phase resistance, and theoretically analyzing the influence mechanism of the unbalanced resistance on the current equation and the angle estimation error of the position sensorless control; on this basis, an adaptive estimator independent of additional hardware is designed to online and real-time estimate the three-phase equivalent resistance value and synchronously correct the non-inductive angle error, realize accurate positioning of the fault phase and angle compensation, and the system comprises a three-phase inverter, a permanent magnet synchronous motor, a speed loop controller, a current loop controller, an inverse Park transformation, an inverse Clark transformation, a space vector pulse width modulation, an adaptive estimator, a phase-locked loop and a three-phase resistance reconstruction module. The application can accurately detect the resistance unbalance degree and position the fault phase in the full working condition range without increasing sensors, and simultaneously eliminates the current frequency multiplication fluctuation caused by not considering the unbalanced resistance.

Description

Technical Field

[0001] This invention belongs to the field of permanent magnet synchronous motor control technology, specifically relating to a method and system for real-time detection and sensorless performance compensation of high resistance faults in the contact terminals of permanent magnet synchronous motors. It is particularly suitable for monitoring poor contact terminal performance and improving control performance under minor faults in sensorless operation conditions. Background Technology

[0002] Permanent magnet synchronous motors (PMSMs) possess high power density, high efficiency, and excellent dynamic response, making them widely used in compressors, electric vehicles, and industrial servo applications. However, in practical applications, the contact terminals of the motor input circuit (such as junction boxes and connectors) are prone to poor contact and high-resistance faults due to manufacturing defects, corrosion, thermal cycling, or mechanical vibration. While high-resistance faults often do not trigger shutdown protection in the early stages, they can cause stator current and voltage imbalances and increased torque pulsation. More alarmingly, localized overheating at the contact points can gradually damage the insulation, eventually leading to a serious fault.

[0003] Taking the compressor industry as an example, the commonly used physical detection method is to run the motor for a period of time and then use a thermistor or infrared thermal imager to find the contact point with abnormal temperature, thereby locating the fault. Although this method can accurately locate the fault, it is an offline or post-fault detection method, which cannot provide real-time early warning and intervention in the early stages of a fault, and thus cannot meet the requirements of high-reliability application scenarios.

[0004] Several methods for online detection of high-resistance faults have been proposed by academia and engineering, among which the zero-sequence voltage detection method and the voltage / current harmonic detection method are representative. The former identifies faults by detecting harmonic components related to phase current in the zero-sequence voltage. Therefore, this method requires the additional construction of a three-phase resistor port network to obtain the zero-sequence voltage, increasing hardware costs. Furthermore, the amplitude of the harmonic components is related to the current magnitude, making it difficult to set a uniform threshold under light load conditions. The latter method is based on voltage / current harmonic detection, utilizing the second harmonic components introduced into the dq-axis voltage or current by resistance imbalance for fault identification. Therefore, the detection effectiveness of this type of method heavily depends on the current loop control gain: when the bandwidth is high, the harmonics manifest in the voltage; when the bandwidth is low, the harmonics manifest in the current, resulting in insufficient detection robustness. In addition, both of these methods can only determine whether a fault exists, but cannot accurately identify which specific phase has poor contact, which is detrimental to subsequent accurate maintenance and fault-tolerant control.

[0005] A search revealed Chinese invention patent CN114528870A, which discloses a method for improving the reliability of early-stage inter-turn short-circuit fault diagnosis in permanent magnet synchronous motors (PMSMs). The method is implemented according to the following steps: Step 1: Establish a mathematical model of the inter-turn short-circuit fault in the PMSM; Step 2: Design a voltage disturbance observer to compensate for power supply imbalance caused by the inverter based on the mathematical model of the inter-turn short-circuit fault in the PMSM; Step 3: After compensation, extract the high-frequency current response of the PMSM under high-frequency signal injection using a bandpass filter, and then perform coordinate transformation and low-pass filter extraction to extract fault features. This method solves the problem that fault features are not obvious in the early stage of inter-turn short-circuit faults, making accurate fault detection impossible.

[0006] The technical comparison between this application and the aforementioned patent is as follows:

[0007] 1. The patent "Reliability Improvement Method for Early Inter-turn Short Circuit Fault Diagnosis of Permanent Magnet Synchronous Motors" uses an inter-turn short circuit fault model, describing short circuit faults caused by internal insulation damage in the windings by introducing the short-circuit turns ratio and short-circuit current. This invention, however, targets high-resistance faults at the contact terminals, i.e., increased contact resistance and asymmetrical three-phase resistance at external terminals due to corrosion, thermal cycling, or vibration. The two differ fundamentally in their physical nature of the fault, mathematical model establishment, and equivalent circuit structure, and belong to two different types of electrical faults.

