A device fault detection method, apparatus, device, and storage medium

Abstract

The application provides a device fault detection method and device, equipment and storage medium, by acquiring temperature field data and current operation condition of the to-be-detected device, temperature change rate analysis is performed on the temperature field data, the singularity region in the temperature field is determined, and the singularity distribution information is determined based on the singularity region; the to-be-analyzed feature is determined according to the singularity distribution information and the temperature field data; and fault recognition is performed based on the to-be-analyzed feature and the current operation condition, so as to obtain the fault detection result of the to-be-detected device; by introducing temperature change rate analysis, the singularity region is used to represent the local mutation feature in the temperature field, so that weak thermal abnormalities caused by hidden problems such as early failure or poor contact of components can be captured; adaptive fault recognition is performed in combination with the current operation condition, the test influence caused by single condition standard misjudgment is reduced, and the fault detection efficiency is effectively improved by the flow detection logic.

Description

Technical Field

[0001] This application relates to the field of equipment operation monitoring, and in particular to a method, apparatus, equipment and storage medium for equipment fault detection. Background Technology

[0002] In the research and testing of electric drive systems for vertical takeoff and landing (VTOL) aircraft, temperature rise characteristics are a key indicator for evaluating the reliability and safety of the electric drive system. Under different flight conditions, the core components of the electric drive system, such as power modules, bus capacitors, power terminals, and drive chips, will experience varying degrees of temperature rise. The distribution characteristics and variation patterns of the temperature rise directly reflect its operating status and potential failure risks.

[0003] Existing testing methods for electric drive systems typically rely on manual operation or simple threshold monitoring. On the one hand, traditional testing methods have low automation levels, often using point thermometers, thermocouples, or handheld infrared thermometers for manual point-by-point measurements. This requires testers to manually switch measurement points and record data under different operating conditions, making it impossible to collect and analyze temperature field image data under all operating conditions in real time, resulting in low testing efficiency. On the other hand, existing monitoring methods lack the ability to deeply analyze the spatial distribution characteristics of the temperature field, and can only judge overheating based on a single temperature threshold. They cannot effectively identify subtle temperature changes caused by poor contact, insulation aging, or local device failure, leading to inaccurate fault location and delayed early warning, increasing the risk of equipment damage and testing safety hazards. Summary of the Invention

[0004] The purpose of this application is to provide a method, apparatus, device, and storage medium for detecting equipment faults, so as to solve the above-mentioned technical problems.

[0005] Firstly, this application provides a method for equipment fault detection, comprising: acquiring temperature field data and current operating conditions of the equipment to be tested; performing temperature change rate analysis on the temperature field data to determine singular regions in the temperature field, and determining singular distribution information based on the singular regions, wherein the singular regions are temperature abrupt change regions in the temperature field; determining features to be analyzed based on the singular distribution information and the temperature field data, and performing fault identification based on the features to be analyzed and the current operating conditions to obtain the fault detection result of the equipment to be tested. The above embodiment, by introducing temperature change rate analysis to replace the traditional absolute temperature threshold judgment, utilizes singular regions to characterize local abrupt change features in the temperature field, effectively capturing weak thermal anomalies caused by early component failure or poor contact. Simultaneously, it combines current operating conditions for adaptive fault identification, avoiding the problem of misjudgment based on a single operating condition standard.

[0006] In one embodiment of this application, analyzing the temperature change rate of the temperature field data to determine singular regions in the temperature field includes: performing multi-scale wavelet transform processing on the temperature field data to obtain signals in different frequency domains, the different frequency domain signals including multiple high-frequency signals; extracting the wavelet coefficient modulus maxima of each high-frequency signal at different scales, and determining the pixels in the temperature field data corresponding to the wavelet coefficient modulus maxima as candidate pixels; calculating the Lipschitz exponent of each candidate pixel based on the wavelet coefficient modulus maxima of each candidate pixel at multiple scales, the Lipschitz exponent being used to characterize the temperature change rate characteristics of the candidate pixels; filtering according to preset singularity point screening conditions and the Lipschitz exponents of each candidate pixel to obtain target pixels, and determining the connected regions formed by the target pixels as the singular regions. The above embodiments utilize multi-scale wavelet transform to decompose temperature field data, accurately capture temperature change edges through high-frequency signals, and further quantify the regularity of each candidate pixel by calculating the Lipschitz exponent. This enables the separation of real fault heat source features from complex background thermal radiation noise, thereby improving the detection rate of small fault areas.

[0007] In one embodiment of this application, calculating the Lipshitz exponent of each candidate pixel based on the wavelet coefficient modulus maxima at multiple scales includes: for each candidate pixel, taking the logarithm of the wavelet coefficient modulus maxima at each scale to obtain multiple logarithmic modulus values, and taking the logarithm of the scale values ​​at each scale to obtain multiple scale logarithmic values; fitting the relationship between the logarithmic modulus values ​​and the scale logarithmic values ​​to obtain a fitting slope, and determining the fitting slope as the corresponding Lipshitz exponent of the candidate pixel. The above embodiment, based on the attenuation characteristics of wavelet transform modulus maxima with scale, accurately solves the Lipshitz exponent through linear fitting, transforming the abstract temperature gradient into a quantifiable mathematical index to ensure the accuracy of subsequent fault identification.

[0008] In one embodiment of this application, determining the singularity distribution information based on the singularity region includes: acquiring the Lipschitz index and spatial location information of each pixel in the singularity region; generating a Lipschitz index distribution map using the spatial location of each pixel in the singularity region as coordinates and the corresponding Lipschitz index as a value, and using the Lipschitz index distribution map as the singularity distribution information. The above embodiment constructs a visualized singularity distribution map, which not only preserves the spatial topology of the fault area but also overlays quantitative information on the severity of the fault, intuitively presenting the thermal stress concentration points inside the electric drive system and assisting in quickly understanding the health status of the equipment.

[0009] In one embodiment of this application, determining the feature to be analyzed based on the singularity distribution information and the temperature field data includes: extracting at least one type of feature from singularity features, temperature features, and spatial morphology features as the feature to be analyzed based on the singularity distribution information and the temperature field data; the singularity features include the average Lipschitz index and the minimum Lipschitz index of pixels within the singularity region, and the area ratio of the singularity region in the temperature field; the temperature features include the average temperature and the highest temperature of the singularity region, and the temperature difference between the singularity region and adjacent regions; the spatial morphology features include the shape and boundary clarity of the singularity region, and the distribution position of the singularity region in the device to be tested. The above embodiment constructs a multi-dimensional feature system, which, through the joint representation of multi-dimensional features, solves the problem that a single temperature feature cannot distinguish fault types, and improves the accuracy of fault mode recognition.

[0010] In one embodiment of this application, fault identification based on the feature to be analyzed and the current operating condition to obtain the fault detection result of the device under test includes: retrieving fault identification conditions corresponding to the current operating condition from a preset rule base, wherein the fault identification conditions include the correspondence between multiple fault types and the feature to be analyzed; performing state determination based on the feature to be analyzed and the fault identification conditions to obtain fault state information; if the fault state information indicates that the device under test has no fault, then the fault detection result is determined to be no device fault; if the fault state information indicates that the device under test has at least one fault type, then the fault location information and fault level are determined based on the feature to be analyzed; generating comprehensive fault information based on the fault type, fault location information, and fault level, and determining the comprehensive fault information as the fault detection result. The above embodiment establishes an operating condition adaptive fault discrimination mechanism, reducing the occurrence of missed or false alarms caused by fixed thresholds, and achieving accurate fault location of physical devices by mapping singular region pixels to obtain fault location information.

