Abnormity diagnosis method, device and equipment for vehicle-mounted motor

Abstract

The invention provides an abnormality diagnosis method, device and equipment for a vehicle-mounted motor, and is applied to the technical field of motor control. The method comprises the steps that motor signals of a vehicle-mounted motor in the operation process are collected; the motor signal comprises a voltage signal, a current signal and a temperature signal. And for each acquisition moment, fusing a voltage state feature vector extracted from the voltage signal and an auxiliary state feature vector extracted from the current signal and the temperature signal to obtain a comprehensive state feature vector of the vehicle-mounted motor at the acquisition moment. And determining a time sequence accumulation value corresponding to each fault type in the preset fault types according to the comprehensive state feature vectors at the plurality of acquisition moments. And according to the time sequence accumulated value corresponding to each fault type, determining an abnormality diagnosis result of the vehicle-mounted motor, so that the problem of relatively low accuracy of a traditional abnormality diagnosis method of the vehicle-mounted motor can be solved.

Description

Technical Field

[0001] This application relates to the field of motor control technology, and in particular to a method, apparatus and equipment for diagnosing abnormalities in vehicle motors. Background Technology

[0002] With the rapid development of new energy vehicles, on-board motors have become a core component affecting the safety and performance of the entire vehicle. Traditional motor anomaly diagnosis methods mainly rely on AI models running on high-performance computing platforms, which suffer from high latency, high power consumption, and high cost, making them difficult to deploy in real time in resource-constrained automotive-grade embedded systems. Furthermore, related technologies are mostly based on analysis of a single signal source, lacking adaptability to complex and changing motor operating conditions and unknown fault modes, resulting in low diagnostic accuracy and robustness. In addition, traditional diagnostic models lack online learning and adaptive update capabilities, failing to effectively address feature drift caused by motor aging, environmental changes, and other factors, limiting their long-term reliability.

[0003] Therefore, there is an urgent need for a method, device, and equipment for diagnosing abnormalities in vehicle motors to solve the problem of low accuracy in traditional methods for diagnosing abnormalities in vehicle motors. Summary of the Invention

[0004] The purpose of this application is to provide a method, device, and equipment for diagnosing abnormalities in vehicle motors, thereby solving the problem of low accuracy in traditional methods for diagnosing abnormalities in vehicle motors.

[0005] In a first aspect, embodiments of this application provide a method for diagnosing anomalies in an on-board motor. The method includes: acquiring motor signals from the on-board motor during operation. The motor signals include voltage signals, current signals, and temperature signals. For each acquisition moment, a comprehensive state feature vector of the on-board motor at the acquisition moment is obtained by fusing the voltage state feature vector extracted from the voltage signal and the auxiliary state feature vectors extracted from the current and temperature signals. Based on the comprehensive state feature vectors from multiple acquisition moments, a time-series cumulative value corresponding to each preset fault type is determined. Based on the time-series cumulative value corresponding to each fault type, an anomaly diagnosis result for the on-board motor is determined.

[0006] The vehicle motor anomaly diagnosis method provided in this application collects motor signals during vehicle motor operation. For each collection moment, it fuses the voltage state feature vector extracted from the voltage signal with the auxiliary state feature vectors extracted from the current and temperature signals to obtain a comprehensive state feature vector of the vehicle motor at the collection moment. Then, it accumulates the comprehensive state features from multiple collection moments over time to obtain a time-series cumulative value corresponding to each preset fault type. Finally, it performs confidence calculation and dual threshold judgment based on the cumulative value. One possible implementation involves determining the time-series cumulative value for each fault type among preset fault types based on the comprehensive state feature vectors from multiple acquisition times. This includes: combining the comprehensive state feature vectors from multiple acquisition times according to time sequence to obtain a comprehensive state feature time-series sequence. For any target fault type among the fault types, the time-series cumulative value of the target fault type is determined based on the continuous state of the target fault type in the comprehensive state feature time-series sequence.

[0007] One possible implementation involves determining the time-series cumulative value of the target fault type based on the continuous state of the target fault type in the integrated state characteristic time-series sequence. This includes: when it is determined that the target fault type occurs continuously in the integrated state characteristic time-series sequence, determining the time-series cumulative value of the target fault type based on the initial value configured for the target fault type and the duration of the target fault type.

[0008] One possible implementation of the vehicle motor anomaly diagnosis method provided in this application embodiment may further include: when it is determined that a sudden fault feature appears in the comprehensive state feature time sequence at any acquisition time, adding a time sequence increment to the time sequence cumulative value of the matching target fault type based on the sudden fault feature, and obtaining the incrementally updated time sequence cumulative value.

[0009] One possible implementation involves determining the anomaly diagnosis result of the on-board motor based on the time-series cumulative value corresponding to each fault type. This includes: determining the confidence level of the on-board motor under each fault type based on the time-series cumulative value corresponding to each fault type; and determining the anomaly diagnosis result of the on-board motor based on the confidence level of each fault type.

[0010] One possible implementation involves determining the abnormal diagnosis result of the on-board motor based on the confidence level of each fault type. This includes: determining the highest confidence level, the candidate fault type to which the highest confidence level belongs, and the second highest confidence level, based on the confidence level of each fault type. If the highest confidence level is greater than a first threshold, and the difference between the highest and second highest confidence levels is greater than a second threshold, the abnormal diagnosis result is determined as a candidate fault type.

