A multi-parameter acquisition-based automobile motor fault detection method

Abstract

This invention belongs to the field of motor fault monitoring technology. It provides a method for detecting automotive motor faults based on multi-parameter acquisition, including: starting monitoring only under special coupled conditions of high altitude and high motor temperature; acquiring the absolute air pressure of the motor's sealed cavity and the low-frequency parasitic capacitance of the stator winding; identifying the real-time thermal time constant by combining the motor's heating power and stator temperature response curves; accurately identifying trace coolant vaporization faults by matching the characteristics of the air pressure time sequence platform with the saturated vapor pressure of glycol; and constructing a medium swelling damage index by weighted fusion of the parasitic capacitance drift rate and the thermal time constant increment ratio to quantify the degree of chemical degradation of the potting compound. This invention can achieve accurate early warning of faults in the early, latent stage when the motor has not yet experienced any abnormalities. When the fault exceeds the standard, it can promptly output a micro-leakage alarm and limit the peak temperature of the motor, significantly improving the reliability and safety of motor operation in heavy-duty vehicles on plateaus.

Description

Technical Field

[0001] This invention belongs to the field of motor fault monitoring technology, specifically a method for detecting automotive motor faults based on multi-parameter acquisition. Background Technology

[0002] New energy vehicles generally use liquid-cooled drive motors. The stator windings are insulated and heat-conducting through epoxy resin potting. They rely on water-glycol coolant circulation for heat dissipation. Long-term vibration and alternating hot and cold operation of the motor can easily cause aging of the water channel seals, leading to micro-leakage of coolant.

[0003] Under the special coupled conditions of high-altitude, heavy-load, and uphill operation, the low environmental pressure combined with the high thermal load of the motor causes the slightly leaking coolant to rapidly boil and vaporize, dispersing into the motor's sealed cavity. High-temperature ethylene glycol vapor then continuously penetrates and erodes the stator potting compound, causing chemical swelling, the formation of micropores, and a loose structure. This leads to a decrease in the thermal conductivity of the potting compound, an increase in its equivalent dielectric constant, and gradually results in insulation degradation, winding stripping, and even phase-to-phase short-circuit breakdown, ultimately causing complete motor failure.

[0004] Existing motor leakage and insulation detection technologies have significant shortcomings: traditional liquid level sensors can only detect large-scale coolant leaks and are completely insensitive to early micro-leaks; traditional DC insulation resistance detection can only trigger post-incident alarms after severe insulation damage and cannot detect early, latent degradation caused by the chemical swelling of potting compound. Furthermore, existing technologies do not consider the phase change characteristics of coolant under high-altitude, low-pressure conditions, nor do they integrate the dual degradation characteristics of potting compound's thermal and dielectric parameters for joint judgment. This results in serious detection blind spots, high false negative rates, and delayed early warnings, failing to meet the needs of early fault prediction and safety protection for automotive motors operating at high altitudes and under heavy loads.

[0005] Therefore, the present invention provides a method for detecting automotive motor faults based on multi-parameter acquisition. Summary of the Invention

[0006] In order to overcome the shortcomings of the prior art, at least one technical problem raised in the background art is solved.

[0007] The technical solution adopted by this invention to solve its technical problem is: This invention provides a method for detecting automotive motor faults based on multi-parameter acquisition, comprising: Step S1: Obtain the vehicle's current altitude and motor stator temperature. When the altitude is greater than a preset altitude threshold and the motor stator temperature is greater than a preset temperature threshold, generate a special working condition monitoring signal. Step S2: Under special working condition monitoring signal, the absolute air pressure value in the motor sealed cavity is collected in real time, and a low frequency detection signal is sent to the stator winding to obtain the low frequency parasitic capacitance value of the stator winding to the motor housing. Step S3: Based on the response curve of the motor's real-time heating power and stator temperature, identify the current thermal time constant of the stator winding; Step S4: Establish a time-series curve of absolute pressure value and determine whether the absolute pressure value stabilizes at the plateau segment of ethylene glycol saturated vapor pressure corresponding to the current temperature after the increase; if so, it is determined that coolant vaporization has occurred. Step S5: If coolant vaporization occurs, calculate the drift rate of the low-frequency parasitic capacitance and the increase ratio of the current thermal time constant relative to the factory time constant reference value. Step S6: Weight the drift rate and the increase ratio to calculate the medium swelling damage index of the stator potting compound. When the medium swelling damage index is greater than the preset medium swelling damage index threshold, output a coolant leakage alarm and limit the peak operating temperature of the motor.

[0008] Preferably, the specific process for obtaining the vehicle's current altitude and the motor stator temperature is as follows: The vehicle's real-time three-dimensional coordinate positioning data is obtained through the vehicle-mounted satellite positioning system, and the vehicle's current altitude is obtained from the three-dimensional coordinate positioning data. Multiple thermistor sensors are pre-installed at the ends of the motor stator windings and in the stator core slots to collect multi-point temperature values ​​in real time. After performing a moving average filtering process on the multi-point temperature values, the maximum value among the multi-point temperature values ​​is extracted as the motor stator temperature.