[0011] In one embodiment of this application, the fault level includes minor fault, moderate fault, and severe fault. After obtaining the fault detection result of the device under test, the device fault detection method further includes: if the fault level indication in the fault detection result is a minor fault, generating and displaying fault prompt information based on the fault type and fault location information in the fault detection result; if the fault level indication in the fault detection result is a moderate or severe fault, issuing a warning signal and saving the current operating data of the device under test. The above embodiment implements a hierarchical closed-loop control strategy, which not only prevents minor anomalies from being excessively amplified and affecting the test progress, but also ensures that losses can be stopped in time and traceability evidence can be preserved when a serious fault occurs, effectively protecting the safety of the test process and the integrity of the data.

[0012] On the other hand, this application embodiment also provides a device for detecting equipment faults, the device for detecting equipment faults includes: a data acquisition module, used to acquire temperature field data and current operating conditions of the device to be tested; a singularity analysis module, used to perform temperature change rate analysis on the temperature field data, determine singular regions in the temperature field, and determine singularity distribution information based on the singular regions, wherein the singular regions are temperature abrupt change regions in the temperature field; and a fault identification module, used to determine the features to be analyzed based on the singularity distribution information and the temperature field data, and perform fault identification based on the features to be analyzed and the current operating conditions, to obtain the fault detection result of the device to be tested.

[0013] This application also provides an electronic device, including: one or more processors; and a storage device for storing one or more programs, which, when executed by the one or more processors, cause the electronic device to implement the device fault detection method as described in any of the above embodiments.

[0014] This application also provides a computer-readable storage medium storing computer-readable instructions thereon, which, when executed by a computer's processor, cause the computer to perform the device fault detection method as described in any of the above embodiments.

[0015] The beneficial effects of this application are as follows: This application provides a method, apparatus, device, and storage medium for equipment fault detection. By acquiring the temperature field data and current operating conditions of the equipment under test, it performs temperature change rate analysis on the temperature field data to determine singular regions in the temperature field, and determines singular distribution information based on the singular regions. Based on the singular distribution information and temperature field data, it determines the features to be analyzed, and performs fault identification based on the features to be analyzed and the current operating conditions to obtain the fault detection results of the equipment under test. This application introduces temperature change rate analysis for region determination and uses singular regions to characterize local abrupt changes in the temperature field. This can capture weak thermal anomalies caused by hidden problems such as early component failure or poor contact. Furthermore, it combines the current operating conditions for adaptive fault identification, reducing the test impact caused by misjudgment of a single operating condition standard. The streamlined detection logic can improve fault detection efficiency.

[0016] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and do not limit this application. Attached Figure Description

[0017] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application. It is obvious that the drawings described below are merely some embodiments of this application, and those skilled in the art can obtain other drawings based on these drawings without any inventive effort. In the drawings: Figure 1 This is a schematic diagram illustrating an exemplary system architecture as shown in an exemplary embodiment of this application; Figure 2 This is a flowchart illustrating an exemplary embodiment of the present application of a device fault detection method; Figure 3 This is a schematic diagram of an exemplary embodiment of the present application illustrating a device for detecting equipment failure; Figure 4 This is a schematic diagram of the structure of a computer system for an electronic device, as illustrated in an exemplary embodiment of this application. Detailed Implementation

[0018] The embodiments of this application will be described below with reference to the accompanying drawings and specific examples. Those skilled in the art can easily understand other advantages and effects of this application from the content disclosed in this specification. This application can also be implemented or applied through other different specific embodiments, and various details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of this application. It should be understood that the preferred embodiments are only for illustrating this application and are not intended to limit the scope of protection of this application.

[0019] It should be noted that the illustrations provided in the following embodiments are only schematic representations of the basic concept of this application. Therefore, the illustrations only show the components related to this application and are not drawn according to the number, shape and size of the components in actual implementation. In actual implementation, the form, quantity and proportion of each component can be arbitrarily changed, and the layout of the components may also be more complex.

[0020] In the following description, numerous details are explored to provide a more thorough explanation of embodiments of the present application. However, it will be apparent to those skilled in the art that embodiments of the present application may be practiced without these specific details. In other embodiments, well-known structures and devices are shown in block diagram form rather than in detail to avoid obscuring embodiments of the present application.

[0021] In this application, unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains, and the terminology used in the specification is for the purpose of describing particular embodiments only and does not limit the application.

[0022] Please see Figure 1 , Figure 1 This is a schematic diagram illustrating an exemplary system architecture as shown in an exemplary embodiment of this application.

[0023] Reference Figure 1 As shown, the system architecture may include a device under test 110 and a computer device 120. The computer device 120 acquires temperature field data and current operating conditions of the device under test 110, performs temperature change rate analysis on the temperature field data, identifies singular regions in the temperature field, determines singularity distribution information based on these regions, and then determines the features to be analyzed based on the singularity distribution information and the temperature field data. Fault identification is then performed based on these features and the current operating conditions to obtain the fault detection result of the device under test. The device under test 110 includes at least an electric drive system. When the device under test 110 is started, its electric drive system is monitored, and after acquiring temperature field data and current operating conditions, the data is transmitted to the computer device 120. The computer device 120 refers to a computing power support terminal device used to carry the program and implementation environment for the device fault detection method, including but not limited to microcomputers, tablet devices, embedded computers, server clusters, and cloud servers.

[0024] In one feasible environment of this application, the computer device 120 includes a data interface that is connected to at least an infrared thermal imaging sensor. This infrared thermal imaging sensor is used to acquire infrared thermal image data of the electric drive system in the device under test 110. The computer device 120 includes at least a data processing unit to determine temperature field data based on the acquired infrared thermal image data.

[0025] This technical solution can be further discussed and analyzed in multiple steps; for specific steps, please refer to [link / reference needed]. Figure 2 , Figure 2 This is a flowchart illustrating an exemplary embodiment of a device fault detection method according to this application. This device fault detection method can be executed in implementation environments supported by various operating systems, and no specific limitation is made to the implementation environment herein. (Refer to...) Figure 2 As shown, the flowchart of the equipment fault detection method includes at least steps S210 to S230, which are described in detail below: In step S210, the temperature field data and current operating conditions of the device under test are acquired.

[0026] In one embodiment of this application, the device under test is specifically embodied in the electric drive system of a vertical take-off and landing aircraft. However, in other application scenarios, the device under test may also be a power electronic device with the risk of thermal failure, such as a new energy vehicle motor controller, an industrial frequency converter, or a high-power energy storage converter. This application does not limit this.

[0027] In one embodiment of this application, temperature field data is used to characterize the thermal distribution state of the device surface. In one possible implementation, this data is acquired in real-time using an infrared thermal imaging sensor to obtain a high-resolution global temperature distribution in a non-contact manner, enabling the capture of transient thermal anomalies. In other implementations, temperature field data can also be obtained manually by deploying thermocouple arrays or fiber optic grating sensors on the surface of key components of the device under test, and then interpolating the collected point array temperature data to generate a high-precision monitoring of specific hot spots.