[0008] 2. The technical objective of the patent "Reliability Improvement Method for Early Inter-turn Short Circuit Fault Diagnosis in Permanent Magnet Synchronous Motors" is to amplify fault characteristics through high-frequency signal injection to diagnose early inter-turn short circuit faults, which is a pure fault detection method. In contrast, this invention not only achieves real-time detection and fault phase location of high-resistance faults at contact terminals, but also actively compensates for angle errors and current frequency fluctuations in sensorless control under minor fault conditions, improving the system's sensorless operation performance. This invention combines fault detection and fault-tolerant control functions, and its technical objectives and solutions differ from existing patents.

[0009] A search revealed that Chinese invention patent CN107482989A discloses a phase-loss-tolerant control method for a non-ideal sinusoidal back-EMF permanent magnet synchronous motor. This method involves transforming the three-phase back EMF of the offline-acquired permanent magnet synchronous motor into a coordinate system to obtain the unit back EMF in a synchronous rotating coordinate system. Based on the unit back EMF and the electromagnetic torque formula in the synchronous rotating coordinate system, the reference q-axis current corresponding to the unit torque is derived. When a single-phase armature winding of the motor is disconnected, based on the principle that the electromagnetic torque of the motor remains unchanged under normal and phase-loss conditions, the equivalent relationship between the non-faulty phase compensation current and the zero-axis compensation current reference value is obtained. Zero-axis voltage feedforward and current closed-loop compensation are used to compensate for the torque pulsation caused by the phase-loss fault, thus achieving fault-tolerant control of the non-ideal sinusoidal back-EMF permanent magnet synchronous motor under phase-loss fault conditions.

[0010] The technical comparison between this application and the aforementioned patent is as follows:

[0011] 1. The patent "A Non-Ideal Sinusoidal Back EMF Permanent Magnet Synchronous Motor Phase Loss-Tolerant Fault Control Method" targets phase loss faults in permanent magnet synchronous motors (one phase armature winding is completely disconnected), which is an open circuit fault. The fault phase current is zero, and the current path is completely cut off. In contrast, this invention targets high resistance faults at the contact terminals, which is a resistive fault. Poor terminal contact leads to asymmetrical resistance in the three phases, but the current path is still maintained. The two are fundamentally different in terms of the physical nature and electrical characteristics of the faults.

[0012] 2. The technical objective of the patent "A Phase-Loss Fault-Tolerant Control Method for a Non-Ideal Sinusoidal Back EMF Permanent Magnet Synchronous Motor" is to compensate for torque ripple and maintain motor operation after a phase loss by using four-arm topology reconstruction, zero-axis voltage feedforward, and zero-axis current closed-loop. In contrast, the technical objective of this invention is to detect the degree of contact terminal defects in real time, compensate for angle errors and current frequency fluctuations in sensorless control, and proactively improve sensorless operation performance under minor faults. The former relies on four-arm hardware, offline back EMF data, and position sensors, while the latter requires no additional hardware, performs fully online adaptive estimation, and does not rely on position sensors; their technical paths and implementation methods differ. Summary of the Invention

[0013] To address the aforementioned technical problems, this invention proposes a method and system for real-time detection and sensorless performance compensation of high-resistance faults at the contact terminals of permanent magnet synchronous motors. The method first establishes a motor model with three-phase resistance asymmetry, analyzing how unbalanced resistance affects the current equation and sensorless control. It points out that resistance asymmetry causes DC bias errors and second-harmonic frequency fluctuations in rotor position estimation. Based on this, an adaptive estimator is designed, employing a recursive least squares method to synchronously estimate the three-phase resistance and sensorless angle error online, thereby determining the faulty phase and its severity. The estimated angle error is then fed into a phase-locked loop (PLL), and closed-loop adjustment is used to obtain the compensated rotor position and speed, thus eliminating the angle estimation error caused by unbalanced resistance and improving sensorless operation performance under minor faults. Compared to existing high-resistance fault detection methods, this invention requires no additional hardware sensors, accurately detects the degree of resistance imbalance and locates the faulty phase across the entire operating range, and effectively suppresses current frequency fluctuations through angle compensation, significantly improving the reliability and control accuracy of the sensorless control system.

[0014] To achieve the above objectives, the technical solution adopted by the present invention is as follows:

[0015] A method for real-time detection and sensorless performance compensation of high resistance faults in the contact terminals of permanent magnet synchronous motors includes the following steps:

[0016] S1. Establish a mathematical model of a three-phase resistive asymmetric permanent magnet synchronous motor and derive the voltage equation under unbalanced resistance.

[0017] S2. Analyze the impact of unbalanced resistance on sensorless control, and derive the voltage equation and position error mechanism in the estimated coordinate system.

[0018] S3. Based on the estimated coordinate system voltage equation obtained in S2, transform it into an adaptively estimable linear regression equation and design an adaptive parameter estimator.

[0019] S4. Based on the three-phase resistance deviation and position error estimated in S3, reconstruct the actual three-phase resistance and obtain the compensated rotor position and speed.

[0020] S5. Based on the rotor position and speed estimated by S4, perform closed-loop vector control on the permanent magnet synchronous motor.

