A pantograph arcing anomaly detection and early warning method using ultraviolet light and infrared imaging fusion

Abstract

The present invention relates to the field of anomaly detection technology, and specifically to a pantograph arcing anomaly detection and early warning method using the fusion of ultraviolet light and infrared imaging, comprising real-time acquisition of ultraviolet spectrum signals in the pantograph area, extraction of arcing characteristic data; acquisition of infrared temperature distribution images of the pantograph and the contact network, and extraction of temperature anomaly area data; synchronization of arcing characteristic data and temperature anomaly area data with time stamps and spatial coordinate mapping to generate a fusion data set aligned in time and space; calculation of the spatiotemporal similarity between arcing characteristic data and temperature anomaly area data based on the fusion data set, and output of an arcing-temperature correlation index; construction of a Gaussian mixture model to dynamically determine the anomaly level; when the anomaly level exceeds a preset threshold, a graded early warning signal is triggered, and a fault location report including the arcing position and the temperature anomaly range is generated. The present invention has high engineering practicality and is of great significance for ensuring the safe and stable operation of the rail transit power supply system.

Description

Technical Field

[0001] The present invention relates to the field of anomaly detection technology, and in particular to a pantograph arcing anomaly detection and early warning method using the fusion of ultraviolet light and infrared imaging. Background Art

[0002] In rail transit power supply systems, the contact state between the pantograph and the catenary directly impacts train power continuity and equipment safety. Due to factors such as complex operating environments, current fluctuations, and changes in contact pressure, pantographs are prone to arcing during operation, creating high-temperature arcing zones and accompanied by localized thermal anomalies. These arcing events not only accelerate contact surface wear and erosion but can also cause safety incidents such as power outages and catenary damage. Therefore, accurate detection and early warning of arcing anomalies are critical to ensuring electrical system reliability.

[0003] In existing pantograph fault detection technologies, arc detection mostly relies on a single ultraviolet or current signal, which is easily affected by environmental interference and difficult to quantify the arc intensity. Although infrared imaging can monitor temperature, it is not deeply associated with arc characteristics, resulting in a high false alarm rate. In addition, existing methods lack dynamic analysis of the coupling relationship between arcing and temperature anomalies, and are unable to predict potential faults. The innovation of the present invention lies in: through the spatiotemporal synchronous acquisition of ultraviolet and infrared data, combined with arc feature extraction and temperature field modeling, a dynamic threshold judgment model is constructed to achieve fusion decision-making of multimodal data, thereby solving the problems of inaccurate arc detection and delayed warning in the existing technology. Summary of the Invention

[0004] The present invention provides a pantograph arcing anomaly detection and early warning method utilizing the fusion of ultraviolet light and infrared imaging.

[0005] A pantograph arcing anomaly detection and early warning method using ultraviolet light and infrared imaging fusion includes the following linear progressive steps:

[0006] S1: collecting ultraviolet spectrum signals in the pantograph area in real time through an ultraviolet sensor array to extract arcing characteristic data, wherein the arcing characteristic data includes arcing duration, spectrum peak value and frequency domain energy ratio;

[0007] S2: Use an infrared thermal imager to obtain infrared temperature distribution images of the pantograph and catenary, and extract data on abnormal temperature areas based on edge detection and morphological closing operations to eliminate interference from environmental heat sources;

[0008] S3: synchronize the arcing characteristic data of S1 with the temperature anomaly area data of S2 by time stamp and spatial coordinate mapping to generate a fused dataset aligned in time and space;

[0009] S4: Based on the fused data set, the temporal and spatial similarity between the arcing feature data and the temperature anomaly area data is calculated using the weighted Euclidean distance, and the arcing-temperature correlation index is output;

[0010] S5: Build a Gaussian mixture model based on the historical fault case library, and dynamically determine the abnormality level by inputting the arcing-temperature correlation index, temperature gradient, and historical fault probability;

[0011] S6: When the abnormality level exceeds the preset threshold, a graded warning signal is triggered and a fault location report including the arcing position and temperature abnormality range is generated.