[0028] In one implementation, taking the infrared thermal image acquired in real time by the infrared thermal imaging sensor as an example, the acquired infrared thermal image is subjected to denoising, enhancement, and normalization processing. Among them, the denoising processing adopts the Gaussian filtering algorithm to filter the original infrared thermal image, suppress the pixel temperature fluctuation caused by noise generated by environmental thermal radiation and equipment magnetic field during the acquisition process, and retain the detailed information and singularity features of the infrared thermal image; secondly, in order to highlight the key feature information of the infrared thermal image, the gray value range of the infrared thermal image is stretched by the histogram equalization algorithm, which improves the contrast between weak hot spots and background areas, making subtle temperature anomalies easier to identify, while specifically enhancing the required detailed information in the image and eliminating the situation of being too bright or too dark; finally, in order to ensure the accuracy and fast convergence characteristics during analysis, the preprocessed infrared thermal image data needs to be normalized to map the temperature value to the [0,1] interval, so as to eliminate the influence of the absolute temperature dimension on the algorithm convergence and threshold setting. Its calculation formula can be represented as formula (1): Formula (1) in, The normalized temperature value, This is the original temperature value. This represents the lowest temperature value in the infrared thermal image. This represents the highest temperature value in the infrared thermal image. After normalization, a temperature matrix is ​​generated, and this temperature matrix is ​​used as the temperature field data. Through the above specific processing, the temperature field data can more accurately reflect the relative heat distribution characteristics, providing a standardized data foundation for subsequent wavelet transform.

[0029] In the embodiments of this application, the current operating condition refers to the load state and environmental conditions of the device under test during testing or operation. For the electric drive system of a vertical takeoff and landing aircraft, the current operating condition includes, but is not limited to, takeoff, climb, hover, cruise, and landing phases. Since the normal temperature rise benchmark of the device under test varies significantly under different operating conditions—for example, the allowable temperature rise under takeoff conditions is higher than that under cruise conditions—only by clearly defining the current operating condition can a correct dynamic reference benchmark be provided for subsequent fault identification, reducing false alarms or missed alarms caused by fixed thresholds.

[0030] In step S220, temperature change rate analysis is performed on the temperature field data to determine the singularity region in the temperature field, and singularity distribution information is determined based on the singularity region.

[0031] In one embodiment of this application, temperature change rate analysis is a mathematical characterization of the local regularity or smoothness of the temperature field. During normal heat conduction, the temperature field typically changes continuously and smoothly. However, when early faults occur inside the equipment, such as insulation aging, poor contact, or localized device breakdown, drastic fluctuations in the temperature gradient at the microscopic scale occur before the macroscopic temperature reaches the alarm threshold. In this application, such anomalies in local regularity are defined as singularities. Based on this, this embodiment uses temperature change rate analysis to capture weak abrupt change signals, thereby identifying singularity regions.

[0032] The aforementioned singularity regions are temperature abrupt change regions in the temperature field. Their physical meaning corresponds to potential thermal stress concentration points or fault heat source centers. The singularity distribution information determined based on these singularity regions provides a structured and quantitative description of these abrupt change characteristics, including not only the spatial location of singular points but also numerical information characterizing the intensity of the abrupt change. Through this method, this application can transform abstract temperature field data into a computable singularity topology, providing a reliable physical basis for distinguishing between real faults and environmental thermal noise.

[0033] Although this embodiment emphasizes the importance of temperature change rate analysis, the specific analysis algorithm is not limited to a specific transformation algorithm. Any mathematical tool that can characterize the local singular characteristics of a signal, such as multi-scale wavelet transform, gradient operator analysis, or frequency domain energy spectrum analysis, can be used as a specific means to implement this step.

[0034] In one embodiment of this application, the following details the implementation method of performing temperature change rate analysis on temperature field data to determine singular regions in the temperature field, including performing multi-scale wavelet transform processing on the temperature field data to obtain signals in different frequency domains, including multiple high-frequency signals; extracting the wavelet coefficient modulus maxima of each high-frequency signal at different scales, and determining the pixels in the temperature field data corresponding to the wavelet coefficient modulus maxima as candidate pixels; calculating the Lipschitz exponent of each candidate pixel based on the wavelet coefficient modulus maxima of each candidate pixel at multiple scales, the Lipschitz exponent being used to characterize the temperature change rate characteristics of the candidate pixels; filtering according to preset singular point screening conditions and the Lipschitz exponents of each candidate pixel to obtain target pixels, and determining the connected regions formed by the target pixels as singular regions.

[0035] In one embodiment of this application, during the implementation of multi-scale wavelet transform, the db4 tightly supported wavelet can be used as the basis function, and a two-dimensional discrete wavelet transform of 3 to 5 levels is performed on the normalized temperature matrix. The db4 wavelet basis, with its moderate vanishing moment and tightly supported characteristics, exhibits high sensitivity to abrupt changes in the temperature field, while possessing moderate computational complexity, meeting the real-time requirements of electric drive system testing. Through multi-level decomposition, the temperature field is separated into one low-frequency signal and multiple high-frequency signals. The low-frequency signal characterizes the overall background trend of the temperature field, while the high-frequency signals accurately reflect the local abrupt changes in the temperature field. Based on this, the modulus maxima of the wavelet coefficients of each high-frequency component at different scales are extracted. These modulus maxima mathematically correspond to the extreme locations of temperature gradient changes, i.e., potential fault heat sources, and are marked as candidate pixel points.

[0036] Although this embodiment preferably uses the db4 wavelet basis and 3 to 5 layers of decomposition, in other embodiments, other wavelet basis functions such as symlets can be selected according to the thermal conductivity characteristics of the material of the device to be tested or the sensor resolution, or the number of decomposition layers can be adjusted to balance frequency resolution and spatial positioning accuracy, as long as the purpose of separating the temperature change features from the background noise is achieved.

[0037] In one embodiment of this application, in order to quantify the intensity of temperature change in candidate pixels, the Lipschitz exponent of each candidate pixel is calculated based on the maximum values ​​of wavelet coefficient moduli at multiple scales. This includes: for each candidate pixel, taking the logarithm of the maximum values ​​of wavelet coefficient moduli at each scale to obtain multiple logarithmic moduli, and taking the logarithm of the scale values ​​at each scale to obtain multiple scale logarithmic values; fitting the data based on the correspondence between the logarithmic moduli and the scale logarithmic values ​​to obtain the fitting slope, and determining the fitting slope as the Lipschitz exponent of the corresponding candidate pixel.

[0038] In one embodiment of this application, the physical essence of the above calculation process is to characterize the local regularity of the signal by utilizing the attenuation characteristic of the wavelet transform modulus maxima with scale variation. Specifically, for each candidate pixel, a coordinate system is constructed with the logarithmic modulus of the wavelet coefficients as the ordinate and the logarithmic value of the scale as the abscissa. The slope of the straight line is solved using linear fitting methods such as least squares, and this slope is the Lipschitz exponent of the candidate pixel. Physically, the Lipschitz exponent is a key indicator for measuring the local smoothness of the temperature field: when the Lipschitz exponent ≥ 1, it indicates that the temperature change in the region is continuous and smooth, belonging to a normal heat conduction region; when the Lipschitz exponent < 1, it indicates that there is a disruption of the local regularity of the temperature field in the region, i.e., a temperature abrupt change has occurred. Furthermore, the smaller the Lipschitz exponent value, the stronger the singularity of the point, and the higher the probability and severity of the corresponding fault. For example, instantaneous hot spots caused by IGBT (Insulated Gate Bipolar Transistor) breakdown typically exhibit an extremely low Lipschitz exponent, while temperature rises due to poor heat dissipation exhibit a relatively high Lipschitz exponent. Through the aforementioned mathematical mapping, this application transforms abstract thermal image features into precisely calculable scalar values, providing an objective basis for subsequent fault classification.