[0021] As a preferred technical solution of the present invention, the specific process of step S1 is as follows:

[0022] S11. First, establish a static coordinate system model under the condition of contact fault, i.e., three-phase resistance asymmetry:

[0023] Let the three-phase stator resistances be respectively , , Define the average resistance And the deviation of each phase resistance from the average value. , , ,satisfy ,

[0024] Then, in the stationary abc coordinate system, the voltage equation is:

[0025] ;

[0026] in , , This refers to the three-phase stator voltage; , , This refers to the three-phase stator current; , , For back electromotive force; L s For stator inductance;

[0027] S12. Next, transform to the rotated dq coordinate system and derive the voltage equations containing resistive asymmetry:

[0028] For rotor position Using the equal-amplitude Park transformation matrix Transforming the voltage and current into a rotating dq coordinate system, the resistance matrix becomes:

[0029] ;

[0030] After expansion, the resistance matrix components in the rotated dq coordinate system are obtained:

[0031] ;

[0032] S13. Finally, substituting the average resistance and deviation, and using the property that the sum of the three-phase cosines is zero, we can simplify to get:

[0033] ;

[0034] , and As second harmonic components, they will generate harmonic disturbances in the current loop;

[0035] Therefore, the voltage equation in the synchronously rotating coordinate system dq is:

[0036] ;

[0037] Where j is defined as the imaginary unit of a complex vector, and the complex vector... , , Let ψ be the electric speed of the motor, and ψf be the rotor flux linkage.

[0038] As a preferred technical solution of the present invention, the specific process of step S2 is as follows:

[0039] S21. Let the estimated angular error between the dq coordinate system and the true dq coordinate system be... Transforming the actual voltage equation to the estimated dq coordinate system, we obtain:

[0040] ;

[0041] in To estimate the current in the dq coordinate system, To estimate the voltage in the dq coordinate system, To estimate the rotational speed;

[0042] The resistance matrix in the dq coordinate system is estimated by similarity transformation:

[0043] ;

[0044] S22. Using Pauli matrix decomposition and rotation transformation rules, decompose the asymmetric part of the resistance, and define:

[0045] ;

[0046] in , The Pauli matrix, after similarity transformation in the estimated dq coordinate system, is obtained as follows:

[0047] ;

[0048] Therefore, the resistance asymmetry matrix in the dq coordinate system can be estimated as follows:

[0049] ;

[0050] in:

[0051] ;

[0052] S23. Substituting the above results into the estimated voltage equation in the dq coordinate system, we obtain the complete model considering resistance asymmetry:

[0053] ;

[0054] S24. From the complete model considering resistance asymmetry, it can be seen that resistance asymmetry generates an additional voltage term in the estimated dq coordinate system. Meanwhile, nominal resistance and actual average resistance The difference between them is equivalent to parameter mismatch. Both of these factors together lead to a deviation in the back EMF observation, which in turn causes DC bias error and second harmonic fluctuation error in the rotor position estimated without a position sensor. Since the mechanical time constant of the motor is much larger than the electrical time constant, and the bandwidth of the adaptive estimator is usually higher than the speed change frequency, it can be assumed that within one control cycle, the actual speed... Compared with the estimated speed Approximately equal, that is Ultimately, the voltage equation is transformed into a regression equation that can be adaptively estimated.

[0055] As a preferred technical solution of the present invention, the specific process of step S3 is as follows:

[0056] S31, General Style Expand, separate the parameters to be estimated, and define:

[0057] ;

[0058] in , This is the dq-axis voltage. , Estimate the current for the dq axis;

[0059] Substitution ,get:

[0060] ;

[0061] Where the coefficient , , , It is given by the following formula:

[0062] ;

[0063] S32. Rewrite the above equations in matrix form and define the vector of parameters to be estimated. ,but:

[0064] ;

[0065] in For actual rotational speed, regression matrix It consists of current and C coefficient;

[0066] S33. Finally, the parameter vector can be updated online using the recursive least squares method. In each control cycle, three-phase current information is collected, and the regression matrix is ​​calculated. and output vector Update according to the following recursive formula:

[0067] ;

[0068] in Let P be the forgetting factor and P be the covariance matrix.

[0069] As a preferred technical solution of the present invention: in order to ensure the continuous excitation conditions of the system, the inherent second harmonic component in the current is used as a natural excitation, without the need to inject an additional signal.

[0070] As a preferred technical solution of the present invention, the specific process of step S4 is as follows:

[0071] S41, obtained from estimation , , Calculate the actual resistance of the three phases:

[0072] ;

[0073] At the same time, by the estimated The estimated rotational speed and position are obtained through a phase-locked loop:

[0074] ;

[0075] in and represent the proportional coefficient and integral coefficient of the phase-locked loop, respectively, and s represents the Laplace operator;

[0076] S42. This provides a real-time estimate of the unbalanced resistance value and position, which serves as a basis for subsequent fault location and sensorless operation of the motor.