[0011] One possible implementation involves fusing the voltage state feature vector extracted from the voltage signal with auxiliary state feature vectors extracted from the current and temperature signals to obtain a comprehensive state feature vector of the on-board motor at the time of data acquisition. This includes: converting the voltage signal to obtain two-dimensional voltage data containing a first voltage component and a second voltage component; determining the voltage state feature vector based on the two-dimensional voltage data; extracting the fundamental amplitude of the current signal from the current signal; determining the measured temperature of the on-board motor based on the temperature signal and a temperature compensation coefficient matching the current environment of the on-board motor; determining the auxiliary state feature vector based on the fundamental amplitude of the current signal and the measured temperature; and concatenating the voltage state feature vector and the auxiliary state feature vector to obtain the comprehensive state feature vector.

[0012] Secondly, embodiments of this application provide an abnormal diagnosis device for an on-board motor, the device including a data acquisition module, a fusion module, and a determination module.

[0013] The acquisition module is used to collect motor signals from the vehicle's motor during operation. These motor signals include voltage, current, and temperature signals.

[0014] The fusion module is used to fuse the voltage state feature vector extracted from the voltage signal and the auxiliary state feature vector extracted from the current signal and temperature signal for each acquisition time to obtain the comprehensive state feature vector of the vehicle motor at the acquisition time.

[0015] The determination module is used to determine the time-series cumulative value corresponding to each fault type in the preset fault types based on the comprehensive state feature vector at multiple acquisition times.

[0016] The determination module is also used to determine the abnormal diagnosis result of the on-board motor based on the timing cumulative value corresponding to each fault type.

[0017] One possible implementation involves determining the time-series cumulative value for each fault type among preset fault types based on the comprehensive state feature vectors from multiple acquisition times. This includes: combining the comprehensive state feature vectors from multiple acquisition times according to time sequence to obtain a comprehensive state feature time-series sequence. For any target fault type among the fault types, the time-series cumulative value of the target fault type is determined based on the continuous state of the target fault type in the comprehensive state feature time-series sequence.

[0018] One possible implementation is that, when determining the time-series cumulative value of a target fault type based on the continuous state of the target fault type in the comprehensive state characteristic time-series sequence, the determining module is specifically used to: determine the time-series cumulative value of the target fault type based on the initial value configured for the target fault type and the duration of the target fault type when the target fault type is determined to occur continuously in the comprehensive state characteristic time-series sequence.

[0019] In one possible implementation, the vehicle motor anomaly diagnosis device provided in this application embodiment can also be used to: when it is determined that a sudden fault feature appears in the comprehensive state feature time sequence at any acquisition time, add a time sequence increment to the time sequence cumulative value of the matching target fault type based on the sudden fault feature, and obtain the time sequence cumulative value after incremental update.

[0020] One possible implementation involves the module determining the anomaly diagnosis result of the on-board motor based on the time-series cumulative value corresponding to each fault type. Specifically, this involves: determining the confidence level of the on-board motor under each fault type based on the time-series cumulative value corresponding to each fault type; and determining the anomaly diagnosis result of the on-board motor based on the confidence level of each fault type.

[0021] One possible implementation involves a module that determines the abnormal diagnosis result of the on-board motor based on the confidence level of each fault type. Specifically, this involves: determining the highest confidence level, the candidate fault type to which the highest confidence level belongs, and the second highest confidence level for each fault type. If the highest confidence level is greater than a first threshold, and the difference between the highest and second highest confidence levels is greater than a second threshold, the abnormal diagnosis result is determined as a candidate fault type.

[0022] One possible implementation involves a fusion module that fuses voltage state feature vectors extracted from voltage signals with auxiliary state feature vectors extracted from current and temperature signals to obtain a comprehensive state feature vector of the on-board motor at the time of data acquisition. Specifically, this involves: converting the voltage signal to obtain two-dimensional voltage data containing a first voltage component and a second voltage component; determining the voltage state feature vector based on the two-dimensional voltage data; extracting the fundamental amplitude of the current signal from the current signal; determining the measured temperature of the on-board motor based on the temperature signal and a temperature compensation coefficient matching the current environment of the on-board motor; determining the auxiliary state feature vector based on the fundamental amplitude of the current signal and the measured temperature; and concatenating the voltage state feature vector and the auxiliary state feature vector to obtain the comprehensive state feature vector.

[0023] Thirdly, embodiments of this application provide an on-board motor anomaly diagnosis device, which has the function of implementing the on-board motor anomaly diagnosis method of the first aspect or any possible implementation thereof. This function can be implemented by hardware or by hardware executing corresponding software. The hardware or software includes one or more modules corresponding to the above-described function.

[0024] Fourthly, embodiments of this application provide a computer-readable storage medium storing instructions that, when executed on a computer, enable the computer to perform the on-board motor anomaly diagnosis method of the first aspect or any possible implementation thereof.

[0025] Fifthly, embodiments of this application provide a computer program product containing instructions that, when run on a computer, enable the computer to execute the on-board motor anomaly diagnosis method described in the first aspect or any possible implementation thereof.