[0039] In one embodiment of this application, after calculating the Lipschitz index, each candidate pixel needs to be screened according to preset singularity screening conditions to determine the final singularity region. In this embodiment, the preset singularity screening conditions include threshold determination. First, a Lipschitz index threshold is set, and candidate pixels that meet the condition are marked as target pixels, eliminating normal fluctuation points. Based on this, isolated noise points generated by environmental thermal radiation or electromagnetic interference can be effectively filtered out, ensuring that the finally determined singularity region truly corresponds to the physical fault source inside the device, significantly improving the signal-to-noise ratio and reliability of the detection results.

[0040] Secondly, in some other possible implementations, the above-mentioned preset singularity screening conditions may also include spatial connectivity analysis. Considering that the anomaly of a single pixel may be caused by sensor noise, it is necessary to perform connected component analysis on the target pixel and retain only the connected regions with an area greater than the preset minimum pixel cluster threshold as effective singularity regions.

[0041] In one embodiment of this application, after determining the singularity region and the quantization index of each point within it, the discrete data is further transformed into a structured intermediate representation, namely, singularity distribution information. Determining the singularity distribution information based on the singularity region includes obtaining the Lipshitz index and spatial location information of each pixel in the singularity region. Using the spatial location of each pixel in the singularity region as coordinates and the corresponding Lipshitz index as a value, a Lipshitz index distribution map is generated, and this map is used as the singularity distribution information.

[0042] In one embodiment of this application, the singularity distribution information constructed above is a new data layer independent of the original temperature field. Although the original infrared thermal image contains absolute temperature information, it is also mixed with a large amount of background noise caused by environmental radiation, material emissivity differences, and steady-state heat conduction. Directly performing fault mode recognition based on the original infrared thermal image is easily interfered with by non-fault factors. However, the singularity distribution information based on the Lipschitz exponent distribution map removes the single-order limitation of temperature through mathematical transformation, retaining only the topological structure that characterizes the degree of local regularity disruption. In this distribution map, the gray value or color depth of a pixel directly maps the probability density and severity of the fault. For example, in one embodiment, regions with a Lipschitz exponent less than 1 can be marked as singular regions, where the lower the exponent value, the higher the brightness or saturation in the distribution map, which can intuitively characterize the drastic temperature field change; while normal regions with an exponent greater than or equal to 1 are suppressed as low-brightness or low-saturation background colors. This allows subsequent fault feature extraction algorithms to focus on analyzing the geometric shape and intensity distribution of the abrupt change signal, improving the convergence speed and robustness of the diagnostic process.

[0043] In one embodiment of this application, the accuracy of spatial location information is a crucial prerequisite in generating the Lipschitz index distribution map. During implementation, the accuracy of this spatial location information is ensured by establishing a precise mapping relationship between the image pixel coordinate system and the physical coordinate system of the device under test. For example, for a planar motor controller cover, this mapping relationship is typically a linear affine transformation; however, for electric drive systems with curved shells or complex internal structures, pre-calibrated distortion correction parameters or a three-dimensional projection model are required to ensure that each pixel in the distribution map uniquely corresponds to its actual physical location inside the device. Without this mapping relationship, even if a highly singular hotspot cluster is detected, it is impossible to quickly determine which part of the device it corresponds to, thus rendering the diagnostic results meaningless and indirectly reducing diagnostic efficiency. Therefore, the aforementioned Lipschitz index distribution map serves as a feature carrier of a two-dimensional numerical matrix and the spatial topological semantics of the device.

[0044] In some other possible implementations, to enhance the perceptibility of the technical solution at the human-computer interaction level, the aforementioned Lipschitz index distribution map can also serve as a core visualization component of the real-time monitoring interface. In practical applications, the calculated index distribution map can be overlaid on a layer of the original infrared thermal image, or presented in a separate window. Furthermore, through a preset pseudo-color mapping strategy, severely singular areas, moderately singular areas, and normal areas can be rendered as progressively changing colors. The health status of the device can be easily grasped through the distribution and intensity of the colors. This visualization method not only assists in manual judgment but also provides an intuitive triggering basis for automated alarm systems.

[0045] Although this embodiment uses the Lipschitz exponent distribution map as the preferred form of singularity distribution information, in other embodiments, this information can also be represented as a structured list containing coordinates and exponent values, a vectorized contour map, or a feature vector that has undergone dimensionality reduction. As long as it fully retains the dual attributes of spatial topology and quantization intensity of the singularity region and can serve as a direct input source for subsequent multidimensional feature extraction, it falls within the protection scope of this application.

[0046] In step S230, the features to be analyzed are determined based on the singularity distribution information and temperature field data, and fault identification is performed based on the features to be analyzed and the current operating conditions to obtain the fault detection results of the equipment to be tested.

[0047] In the embodiments of this application, the feature to be analyzed is a multi-dimensional feature vector that integrates singularity indicators and temperature field data. While singularity distribution information alone can locate abrupt change points, it is difficult to directly determine the fault type; while raw temperature field data alone can reflect heat, it lacks sensitivity to the fault mechanism. Therefore, this application constructs a comprehensive feature that includes both abrupt change features and temperature attribute features, i.e., the feature to be analyzed. During fault identification, the extracted feature to be analyzed is associated with the current operating condition, and the corresponding discrimination conditions are matched in a preset rule base or model. The final fault detection result not only includes a status determination of whether a fault exists, but also further includes detailed information such as fault type, fault location, and severity level.

[0048] In one embodiment of this application, determining the feature to be analyzed based on singularity distribution information and temperature field data includes: extracting at least one type of feature from singularity features, temperature features, and spatial morphology features as the feature to be analyzed, based on singularity distribution information and temperature field data. By constructing a multi-dimensional feature that integrates mathematical indicators and physically measured attributes, the problem of insufficient discriminative power of single-dimensional features under complex operating conditions can be solved. Furthermore, by combining the above three types of features, the thermal state of the equipment can be determined from different aspects, forming highly specific fault-related information, thereby supporting subsequent fault classification and location.

[0049] The singularity features include the average Lipschitz exponent, the minimum Lipschitz exponent, and the area ratio of the singular region within the temperature field. The minimum Lipschitz exponent reflects the most drastic temperature abrupt change at a single point within the singular region, typically corresponding to the core location of the fault source. A lower Lipschitz exponent indicates a more complete disruption of local regularity at that point, increasing the likelihood of hard breakdown or short circuit. The average Lipschitz exponent characterizes the overall irregularity of the temperature field within the singular region, used to assess the average deterioration level of the fault area. The area ratio, the percentage of pixels in the singular region out of the total number of pixels in the temperature field, directly reflects the physical extent of the fault's impact. In practical applications, if a region has an extremely low minimum Lipschitz exponent but a small area ratio, it may indicate an early point defect; conversely, if the average exponent is low and the area ratio is significantly increased, it may indicate that the fault has entered the diffusion stage or that a large-scale systemic thermal anomaly exists. Based on these singularity features, different types of thermal failure modes can be distinguished across both intensity and extent dimensions.