[0077] As a preferred technical solution of the present invention, the specific process of step S5 is as follows:

[0078] S51. Define parameters:

[0079] Speed ​​outer loop: Reference speed Compared with the estimated speed The difference is input to the PI regulator, and the output is the q-axis current reference value:

[0080] ;

[0081] in and These represent the proportional coefficient and integral coefficient of the outer velocity loop, respectively.

[0082] Inner current loop: The reference current... , With estimation of dq axis current , The difference is input to the PI regulator, and the output is the dq-axis voltage reference value:

[0083] ;

[0084] in and These represent the proportional and integral coefficients of the inner loop of the d-axis current, respectively. and These represent the proportional coefficient and integral coefficient of the inner loop of the q-axis current, respectively.

[0085] S52, Utilization The obtained dq-axis reference voltage is used to obtain the three-phase reference voltage through inverse Park transformation. Then, combined with space vector modulation technology, the drive signal of the inverter switching transistor is generated to realize the control of the three-phase inverter.

[0086] A real-time detection and sensorless performance compensation system for high-resistance faults at the contact terminals of a permanent magnet synchronous motor is characterized by comprising a three-phase inverter, a permanent magnet synchronous motor, a speed loop controller, a current loop controller, an inverse Park transform, an inverse Clark transform, a space vector pulse width modulation, an adaptive estimator, a phase-locked loop, and a three-phase resistance reconstruction module, wherein:

[0087] The permanent magnet synchronous motor is connected to an adaptive estimator, which collects phase current. , And the bus voltage, and through coordinate transformation, the estimated feedback current in the dq coordinate system is obtained. , The voltage command value and feedback current are input into the adaptive estimator to estimate the three-phase resistance deviation online. , and average resistance Simultaneously estimate the angle error ;

[0088] The adaptive estimator is connected to a phase-locked loop to convert the angle error... Input a phase-locked loop, and obtain the compensated rotor position through closed-loop regulation. With rotational speed ;

[0089] The phase-locked loop is connected to the three-phase resistance reconstruction module. The estimated three-phase resistance deviation is input into the three-phase resistance reconstruction module to realize the location of the faulty phase and the quantification of the degree of high resistance fault.

[0090] The three-phase resistance reconstruction module is connected to the speed loop controller and the current loop controller respectively, and uses the estimated rotor speed... With reference speed The difference is used to obtain the q-axis reference current via the speed loop controller. Using the estimated dq-axis current , With reference current , The difference is used by the current loop controller to obtain the voltage reference in the rotating coordinate system. , ;

[0091] The output terminals of the speed loop controller and the current loop controller are sequentially connected to the inverse Park transform, the inverse Clark transform, and the space vector pulse width modulation. Switching signals are obtained through the inverse Park transform, the inverse Clark transform, and the space vector pulse width modulation, which are used to control the three-phase inverter.

[0092] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0093] The real-time detection and compensation method and system for high-resistance faults at the contact terminals of permanent magnet synchronous motors proposed in this invention do not require additional hardware sensors and fully utilize the current feedback signal in the existing driver, significantly reducing system costs. Existing methods are greatly affected by operating conditions and cannot locate the faulty phase. This invention uses an adaptive estimator to synchronously estimate the three-phase resistance and angle error online, which can accurately detect resistance imbalance and determine the faulty phase in the entire operating range. The detection accuracy does not depend on the current magnitude and control gain, and it has good robustness. At the same time, the estimated angle error is fed into the phase-locked loop to dynamically correct the rotor position, which effectively suppresses the angle estimation deviation and current frequency fluctuation caused by unbalanced resistance, and significantly improves the sensorless operation performance under minor faults. Attached Figure Description

[0094] Figure 1 This is a general block diagram of the method for real-time detection and sensorless performance compensation of high resistance faults in the contact terminals of permanent magnet synchronous motors in this embodiment of the invention.

[0095] Figure 2 This is a graph showing the estimation results of the uncompensated rotor position error and unbalanced resistance coefficient when there is a minor terminal fault in phase a in this embodiment of the invention.

[0096] Figure 3 This is a graph showing the estimation results of rotor position error and unbalanced resistance coefficient when a phase a minor terminal fault is compensated using the method of this invention in an embodiment of the invention.

[0097] Figure 4 This is a graph showing the estimation results of the unbalanced resistance coefficient in an embodiment of the present invention;

[0098] Figure 5 This is a diagram showing the estimation results of uncompensated rotor position error and unbalanced resistance coefficient when there are minor terminal faults in phases a and b in an embodiment of the present invention.

[0099] Figure 6 This is a graph showing the estimation results of rotor position error and unbalanced resistance coefficient when a and b phases have minor terminal faults, compensated by the method of this invention in an embodiment of the invention.

[0100] Figure 7 This is a diagram showing the estimation results of the unbalanced resistance coefficient in an embodiment of the present invention. Detailed Implementation

[0101] The present invention will be further illustrated below with reference to the accompanying drawings and specific embodiments. It should be understood that the following specific embodiments are for illustrative purposes only and are not intended to limit the scope of the invention.