[0026] The technical effects of any of the design methods in aspects two through five can be found in aspect one or in different possible implementations of aspect one, and will not be repeated here. Attached Figure Description

[0027] To more clearly illustrate the technical solutions in the specific embodiments of this application or the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0028] Figure 1 A system structure diagram of an abnormal diagnosis system for an on-board motor provided in this application embodiment; Figure 2 Another structural schematic diagram of an abnormal diagnosis system for an on-board motor provided in this application embodiment; Figure 3 A flowchart illustrating a method for diagnosing anomalies in an on-board motor, as provided in an embodiment of this application; Figure 4 A schematic diagram of a fault diagnosis device for an on-board motor provided in an embodiment of this application; Figure 5 Another system architecture diagram of an abnormal diagnosis system for an on-board motor provided in this application embodiment. Detailed Implementation

[0029] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. The components of the embodiments of this application described and shown in the accompanying drawings can generally be arranged and designed in various different configurations.

[0030] Therefore, the following detailed description of the embodiments of this application provided in the accompanying drawings is not intended to limit the scope of the claimed application, but merely to illustrate selected embodiments of the application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without inventive effort are within the scope of protection of this application.

[0031] With the continuous improvement of the intelligence level of new energy vehicles, the real-time monitoring and anomaly diagnosis of the on-board motor, as the core actuator of the power system, has become a key technology for ensuring vehicle safety and reliability. Currently, anomaly diagnosis technologies for on-board motors are mainly divided into the following categories: The first category is based on traditional signal analysis methods, such as time-domain analysis of current and voltage waveforms, or frequency-domain analysis using Fourier transform (FFT) to identify characteristic frequency components. While these methods are relatively simple to implement, their diagnostic capabilities heavily rely on expert experience and prior knowledge. They are not sensitive enough to complex, coupled, or early-stage faults, and are difficult to handle non-stationary operating conditions of motors.

[0032] The second category is based on physical models. This method establishes a precise mathematical model of the motor and compares the residuals between the model output and the actual measurements to determine anomalies. This type of method has high diagnostic accuracy for known fault modes, but it is highly dependent on the accuracy of the model and has difficulty modeling motor aging, manufacturing tolerances, and unknown fault types, resulting in poor adaptability in real-world complex vehicle environments.

[0033] The third category is rule-based or threshold-based methods, which use manually set alarm thresholds for parameters such as current and temperature to determine faults. This method is logically simple and has a fast response, but the rules are often too rigid, making it difficult to effectively distinguish between transient interference and real faults, resulting in a high false alarm rate, and it cannot achieve early warning and fine-grained classification of faults.

[0034] The fourth category of artificial intelligence-based methods (especially deep learning) demonstrates significant advantages, enabling them to automatically learn fault characteristics from high-dimensional data. However, existing AI diagnostic solutions typically rely on high-performance computing platforms (such as GPUs or high-end automotive ECUs) for model inference, resulting in high system costs, high power consumption, and inference latency often exceeding 100 milliseconds, failing to meet the stringent real-time requirements of high-speed automotive motor operation. Furthermore, existing models are mostly trained offline and deployed in a fixed manner, lacking the ability to continuously learn and adaptively update on the vehicle side, making it difficult to address the differences between different vehicle models and individual motors, as well as performance degradation issues during long-term operation.

[0035] Based on this, this application provides a method for abnormal diagnosis of an on-board motor. The method includes acquiring motor signals from the on-board motor during operation. The motor signals include voltage signals, current signals, and temperature signals. For each acquisition moment, a comprehensive state feature vector of the on-board motor at the acquisition moment is obtained by fusing the voltage state feature vector extracted from the voltage signal and the auxiliary state feature vectors extracted from the current and temperature signals. Based on the comprehensive state feature vectors from multiple acquisition moments, a time-series cumulative value corresponding to each preset fault type is determined. Based on the time-series cumulative value corresponding to each fault type, the abnormal diagnosis result of the on-board motor is determined.

[0036] The abnormal diagnosis method for vehicle-mounted motors provided in this application collects motor signals during operation. For each collection moment, it fuses the voltage state feature vector extracted from the voltage signal with auxiliary state feature vectors extracted from the current and temperature signals to obtain a comprehensive state feature vector of the vehicle-mounted motor at that moment. Then, it accumulates the comprehensive state features from multiple collection moments over time to obtain a time-series cumulative value for each preset fault type. Finally, it calculates confidence and performs dual threshold determination based on the cumulative value to obtain the diagnostic result for the vehicle-mounted motor, significantly improving the robustness and reliability of the diagnostic decision.

[0037] The methods provided in the embodiments of this application will now be described in conjunction with the specific accompanying drawings.

[0038] On the one hand, embodiments of this application provide an abnormality diagnosis system for vehicle-mounted motors. For example... Figure 1 As shown, the vehicle motor anomaly diagnosis system 100 may include a vehicle motor 110, a sensor array 120, a microcontroller 130, a vehicle communication network 140, and a vehicle controller 150.

[0039] Among them, the vehicle motor 110 is the object to be diagnosed. For example, the vehicle motor 110 may include a drive motor, a power steering motor, or a compressor motor.