[0012] Optionally, the S1 includes:

[0013] S11, collecting ultraviolet spectrum signals of the pantograph area by using an ultraviolet sensor array arranged on the top and both sides of the pantograph, wherein the ultraviolet sensor array covers a wavelength range of 200 nm to 400 nm and a sampling frequency of not less than 100 Hz;

[0014] S12, preprocessing the ultraviolet spectrum signal, including background noise filtering and outlier removal;

[0015] S13, identifying the arcing start and end time periods based on a dynamic threshold of the background ultraviolet intensity, and calculating the length of time that continuously exceeds the threshold as the arcing duration, wherein the dynamic threshold is determined by the background mean and the weighted standard deviation;

[0016] S14, spectrum peak and frequency domain feature extraction: within the arcing duration, the maximum value of the UV intensity is extracted as the spectrum peak;

[0017] S15, frequency domain energy ratio calculation: perform fast Fourier transform on the ultraviolet signal during the arcing period to calculate the ratio of high frequency energy to the total energy.

[0018] Optionally, the S2 includes:

[0019] S21, infrared image acquisition: using an infrared thermal imager placed on the pantograph working path to continuously acquire temperature distribution images of the pantograph and the contact network area;

[0020] S22, temperature anomaly area extraction: Based on the temperature image obtained by the infrared thermal imager, the edge detection algorithm Canny is used to calculate the temperature gradient map, extract the hot zone boundary information, and perform morphological closing operation on the edge image to obtain the temperature anomaly area with clear contours;

[0021] S23, elimination of environmental heat source interference: Combined with the static reference background temperature model, eliminate constant heat source interference and only retain motion-related or sudden abnormal temperature rise areas.

[0022] Optionally, the S22 includes:

[0023] S221, High Temperature Edge Extraction: Based on the temperature gradient of infrared images, the Canny edge detection algorithm is applied to extract the boundary contours of temperature mutation areas and identify potential high temperature anomaly areas;

[0024] S222, morphological closing operation processing: performing a morphological closing operation on the extracted edge image to fill regional holes and eliminate noise breakpoints.

[0025] Optionally, the S23 includes:

[0026] S231, constructing a background temperature model: collecting infrared thermal image data sequences when the pantograph is in a stationary or non-arcing operating state over a period of time;

[0027] Calculate the statistical characteristics of the temperature distribution of each pixel, including the mean and standard deviation;

[0028] S232, screening out constant heat sources: comparing the dynamic temperature map with the background model, performing differential analysis on the infrared image frame during runtime and the background temperature mean map, and defining a dynamic temperature rise mask.

[0029] Optionally, the S3 includes:

[0030] S31, timestamp synchronization: Time-align the arcing characteristic data collected by the UV sensor with the temperature image frame sequence collected by the infrared thermal imager, and use linear interpolation to complete the asynchronous signal alignment;

[0031] S32, spatial coordinate mapping: Based on the relatively fixed installation geometric relationship between the UV and infrared sensors, a spatial mapping function is constructed to map the UV feature point coordinates to the infrared image coordinate system;

[0032] S33, generating a fused data set: merging the time-synchronized arcing characteristics with the temperature anomaly data in the mapped coordinates to form a fused data set in a unified spatiotemporal reference system.

[0033] Optionally, the S4 includes:

[0034] S41, feature vector construction: based on the corresponding time and space positions in the time-space aligned fusion data set, the arcing feature vector and infrared temperature feature vector are extracted respectively;

[0035] S42, spatiotemporal similarity calculation and correlation index output: weighted Euclidean distance is used to calculate the similarity between the normalized ultraviolet and infrared feature vectors, and the similarity distance is converted into an "arcing-temperature correlation index" through an inverse function.

[0036] Optionally, the S42 includes:

[0037] S421, normalization processing: use Min-Max normalization to map each feature to interval;

[0038] S422, weighted Euclidean distance calculation: using normalized weighted Euclidean distance to measure the difference between arcing characteristics and temperature characteristics;

[0039] S423, correlation index calculation: reflecting the spatiotemporal similarity as an arcing-temperature correlation index.

[0040] Optionally, the S5 includes:

[0041] S51, Gaussian mixture model training: Based on the historical pantograph arcing fault case library, a Gaussian mixture model is constructed with arcing-temperature correlation index, temperature gradient and historical fault probability as feature inputs;

[0042] S52, dynamic determination of abnormality level: Input the three-dimensional features of the current monitoring moment into the trained Gaussian mixture model, calculate the posterior probability of each Gaussian component, and output the abnormality level at that moment;

[0043] According to the Gaussian component Danger level label, output abnormal level .