[0102] like Figure 1As shown, this invention provides a method for real-time detection and sensorless performance compensation of high-resistance faults in the contact terminals of a permanent magnet synchronous motor. This method does not require additional hardware sensors and fully utilizes the current feedback signal in the existing driver. It uses an adaptive observer to synchronously estimate the three-phase equivalent resistance and sensorless angle error online, achieving accurate fault phase location and sensorless performance compensation under minor faults. The specific steps are as follows:

[0103] S1. Establish a mathematical model of a three-phase resistive asymmetric permanent magnet synchronous motor and derive the voltage equation under unbalanced resistance.

[0104] The specific process is as follows:

[0105] S11. First, establish a static coordinate system model under the condition of contact fault, i.e., three-phase resistance asymmetry:

[0106] Let the three-phase stator resistances be respectively , , Define the average resistance And the deviation of each phase resistance from the average value. , , ,satisfy ,

[0107] Then, in the stationary abc coordinate system, the voltage equation is:

[0108] ;

[0109] in , , This refers to the three-phase stator voltage; , , This refers to the three-phase stator current; , , For back electromotive force; L s For stator inductance;

[0110] S12. Next, transform to the rotated dq coordinate system and derive the voltage equations containing resistive asymmetry:

[0111] For rotor position Using the equal-amplitude Park transformation matrix Transforming the voltage and current into a rotating dq coordinate system, the resistance matrix becomes:

[0112] ;

[0113] After expansion, the resistance matrix components in the rotated dq coordinate system are obtained:

[0114] ;

[0115] S13. Finally, substituting the average resistance and deviation, and using the property that the sum of the three-phase cosines is zero, we can simplify to get:

[0116] ;

[0117] , and As second harmonic components, they will generate harmonic disturbances in the current loop;

[0118] Therefore, the voltage equation in the synchronously rotating coordinate system dq is:

[0119] ;

[0120] Where j is defined as the imaginary unit of a complex vector, and the complex vector... , , Let ψ be the electric speed of the motor, and ψf be the rotor flux linkage.

[0121] It is evident that resistance asymmetry introduces a relationship with rotor position. Related second harmonic components and They will generate frequency harmonic disturbances in the current loop, which is the physical basis of existing harmonic detection methods.

[0122] S2. Analyze the impact of unbalanced resistance on sensorless control, and derive the voltage equation and position error mechanism in the estimated coordinate system.

[0123] The specific process is as follows:

[0124] S21. In sensorless control, an extended back EMF model is typically used for rotor position estimation. Taking a surface-mounted permanent magnet synchronous motor as an example, the voltage equation in the actual dq coordinate system is as described above. Let the angular error between the estimated dq coordinate system and the actual dq coordinate system be... Transforming the actual voltage equation to the estimated dq coordinate system, we obtain:

[0125] ;

[0126] in To estimate the current in the dq coordinate system, To estimate the voltage in the dq coordinate system, To estimate the rotational speed;

[0127] The resistance matrix in the dq coordinate system is estimated by similarity transformation:

[0128] ;

[0129] S22. Using Pauli matrix decomposition and rotation transformation rules, decompose the asymmetric part of the resistance, and define:

[0130] ;

[0131] in , The Pauli matrix, after similarity transformation in the estimated dq coordinate system, is obtained as follows:

[0132] ;

[0133] Therefore, the resistance asymmetry matrix in the dq coordinate system can be estimated as follows:

[0134] ;

[0135] in:

[0136] ;;

[0137] S23. Substituting the above results into the estimated voltage equation in the dq coordinate system, we obtain the complete model considering resistance asymmetry:

[0138] ;

[0139] S24. From the complete model considering resistance asymmetry, it can be seen that resistance asymmetry generates an additional voltage term in the estimated dq coordinate system. Meanwhile, nominal resistance and actual average resistance The difference between them is equivalent to parameter mismatch. Both of these factors together lead to a deviation in the back EMF observation, which in turn causes DC bias error and second harmonic fluctuation error in the rotor position estimated without a position sensor. Since the mechanical time constant of the motor is much larger than the electrical time constant, and the bandwidth of the adaptive estimator is usually higher than the speed change frequency, it can be assumed that within one control cycle, the actual speed... Compared with the estimated speed Approximately equal, that is Ultimately, the voltage equation is transformed into a regression equation that can be adaptively estimated.

[0140] S3. Based on the estimated coordinate system voltage equation obtained in S2, transform it into an adaptively estimable linear regression equation and design an adaptive parameter estimator.