[0009] Preferably, the specific process for generating the special working condition monitoring signal is as follows: The obtained altitude is compared with the preset altitude threshold, and the obtained motor stator temperature is compared with the preset temperature threshold. When the altitude is greater than a preset altitude threshold and the motor stator temperature is greater than a preset temperature threshold, a special operating condition monitoring signal is generated.

[0010] Preferably, the specific process for collecting the absolute pressure value is as follows: Under special operating conditions, the absolute air pressure value is collected in real time by a miniature absolute pressure sensor installed on the inner wall of the motor's sealed cavity.

[0011] Preferably, the specific process for obtaining the low-frequency parasitic capacitance value of the stator winding to the motor housing is as follows: The inverter inside the motor controller MCU sends a zero-sequence AC voltage signal with a preset frequency and preset amplitude in a three-phase pulse width modulation (PWM) waveform, and sends the zero-sequence AC voltage signal as a low-frequency detection signal to the stator winding. The zero-sequence leakage current signal flowing from the stator winding to the motor housing is collected. The phase of the zero-sequence leakage current signal is compared with the low-frequency detection signal to obtain the capacitive reactive current component with a phase difference of 90 degrees from the low-frequency detection signal. The root mean square (RMS) operation is then performed to obtain the effective value of the capacitive reactive current component. Based on the effective value of the capacitive reactive current component, combined with Calculate the low-frequency parasitic capacitance of the stator windings to the motor housing. This is the low-frequency parasitic capacitance value. This represents the effective value of the capacitive reactive current component. The preset frequency for low-frequency detection signals. This is the preset effective value of the amplitude of the low-frequency detection signal.

[0012] Preferably, the specific process for constructing the response curve is as follows: The effective value of the three-phase stator current during motor operation is collected in real time. Combined with the phase resistance value of the stator winding at the current temperature, the real-time copper loss heat generation of the motor is calculated as the real-time heating power of the motor. When the rate of change of the real-time heat power of the motor is greater than or equal to the threshold of heat power change, the time when the rate of change reaches the threshold of heat power change is taken as the zero point, and the stator temperature of the motor is recorded within the preset time window. If the fluctuation range of the real-time heating power of the motor is less than the steady-state tolerance range within the preset time window, the heating power is determined to be stable, and a response curve is constructed with time as the horizontal axis and motor stator temperature as the vertical axis.

[0013] Preferably, the specific process for identifying the current thermal time constant of the stator winding is as follows: Establish the transient temperature rise equation for the stator winding: , for The stator temperature of the motor at any given time. The initial stator temperature of the motor at time zero. This refers to the real-time heating power of the motor. The equivalent thermal resistance from the stator winding to the cooling system. The thermal time constant; The least squares curve fitting algorithm is used to substitute the constructed response curve into the transient temperature rise equation for solving, and the solved thermal time constant is obtained. As the current thermal time constant of the stator winding.

[0014] Preferably, the specific process for determining that coolant vaporization has occurred is as follows: Using time as the horizontal axis and the real-time collected absolute air pressure value as the vertical axis, a time-series variation curve of the absolute air pressure value is constructed to obtain the current motor stator temperature and the Antoine constant. The motor stator temperature and the Antoine constant are substituted into the Antoine equation to calculate the theoretical saturated vapor pressure of ethylene glycol at the current temperature. The first derivative of the time-series change curve is calculated to obtain the real-time rate of change of the absolute pressure value. When the real-time change first exceeds the rate of increase threshold, and within the observation time window, the absolute value of the real-time rate of change is less than the pressure fluctuation threshold, and the absolute value of the difference between the absolute pressure value and the theoretical ethylene glycol saturated vapor pressure is less than the matching error threshold, it is determined that the absolute pressure value has risen and stabilized at the ethylene glycol saturated vapor pressure corresponding to the current temperature, forming a plateau segment, i.e., coolant vaporization has occurred.

[0015] Preferably, the specific process for calculating the drift rate of the low-frequency parasitic capacitance value and the increase ratio of the current thermal time constant relative to the factory time constant reference value is as follows: When coolant vaporization is determined to have occurred: Obtain the factory-set low-frequency parasitic capacitance reference value, calculate the difference between the low-frequency parasitic capacitance value and the factory-set low-frequency parasitic capacitance reference value, and divide the resulting difference by the factory-set low-frequency parasitic capacitance reference value to obtain the drift rate of the low-frequency parasitic capacitance value. Obtain the factory time constant reference value, calculate the difference between the current thermal time constant and the factory time constant reference value, and divide the resulting difference by the factory time constant reference value to obtain the increase ratio of the current thermal time constant relative to the factory time constant reference value.

[0016] Preferably, the specific process of outputting the coolant leakage alarm is as follows: Obtain the pre-calibrated capacitance characteristic weighting coefficient and thermal characteristic weighting coefficient. The sum of the capacitance characteristic weighting coefficient and thermal characteristic weighting coefficient is 1. Multiply the drift rate by the capacitance characteristic weighting coefficient to obtain the capacitance degradation component. Multiply the increase ratio by the thermal characteristic weighting coefficient to obtain the thermal resistance degradation component. The capacitance degradation component and the thermal resistance degradation component are added together to complete the weighted fusion of drift rate and increase ratio. The numerical result obtained after weighted fusion is used as the medium swelling damage index of stator potting compound. When the medium swelling damage index is greater than the preset medium swelling damage index threshold, a coolant leakage alarm is output.