[0050] Secondly, temperature characteristics include the average and maximum temperatures of the singular region, as well as the temperature difference between the singular region and adjacent regions. During electric drive system testing, changes in ambient temperature, the steady-state temperature rise of the equipment itself, and sensor drift can all cause fluctuations in absolute temperature readings. Relying solely on the maximum or average temperature for judgment can easily lead to false alarms. The temperature difference between the singular region and adjacent normal regions extracts the local abnormal heat increment, which can effectively offset the common-mode ambient thermal background noise and highlight the additional heat generated by the fault itself. For example, under takeoff conditions, even if the overall temperature of the IGBT module has risen to the normal high level of 80°C, as long as its temperature difference with the surrounding area remains within the safe threshold, it can still be judged as normal. If, under low-temperature cruise conditions, a small area is detected with an absolute temperature of only 60°C, but the temperature difference with the adjacent area reaches 25°C, an anomaly warning should be triggered immediately. In addition, the average and maximum temperatures serve as auxiliary criteria, mainly used to verify the thermal energy level of the fault, preventing false high-gradient signals generated by sensor defects from being misjudged as real faults.

[0051] The aforementioned spatial morphological features include the shape and boundary sharpness of the singular regions, as well as their distribution within the device under test. This feature establishes a direct mapping chain between image geometric attributes and physical fault mechanisms. Shape parameters can be extracted by labeling connected components and fitting contours to the singular regions, and boundary sharpness can be quantified by calculating the gradient magnitude variance of the region edges. Combined with the device layout diagram of the device under test, the specific component locations corresponding to the singular regions can also be determined. In one feasible environment of this embodiment, different fault types exhibit significantly different combinations of morphological features: for example, when a singular region is detected as a point with extremely clear boundaries and located on the surface of the IGBT device, combined with a low Lipschitz index, it can be identified as an IGBT short circuit or breakdown fault; when the singular region is strip-shaped or planar, with continuous heat traces distributed along the conductive path of the copper busbar or terminals, it corresponds to resistance overheating caused by poor contact of the power terminals or copper busbar; and when the singular region is a large-area diffuse shape with blurred boundaries and covers most of the controller area, even with a relatively high Lipschitz index, it should be judged as a systemic thermal management fault such as poor heat dissipation or airflow blockage. Through the above-mentioned structured feature extraction method, this application transforms unstructured temperature field data into semantically clear diagnostic evidence, significantly improving the interpretability and accuracy of fault identification.

[0052] The above-described correspondence between morphology and fault is merely an illustrative example of this embodiment. In actual deployment, more refined feature-fault mapping rule bases or dedicated pattern recognition models can be established or trained based on the structural characteristics of specific equipment and historical fault data, all of which fall within the protection scope of this application.

[0053] In one embodiment of this application, based on the construction of a multi-dimensional feature system to be analyzed, fault identification is performed based on the features to be analyzed and the current operating conditions to obtain the fault detection result of the device under test. This includes: retrieving fault identification conditions corresponding to the current operating conditions from a preset rule base, whereby the fault identification conditions include the correspondence between various fault types and the features to be analyzed; determining the state based on the features to be analyzed and the fault identification conditions to obtain fault state information; if the fault state information indicates that the device under test has no fault, then the fault detection result is determined to be no equipment fault; if the fault state information indicates that the device under test has at least one fault type, then the fault location information and fault level are determined based on the features to be analyzed; generating comprehensive fault information based on the fault type, fault location information, and fault level, and determining the comprehensive fault information as the fault detection result.

[0054] In one embodiment of this application, the current operating condition is introduced as a dynamic index parameter, solving the technical problem of poor adaptability of fixed thresholds in complex operating scenarios. For the electric drive system of a vertical takeoff and landing (VTOL) aircraft, its operating modes cover multiple typical stages such as takeoff, climb, hovering, cruise, and landing. The motor speed, torque output, and cooling conditions differ significantly in each stage, resulting in drastically different temperature field distribution benchmarks under normal conditions. For example, during takeoff, the electric drive system is under short-term overload, and the temperature of the IGBT module may rapidly rise above 85°C with a high temperature gradient, which is a normal thermal response. However, during cruise, the system operates in a steady state, and the same temperature level and gradient highly likely indicate heat dissipation failure or device malfunction. Therefore, the preset rule base is not a single static threshold table, but a multi-dimensional discrimination matrix with operating conditions as the key. In actual implementation, this rule base stores the characteristic threshold ranges corresponding to different fault types under various typical operating conditions, and the thresholds are obtained based on historical test data and thermal simulation model calibration. Once the current operating condition signal is obtained, the corresponding discrimination logic is automatically loaded, thereby ensuring that the benchmark for fault identification always matches the current physical state of the equipment, effectively reducing false alarms or missed alarms caused by operating condition switching.

[0055] In one embodiment of this application, during the actual execution of the state determination, the correspondence between the features included in the fault identification conditions and the fault type is the key basis for achieving diagnosis. Taking one possible implementation as an example, the minimum Lipschitz index, average Lipschitz index, maximum temperature, temperature difference, and the morphological and location characteristics of the singularity region can be defined in a preset rule base for different types of faults as discrimination criteria: For short circuit or breakdown faults of insulated gate bipolar transistors or power transistors, at least one of the following is included: In the singularity characteristics, the minimum Lipschitz index is usually between 0.2 and 0.6, and the average index is less than 0.7; in the temperature characteristics, the maximum temperature exceeds 85 degrees Celsius and the temperature difference with the surrounding area is greater than 25 degrees Celsius; in the spatial morphology characteristics, the singularity region is dot-shaped with very clear boundaries and is located on the surface of the insulated gate bipolar transistor device. When a bus capacitor bulges or fails, it includes at least one of the following: In singularity characteristics, the minimum Lipschitz index is between 0.5 and 0.8, and the average index is less than 0.8; in temperature characteristics, the highest temperature exceeds 75 degrees Celsius and the temperature difference is greater than 15 degrees Celsius; in spatial morphology characteristics, the singular region is circular or elliptical with relatively clear boundaries and is located at the top of the capacitor or at the pin. Faults with poor contact at power terminals or copper busbars include at least one of the following: In singularity characteristics, the minimum Lipschitz index is between 0.6 and 0.9, and the average index is less than 0.9; in temperature characteristics, the highest temperature exceeds 70 degrees Celsius and the temperature difference is greater than 15 degrees Celsius; in spatial morphology characteristics, the singular region is strip-shaped or planar, with continuous heat traces distributed along the conductive path of the copper busbar or terminal. Abnormalities in the drive integrated circuit or auxiliary power supply include at least one of the following: In singularity characteristics, the minimum Lipschitz index is 0.5 to 0.8, and the average index is less than 0.8; in temperature characteristics, the highest temperature exceeds 70 degrees Celsius and the temperature difference is greater than 10 degrees Celsius; in spatial morphology characteristics, the singular region is small and dense, located in the area of ​​the drive integrated circuit and its surrounding auxiliary devices. Fault characteristics of poor heat dissipation or blocked airflow include at least one of the following: In singularity characteristics, the minimum Lipschitz index is between 0.8 and 1.0, and the average index is greater than 0.8; in temperature characteristics, the average temperature exceeds 65 degrees Celsius but there is no obvious highest temperature peak; in spatial morphology characteristics, the singular region is widely diffused with blurred boundaries, covering most of the controller area. For insulation aging or creepage faults, the minimum Lipschitz index is between 0.7 and 0.9, and the average index is less than 0.9; in temperature characteristics, the highest temperature exceeds 60 degrees Celsius and the temperature difference is greater than 10 degrees Celsius; in spatial morphology characteristics, the singular region is diffusely distributed along the insulation boundary of the printed circuit board with blurred boundaries. Under normal conditions, at least one of the following is included: in singularity features, both the minimum Lipschitz exponent and the average exponent are greater than or equal to 1.0; in temperature features, the average temperature is below 65 degrees Celsius and the temperature difference is less than 10 degrees Celsius; in spatial morphology features, there are no obvious singular regions and the temperature field is uniformly and smoothly distributed.