[0044] Optionally, the S6 includes:

[0045] S61, abnormality level judgment and graded warning triggering: Based on the abnormality level output by the Gaussian mixture model, the corresponding graded warning signal is triggered respectively;

[0046] S62, fault location report generation: according to the frame corresponding to the abnormal level , extract the arc space position in the corresponding UV image Bounding box of temperature anomaly area in infrared image , automatically generate positioning reports .

[0047] Beneficial effects of the present invention:

[0048] By integrating ultraviolet and infrared dual-modal sensing methods, the present invention can extract key features from multi-dimensional data such as arcing duration, spectral characteristics and thermal anomaly areas, construct a unified time-space aligned data set, and significantly improve the recognition accuracy and response speed of arcing anomaly events. Compared with traditional detection methods based on a single signal or image source, the present invention has stronger anti-interference and fault location capabilities in complex environments, and is particularly suitable for real-time monitoring under high-speed operating conditions.

[0049] The present invention further introduces arcing-temperature correlation index modeling and constructs a Gaussian mixture model in combination with historical fault data to achieve intelligent judgment of abnormality levels and graded warning. When the abnormality level exceeds the threshold, the system can automatically generate a fault location report including the arcing position and temperature abnormality range. It has high engineering practicality and promotion value, and is of great significance for ensuring the safe and stable operation of the rail transit power supply system. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only for the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0051] Figure 1 is a flow chart of a method according to an embodiment of the present invention;

[0052] Figure 2 This is a data collection diagram of an embodiment of the present invention. DETAILED DESCRIPTION

[0053] The present invention is described in detail below with reference to the accompanying drawings and specific embodiments. It is also noted that, to provide a more detailed description, the following embodiments are best and preferred embodiments, and those skilled in the art may employ alternative methods for implementing certain known technologies. Furthermore, the accompanying drawings are intended only to provide a more detailed description of the embodiments and are not intended to limit the present invention.

[0054] It should be noted that references in the specification to "one embodiment," "an embodiment," "exemplary embodiments," "some embodiments," etc. indicate that the described embodiments may include specific features, structures, or characteristics, but not necessarily every embodiment will include such specific features, structures, or characteristics. Furthermore, when specific features, structures, or characteristics are described in conjunction with an embodiment, it is within the knowledge of persons skilled in the relevant art to implement such features, structures, or characteristics in conjunction with other embodiments (whether or not explicitly described).

[0055] In general, terms can be understood, at least in part, from their use in context. For example, depending at least in part on the context, the term "one or more" as used herein can be used to describe any feature, structure, or characteristic in the singular sense, or can be used to describe a combination of features, structures, or characteristics in the plural sense. Additionally, the term "based on" can be understood as not necessarily intended to convey an exclusive set of factors, but can instead, depending at least in part on the context, allow for the presence of other factors that are not necessarily explicitly described.

[0056] like Figure 1-Figure 2As shown, a pantograph arcing anomaly detection and early warning method using the fusion of ultraviolet light and infrared imaging includes the following linear progressive steps:

[0057] S1: The UV spectral signal of the pantograph area is collected in real time through the UV sensor array to extract arcing characteristic data, including arcing duration, spectral peak value and frequency domain energy ratio;

[0058] S2: Use an infrared thermal imager to obtain infrared temperature distribution images of the pantograph and catenary, and extract data on abnormal temperature areas based on edge detection and morphological closing operations to eliminate interference from environmental heat sources;

[0059] S3: synchronize the arcing characteristic data of S1 with the temperature anomaly area data of S2 by time stamp and spatial coordinate mapping to generate a fused dataset aligned in time and space;

[0060] S4: Based on the fused data set, the temporal and spatial similarity between the arcing feature data and the temperature anomaly area data is calculated using the weighted Euclidean distance, and the arcing-temperature correlation index is output;

[0061] S5: Build a Gaussian mixture model based on the historical fault case library, and dynamically determine the abnormality level by inputting the arcing-temperature correlation index, temperature gradient, and historical fault probability;

[0062] S6: When the abnormality level exceeds the preset threshold, a graded warning signal is triggered and a fault location report including the arcing position and temperature abnormality range is generated.