[0017] The beneficial effects of this invention are as follows: This invention constrains the fault monitoring logic to the special operating conditions of high altitude coupling, where the altitude exceeds a preset threshold and the motor is under high temperature and heavy load. It specifically addresses the shortcomings of existing multi-parameter motor detection solutions that are only suitable for plains and normal pressure scenarios. Under plain conditions, minor coolant leaks will accumulate liquid, which can be identified by level and water immersion sensors. However, in the low-pressure and high-temperature environment of high altitudes, trace amounts of coolant will vaporize into ethylene glycol vapor, without producing liquid accumulation. Conventional level sensing, insulation resistance monitoring, and single current / temperature diagnosis cannot identify the early deterioration of the potting compound due to vapor erosion. This invention accurately identifies the ethylene glycol vaporization phenomenon through the characteristics of the cavity air pressure timing platform. It can complete the detection in the initial stage of the fault when the winding insulation is not damaged and the normal operating parameters are normal, thus avoiding sudden inter-turn short circuits and motor burnout after the potting compound continues to swell and crack, and significantly improving the driving safety of vehicles under heavy load in high altitudes.

[0018] This invention integrates two independent but complementary degradation characteristics for joint diagnosis: first, the low-frequency parasitic capacitance drift of the winding to ground caused by the swelling micropores of the epoxy potting compound (characterizing the degradation of the dielectric properties of the insulating medium); second, the increase in the thermal time constant due to the decay of the thermal conductivity of the colloid (characterizing the degradation of the overall thermal conductivity). A weighted fusion is used to construct a dedicated dielectric swelling damage index to comprehensively assess the degree of degradation. Unlike single-parameter discrimination schemes, this invention effectively eliminates external interference such as poor heat dissipation due to the thin air at high altitudes, environmental temperature fluctuations, and sensor temperature drift, significantly reducing the probability of false alarms and missed alarms in high-altitude environments. Attached Figure Description

[0019] The invention will now be further described with reference to the accompanying drawings.

[0020] Figure 1 This is a flowchart of the steps of a method for detecting automotive motor faults based on multi-parameter acquisition according to the present invention; Figure 2 This is a flowchart of a method for detecting automotive motor faults based on multi-parameter acquisition, according to the present invention. Detailed Implementation

[0021] To make the technical means, creative features, objectives and effects of this invention easier to understand, the invention will be further described below in conjunction with specific embodiments.