[0040] The sensor array 120 can be mounted on the vehicle motor 110. The sensor array 120 is used to acquire motor signals of the vehicle motor 110 in real time during operation. For example, the sensor array 120 may include a voltage sensor, a current sensor, and a temperature sensor.

[0041] Specifically, voltage sensors can be installed on the DC bus and three-phase output terminals of the motor controller of the on-board motor 110 to collect bus voltage and phase voltage signals. Current sensors can be connected in series with the three-phase windings of the on-board motor 110 to collect current signals. Temperature sensors can be attached to the stator windings and bearing outer rings of the on-board motor 110 to collect temperature signals.

[0042] The microcontroller 130 is electrically connected to the sensor array 120. Specifically, the microcontroller 130 is used to execute the abnormal diagnosis method for the vehicle motor provided in the embodiments of this application, receive motor signals collected by the sensor array 120, fuse voltage, current and temperature signals, extract features, perform time-series cumulative calculation, and execute dual threshold judgment logic based on the calculation results to generate abnormal diagnosis results for the vehicle motor.

[0043] The vehicle communication network 140 is used to send the diagnostic results generated by the microcontroller 130 to the vehicle controller 150 in real time to trigger corresponding audible and visual alarms, power-limited operation, or maintenance prompts. For example, the vehicle communication network 140 can be a controller area network bus.

[0044] Furthermore, such as Figure 2 As shown, the vehicle motor anomaly diagnosis system 100 may also include a cloud platform 160.

[0045] The cloud platform 160 can communicate with the vehicle communication network 140, for example, through the vehicle T-BOX. The cloud platform 160 can receive long-term operating data uploaded from the vehicle, perform model retraining and optimization, and send the updated model parameters to the microcontroller 130 via over-the-air (OTA) download technology to achieve adaptive updates of the diagnostic algorithm.

[0046] It should be noted that the above Figure 1 and Figure 2 The illustrated vehicle motor anomaly diagnosis system 100 is merely an example illustrating the application scenario of the solution in this application, and is not intended to limit the application scenario of the solution in this application.

[0047] On the one hand, embodiments of this application provide a method for diagnosing abnormalities in vehicle-mounted motors, which can be performed by... Figure 1 The on-board motor fault diagnosis system 100 shown is executed. For example... Figure 3 As shown, the method may include the following steps.

[0048] S301 collects motor signals from the vehicle's onboard motor during operation.

[0049] The on-board motor can refer to any type of motor actually installed and operating in the vehicle. For example, on-board motors include, but are not limited to, drive motors (such as permanent magnet synchronous motors and induction motors) that serve as the vehicle's primary power source, electric power steering motors that provide steering assistance, and compressor motors that drive the air conditioning compressor. Motor signals include, but are not limited to, voltage signals, current signals, and temperature signals.

[0050] Specifically, when the vehicle motor is activated after the vehicle is powered on and performs its designed functions, the motor signals during operation are collected by corresponding sensors arranged on the motor and its controller.

[0051] One possible implementation involves capturing the voltage waveform of the on-board motor in real time using voltage sensors (e.g., differential amplifier circuits or isolation amplifiers) installed on the DC bus and three-phase output terminals of the motor controller. The current value of the on-board motor is measured in real time using current sensors (e.g., Hall effect current sensors or shunt resistors) connected in series on the three-phase power supply lines of the motor. Temperature changes in the on-board motor are sensed using temperature sensors (e.g., NTC thermistors or PT100 platinum resistance thermometers) installed at the stator winding ends and bearing outer rings.

[0052] S302, for each acquisition time, fuses the voltage state feature vector extracted from the voltage signal and the auxiliary state feature vector extracted from the current signal and temperature signal to obtain the comprehensive state feature vector of the vehicle motor at the acquisition time.

[0053] The acquisition time can be any point in time where the motor signal is discretized and sampled according to a preset sampling period. The voltage state feature vector is a multi-dimensional feature representation that characterizes the current electrical state of the motor, formed by specific transformations and feature extraction of the original voltage signal. The auxiliary state feature vector refers to the feature representation extracted from the current and temperature signals, used to assist in verification and supplement information.

[0054] One possible implementation involves fusing the voltage state feature vector extracted from the voltage signal with auxiliary state feature vectors extracted from the current and temperature signals to obtain a comprehensive state feature vector of the on-board motor at the time of data acquisition. This includes: converting the voltage signal to obtain two-dimensional voltage data containing a first voltage component and a second voltage component; determining the voltage state feature vector based on the two-dimensional voltage data; extracting the fundamental amplitude of the current signal from the current signal; determining the measured temperature of the on-board motor based on the temperature signal and a temperature compensation coefficient matching the current environment of the on-board motor; determining the auxiliary state feature vector based on the fundamental amplitude of the current signal and the measured temperature; and concatenating the voltage state feature vector and the auxiliary state feature vector to obtain the comprehensive state feature vector.

[0055] Specifically, the voltage signal is first converted to obtain two-dimensional voltage data containing a first voltage component and a second voltage component. For example, the collected three-phase voltage signal (Va, Vb, Vc) is transformed from a three-phase stationary coordinate system to a two-phase stationary α-β coordinate system using the Clarke transform, resulting in mutually orthogonal voltage components Vα and Vβ, which constitute two-dimensional voltage data.