[0141] The specific process is as follows:

[0142] S31, General Style Expand, separate the parameters to be estimated, and define:

[0143] ;

[0144] in , This is the dq-axis voltage. , Estimate the current for the dq axis;

[0145] Substitution ,get:

[0146] ;

[0147] Where the coefficient , , , It is given by the following formula:

[0148] ;

[0149] S32. Rewrite the above equations in matrix form and define the vector of parameters to be estimated. ,but:

[0150] ;

[0151] in For actual rotational speed, regression matrix It consists of current and C coefficient;

[0152] S33. Finally, the parameter vector can be updated online using the recursive least squares method. In each control cycle, three-phase current information is collected, and the regression matrix is ​​calculated. and output vector Update according to the following recursive formula:

[0153] ;

[0154] in Let P be the forgetting factor and P be the covariance matrix.

[0155] To ensure continuous excitation of the system, the inherent second harmonic component in the current is used as a natural excitation, eliminating the need for additional signal injection.

[0156] S4. Based on the three-phase resistance deviation and position error estimated in S3, reconstruct the actual three-phase resistance and obtain the compensated rotor position and speed.

[0157] The specific process is as follows:

[0158] S41, obtained from estimation , , Calculate the actual resistance of the three phases:

[0159] ;

[0160] At the same time, by the estimated The estimated rotational speed and position are obtained through a phase-locked loop:

[0161] ;

[0162] in and represent the proportional coefficient and integral coefficient of the phase-locked loop, respectively, and s represents the Laplace operator;

[0163] S42. This provides a real-time estimate of the unbalanced resistance value and position, offering a basis for subsequent fault location and sensorless motor operation. This estimation method does not rely on additional hardware, and its estimation accuracy is unaffected by the controller bandwidth, making it effective across the entire operating range.

[0164] S5. Based on the rotor position and speed estimated by S4, perform closed-loop vector control on the permanent magnet synchronous motor.

[0165] The specific process is as follows:

[0166] S51. Define parameters:

[0167] Speed ​​outer loop: Reference speed Compared with the estimated speed The difference is input to the PI regulator, and the output is the q-axis current reference value:

[0168] ;

[0169] in and These represent the proportional coefficient and integral coefficient of the outer velocity loop, respectively.

[0170] Inner current loop: The reference current... , With estimation of dq axis current , The difference is input to the PI regulator, and the output is the dq-axis voltage reference value:

[0171] ;

[0172] in and These represent the proportional and integral coefficients of the inner loop of the d-axis current, respectively. and These represent the proportional coefficient and integral coefficient of the inner loop of the q-axis current, respectively.

[0173] S52, Utilization The obtained dq-axis reference voltage is used to obtain the three-phase reference voltage through inverse Park transformation. Then, combined with space vector modulation technology, the drive signal of the inverter switching transistor is generated to realize the control of the three-phase inverter.

[0174] like Figure 1 As shown in the figure, this invention also proposes a real-time detection and sensorless performance compensation system for high-resistance faults at the contact terminals of a permanent magnet synchronous motor, including a three-phase inverter, a permanent magnet synchronous motor, a speed loop controller, a current loop controller, an inverse Park transform, an inverse Clark transform, a space vector pulse width modulation, an adaptive estimator, a phase-locked loop, and a three-phase resistance reconstruction module, as detailed below:

[0175] By collecting phase current , And the bus voltage, and through coordinate transformation, the estimated feedback current in the dq coordinate system is obtained. , The voltage command value and feedback current are input into the adaptive estimator to estimate the three-phase resistance deviation online. , and average resistance Simultaneously estimate the angle error ; angular error Input a phase-locked loop, and obtain the compensated rotor position through closed-loop regulation. With rotational speed The estimated three-phase resistance deviation is input into the three-phase resistance reconstruction module to achieve fault phase location and quantification of high-resistance fault degree. The estimated rotor speed is then used... With reference speed The difference is used to obtain the q-axis reference current via the speed loop controller. Using the estimated dq-axis current , With reference current , The difference is used by the current loop controller to obtain the voltage reference in the rotating coordinate system. , Finally, the switching signal is obtained through inverse Park transform, inverse Clark transform, and space vector pulse width modulation, which is used to control the three-phase inverter and realize closed-loop vector control of the permanent magnet synchronous motor.

[0176] Figure 2-4 The figure shows the simulation results of phase a minor terminal fault (equivalent resistance 0.3Ω) with and without compensation in Embodiment a of the present invention; wherein:

[0177] Figure 2 The waveform of the rotor position estimation error without compensation shows that there is obvious DC bias and second harmonic fluctuation in the error, with a maximum error of about 0.36 rad.

[0178] Figure 3 The rotor position estimation error waveform after compensation using the method of the present invention effectively suppresses DC bias and second harmonic fluctuations, reducing the error to about 0.13 rad, and significantly improving the position estimation accuracy.

[0179] Figure 4 The waveform of the estimated unbalanced resistance coefficient is shown below. This is an estimate of the average resistance. This is the estimated value of the phase a resistance deviation. The estimated value for the phase b resistance deviation converges to near the true value within 6 seconds, verifying the effectiveness of the adaptive estimator. The rated phase resistance of the motor is 0.17Ω, which can be reconstructed from the estimation results. , , This indicates that a slight contact failure occurred in phase a, verifying the effectiveness of the method of the present invention in locating the faulty phase.