[0022] Example 1 Please see Figure 1 and Figure 2 As shown in the embodiment of the present invention, a method for detecting automotive motor faults based on multi-parameter acquisition includes: Step S1: Obtain the vehicle's current altitude and motor stator temperature. When the altitude is greater than a preset altitude threshold and the motor stator temperature is greater than a preset temperature threshold, generate a special working condition monitoring signal. In some specific embodiments, the process of obtaining the vehicle's current altitude and motor stator temperature in step S1 is as follows: The vehicle's real-time three-dimensional coordinate positioning data is obtained by the receiving terminal of the vehicle-mounted satellite positioning system, and the current altitude of the vehicle is extracted from the three-dimensional coordinate positioning data. Multiple thermistor sensors are pre-embedded at the ends of the motor stator windings and in the stator core slots to collect multi-point temperature values ​​in real time according to a preset temperature sampling period. After performing a moving average filtering process on the multi-point temperature values ​​to eliminate electromagnetic interference noise, the maximum value among the multi-point temperature values ​​is extracted as the motor stator temperature. In some specific embodiments, the process of generating a special operating condition monitoring signal in step S1 when the altitude is greater than a preset altitude threshold and the motor stator temperature is greater than a preset temperature threshold is as follows: The obtained altitude is compared with a preset altitude threshold, and the obtained motor stator temperature is compared with a preset temperature threshold. When the altitude is greater than the preset altitude threshold and the motor stator temperature is greater than the preset temperature threshold, the special working condition monitoring signal is generated and output. The special working condition monitoring signal is used for subsequent detection processes for coolant micro-leakage and potting compound swelling. It should be noted that the preset height threshold is set to, for example, 3000 meters. The physical basis for setting the height threshold is that when the altitude reaches 3000 meters or above, the ambient atmospheric pressure will drop significantly, for example, usually to about 70 kPa or below. At this time, the boiling point of the motor coolant will drop significantly, thus providing an external low-pressure environment condition for vaporization at non-extreme high temperatures. It should also be further explained that the preset temperature threshold is set to, for example, 110°C. The basis for setting the temperature threshold is that, under the above-mentioned high-altitude and low-pressure environment, the stator surface temperature of 110°C is sufficient to cause the coolant that has leaked into the motor sealing cavity to vaporize rapidly, generating high-temperature ethylene glycol vapor, which in turn poses a potential risk of chemical swelling to the stator potting compound. Step S2: Under special working condition monitoring signal, the absolute air pressure value in the motor sealed cavity is collected in real time, and a low frequency detection signal is sent to the stator winding to obtain the low frequency parasitic capacitance value of the stator winding to the motor housing. In some specific embodiments, the process of real-time acquisition of the absolute air pressure value inside the motor sealing cavity in step S2 under special operating condition monitoring signals is as follows: Upon receiving the special working condition monitoring signal, the gas pressure inside the cavity is collected in real time by a miniature absolute pressure sensor pre-installed on the inner wall of the motor's sealed cavity, according to a preset air pressure sampling frequency. The collected pressure analog signal is then converted from analog to digital and low-pass filtered to filter out the high-frequency airflow disturbance noise caused by the high-speed rotation of the motor rotor, thus obtaining a smooth absolute pressure value. In some specific embodiments, the specific process of sending a low-frequency detection signal to the stator winding to obtain the low-frequency parasitic capacitance value of the stator winding to the motor housing in step S2 is as follows: The inverter inside the motor controller (MCU) sends a zero-sequence AC voltage signal with a preset frequency and preset amplitude in the basic three-phase pulse width modulation (PWM) waveform that maintains normal motor drive, and sends the zero-sequence AC voltage signal as the low-frequency detection signal to the stator winding. The zero-sequence leakage current signal of the stator winding flowing to the motor housing through the insulating medium (i.e., stator potting compound) is collected synchronously. Using an orthogonal demodulation algorithm, the zero-sequence leakage current signal is compared with the injected low-frequency detection signal to extract the capacitive reactive current component with a phase difference of 90 degrees from the low-frequency detection signal; the root mean square (RMS) operation is performed on the extracted capacitive reactive current component to calculate the effective value of the capacitive reactive current component. Based on the extracted effective value of the capacitive reactive current component, combined with the formula The low-frequency parasitic capacitance of the stator winding to the motor housing was calculated; wherein, The low-frequency parasitic capacitance value is... This is the effective value of the capacitive reactive current component. The preset frequency of the low-frequency detection signal. The preset effective value of the amplitude of the low-frequency detection signal; It should be noted that the preset frequency of the low-frequency detection signal is set, for example, between 10Hz and 50Hz. The reason for this setting is that when evaluating the chemical swelling of the stator potting compound due to the absorption of ethylene glycol vapor, the change in dielectric constant is mainly reflected in the drift of capacitive reactance. Specifically, if a high-frequency signal is injected, the inductive reactance of the stator winding itself will increase, thereby masking the small capacitance change characteristics. If a DC signal is injected, only the insulation resistance can be detected, and capacitance information cannot be obtained. Using a low-frequency AC signal of 10Hz to 50Hz can effectively avoid the interference of the motor's fundamental operating frequency, while maximizing the acquisition of the small parasitic capacitance changes caused by the swelling of the potting compound, ensuring high sensitivity and accuracy of the detection. Step S3: Based on the response curve of the motor's real-time heating power and stator temperature, identify the current thermal time constant of the stator winding; In some specific embodiments, in step S3, based on the response curve of the motor's real-time heating power and stator temperature, the specific process of constructing the response curve is as follows: The effective value of the three-phase stator current during the motor operation is collected in real time, and combined with the phase resistance value of the stator winding at the current temperature, the real-time copper loss heat generation of the motor is calculated, and the real-time copper loss heat generation is used as the real-time heating power of the motor. The rate of change of the real-time heating power of the motor is monitored in real time. When the rate of change of the real-time heating power of the motor is greater than or equal to the preset heating power change threshold, a thermal response recording signal is generated. Under the thermal response recording signal, the moment when the rate of change reaches the threshold of the heating power change is taken as the time zero point. In the subsequent preset time window, the rise value of the motor stator temperature is continuously recorded according to the preset time step. Simultaneously monitor the real-time heating power of the motor within the preset time window. If the fluctuation range of the real-time heating power of the motor within the preset time window is always less than the preset steady-state tolerance range, it is determined that the heating power is stable, and the recorded data is used to construct the response curve with time as the horizontal axis and motor stator temperature as the vertical axis. It should be noted that when the rate of change of the real-time heating power of the motor is greater than or equal to the heating power change threshold, and then remains stable within the preset time window, combined with the high altitude conditions already met in step S1, it indicates that the vehicle is currently in a high heat load continuous operation condition of heavy load long downhill or continuous steep uphill on the plateau. It should be further explained that the threshold for change in heat generation power is set, for example, 30% / second of the rated power of the motor, the preset time window is set, for example, 60 seconds, and the steady-state tolerance range is set, for example, ±5% of the average heat generation power after the step. The physical basis for this setting is that only when the heat generation power experiences a sufficiently large step surge (≥30% / second) and is applied for a sufficiently long time (60 seconds) can the stator winding generate a sufficiently obvious transient temperature rise response, thereby ensuring a sufficiently high signal-to-noise ratio and fitting accuracy when identifying the thermal time constant. At the same time, since the actual output power of the motor is difficult to maintain an absolutely constant value due to the small undulations of the road surface and the driver's fine adjustment of the accelerator pedal during the actual uphill process, setting a steady-state tolerance range of ±5% can cover the small power fluctuations under actual driving conditions, avoid frequent interruptions of the algorithm, and ensure that the overall heat source input within the time window is approximately an ideal step constant heat source. In some specific embodiments, the process of identifying the current thermal time constant of the stator winding in step S3 is as follows: Establish the transient temperature rise equation for the stator winding: ,in, for The stator temperature of the motor at any given time. The initial stator temperature of the motor at time zero. This refers to the real-time heating power of the motor. The equivalent