[0063] S1 includes:

[0064] S11, collecting ultraviolet spectrum signals of the pantograph area by means of ultraviolet sensor arrays arranged on the top and both sides of the pantograph, wherein the ultraviolet sensor array covers a wavelength range of 200 nm to 400 nm and a sampling frequency of not less than 100 Hz;

[0065] S12, preprocessing the UV spectrum signal, including background noise filtering (using sliding window averaging method) and outlier removal. The filtered signal is expressed as:

[0066] is an odd window length,

[0067] in, is the original UV intensity sequence, is the index offset within the local window, is the UV intensity value after filtering;

[0068] S13, based on the dynamic threshold of the background ultraviolet intensity, the arc start and end time periods are identified, and the length of time that the arc continuously exceeds the threshold is calculated as the arc duration. The dynamic threshold is determined by the background mean and the weighted standard deviation, and is expressed as:

[0069] ;

[0070] in, is the dynamic threshold, is the mean UV intensity in the background stage, is the background standard deviation, is the coefficient, and the empirical setting is , used to enhance the anomaly discrimination;

[0071] S14, spectral peak and frequency domain feature extraction: During the arcing duration, the maximum value of the UV intensity is extracted as the spectral peak, expressed as:

[0072] ;

[0073] in, is the arcing time period, is the starting time point of arcing, that is, the time when the UV signal exceeds the arcing judgment threshold for the first time. is the time point when the arc ends, that is, the time when the UV signal is higher than the threshold for the last time. is the ultraviolet wavelength;

[0074] S15, frequency domain energy ratio calculation: Perform fast Fourier transform (FFT) on the UV signal during the arcing period to calculate the ratio of high-frequency energy to the total energy, expressed as:

[0075] ;

[0076] in, is the frequency component, is the spectrum amplitude, It is the high frequency starting threshold, which is set based on the experience of typical arcing frequency (above 25Hz).

[0077] S2 includes:

[0078] S21, infrared image acquisition: The infrared thermal imager placed on the pantograph working path continuously acquires the temperature distribution image of the pantograph and the contact network area. , Represents the spatial pixel coordinates in the image, in units of temperature ;

[0079] S22, temperature anomaly area extraction: Based on the temperature image obtained by the infrared thermal imager, the edge detection algorithm Canny is used to calculate the temperature gradient map, extract the hot zone boundary information, and perform morphological closing operation on the edge image to obtain the temperature anomaly area with clear contours;

[0080] S23, elimination of environmental heat source interference: Combined with the static reference background temperature model, eliminate constant heat source interference and only retain motion-related or sudden abnormal temperature rise areas.

[0081] S22 includes:

[0082] S221, high temperature edge extraction: Based on the temperature gradient of the infrared image, the edge detection algorithm Canny is applied to extract the boundary contour of the temperature mutation area and identify the potential high temperature abnormal area, which is expressed as:

[0083] ;

[0084] in, is the temperature gradient edge intensity, Represent the horizontal and vertical gradients of the temperature graph, is the edge strength of the corresponding pixel point, the larger the value, the more obvious the edge;

[0085] S222, morphological closing operation processing: perform morphological closing operation (dilation followed by erosion) on the extracted edge image to fill regional holes and eliminate noise breakpoints, which is expressed as:

[0086] ;

[0087] in, represents the expansion operation, represents the corrosion operation, is the structural element (a circular core with a radius of 3 is selected), It is the high temperature area mask obtained after closing operation.

[0088] S23 includes:

[0089] S231, build background temperature model (static modeling): collect infrared thermal image data sequence when the pantograph is stationary or in non-arcing operation state for a period of time, expressed as ,in Indicates the Pixels in the frame image The temperature value at

[0090] Calculate the statistical characteristics of the temperature distribution of each pixel, including the mean and standard deviation , expressed as:

[0091] ;

[0092] ;

[0093] S232, Screening out Constant Heat Sources: Comparing the Dynamic Temperature Map with the Background Model, and Running the Infrared Image Frame Perform differential analysis with the background temperature mean map and define the dynamic temperature rise mask, which is expressed as:

[0094] ;

[0095] ;

[0096] in, is the abnormal threshold, ,in is the empirical coefficient (value range is 2 to 3), is the difference between the real-time temperature and the background temperature, It is a binary mask of the temperature rise area, 1 means retaining and 0 means excluding.