[0180] Figure 5-7 The following are simulation results of the present invention embodiment with and without compensation for minor terminal faults in phases a and b (equivalent resistances of 0.3Ω and 0.15Ω, respectively); wherein:

[0181] Figure 5 The waveform of the rotor position estimation error without compensation shows that there is obvious DC bias and second harmonic fluctuation in the error, with a maximum error of about 0.3 rad.

[0182] Figure 6 The rotor position estimation error waveform after compensation using the method of the present invention effectively suppresses DC bias and second harmonic fluctuations, reducing the error to about 0.12 rad, and significantly improving the position estimation accuracy.

[0183] Figure 7 The waveform of the estimated unbalanced resistance coefficient is shown below. This is an estimate of the average resistance. This is the estimated value of the phase a resistance deviation. This is the estimated value for the phase b resistance deviation. The rated phase resistance of the motor is 0.17Ω. The value can be reconstructed from the estimated result. , , This indicates that both phase a and phase b have slight contact failures, verifying the effectiveness of the method of the present invention in fault location and compensation under multiphase fault conditions.

[0184] It should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention in any other way. Any modifications or equivalent changes made based on the technical essence of the present invention shall still fall within the scope of protection claimed by the present invention.

Claims

1. A method for real-time detection and sensorless performance compensation of high resistance faults in the contact terminals of a permanent magnet synchronous motor, characterized in that, Includes the following steps: S1. Establish a mathematical model of a three-phase resistive asymmetric permanent magnet synchronous motor and derive the voltage equation under unbalanced resistance. S2. Analyze the impact of unbalanced resistance on sensorless control, and derive the voltage equation and position error mechanism in the estimated coordinate system. S3. Based on the estimated coordinate system voltage equation obtained in S2, transform it into an adaptively estimable linear regression equation and design an adaptive parameter estimator. S4. Based on the three-phase resistance deviation and position error estimated in S3, reconstruct the actual three-phase resistance and obtain the compensated rotor position and speed. S5. Based on the rotor position and speed estimated by S4, perform closed-loop vector control on the permanent magnet synchronous motor.

2. The method for real-time detection and sensorless performance compensation of high resistance faults in the contact terminals of a permanent magnet synchronous motor according to claim 1, characterized in that: The specific process of step S1 is as follows: S11. First, establish a static coordinate system model under the condition of contact fault, i.e., three-phase resistance asymmetry: Let the three-phase stator resistances be respectively , , Define the average resistance And the deviation of each phase resistance from the average value. , , ,satisfy , Then, in the stationary abc coordinate system, the voltage equation is: ； in , , This refers to the three-phase stator voltage; , , This refers to the three-phase stator current; , , For back electromotive force; L s For stator inductance; S12. Next, transform to the rotated dq coordinate system and derive the voltage equations containing resistive asymmetry: For rotor position Using the equal-amplitude Park transformation matrix Transforming the voltage and current into a rotating dq coordinate system, the resistance matrix becomes: ； After expansion, the resistance matrix components in the rotated dq coordinate system are obtained: ； S13. Finally, substituting the average resistance and deviation, and using the property that the sum of the three-phase cosines is zero, we can simplify to get: ； , and As second harmonic components, they will generate harmonic disturbances in the current loop; Therefore, the voltage equation in the synchronously rotating coordinate system dq is: ； Where j is defined as the imaginary unit of a complex vector, and the complex vector... , , Let ψ be the electric speed of the motor, and ψf be the rotor flux linkage.

3. The method for real-time detection and sensorless performance compensation of high resistance faults in the contact terminals of a permanent magnet synchronous motor according to claim 1, characterized in that: The specific process of step S2 is as follows: S21. Let the estimated angular error between the dq coordinate system and the true dq coordinate system be... Transforming the actual voltage equation to the estimated dq coordinate system, we obtain: ； in To estimate the current in the dq coordinate system, To estimate the voltage in the dq coordinate system, To estimate the rotational speed; The resistance matrix in the dq coordinate system is estimated by similarity transformation: ； S22. Using Pauli matrix decomposition and rotation transformation rules, decompose the asymmetric part of the resistance, and define: ； in , The Pauli matrix, after similarity transformation in the estimated dq coordinate system, is obtained as follows: ； Therefore, the resistance asymmetry matrix in the dq coordinate system can be estimated as follows: ； in: ； S23. Substituting the above results into the estimated voltage equation in the dq coordinate system, we obtain the complete model considering resistance asymmetry: ； S24. From the complete model considering resistance asymmetry, it can be seen that resistance asymmetry generates an additional voltage term in the estimated dq coordinate system. Meanwhile, nominal resistance and actual average resistance The difference between them is equivalent to parameter mismatch. Both of these factors together lead to a deviation in the back EMF observation, which in turn causes DC bias error and second harmonic fluctuation error in the rotor position estimated without a position sensor. Since the mechanical time constant of the motor is much larger than the electrical time constant, and the bandwidth of the adaptive estimator is usually higher than the speed change frequency, it can be assumed that within one control cycle, the actual speed... Compared with the estimated speed Approximately equal, that is Ultimately, the voltage equation is transformed into a regression equation that can be adaptively estimated.