thermal resistance from the stator winding to the cooling system. The thermal time constant; The least squares curve fitting algorithm is used to substitute the constructed response curve into the transient temperature rise equation for parameter identification, and the parameters in the equation are solved. ; Extract the solved parameters That is, the time required for the stator temperature to rise to a steady-state temperature rise of 63.2%, which is used as the current thermal time constant of the stator winding; For example, the specific process for calculating the current thermal time constant is as follows: From the constructed response curve, N measured data points are extracted within the preset time window (e.g., 60 seconds) according to a preset time step of 1 second, and denoted as N. ,in Sampling time, This represents the measured stator temperature of the motor at that moment. Constructing the objective error function of the nonlinear least squares method The specific expression is: ,in, Given the initial stator temperature of the motor, Given the known real-time heating power of the motor. and These are the unknown parameters to be identified; Retrieve the factory-stored equivalent thermal resistance reference value and factory-stored time constant reference value from the system, respectively, as parameters. and The initial estimate of the iteration is obtained by using the Gauss-Newton iterative algorithm or gradient descent method to calculate the target error function. Perform differentiation and matrix operations, and continuously update. and The value of makes the target error function The value converges in the decreasing direction when the result is obtained after two consecutive iterations. The absolute value of the difference is less than the preset convergence accuracy threshold (e.g., If the iteration count reaches the preset maximum number of iterations (e.g., 100), stop the iteration and extract the final converged parameters at the time of stopping the iteration. The value of is used as the current thermal time constant of the stator winding; It should be noted that the physical basis for evaluating the potting compound state by identifying the current thermal time constant is as follows: Under normal conditions, the stator potting compound is dense and has good thermal conductivity, allowing heat to be quickly conducted to the motor housing and coolant, resulting in a small thermal time constant. When coolant micro-leakage and vaporization occur, the high-temperature ethylene glycol vapor causes the stator potting compound to chemically swell, creating micropores or loosening the structure within the compound, leading to a decrease in thermal conductivity. According to thermodynamic principles, an increase in thermal resistance directly affects the thermal time constant. The extension of the time constant. By identifying the thermal time constant, the physical structure degradation of the potting compound due to swelling can be obtained from a thermodynamic perspective; Step S4: Establish a time-series curve of absolute pressure value and determine whether the absolute pressure value stabilizes at the plateau segment of ethylene glycol saturated vapor pressure corresponding to the current temperature after the increase; if so, it is determined that coolant vaporization has occurred. In some specific embodiments, the process of establishing the time-series variation curve of the absolute pressure value and determining whether the absolute pressure value stabilizes at the plateau segment of the ethylene glycol saturated vapor pressure corresponding to the current temperature in step S4 is as follows: Using time as the horizontal axis and the absolute pressure value collected in real time in step S2 as the vertical axis, a time-series variation curve of the absolute pressure value is constructed. The current stator temperature of the motor is obtained synchronously, and the pre-calibrated and stored Antoine constant is retrieved; the motor stator temperature and the Antoine constant are substituted into the preset Antoine equation to calculate the theoretical saturated vapor pressure of ethylene glycol at the current temperature. For example, the specific process for calculating the theoretical saturated vapor pressure of ethylene glycol at the current temperature is as follows: Retrieve Antoine constants A, B, and C, which are pre-stored in the non-volatile memory of the motor controller; It should be noted that the Antoine constant A, Antoine constant B, and Antoine constant C are a set of thermodynamic empirical constants obtained by fitting and calibrating the coolant of a specific concentration (e.g., a 50% volume ratio ethylene glycol aqueous solution) actually added to the motor before it leaves the factory through a vaporization experiment in a constant temperature sealed container and using the least squares method. The current stator temperature of the motor is obtained as the independent variable. Substituting into the pre-defined Antoine equation, the standard mathematical expression of the Antoine equation is: in, For the target vapor pressure, The current stator temperature of the motor (in degrees Celsius, °C); Performing the antilogarithmic operation on the Antoine equation, i.e., using the formula Calculate the target vapor pressure The value; Based on the pressure unit system used when calibrating the Antoine constant, if the target vapor pressure is obtained... If the unit is millimeters of mercury (mmHg) or bar, multiply it by the corresponding standard physical conversion factor to convert it to kilopascals (kPa), and use the converted value as the theoretical saturated vapor pressure of ethylene glycol at the current temperature. The first derivative of the time-series variation curve is calculated to obtain the real-time rate of change of the absolute air pressure value. When the real-time change is detected to first exceed a preset rate of increase threshold (indicating that the air pressure is rising rapidly), and then within a continuous preset observation time window, the absolute value of the real-time rate of change is continuously less than a preset pressure fluctuation threshold (e.g., set to ±2 kPa), and the absolute value of the difference between the absolute air pressure value and the calculated theoretical ethylene glycol saturated vapor pressure is less than a preset matching error threshold (e.g., set to 5 kPa), it is determined that the absolute air pressure value has stabilized at the ethylene glycol saturated vapor pressure corresponding to the current temperature after the increase, thus forming the plateau segment; In some specific embodiments, in step S4, if so, the specific process of determining that coolant vaporization has occurred is as follows: When the logical judgment confirms that the platform segment with the above characteristics appears in the time-series change curve, the system determines that the gas pressure change pattern in the current motor sealing cavity conforms to the phase change thermodynamic characteristics of liquid-to-gas phase transition. It should be noted that if the increase in air pressure inside the motor's sealed cavity is solely due to thermal expansion and contraction caused by the heating of the internal dry air, the air pressure will rise linearly or quasi-linearly with the increase in temperature, and will not stagnate at a specific pressure value. Only when there is a slight leakage of coolant inside the cavity will the liquid boil and vaporize as the temperature rises, causing the air pressure to rise sharply. When the vapor partial pressure inside the cavity reaches the saturated vapor pressure at that temperature, vaporization and liquefaction reach a dynamic equilibrium, and the air pressure will no longer continue to rise, thus exhibiting a unique plateau segment characteristic on the curve. When the platform segment is identified, the fault diagnosis module of the motor controller sets the internal vaporization flag to a valid state (i.e., Boolean truth value), and formally determines that coolant vaporization has occurred inside the motor. Step S5: If coolant vaporization occurs, calculate the drift rate of the low-frequency parasitic capacitance and the increase ratio of the current thermal time constant relative to the factory time constant reference value. In some specific embodiments, in step S5, if coolant vaporization occurs, the specific process for calculating the drift rate of the low-frequency parasitic capacitance value and the increase ratio of the current thermal time constant relative to the factory time constant reference value is as follows: When the output of step S4 indicates that coolant vaporization has occurred inside the motor, the factory low-frequency parasitic capacitance reference value is obtained. This reference value is the healthy capacitance value measured when the motor is in the factory calibration stage, without any coolant leakage and at the same current motor stator temperature. The low-frequency parasitic capacitance value obtained in real time in step S2 is compared with the factory low-frequency parasitic capacitance reference value. The difference is then divided by the factory low-frequency parasitic capacitance reference value to calculate the drift rate of the low-frequency parasitic capacitance value. It should be noted that the physical basis for calculating the drift rate is as follows: when the stator potting compound is exposed to high-temperature ethylene glycol vapor for a long time and undergoes chemical swelling, the internal polymer cross-linking network will loosen, the inter-chain segment spacing will increase, and polar molecules, namely ethylene glycol and water, will penetrate into the interior of the colloid in large quantities. Since the dielectric constant of polar molecules is much greater than that of dry epoxy resin, this will lead to an increase in the overall equivalent dielectric constant of the potting compound, which will manifest as a significant positive drift in the low-frequency parasitic capacitance value of the stator winding to the motor housing in the macroscopic electrical characteristics. Synchronously acquire the factory time constant reference value, which is the standard thermal time constant identified by the same first-order lumped parameter thermodynamic model when the motor is in factory healthy condition; The difference between the current thermal time constant identified in step S3 and the factory time constant reference value is calculated, and the difference is divided by the factory time constant reference value to calculate the increase ratio of the current thermal time constant relative to the factory time constant reference value.