[0056] This feature threshold table serves as a rule base for fault identification. In actual testing, temperature field data is first collected through infrared thermal imaging. The Lipschitz index of each candidate pixel is calculated through multi-scale wavelet transform and modulus maxima analysis. Pixels with indices below a preset threshold are selected to form singularity regions. Then, features such as minimum index, average index, maximum temperature, temperature difference, area ratio, shape, boundary clarity, and location of these regions are extracted. Finally, the extracted features are matched with the thresholds and morphological conditions corresponding to the fault types in the table to determine the specific fault type, fault location, and fault severity.

[0057] The thresholds mentioned above are for illustrative purposes only. In practical applications, rule bases can also be constructed using machine learning models, decision trees, or fuzzy logic.

[0058] In one embodiment of this application, when the status determination result indicates the presence of a fault, it is further necessary to determine the fault location information and fault level to generate comprehensive fault information with engineering guidance significance. The acquisition of fault location information relies on precise registration between image space and physical space. In actual implementation, the geometric center coordinates of the target pixel cluster within the singularity region are first calculated and used as the image location point of the fault heat source. Subsequently, through a pre-calibrated coordinate transformation matrix, these image coordinates are mapped onto the device layout diagram of the device under test, thereby automatically identifying the component name and physical installation location corresponding to the hot spot. This automated traceability capability significantly shortens the fault diagnosis time.

[0059] Meanwhile, the classification of fault levels comprehensively considers the severity and scope of the anomaly, for example, based on the minimum Lipschitz index. and the proportion of the area of ​​the strange zone The combination is classified into levels: when and The fault was initially determined to be minor. when and The fault was initially determined to be moderate. when and The fault was determined to be a severe one.

[0060] The final comprehensive fault information not only includes the fault status identifier, but also integrates the fault type, precise location, severity level, and key characteristic values, providing complete data support for subsequent graded response strategies.

[0061] Furthermore, for cases where the status is determined to be without equipment faults, the output fault detection result serves as a health status confirmation, providing positive feedback for the continuity of the testing process. Through the aforementioned adaptive identification mechanism based on operating conditions, this application can improve the safety of electric drive system testing and the depth and effectiveness of data analysis.

[0062] In one embodiment of this application, the fault level includes minor fault, moderate fault and severe fault. After obtaining the fault detection result of the device under test, the device fault detection method further includes: if the fault level in the fault detection result indicates a minor fault, generating and displaying fault prompt information based on the fault type and fault location information in the fault detection result; if the fault level in the fault detection result indicates a moderate fault or a severe fault, issuing a warning signal and saving the current operating data of the device under test.

[0063] In actual implementation, the determination of the fault level can be based on the quantitative standard in the aforementioned embodiments, namely, the minimum Lipschitz index, which comprehensively considers the severity of temperature change. and the proportion of singular region area representing the scope of the fault's impact. The subsequent hierarchical system.

[0064] For minor faults, the generated fault message not only includes a textual description of the fault type but also integrates targeted operational suggestions. For example, the message is displayed as a yellow warning on the human-machine interface, clearly indicating the location of the abnormal device, and simultaneously pops up operational instructions to remind test personnel to remain vigilant and pay close attention to the test process. It also suggests canceling subsequent high-power tests or inspections to avoid exacerbating controller failures. Since minor faults typically correspond to early performance degradation of devices or slight local contact defects, their thermal runaway risk is still within a controllable range. This avoids immediately terminating the entire test process due to minor anomalies, significantly improving the data acquisition efficiency and resource utilization of electric drive system R&D testing. It also provides a valuable window of data on operation with faults for studying fault evolution patterns.

[0065] For moderate or severe faults, the response strategy shifts to mandatory safety protection and evidence preservation. When such high-level faults are detected, the control unit immediately outputs a warning signal. This signal can take various forms, including but not limited to triggering the test bench's audible and visual alarms, sending high-priority pop-up messages to remote monitoring terminals, or pushing emergency vibration alerts to on-site handheld terminals via the communication module. This ensures that test personnel can perceive the risk and take emergency measures immediately. Simultaneously, the system automatically triggers the synchronous saving of the current operating data of the device under test. This includes not only the original infrared thermal image sequence and Lipschitz index distribution map within a preset time period before and after the fault occurrence, but also the coordinates of the located fault area, the identified fault category, and corresponding key operating parameters such as voltage, current, speed, and ambient temperature. This ensures the safety of the test site, and the fully preserved fault snapshot data provides tracing evidence for subsequent root cause analysis and algorithm model iteration optimization. Through the hierarchical closed-loop control of this embodiment, this application successfully achieves a leap from passive safety defense to proactive and refined test management, maximizing the value density of test data while ensuring the bottom line of test safety.

[0066] The specific threshold ranges and actions in the above-mentioned graded response strategy are only illustrative examples. In practical applications, they can be flexibly adjusted according to the tolerance characteristics of the device under test, the test purpose, and safety specifications. In addition, the transmission method of the warning signal is not limited to wired or wireless communication, but may also include hardware cascade actions such as directly cutting off the main circuit contactor.

[0067] Based on the above embodiments, this application further proposes a specific embodiment. This specific embodiment integrates the scattered algorithms, feature extraction logic and hierarchical response strategies in the aforementioned embodiments into a specific hardware system. In this application scenario, the device to be tested is the electric drive system of a vertical take-off and landing aircraft, and the testing device includes an infrared thermal imaging sensor module, a data acquisition and processing module, an intelligent analysis module, a communication module, a control and alarm module, a test state switching module, and a human-machine interface module.

[0068] The data acquisition and processing module first addresses pixel-level temperature fluctuations in the original infrared thermal image caused by environmental thermal radiation or equipment magnetic field interference by employing a Gaussian filtering algorithm for noise reduction. This step effectively suppresses random noise while preserving image edge details and singular features to the greatest extent possible, avoiding the loss of weak fault signals due to excessive smoothing. Secondly, to highlight key features in the temperature field, a histogram equalization algorithm is used to stretch the grayscale range of the temperature field, improving the contrast between weak hot spots and the background area. This makes subtle temperature anomalies easier for subsequent algorithms to identify, while also eliminating overexposure or underexposure caused by uneven sensor response. Finally, to ensure the convergence speed and numerical stability of the deep learning or wavelet transform algorithms in the intelligent analysis module, the enhanced temperature field data is normalized. This not only provides standardized data input for the accurate calculation of the Lipschitz exponent but also improves the system's robustness under complex electromagnetic and thermal environments.