4. The method for real-time detection and sensorless performance compensation of high resistance faults in the contact terminals of a permanent magnet synchronous motor according to claim 1 or 3, characterized in that: The specific process of step S3 is as follows: S31, General Style Expand, separate the parameters to be estimated, and define: ； in , This is the dq-axis voltage. , Estimate the current for the dq axis; Substitution ,get: ； Where the coefficient , , , It is given by the following formula: ； S32. Rewrite the above equations in matrix form and define the vector of parameters to be estimated. ,but: ； in For actual rotational speed, regression matrix It consists of current and C coefficient; S33. Finally, the parameter vector can be updated online using the recursive least squares method. In each control cycle, three-phase current information is collected, and the regression matrix is ​​calculated. and output vector Update according to the following recursive formula: ； in Let P be the forgetting factor and P be the covariance matrix.

5. The method for real-time detection and sensorless performance compensation of high resistance faults in the contact terminals of a permanent magnet synchronous motor according to claim 4, characterized in that: To ensure continuous excitation of the system, the inherent second harmonic component in the current is used as a natural excitation, eliminating the need for additional signal injection.

6. The method for real-time detection and sensorless performance compensation of high resistance faults in the contact terminals of a permanent magnet synchronous motor according to claim 1, characterized in that: The specific process of step S4 is as follows: S41, obtained from estimation , , Calculate the actual resistance of the three phases: ； At the same time, by the estimated The estimated rotational speed and position are obtained through a phase-locked loop: ； in and represent the proportional coefficient and integral coefficient of the phase-locked loop, respectively, and s represents the Laplace operator; S42. This provides a real-time estimate of the unbalanced resistance value and position, which serves as a basis for subsequent fault location and sensorless operation of the motor.

7. The method for real-time detection and sensorless performance compensation of high resistance faults in the contact terminals of a permanent magnet synchronous motor according to claim 1, characterized in that: The specific process of step S5 is as follows: S51. Define parameters: Speed ​​outer loop: Reference speed Compared with the estimated speed The difference is input to the PI regulator, and the output is the q-axis current reference value: ； in and These represent the proportional coefficient and integral coefficient of the outer velocity loop, respectively; Inner current loop: The reference current... , With estimation of dq axis current , The difference is input to the PI regulator, and the output is the dq-axis voltage reference value: ； in and These represent the proportional and integral coefficients of the inner loop of the d-axis current, respectively. and These represent the proportional coefficient and integral coefficient of the inner loop of the q-axis current, respectively; S52, Utilization The obtained dq-axis reference voltage is used to obtain the three-phase reference voltage through inverse Park transformation. Then, combined with space vector modulation technology, the drive signal of the inverter switching transistor is generated to realize the control of the three-phase inverter.

8. The real-time detection and sensorless performance compensation system for high resistance faults in the contact terminals of a permanent magnet synchronous motor according to any one of claims 1-7, characterized in that, It includes a three-phase inverter, a permanent magnet synchronous motor, a speed loop controller, a current loop controller, inverse Park transform, inverse Clark transform, space vector pulse width modulation, an adaptive estimator, a phase-locked loop, and a three-phase resistance reconfiguration module, wherein: The permanent magnet synchronous motor is connected to an adaptive estimator, which collects phase current. , And the bus voltage, and through coordinate transformation, the estimated feedback current in the dq coordinate system is obtained. , The voltage command value and feedback current are input into the adaptive estimator to estimate the three-phase resistance deviation online. , and average resistance Simultaneously estimate the angle error ; The adaptive estimator is connected to a phase-locked loop to convert the angle error. Input a phase-locked loop, and obtain the compensated rotor position through closed-loop regulation. With rotational speed ; The phase-locked loop is connected to the three-phase resistance reconstruction module. The estimated three-phase resistance deviation is input into the three-phase resistance reconstruction module to realize the location of the faulty phase and the quantification of the degree of high resistance fault. The three-phase resistance reconstruction module is connected to the speed loop controller and the current loop controller respectively, and uses the estimated rotor speed... With reference speed The difference is used to obtain the q-axis reference current via the speed loop controller. Using the estimated dq-axis current , With reference current , The difference is used by the current loop controller to obtain the voltage reference in the rotating coordinate system. , ; The output terminals of the speed loop controller and the current loop controller are sequentially connected to the inverse Park transform, the inverse Clark transform, and the space vector pulse width modulation. Switching signals are obtained through the inverse Park transform, the inverse Clark transform, and the space vector pulse width modulation, which are used to control the three-phase inverter.

Images