[0023] Step S6: Weight the drift rate and the increase ratio to calculate the medium swelling damage index of the stator potting compound. When the medium swelling damage index is greater than the preset medium swelling damage index threshold, output a coolant leakage alarm and limit the peak operating temperature of the motor.

[0024] In some specific embodiments, the specific process of weighted fusion of the drift rate and the increase ratio in step S6 is as follows: Obtain pre-calibrated capacitance feature weighting coefficients and thermal feature weighting coefficients, wherein the sum of the capacitance feature weighting coefficients and the thermal feature weighting coefficients is 1; Multiply the drift rate calculated in step S5 by the capacitor characteristic weighting coefficient to obtain the capacitor degradation component; at the same time, multiply the increase ratio calculated in step S5 by the thermal characteristic weighting coefficient to obtain the thermal resistance degradation component. The capacitance degradation component and the thermal resistance degradation component are linearly added to complete the weighted fusion of the drift rate and the increase ratio. It should be noted that the capacitance characteristic weighting coefficient is set to, for example, 0.6, and the thermal characteristic weighting coefficient is set to, for example, 0.4. The physical basis for the weighting allocation is as follows: In the early stage of chemical swelling of the stator potting compound, the infiltration of polar molecules will rapidly cause a sudden change in the dielectric constant. Therefore, the drift rate of low-frequency parasitic capacitance is more sensitive to early micro-swelling and should be given a higher weight. As the swelling intensifies, the macroscopic physical structure of the colloid becomes loose, leading to a decrease in thermal conductivity. At this time, the increase in the thermal time constant begins to show a significant effect. Therefore, it is given a relatively minor but indispensable weight. The combination of the two can achieve accurate quantification of the entire swelling life cycle. In some specific embodiments, in step S6, the process of calculating the medium swelling damage index of the stator potting compound, and outputting a coolant leakage alarm when the medium swelling damage index is greater than a preset medium swelling damage index threshold, and limiting the peak operating temperature of the motor, is as follows: The numerical result obtained after the above weighted fusion is used as the medium swelling damage index of the stator potting compound; The medium swelling damage index is compared with the preset medium swelling damage index threshold in real time. It should be noted that the preset dielectric swelling damage index threshold is set to, for example, 0.15 (i.e., the overall deterioration degree reaches 15%). The basis for setting the threshold is that when the overall swelling deterioration degree of epoxy resin potting compound exceeds 15%, its mechanical adhesion and insulation dielectric strength will enter a nonlinear accelerated decay period. If it continues to operate under high load and high vibration conditions, it is very easy to cause large-area insulation enamel wire peeling or phase-to-phase short circuit breakdown. When the medium swelling damage index is greater than the preset medium swelling damage index threshold, the motor controller sends the coolant leakage alarm to the vehicle controller (VCU) and instrument panel via the vehicle CAN bus to remind the driver. At the same time, the motor controller actively modifies the internal thermal derating control strategy to forcibly reduce the peak operating temperature of the motor from the rated upper limit under normal operating conditions (e.g., 150°C) and limit it to the safe protection temperature (e.g., 90°C). It should be noted that the physical basis for limiting the peak operating temperature to 90°C is that 90°C is lower than the boiling point of the motor coolant (ethylene glycol aqueous solution) under standard atmospheric pressure and slightly low pressure. By limiting the motor's heating, the continued boiling and vaporization of residual leaked coolant can be cut off from the thermodynamic source, thereby effectively curbing the continuous chemical swelling and damage of the stator potting compound by high-temperature steam.