[0097] S3 includes:

[0098] S31, timestamp synchronization: Time-align the arcing characteristic data collected by the UV sensor with the temperature image frame sequence collected by the infrared thermal imager. Linear interpolation is used to complete the asynchronous signal alignment, which is expressed as:

[0099] ;

[0100] in, is the temperature value of the abnormal area in the infrared image, are the adjacent sampling moments of the infrared image, is the time point of the UV arcing event, is the interpolated infrared temperature value used to align with the arcing feature, For Infrared temperature image data at the moment, For Infrared temperature image data at the moment;

[0101] S32, spatial coordinate mapping: Based on the relatively fixed installation geometric relationship between the UV and infrared sensors, a spatial mapping function is constructed , the coordinates of the ultraviolet feature points Mapped to the infrared image coordinate system, it is expressed as:

[0102] ( x ir , y ir )=M( x uv , y uv )=H⋅[ x uv y uv 1 ] ;

[0103] in, The pixel coordinates where arcing occurs in the UV image, is the coordinate corresponding to the infrared image, is the 2D homography matrix obtained by camera calibration, is a mapping function used for image registration;

[0104] S33, fusion data set generation: arcing characteristics after time synchronization Temperature anomaly data under mapping coordinates Merge to form a fused dataset under a unified spatiotemporal reference system, expressed as:

[0105] ;

[0106] in, is a spatiotemporally aligned fused dataset, is the characteristic vector of ultraviolet arc (including duration, spectral peak, and frequency domain energy ratio).

[0107] S4 includes:

[0108] S41, feature vector construction: Based on the corresponding time and space positions in the time-space aligned fusion data set, the arc feature vector (including arc intensity, duration, frequency domain energy ratio) and infrared temperature feature vector (including maximum temperature, temperature gradient, temperature variance) are extracted respectively, expressed as:

[0109] ;

[0110] in, is the peak intensity of the UV arc, is the arcing duration, is the proportion of high-frequency energy in the arcing frequency domain, is the highest temperature in the temperature anomaly area, is the temperature gradient (edge ​​slope), is the variance within the temperature region, is the ultraviolet eigenvector (arcing feature), is the infrared feature vector (temperature feature);

[0111] S42, spatiotemporal similarity calculation and correlation index output: weighted Euclidean distance is used to calculate the similarity between the normalized ultraviolet and infrared feature vectors, and the similarity distance is converted into an "arcing-temperature correlation index" through an inverse function.

[0112] S42 includes:

[0113] S421, normalization processing: use Min-Max normalization to map each feature to The interval is expressed as:

[0114] ;

[0115] ;

[0116] ;

[0117] in, is the original eigenvalue, It is a feature The minimum and maximum values ​​among all samples, is the normalized eigenvalue, is the normalized peak intensity of the UV arc, is the normalized arcing duration, is the normalized proportion of high-frequency energy in the arcing frequency domain, is the maximum temperature in the normalized temperature anomaly area, is the normalized temperature gradient (edge ​​slope), is the variance within the normalized temperature region, is the normalized UV feature vector (arcing feature), is the normalized infrared feature vector (temperature feature);

[0118] S422, weighted Euclidean distance calculation: The normalized weighted Euclidean distance is used to measure the difference between the arcing characteristics and the temperature characteristics, expressed as:

[0119] ;

[0120] in, is the weighted Euclidean distance, which is the similarity between the ultraviolet and infrared feature vectors, For the The weighting coefficients of the feature channels, ;

[0121] S423, correlation index calculation: the spatiotemporal similarity Reflected as arc-temperature correlation index , expressed as:

[0122] ;

[0123] in, is the arc-temperature correlation index, The larger the value, the higher the matching degree between arc burning and temperature anomaly.