[0069] After data standardization, the intelligent analysis module performs a 3-level decomposition using the db4 wavelet basis, calculates the Lipschitz exponent for each pixel, and generates an exponent distribution map. Assuming a connected region is detected on the surface of the IGBT device, its minimum Lipschitz exponent... The average Lipschitz index is 0.35. The ratio is 0.58, representing the area of ​​the singular region. The figure is 2%, while the highest temperature in the region is... Reaching 92℃, with a temperature difference compared to neighboring areas The temperature is 35℃, and the morphology is a dotted pattern with extremely clear boundaries. The above multidimensional features to be analyzed are matched with the discrimination conditions under the corresponding operating conditions in the preset rule base to determine the fault type. Further, the fault is classified according to the grading standard to generate comprehensive fault information including fault type, location, and level.

[0070] In response to the fault detection results, the control and alarm module immediately executes a tiered response strategy. For example, at the medium level, the control unit sends an emergency stop command to the test bench, cutting off the power supply to the electric drive system to prevent the fault from escalating; simultaneously, the alarm unit triggers a red audible and visual alarm and pushes a high-priority warning signal to the remote monitoring terminal via the communication module. At the same time, all operating data within a preset time period before and after the fault occurrence is automatically and synchronously saved, including the original infrared thermal image sequence, Lipschitz index distribution map, the coordinates of the identified fault area, the fault category, and corresponding parameters such as voltage, current, speed, and ambient temperature.

[0071] In terms of human-machine interaction and data transmission, the human-machine interface module provides intuitive real-time monitoring functions, overlaying and displaying the original temperature field, Lipschitz index, and fault diagnosis results, enabling test personnel to immediately grasp the health status of the equipment. The communication module supports multiple methods such as 4G, 5G mobile communication, or satellite communication to achieve remote real-time transmission of test data.

[0072] This application provides a method for equipment fault detection. It acquires temperature field data and current operating conditions of the equipment under test, analyzes the rate of temperature change in the temperature field data to identify singular regions within the temperature field, and determines singularity distribution information based on these regions. The method then uses the singularity distribution information and temperature field data to determine the features to be analyzed, and combines this with the current operating conditions to identify faults, thus obtaining fault detection results. Through this process, this application achieves a complete closed loop from data acquisition and feature extraction to intelligent diagnosis, solving the technical problem of insufficient early-stage, subtle fault identification under complex operating conditions, and improving the safety and diagnostic efficiency of equipment testing.

[0073] The following describes an embodiment of the apparatus described in this application, which can be used to execute the device fault detection method described in the above embodiments of this application. For details not disclosed in the device embodiments of this application, please refer to the embodiments of the device fault detection method described in the above applications.

[0074] Figure 3 This is a schematic diagram illustrating a device for detecting equipment failures, as shown in an exemplary embodiment of this application. This device can be applied to… Figure 1 The method described is implemented in a device that has the necessary conditions for execution. This embodiment does not impose specific limitations on the devices to which the device is applicable.

[0075] like Figure 3 As shown, the exemplary device fault detection apparatus includes: a data acquisition module 301, a singularity analysis module 302, and a fault identification module 303.

[0076] The data acquisition module 301 is used to acquire the temperature field data and current operating conditions of the device under test; the singularity analysis module 302 is used to perform temperature change rate analysis on the temperature field data, determine the singularity region in the temperature field, and determine the singularity distribution information based on the singularity region, which is the temperature change region in the temperature field; the fault identification module 303 is used to determine the feature to be analyzed based on the singularity distribution information and temperature field data, and perform fault identification based on the feature to be analyzed and the current operating conditions to obtain the fault detection result of the device under test.

[0077] Embodiments of this application also provide an electronic device, including: one or more processors; and a storage device for storing one or more programs, which, when executed by one or more processors, cause the electronic device to implement the device fault detection method provided in the above embodiments.

[0078] Figure 4 This is a schematic diagram illustrating the structure of a computer system for an electronic device, as shown in an exemplary embodiment of this application. It should be noted that... Figure 4 The computer system 400 of the electronic device shown is merely an example and should not impose any limitation on the functionality and scope of use of the embodiments of this application.

[0079] like Figure 4 As shown, the computer system 400 includes a Central Processing Unit (CPU) 401, which can perform various appropriate actions and processes, such as executing the methods described in the above embodiments, based on programs stored in Read-Only Memory (ROM) 402 or programs loaded from storage into Random Access Memory (RAM) 403. The RAM 403 also stores various programs and data required for system operation. The CPU 401, ROM 402, and RAM 403 are interconnected via a bus. An I / O interface 405 is also connected to the bus 404, where the I / O interface 405 refers to an input / output interface.

[0080] The following components are connected to I / O interface 405: an input section 406 including a keyboard, mouse, etc.; an output section 407 including a cathode ray tube (CRT), liquid crystal display (LCD), etc., and speakers, etc.; a storage section 408 including a hard disk, etc.; and a communication section 409 including a network interface card such as a LAN (Local Area Network) card, modem, etc. The communication section performs communication processing via a network such as the Internet. A drive is also connected to I / O interface 405 as needed. Removable media 411, such as a disk, optical disk, magneto-optical disk, semiconductor memory, etc., are installed on drive 410 as needed so that computer programs read from them can be installed into storage section 408 as needed.

[0081] Specifically, according to embodiments of this application, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments of this application include a computer program product comprising a computer program carried on a computer-readable medium, the computer program including a computer program for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via communication section 409, and / or installed from removable medium 411. When the computer program is executed by central processing unit (CPU) 401, it performs various functions defined in the system of this application.

[0082] It should be noted that the computer-readable medium shown in the embodiments of this application can be a computer-readable signal medium or a computer-readable storage medium, or any combination of the two. A computer-readable storage medium can be, for example, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of a computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM), flash memory, optical fiber, portable compact disc read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In this application, a computer-readable signal medium may include a data signal propagated in baseband or as part of a carrier wave, carrying a computer-readable computer program. Such propagated data signals can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. Computer-readable signal media can also be any computer-readable medium other than computer-readable storage media, which can send, propagate, or transmit a program for use by or in connection with an instruction execution system, apparatus, or device. The computer program contained on the computer-readable medium can be transmitted using any suitable medium, including but not limited to wireless, wired, etc., or any suitable combination thereof.

[0083] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this application. Each block in a flowchart or block diagram may represent a module, segment, or portion of code, which contains one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in a block diagram or flowchart, and combinations of blocks in a block diagram or flowchart, may be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0084] In the corresponding figures of the above embodiments, connecting lines can represent the connection relationship between various components, indicating more constitutive signal paths and / or one or more ends of some lines having arrows to indicate the main information flow direction. Connecting lines are an identifier and are not a limitation on the scheme itself, but rather, using these lines in conjunction with one or more exemplary embodiments helps to more easily connect circuits or logic units. Any signal represented (determined by design requirements or preferences) can actually include one or more signals that can be transmitted in any direction and can be implemented in any suitable type of signal scheme.