[0056] Then, based on the two-dimensional voltage data, the voltage state feature vector is determined. For example, the two-dimensional voltage data (Vα, Vβ) is input into a pre-trained lightweight convolutional neural network. Through its preset convolutional kernels (e.g., designed to specifically capture harmonics of a specific order, negative sequence components, or voltage distortion patterns), it automatically extracts high-level, discriminative local feature patterns from the data and organizes the activation values ​​of these features into a fixed-dimensional vector to obtain the voltage state feature vector.

[0057] Simultaneously, the fundamental amplitude of the current signal is extracted based on the current signal. For example, a fast Fourier transform is performed on the acquired instantaneous current signal, or a phase-locked loop (PLL) technique is used to separate its fundamental component (usually 50Hz or a component corresponding to the motor's electrical frequency), and the amplitude of the fundamental component of the current signal is calculated.

[0058] Then, based on the temperature signal and the temperature compensation coefficient that matches the current environment of the vehicle motor, the measured temperature value of the vehicle motor is determined. For example, the original ADC value of the temperature sensor is read, combined with the temperature-voltage characteristic curve calibrated for the sensor model and installation location and the ambient temperature compensation coefficient (used to compensate for the drift of the sensor itself caused by changes in ambient temperature) stored in the memory, and the compensated, more accurate measured temperature value of the motor component is obtained through interpolation or formula calculation.

[0059] Next, an auxiliary state feature vector is determined based on the fundamental current amplitude and the measured temperature value. For example, the fundamental current amplitude and the measured temperature value can be directly combined into a low-dimensional vector as the auxiliary state feature vector.

[0060] Finally, the voltage state feature vector and the auxiliary state feature vector are concatenated to obtain the comprehensive state feature vector. For example, array or matrix concatenation operations (such as NumPy's concatenate function) can be used to link the high-dimensional voltage state feature vector with the low-dimensional auxiliary state feature vector, combining them into a higher-dimensional feature vector. This comprehensive state feature vector integrates the electrical details, current information, and temperature information of the motor, providing a comprehensive state description foundation for subsequent steps.

[0061] S303, based on the comprehensive state feature vector at multiple acquisition times, determines the time-series cumulative value corresponding to each fault type in the preset fault types.

[0062] The preset fault type is a predefined category of motor abnormality that needs to be diagnosed. For example, the preset fault type includes, but is not limited to, bearing fault, jamming fault, stator winding short circuit, over-temperature and overload. This application does not limit the preset fault type.

[0063] One possible implementation involves determining the time-series cumulative value for each fault type among preset fault types based on the comprehensive state feature vectors from multiple acquisition times. This includes: combining the comprehensive state feature vectors from multiple acquisition times according to time sequence to obtain a comprehensive state feature time-series sequence. For any target fault type among the fault types, the time-series cumulative value of the target fault type is determined based on the continuous state of the target fault type in the comprehensive state feature time-series sequence.

[0064] Specifically, the comprehensive state feature vectors from multiple acquisition times are first combined according to their temporal order to obtain a comprehensive state feature time series sequence. For any target fault type among the preset fault types, the temporal cumulative value of the target fault type is determined based on the degree of its continuous occurrence in the comprehensive state feature time series sequence.

[0065] For example, if the feature vector consistently exhibits a pattern highly correlated with a certain fault type over a period of time, then that fault type is considered to be in a continuously active state, and its cumulative value will gradually increase as this state continues. This process can simulate the development of progressive faults in vehicle motors (such as bearing wear and insulation aging). If the comprehensive state feature vector exhibits instantaneous and drastic abnormal changes, it may trigger a rapid adjustment of the cumulative value to respond to sudden faults.

[0066] Furthermore, the step of determining the time-series cumulative value of the target fault type based on the continuous state of the target fault type in the comprehensive state feature time-series includes: when it is determined that the target fault type occurs continuously in the comprehensive state feature time-series, determining the time-series cumulative value of the target fault type based on the initial value configured for the target fault type and the duration of the target fault type.

[0067] Specifically, when a target fault type appears consecutively in the time series of integrated state characteristics, its time-series cumulative value is calculated based on preset rules. For example, an initial cumulative value (initial value) can be configured for each fault type. When the integrated state feature vectors at multiple consecutive time points are analyzed and determined to indicate the same target fault type, it is considered that the fault type is in a state of continuous occurrence. Subsequently, based on the duration of this state, the time-series cumulative value of the target fault type is dynamically updated and determined according to a predetermined accumulation algorithm (e.g., adding a fixed step value to the initial value for each unit of time, or calculating according to an increment proportional to the duration).

[0068] Furthermore, when it is determined that a sudden fault feature appears in the comprehensive state feature time series at any acquisition time, a time series increment is added to the time series cumulative value of the matching target fault type based on the sudden fault feature to obtain the incrementally updated time series cumulative value.