[0124] S5 includes:

[0125] S51, Gaussian mixture model training: Based on the historical pantograph arcing fault case library, a Gaussian mixture model is constructed with arcing-temperature correlation index, temperature gradient and historical fault probability as feature inputs, expressed as:

[0126] ;

[0127] ;

[0128] in, For the The arc-temperature correlation index of the samples, is the normalized temperature gradient, For the The historical failure probability corresponding to the sample (obtained from historical annotation statistics), is the number of mixed Gaussian distributions (i.e., the number of clusters), which is generally 2-5. is the multidimensional Gaussian probability density function, For the The covariance matrix of the Gaussian distribution describes the correlation between the features of each dimension;

[0129] S52, dynamic determination of abnormality level: The three-dimensional features at the current monitoring moment are input into the trained Gaussian mixture model. By calculating the posterior probability of each Gaussian component, the abnormality level at the moment is comprehensively output, which is expressed as:

[0130] ;

[0131] in, is the new input feature vector, , is the corresponding posterior probability in the GMM model;

[0132] According to the Gaussian component Danger level label, output abnormal level , expressed as:

[0133] .

[0134] S6 includes:

[0135] S61, abnormal level judgment and graded warning triggering: Based on the abnormal level output by the Gaussian mixture model, the corresponding graded warning signal is triggered respectively , expressed as:

[0136] ;

[0137] S62, fault location report generation: according to the frame corresponding to the abnormal level , extract the arc space position in the corresponding UV image Bounding box of temperature anomaly area in infrared image , automatically generate positioning reports , expressed as:

[0138] .

[0139] The present invention encompasses any alternatives, modifications, equivalents, and solutions that fall within the spirit and scope of the present invention. To provide a thorough understanding of the present invention, specific details are described in detail below in connection with the preferred embodiments of the present invention, but those skilled in the art will be able to fully understand the present invention without these detailed descriptions. Furthermore, to avoid unnecessary confusion regarding the essence of the present invention, well-known methods, processes, procedures, components, and circuits have not been described in detail.

[0140] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present invention. These improvements and modifications should also be regarded as within the scope of protection of the present invention.

Claims

1. A pantograph arcing anomaly detection and early warning method using ultraviolet light and infrared imaging fusion, characterized in that: It includes the following linear progressive steps: S1: collecting ultraviolet spectrum signals in the pantograph area in real time through an ultraviolet sensor array to extract arcing characteristic data, wherein the arcing characteristic data includes arcing duration, spectrum peak value and frequency domain energy ratio; S2: Use an infrared thermal imager to obtain infrared temperature distribution images of the pantograph and catenary, and extract data on abnormal temperature areas based on edge detection and morphological closing operations to eliminate interference from environmental heat sources; S3: synchronize the arcing characteristic data of S1 with the temperature anomaly area data of S2 by time stamp and spatial coordinate mapping to generate a fused dataset aligned in time and space; S4: Based on the fused data set, the temporal and spatial similarity between the arcing feature data and the temperature anomaly area data is calculated using the weighted Euclidean distance, and the arcing-temperature correlation index is output; S5: Build a Gaussian mixture model based on the historical fault case library, and dynamically determine the abnormality level by inputting the arcing-temperature correlation index, temperature gradient, and historical fault probability; S6: When the abnormality level exceeds the preset threshold, a graded warning signal is triggered and a fault location report including the arcing position and temperature abnormality range is generated; The S4 includes: S41, feature vector construction: based on the corresponding time and space positions in the time-space aligned fusion data set, the arcing feature vector and infrared temperature feature vector are extracted respectively; S42, spatiotemporal similarity calculation and correlation index output: weighted Euclidean distance is used to calculate the similarity between the normalized UV and IR feature vectors, and the similarity distance is converted into an "arcing-temperature correlation index" through an inverse function; The S5 includes: S51, Gaussian mixture model training: Based on the historical pantograph arcing fault case library, a Gaussian mixture model is constructed with arcing-temperature correlation index, temperature gradient and historical fault probability as feature inputs; S52, dynamic determination of abnormality level: Input the three-dimensional features of the current monitoring moment into the trained Gaussian mixture model, calculate the posterior probability of each Gaussian component, and output the abnormality level at that moment; According to the Gaussian component Danger level label, output abnormal level .