[0085] The units described in the embodiments of this application can be implemented in software or hardware, and the described units can also be located in a processor. The names of these units do not necessarily limit the specific unit itself.

[0086] Another aspect of this application provides a computer-readable storage medium having a computer program stored thereon that, when executed by a processor, implements the method as described above. This computer-readable storage medium may be included in the electronic device described in the above embodiments, or it may exist independently and not assembled into the electronic device.

[0087] Embodiments of this application also provide a computer program product, including a computer program that, when executed by a processor, implements a device fault detection method as described in any of the above embodiments.

[0088] It should be noted that although several modules or units for the device used to perform actions have been mentioned in the detailed description above, this division is not mandatory. In fact, according to the embodiments of this application, the features and functions of two or more modules or units described above can be embodied in one module or unit. Conversely, the features and functions of one module or unit described above can be further divided and embodied by multiple modules or units.

[0089] Through the above description of the embodiments, those skilled in the art will readily understand that the exemplary embodiments described herein can be implemented by software or by combining software with necessary hardware. Therefore, the technical solutions according to the embodiments of this application can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (such as a CD-ROM, USB flash drive, external hard drive, etc.) or on a network, including several instructions to cause a computing device (such as a personal computer, server, touch terminal, or network device, etc.) to execute the method according to the embodiments of this application.

[0090] This application can be used in a wide range of general-purpose or special-purpose computing system environments or configurations. For example: personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer electronics devices, network PCs, minicomputers, mainframe computers, distributed computing environments including any of the above systems or devices, etc.

[0091] Other embodiments of this application will readily occur to those skilled in the art upon consideration of the specification and practice of the embodiments disclosed herein. This application is intended to cover any variations, uses, or adaptations of this application that follow the general principles of this application and include common knowledge or customary techniques in the art not disclosed herein.

[0092] The above embodiments are merely illustrative of the principles and effects of this application and are not intended to limit this application. Any person skilled in the art can modify or alter the above embodiments without departing from the spirit and scope of this application. Therefore, all equivalent modifications or alterations made by those skilled in the art without departing from the spirit and technical concept disclosed in this application should still be covered by the claims of this application.

Claims

1. A method for detecting equipment faults, characterized in that, The equipment fault detection method includes: Acquire temperature field data and current operating conditions of the equipment under test; Temperature change rate analysis is performed on the temperature field data to identify singular regions in the temperature field, and singularity distribution information is determined based on the singular regions, wherein the singular regions are temperature abrupt change regions in the temperature field. Based on the singularity distribution information and the temperature field data, the features to be analyzed are determined, and based on the features to be analyzed and the current operating conditions, fault identification is performed to obtain the fault detection results of the equipment to be tested.

2. The equipment fault detection method according to claim 1, characterized in that, Temperature change rate analysis was performed on the temperature field data to identify singular regions in the temperature field, including: The temperature field data is subjected to multi-scale wavelet transform processing to obtain signals in different frequency domains, including multiple high-frequency signals. Extract the modulus maxima of wavelet coefficients of each high-frequency signal at different scales, and determine the pixels in the temperature field data that correspond to the modulus maxima of wavelet coefficients as candidate pixels. Based on the maximum modulus of the wavelet coefficients of each candidate pixel at multiple scales, the Lipschitz exponent of each candidate pixel is calculated, and the Lipschitz exponent is used to characterize the temperature change rate of the candidate pixel. The target pixels are obtained by filtering according to the preset singularity filtering conditions and the Lipschitz index of each candidate pixel, and the connected region formed by the target pixels is determined as the singularity region.

3. The equipment fault detection method according to claim 2, characterized in that, The Lipschitz exponent for each candidate pixel is calculated based on the maximum modulus of the wavelet coefficients at multiple scales, including: For each candidate pixel, take the logarithm of the maximum value of the wavelet coefficient modulus at each scale to obtain multiple logarithmic modulus values, and take the logarithm of the scale value at each scale to obtain multiple scale logarithmic values. The fitting slope is obtained by fitting the logarithmic modulus and the scale logarithmic value, and the fitting slope is determined as the Lipschitz index of the corresponding candidate pixel.

4. The equipment fault detection method according to claim 1, characterized in that, Determining the singularity distribution information based on the singularity region includes: Obtain the Lipschitz index and spatial location information of each pixel in the singularity region; Using the spatial position of each pixel in the singularity region as coordinates and the corresponding Lipschitz index as value, a Lipschitz index distribution map is generated, and the Lipschitz index distribution map is used as the singularity distribution information.

5. The equipment fault detection method according to claim 4, characterized in that, Based on the singularity distribution information and the temperature field data, the features to be analyzed include: Based on the singularity distribution information and the temperature field data, at least one type of feature among singularity features, temperature features, and spatial morphology features is extracted as the feature to be analyzed. The singularity features include the average Lipschitz index and the minimum Lipschitz index of pixels within the singularity region, as well as the area ratio of the singularity region in the temperature field. The temperature characteristics include the average temperature and the highest temperature of the singularity region, as well as the temperature difference between the singularity region and the adjacent region. The spatial morphological features include the shape and boundary clarity of the singularity region, as well as the distribution location of the singularity region in the device under test.

6. The equipment fault detection method according to claim 5, characterized in that, Based on the features to be analyzed and the current operating conditions, fault identification is performed to obtain the fault detection results of the device under test, including: Based on the current operating condition, fault identification conditions corresponding to the current operating condition are retrieved from the preset rule base. The fault identification conditions include the correspondence between various fault types and the features to be analyzed. Based on the features to be analyzed and the fault identification conditions, a state determination is performed to obtain fault state information; If the fault status information indicates that the device under test is fault-free, then the fault detection result is determined to be no device fault. If the fault status information indicates that the device under test has at least one type of fault, then the fault location information and fault level are determined according to the features to be analyzed; based on the fault type, fault location information and fault level, comprehensive fault information is generated, and the comprehensive fault information is determined as the fault detection result.

7. The equipment fault detection method according to any one of claims 1-6, characterized in that, The fault levels include minor faults, moderate faults, and severe faults. After obtaining the fault detection results of the equipment under test, the equipment fault detection method further includes: If the fault level indication in the fault detection result is a minor fault, a fault prompt message is generated and displayed based on the fault type and fault location information in the fault detection result; If the fault level indication in the fault detection result is a moderate or severe fault, a warning signal is issued and the current operating data of the device under test is saved.

8. A device for detecting equipment faults, characterized in that, The equipment fault detection device includes: The data acquisition module is used to acquire temperature field data and current operating conditions of the equipment under test; The singularity analysis module is used to perform temperature change rate analysis on the temperature field data, identify singular regions in the temperature field, and determine singularity distribution information based on the singular regions, wherein the singular regions are temperature abrupt change regions in the temperature field. The fault identification module is used to determine the feature to be analyzed based on the singularity distribution information and the temperature field data, and to perform fault identification based on the feature to be analyzed and the current operating conditions, so as to obtain the fault detection result of the device to be tested.

9. An electronic device, characterized in that, It includes a processor, a memory, and a communication bus; the communication bus is used to connect the processor and the memory; the processor is used to execute a computer program stored in the memory to implement the device fault detection method as described in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, It stores a computer program that enables the computer to perform the equipment fault detection method as described in any one of claims 1-7.

Images