[0025] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely illustrative of the principles of the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the present invention as claimed. The scope of protection of the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for detecting automotive motor faults based on multi-parameter acquisition, characterized in that: include: Step S1: Obtain the vehicle's current altitude and motor stator temperature. When the altitude is greater than a preset altitude threshold and the motor stator temperature is greater than a preset temperature threshold, generate a special working condition monitoring signal. Step S2: Under special working condition monitoring signal, the absolute air pressure value in the motor sealed cavity is collected in real time, and a low frequency detection signal is sent to the stator winding to obtain the low frequency parasitic capacitance value of the stator winding to the motor housing. Step S3: Based on the response curve of the motor's real-time heating power and stator temperature, identify the current thermal time constant of the stator winding; Step S4: Establish a time-series curve of absolute pressure value and determine whether the absolute pressure value stabilizes at the plateau segment of ethylene glycol saturated vapor pressure corresponding to the current temperature after the increase; if so, it is determined that coolant vaporization has occurred. Step S5: If coolant vaporization occurs, calculate the drift rate of the low-frequency parasitic capacitance and the increase ratio of the current thermal time constant relative to the factory time constant reference value. Step S6: Weight the drift rate and the increase ratio to calculate the medium swelling damage index of the stator potting compound. When the medium swelling damage index is greater than the preset medium swelling damage index threshold, output a coolant leakage alarm and limit the peak operating temperature of the motor.

2. The method for detecting automotive motor faults based on multi-parameter acquisition according to claim 1, characterized in that, The specific process for obtaining the vehicle's current altitude and the motor stator temperature is as follows: The vehicle's real-time three-dimensional coordinate positioning data is obtained through the vehicle-mounted satellite positioning system, and the vehicle's current altitude is obtained from the three-dimensional coordinate positioning data. Multiple thermistor sensors are pre-installed at the ends of the motor stator windings and in the stator core slots to collect multi-point temperature values ​​in real time. After performing a moving average filtering process on the multi-point temperature values, the maximum value among the multi-point temperature values ​​is extracted as the motor stator temperature.

3. The method for detecting automotive motor faults based on multi-parameter acquisition according to claim 1, characterized in that, The specific process for generating the special working condition monitoring signal is as follows: The obtained altitude is compared with the preset altitude threshold, and the obtained motor stator temperature is compared with the preset temperature threshold. When the altitude is greater than a preset altitude threshold and the motor stator temperature is greater than a preset temperature threshold, a special operating condition monitoring signal is generated.

4. The method for detecting automotive motor faults based on multi-parameter acquisition according to claim 1, characterized in that, The specific process for collecting the absolute pressure value is as follows: Under special operating conditions, the absolute air pressure value is collected in real time by a miniature absolute pressure sensor installed on the inner wall of the motor's sealed cavity.