[0069] Specifically, in order to capture sudden and severe faults, the system also monitors whether the integrated state characteristic time series exhibits abrupt fault characteristics at any acquisition time. These abrupt fault characteristics can be patterns of drastic and significant abnormal changes in the feature vector at a single or adjacent moment compared to normal operating conditions, such as a sudden drop or rise in voltage signal, or a sudden increase in the energy of a specific high-frequency component.

[0070] When such abrupt fault characteristics are detected, the target fault type matched by the abrupt characteristic can be determined according to a preset mapping relationship (e.g., voltage clamping abruptness matches an "inverter IGBT open circuit" fault). Subsequently, a preset timing increment is directly added to the current timing accumulation value of the matched fault type. This process can quickly improve the cumulative index of the fault type by assigning a timing increment to the abrupt fault characteristics, thereby achieving a sensitive response to sudden faults.

[0071] S304. Based on the timing cumulative value corresponding to each fault type, determine the abnormal diagnosis result of the vehicle motor.

[0072] One possible implementation involves determining the anomaly diagnosis result of the on-board motor based on the time-series cumulative value corresponding to each fault type. This includes: determining the confidence level of the on-board motor under each fault type based on the time-series cumulative value corresponding to each fault type; and determining the anomaly diagnosis result of the on-board motor based on the confidence level of each fault type.

[0073] Furthermore, based on the confidence level of each fault type, the abnormal diagnosis result of the on-board motor is determined, including: based on the confidence level of each fault type, determining the highest confidence level, the candidate fault type to which the highest confidence level belongs, and the second highest confidence level. If the highest confidence level is greater than a first threshold, and the difference between the highest confidence level and the second highest confidence level is greater than a second threshold, the abnormal diagnosis result is determined as a candidate fault type.

[0074] Specifically, the time-series cumulative values ​​(or their converted logit values) for each fault type are input into a temperature-compensated Softmax function to calculate the confidence level for each fault type. The specific calculation formula is as follows:

[0075] in, This represents the confidence level for the i-th fault type; This represents the time-series cumulative value for the i-th fault type; β is the preset temperature compensation coefficient, in units of 1 / ℃. ΔT is the difference between the current ambient temperature and the standard reference temperature (e.g., 25℃), in units of ℃. This represents the summation over all j preset fault types. This is the time-series cumulative value corresponding to the fault type. This calculation process ensures all confidence levels are met. The sum of these terms is 1, and the thermal stability of the model can be improved by dynamically adapting to temperature changes through the β·ΔT term.

[0076] Then, based on the confidence set for each fault type calculated in Phase 1 { }, to determine the final abnormal diagnosis result.

[0077] For example, first from { Identify the highest confidence level in} and their corresponding candidate fault types, and the second highest confidence level Then, only if both conditions are met ,and Only in this case will the candidate fault type be determined as the current anomaly diagnosis result. The first threshold can be set, for example, 0.7. This is the second threshold, for example, it can be 0.3.

[0078] If any condition is not met, for example, if the following occurs This indicates that the confidence level for the candidate fault type is insufficient. Or, if... In this situation, it indicates that the candidate fault type is not sufficiently distinguishable from other fault types. At this point, the current status can be determined as "normal," "status unclear," or "further monitoring required," thus avoiding the output of unreliable specific fault conclusions.

[0079] The above primarily describes the solutions provided in this application from the perspective of the device's working principle. It is understood that, in order to achieve the aforementioned functions, the on-board motor anomaly diagnosis device includes corresponding hardware structures and / or software modules for executing each function. Those skilled in the art should readily recognize that, in conjunction with the algorithm steps of the examples described in the embodiments disclosed herein, this application can be implemented in hardware or a combination of hardware and computer software. Whether a function is executed in hardware or by computer software driving hardware depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0080] This application embodiment can divide the vehicle motor anomaly diagnosis device into functional modules according to the above method example. For example, each function can be divided into a separate functional module, or two or more functions can be integrated into one processing module. The integrated module can be implemented in hardware or as a software functional module.

[0081] It should be noted that the module division in this embodiment is illustrative and represents only one logical functional division; in actual implementation, other division methods may be used. When dividing functional modules according to their respective functions, Figure 4 A schematic diagram of a possible configuration of the on-board motor fault diagnosis device involved in the above and embodiment is shown. Figure 4 As shown, the vehicle motor abnormality diagnosis device 400 may include: a data acquisition module 401, a fusion module 402, and a determination module 403.

[0082] The acquisition module 401 is used to support the execution of the on-board motor fault diagnosis device 400. Figure 3 S301 in the illustrated method for diagnosing abnormalities in vehicle-mounted motors.

[0083] Fusion module 402 is used to support the execution of the on-board motor fault diagnosis device 400. Figure 3 S302 in the illustrated method for diagnosing abnormalities in vehicle-mounted motors.

[0084] Determining module 403, used to support the execution of the on-board motor fault diagnosis device 400. Figure 3 S303 and S304 are illustrated in the abnormal diagnosis method of the vehicle motor.

[0085] One possible implementation involves determining the time-series cumulative value for each fault type among preset fault types based on the comprehensive state feature vectors from multiple acquisition times. This includes: combining the comprehensive state feature vectors from multiple acquisition times according to time sequence to obtain a comprehensive state feature time-series sequence. For any target fault type among the fault types, the time-series cumulative value of the target fault type is determined based on the continuous state of the target fault type in the comprehensive state feature time-series sequence.