2. The method for detecting and warning pantograph arcing anomalies by combining ultraviolet light and infrared imaging according to claim 1, characterized in that: Said S1 comprises: S11, collecting ultraviolet spectrum signals of the pantograph area by means of ultraviolet sensor arrays arranged on the top and both sides of the pantograph, wherein the ultraviolet sensor array covers a wavelength range of 200 nm to 400 nm and a sampling frequency of not less than 100 Hz; S12, preprocessing the ultraviolet spectrum signal, including background noise filtering and outlier removal; S13, identifying the arcing start and end time periods based on a dynamic threshold of the background ultraviolet intensity, and calculating the length of time that continuously exceeds the threshold as the arcing duration, wherein the dynamic threshold is determined by the background mean and the weighted standard deviation; S14, spectrum peak and frequency domain feature extraction: within the arcing duration, the maximum value of the UV intensity is extracted as the spectrum peak; S15, frequency domain energy ratio calculation: perform fast Fourier transform on the ultraviolet signal during the arcing period to calculate the ratio of high frequency energy to the total energy.

3. The method for detecting and warning pantograph arcing anomalies by combining ultraviolet light and infrared imaging according to claim 2, characterized in that: The S2 includes: S21, infrared image acquisition: using an infrared thermal imager placed on the pantograph working path to continuously acquire temperature distribution images of the pantograph and the contact network area; S22, temperature anomaly area extraction: Based on the temperature image obtained by the infrared thermal imager, the edge detection algorithm Canny is used to calculate the temperature gradient map, extract the hot zone boundary information, and perform morphological closing operation on the edge image to obtain the temperature anomaly area with clear contours; S23, elimination of environmental heat source interference: Combined with the static reference background temperature model, eliminate constant heat source interference and only retain motion-related or sudden abnormal temperature rise areas.

4. The method for detecting and warning pantograph arcing anomalies by combining ultraviolet light and infrared imaging according to claim 3, characterized in that: The S22 includes: S221, High Temperature Edge Extraction: Based on the temperature gradient of infrared images, the Canny edge detection algorithm is applied to extract the boundary contours of temperature mutation areas and identify potential high temperature anomaly areas; S222, morphological closing operation processing: performing a morphological closing operation on the extracted edge image to fill regional holes and eliminate noise breakpoints.

5. The method for detecting and warning pantograph arcing anomalies by combining ultraviolet light and infrared imaging according to claim 3, characterized in that: The S23 includes: S231, constructing a background temperature model: collecting infrared thermal image data sequences of the pantograph in a stationary or non-arcing operating state over a period of time, and calculating the statistical characteristics of the temperature distribution of each pixel, including the mean and standard deviation; S232, screening out constant heat sources: comparing the dynamic temperature map with the background model, performing differential analysis on the infrared image frame during runtime and the background temperature mean map, and defining a dynamic temperature rise mask.

6. The method for detecting and warning pantograph arcing anomalies by combining ultraviolet light and infrared imaging according to claim 3, characterized in that: The S3 includes: S31, timestamp synchronization: Time-align the arcing characteristic data collected by the UV sensor with the temperature image frame sequence collected by the infrared thermal imager, and use linear interpolation to complete the asynchronous signal alignment; S32, spatial coordinate mapping: Based on the relatively fixed installation geometric relationship between the UV and infrared sensors, a spatial mapping function is constructed to map the UV feature point coordinates to the infrared image coordinate system; S33, generating a fused data set: merging the time-synchronized arcing characteristics with the temperature anomaly data in the mapped coordinates to form a fused data set in a unified spatiotemporal reference system.

7. The method for detecting and warning pantograph arcing anomalies by combining ultraviolet light and infrared imaging according to claim 1, characterized in that: The S42 includes: S421, normalization processing: use Min-Max normalization to map each feature to interval; S422, weighted Euclidean distance calculation: using normalized weighted Euclidean distance to measure the difference between arcing characteristics and temperature characteristics; S423, correlation index calculation: reflecting the spatiotemporal similarity as an arcing-temperature correlation index.

8. The method for detecting and warning pantograph arcing anomalies by combining ultraviolet light and infrared imaging according to claim 1, characterized in that: The S6 includes: S61, abnormality level judgment and graded warning triggering: Based on the abnormality level output by the Gaussian mixture model, the corresponding graded warning signal is triggered respectively; S62, fault location report generation: according to the frame corresponding to the abnormal level , extract the arc space position in the corresponding UV image Bounding box of temperature anomaly area in infrared image , automatically generate positioning reports .

Images