5. The method for detecting automotive motor faults based on multi-parameter acquisition according to claim 1, characterized in that, The specific process for obtaining the low-frequency parasitic capacitance value of the stator winding to the motor housing is as follows: The inverter inside the motor controller MCU sends a zero-sequence AC voltage signal with a preset frequency and preset amplitude in a three-phase pulse width modulation (PWM) waveform, and sends the zero-sequence AC voltage signal as a low-frequency detection signal to the stator winding. The zero-sequence leakage current signal flowing from the stator winding to the motor housing is collected. The phase of the zero-sequence leakage current signal is compared with the low-frequency detection signal to obtain the capacitive reactive current component with a phase difference of 90 degrees from the low-frequency detection signal. The root mean square (RMS) operation is then performed to obtain the effective value of the capacitive reactive current component. Based on the effective value of the capacitive reactive current component, combined with Calculate the low-frequency parasitic capacitance of the stator windings to the motor housing. This is the low-frequency parasitic capacitance value. This represents the effective value of the capacitive reactive current component. The preset frequency for low-frequency detection signals. This is the preset effective value of the amplitude of the low-frequency detection signal.

6. The method for detecting automotive motor faults based on multi-parameter acquisition according to claim 1, characterized in that, The specific process for constructing the response curve is as follows: The effective value of the three-phase stator current during motor operation is collected in real time. Combined with the phase resistance value of the stator winding at the current temperature, the real-time copper loss heat generation of the motor is calculated as the real-time heating power of the motor. When the rate of change of the real-time heat power of the motor is greater than or equal to the threshold of heat power change, the time when the rate of change reaches the threshold of heat power change is taken as the zero point, and the stator temperature of the motor is recorded within the preset time window. If the fluctuation range of the real-time heating power of the motor is less than the steady-state tolerance range within the preset time window, the heating power is determined to be stable, and a response curve is constructed with time as the horizontal axis and motor stator temperature as the vertical axis.

7. The method for detecting automotive motor faults based on multi-parameter acquisition according to claim 1, characterized in that, The specific process for identifying the current thermal time constant of the stator winding is as follows: Establish the transient temperature rise equation for the stator winding: , for The stator temperature of the motor at any given time. The initial stator temperature of the motor at time zero. This refers to the real-time heating power of the motor. The equivalent thermal resistance from the stator winding to the cooling system. The thermal time constant; The least squares curve fitting algorithm is used to substitute the constructed response curve into the transient temperature rise equation for solving, and the solved thermal time constant is obtained. As the current thermal time constant of the stator winding.

8. The method for detecting automotive motor faults based on multi-parameter acquisition according to claim 1, characterized in that, The specific process for determining that coolant vaporization has occurred is as follows: Using time as the horizontal axis and the real-time collected absolute air pressure value as the vertical axis, a time-series variation curve of the absolute air pressure value is constructed to obtain the current motor stator temperature and the Antoine constant. The motor stator temperature and the Antoine constant are substituted into the Antoine equation to calculate the theoretical saturated vapor pressure of ethylene glycol at the current temperature. The first derivative of the time-series change curve is calculated to obtain the real-time rate of change of the absolute pressure value. When the real-time change first exceeds the rate of increase threshold, and within the observation time window, the absolute value of the real-time rate of change is less than the pressure fluctuation threshold, and the absolute value of the difference between the absolute pressure value and the theoretical ethylene glycol saturated vapor pressure is less than the matching error threshold, it is determined that the absolute pressure value has risen and stabilized at the ethylene glycol saturated vapor pressure corresponding to the current temperature, forming a plateau segment, i.e., coolant vaporization has occurred.

9. The method for detecting automotive motor faults based on multi-parameter acquisition according to claim 1, characterized in that, The specific process for calculating the drift rate of the low-frequency parasitic capacitance and the increase ratio of the current thermal time constant relative to the factory time constant reference value is as follows: When coolant vaporization is determined to have occurred: Obtain the factory-set low-frequency parasitic capacitance reference value, calculate the difference between the low-frequency parasitic capacitance value and the factory-set low-frequency parasitic capacitance reference value, and divide the resulting difference by the factory-set low-frequency parasitic capacitance reference value to obtain the drift rate of the low-frequency parasitic capacitance value. Obtain the factory time constant reference value, calculate the difference between the current thermal time constant and the factory time constant reference value, and divide the resulting difference by the factory time constant reference value to obtain the increase ratio of the current thermal time constant relative to the factory time constant reference value.

10. The method for detecting automotive motor faults based on multi-parameter acquisition according to claim 1, characterized in that, The specific process for the output coolant leakage alarm is as follows: Obtain the pre-calibrated capacitance characteristic weighting coefficient and thermal characteristic weighting coefficient. The sum of the capacitance characteristic weighting coefficient and thermal characteristic weighting coefficient is 1. Multiply the drift rate by the capacitance characteristic weighting coefficient to obtain the capacitance degradation component. Multiply the increase ratio by the thermal characteristic weighting coefficient to obtain the thermal resistance degradation component. The capacitance degradation component and the thermal resistance degradation component are added together to complete the weighted fusion of drift rate and increase ratio. The numerical result obtained after weighted fusion is used as the medium swelling damage index of stator potting compound. When the medium swelling damage index is greater than the preset medium swelling damage index threshold, a coolant leakage alarm is output.

Images