[0086] One possible implementation is that, when determining the time-series cumulative value of a target fault type based on the continuous state of the target fault type in the comprehensive state characteristic time-series sequence, the determining module is specifically used to: determine the time-series cumulative value of the target fault type based on the initial value configured for the target fault type and the duration of the target fault type when the target fault type is determined to occur continuously in the comprehensive state characteristic time-series sequence.

[0087] In one possible implementation, the vehicle motor anomaly diagnosis device provided in this application embodiment can also be used to: when it is determined that a sudden fault feature appears in the comprehensive state feature time sequence at any acquisition time, add a time sequence increment to the time sequence cumulative value of the matching target fault type based on the sudden fault feature, and obtain the time sequence cumulative value after incremental update.

[0088] One possible implementation involves the module determining the anomaly diagnosis result of the on-board motor based on the time-series cumulative value corresponding to each fault type. Specifically, this involves: determining the confidence level of the on-board motor under each fault type based on the time-series cumulative value corresponding to each fault type; and determining the anomaly diagnosis result of the on-board motor based on the confidence level of each fault type.

[0089] One possible implementation involves a module that determines the abnormal diagnosis result of the on-board motor based on the confidence level of each fault type. Specifically, this involves: determining the highest confidence level, the candidate fault type to which the highest confidence level belongs, and the second highest confidence level for each fault type. If the highest confidence level is greater than a first threshold, and the difference between the highest and second highest confidence levels is greater than a second threshold, the abnormal diagnosis result is determined as a candidate fault type.

[0090] One possible implementation involves a fusion module that fuses voltage state feature vectors extracted from voltage signals with auxiliary state feature vectors extracted from current and temperature signals to obtain a comprehensive state feature vector of the on-board motor at the time of data acquisition. Specifically, this involves: converting the voltage signal to obtain two-dimensional voltage data containing a first voltage component and a second voltage component; determining the voltage state feature vector based on the two-dimensional voltage data; extracting the fundamental amplitude of the current signal from the current signal; determining the measured temperature of the on-board motor based on the temperature signal and a temperature compensation coefficient matching the current environment of the on-board motor; determining the auxiliary state feature vector based on the fundamental amplitude of the current signal and the measured temperature; and concatenating the voltage state feature vector and the auxiliary state feature vector to obtain the comprehensive state feature vector.

[0091] It should be noted that all relevant content of each step involved in the above method embodiments can be referenced from the functional description of the corresponding functional module, and will not be repeated here.

[0092] The vehicle motor fault diagnosis device 400 provided in this application embodiment is used to perform the above-mentioned... Figure 2 The method for diagnosing abnormalities in vehicle motors shown can achieve the same effect as the method for diagnosing abnormalities in vehicle motors described above.

[0093] This application embodiment also provides an abnormal diagnosis device for vehicle motors, which can perform the abnormal diagnosis method and related steps of vehicle motors in the above method embodiments.

[0094] This application also provides a computer-readable storage medium storing instructions that, when executed, perform the abnormal diagnosis method and related steps for the vehicle motor in the above method embodiments.

[0095] This application also provides a computer program product that, when run on a computer, causes the computer to execute the abnormal diagnosis method and related steps of the vehicle motor described in the above method embodiments.

[0096] In some embodiments, the methods shown in this application can be implemented as computer program instructions encoded in a machine-readable format on a computer-readable storage medium or on other non-transitory media or articles of art.

[0097] This application embodiment also provides an abnormal diagnosis system 100 for vehicle motors, such as... Figure 5 As shown, the vehicle motor fault diagnosis system 100 includes at least one processor 501 and at least one interface circuit 502.

[0098] As an example, when the vehicle motor fault diagnosis system 100 includes a processor and an interface circuit, the processor can be... Figure 5 The processor 501 shown in the solid box (or the processor 501 shown in the dashed box) can be an interface circuit. Figure 5 The interface circuit 502 is shown in the solid box (or the dashed box). When the on-board motor fault diagnosis system 100 includes two processors and two interface circuits, then the two processors include... Figure 5 The processor 501 shown in the solid box and the processor 501 shown in the dashed box, these two interface circuits include Figure 5 Interface circuit 502 is shown in both solid and dashed boxes. No limitations are imposed on this.

[0099] The processor 501 and the interface circuit 502 can be interconnected via a line. For example, the interface circuit 502 can be used to receive signals. Alternatively, the interface circuit 502 can be used to send signals to other devices (such as the processor 501). For instance, the interface circuit 502 can read computer instructions stored in memory and send those instructions to the processor 501. The processor 501 executes the instructions and, in conjunction with input / output devices, implements the various steps in the above embodiments, such as implementing... Figure 2 and Figure 3 The methods illustrated are the steps performed in the embodiments shown. Of course, this on-board motor fault diagnosis system may also include other discrete components, and this application embodiment does not specifically limit this.

[0100] Through the above description of the embodiments, those skilled in the art can clearly understand that, for the sake of convenience and brevity, only the division of the above functional modules is used as an example. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above.

[0101] In the several embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of modules or units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another device, or some features may be ignored or not executed. Furthermore, the mutual coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between devices or units may be electrical, mechanical, or other forms.

[0102] The units described as separate components may or may not be physically separate. A component shown as a unit can be one or more physical units; that is, it can be located in one place or distributed in multiple different locations. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0103] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0104] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a readable storage medium. Based on this understanding, the technical solution of the embodiments of this application, or the part that contributes to it, or all or part of the technical solution, can be embodied in the form of a software product. This software product is stored in a storage medium and includes several instructions to cause a device (which may be a microcontroller, chip, etc.) or processor to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0105] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions within the technical scope disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A method for diagnosing abnormalities in an on-board motor, characterized in that, The method includes: The system collects motor signals from the vehicle's motor during operation; these motor signals include voltage signals, current signals, and temperature signals. For each acquisition moment, the voltage state feature vector extracted from the voltage signal and the auxiliary state feature vector extracted from the current signal and the temperature signal are fused to obtain the comprehensive state feature vector of the vehicle motor at the acquisition moment. Based on the comprehensive state feature vector at multiple acquisition times, determine the time-series cumulative value corresponding to each fault type in the preset fault types; The abnormal diagnosis result of the vehicle motor is determined based on the time-series cumulative value corresponding to each of the fault types.

2. The method according to claim 1, characterized in that, The step of determining the time-series cumulative value corresponding to each fault type in the preset fault types based on the comprehensive state feature vector at multiple acquisition times includes: The comprehensive state feature vectors of multiple acquisition times are combined in time sequence to obtain a comprehensive state feature time sequence; For any target fault type among the fault types, the time-series cumulative value of the target fault type is determined based on the continuous state of the target fault type in the time-series of the comprehensive state characteristics.

3. The method according to claim 2, characterized in that, The step of determining the time-series cumulative value of the target fault type based on the persistent state of the comprehensive state feature time-series sequence includes: If the target fault type is found to occur continuously in the time series of the integrated state characteristics, the time series cumulative value of the target fault type is determined according to the initial value configured for the target fault type and the duration of the target fault type.

4. The method according to claim 3, characterized in that, The method further includes: When it is determined that the integrated state feature time series has a sudden fault feature at any of the acquisition times, a time series increment is added to the time series cumulative value of the matching target fault type based on the sudden fault feature to obtain the incrementally updated time series cumulative value.

5. The method according to claim 1, characterized in that, The step of determining the abnormal diagnosis result of the on-board motor based on the time-series cumulative value corresponding to each of the fault types includes: Based on the time-series cumulative value corresponding to each of the fault types, the confidence level of the on-board motor in each of the fault types is determined; The abnormal diagnosis result of the vehicle motor is determined based on the confidence level of each of the aforementioned fault types.

6. The method according to claim 5, characterized in that, The step of determining the abnormal diagnosis result of the on-board motor based on the confidence level of each of the fault types includes: Based on the confidence level of each of the aforementioned fault types, determine the highest confidence level, the candidate fault type to which the highest confidence level belongs, and the second highest confidence level. If the highest confidence level is greater than a first threshold and the difference between the highest confidence level and the second highest confidence level is greater than a second threshold, the abnormal diagnosis result is determined to be the candidate fault type.

7. The method according to claim 1, characterized in that, The step of fusing the voltage state feature vector extracted from the voltage signal and the auxiliary state feature vectors extracted from the current signal and the temperature signal to obtain the comprehensive state feature vector of the vehicle motor at the acquisition time includes: The voltage signal is converted to obtain two-dimensional voltage data containing a first voltage component and a second voltage component; Based on the two-dimensional voltage data, the voltage state feature vector is determined; Based on the current signal, the fundamental amplitude of the current signal is extracted; The measured temperature value of the vehicle motor is determined based on the temperature signal and the temperature compensation coefficient that matches the current environment of the vehicle motor. The auxiliary state feature vector is determined based on the current fundamental amplitude and the measured temperature value. The comprehensive state feature vector is obtained by concatenating the voltage state feature vector and the auxiliary state feature vector.

8. A diagnostic device for abnormalities in a vehicle-mounted motor, characterized in that, The device includes: The acquisition module is used to acquire motor signals of the vehicle motor during operation; the motor signals include voltage signals, current signals, and temperature signals. The fusion module is used to fuse the voltage state feature vector extracted from the voltage signal and the auxiliary state feature vector extracted from the current signal and the temperature signal for each acquisition time to obtain the comprehensive state feature vector of the vehicle motor at the acquisition time. The determination module is used to determine the time-series cumulative value corresponding to each fault type in the preset fault types based on the comprehensive state feature vector at multiple acquisition times; The determining module is further configured to determine the abnormal diagnosis result of the vehicle motor based on the time-series cumulative value corresponding to each of the fault types.

9. A diagnostic device for abnormalities in vehicle-mounted motors, characterized in that, The vehicle motor anomaly diagnosis device includes a processor and a memory. The memory stores machine-executable instructions that can be executed by the processor. The processor executes the machine-executable instructions to implement the vehicle motor anomaly diagnosis method according to any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the abnormal diagnosis method for the vehicle motor as described in any one of claims 1 to 7.

Images