A control method for intelligent inspection equipment for subway track line inspection

Abstract

The present invention relates to the technical field of inspection equipment control, in particular to an intelligent inspection equipment control method for subway track line inspection. It acquires the geographical location information of each subway track sub-region to be inspected, generates an inspection route for the inspection equipment according to the geographical location information of each subway track sub-region to be inspected; controls the inspection equipment to inspect the subway track to be inspected according to the inspection route; obtains the subway track feature image of the subway track sub-region to be inspected through the camera mounted on the inspection equipment, and performs feature extraction on the subway track feature image; analyzes the subway track sub-region to be inspected according to the LBP feature matrix of the subway track sub-region to be inspected, and generates an inspection result. The present invention improves the inspection and scheduling capabilities of the inspection equipment, effectively saves inspection resources, improves inspection efficiency and accuracy, and ensures the safety and normal operation of the subway track.

Description

Technical Field

[0001] The present invention relates to the technical field of inspection equipment control, and particularly to an intelligent inspection equipment control method for subway track line inspection. Background Art

[0002] With the acceleration of the urbanization process, the subway, as the core carrier of urban rail transit, faces severe challenges in its safe operation. The deterioration problem of track lines is significantly affected by the dynamic coupling effect of environmental factors. Traditional inspections mostly adopt fixed-cycle manual inspections or single-sensor monitoring, which have problems such as response lag, resource waste, and low hidden danger identification rate. In the prior art, although there are studies applying Internet of Things technology to track monitoring, most focus on the collection of static environmental parameters and lack the ability to dynamically model the synergistic effects of multiple factors such as temperature and humidity, geological deformation, and electromagnetic interference, resulting in insufficient deterioration prediction accuracy. Especially in the trend of subway network complexity, conventional inspection route planning algorithms are difficult to balance real-time risk distribution and equipment scheduling efficiency, and often adopt full-line coverage inspections, causing many unnecessary detection operations and serious waste of inspection equipment resources. In addition, the track surface defect detection technology based on visible light images generally has defects such as single feature extraction dimension and poor texture analysis robustness. Traditional algorithms such as LBP (Local Binary Pattern) have a high false detection rate under dynamic lighting conditions, resulting in low reliability of the inspection results obtained by inspection equipment. Summary of the Invention

[0003] The present invention overcomes the deficiencies of the prior art and provides an intelligent inspection equipment control method for subway track line inspection.

[0004] The technical solution adopted by the present invention to achieve the above object is as follows:

[0005] The present invention discloses an intelligent inspection equipment control method for subway track line inspection, including the following steps:

[0006] Obtain the deterioration index of the subway track to be inspected under the combined action of various environmental factor data, and construct a deterioration index hash dictionary of the subway track to be inspected according to the deterioration index of the subway track to be inspected under the combined action of various environmental factor data;

[0007] Divide the subway track to be inspected into several sub-subway track areas, obtain the historical dynamic environmental factor data sets of each sub-subway track area within a preset time period, and construct a deterioration index evolution curve of each sub-subway track area that decays and aggregates over time according to the historical dynamic environmental factor data sets and in combination with the deterioration index hash dictionary;

[0008] Evaluate each sub - subway track area according to the evolution curve of the deterioration index aggregated over time for each sub - subway track area, and obtain the dynamic risk score value of each sub - subway track area; Determine the sub - subway track areas that need to be inspected and those that do not need to be inspected according to the dynamic risk score values of each sub - subway track area;

[0009] Obtain the geographical location information of each sub - subway track area that needs to be inspected, and generate the inspection route of the inspection equipment according to the geographical location information of each sub - subway track area that needs to be inspected; Control the inspection equipment to inspect the subway track to be inspected according to the inspection route;

[0010] When the inspection equipment moves to the sub - subway track area that needs to be inspected, obtain the subway track feature image of the sub - subway track area that needs to be inspected through the camera mounted on the inspection equipment, and perform feature extraction on the subway track feature image to obtain the LBP feature matrix of the sub - subway track area that needs to be inspected;

[0011] Analyze the sub - subway track area that needs to be inspected according to the LBP feature matrix of the sub - subway track area that needs to be inspected, generate an inspection result, and send the inspection result to the memory of the inspection equipment for storage.

[0012] Preferably, obtain the deterioration index of the subway track to be inspected under the action of each environmental factor data combination, and construct a deterioration index hash dictionary of the subway track to be inspected according to the deterioration index of the subway track to be inspected under the action of each environmental factor data combination. Specifically:

[0013] Obtain the rail material information of the subway track to be inspected, and retrieve the deterioration index of the subway track to be inspected under the action of each environmental factor data combination in the big data network; And construct a dynamic hash mapping table;

[0014] Discretize each environmental factor data combination to obtain the factor discrete value of each environmental factor data combination; And combine each factor discrete value according to a preset weight into a binary feature code to generate a unique hash identification key for each environmental factor data combination;

[0015] Generate a number of key - value pairs in the dynamic hash mapping table according to the hash identification key corresponding to each environmental factor data combination; Map the deterioration index under the action of each environmental factor data combination and the corresponding hash identification key to form a key - value pair association, and obtain the linked - list - type storage slot mapped by each hash identification key;

[0016] Write the corresponding deterioration index into the linked - list - type storage slot mapped by the corresponding hash identification key in the dynamic hash mapping table through a double - hash conflict resolution mechanism to obtain the deterioration index hash dictionary of the subway track to be inspected.

[0017] Preferably, the subway tracks to be inspected are divided into several sub - subway - track areas, and a historical dynamic environmental factor data set of each sub - subway - track area within a preset time period is obtained. According to the historical dynamic environmental factor data set and in combination with the deterioration index hash dictionary, a deterioration index evolution curve that decays and aggregates over time for each sub - subway - track area is constructed. Specifically:

[0018] Perform temporal feature analysis on the historical dynamic environmental factor data set of each sub - subway - track area within a preset time period, and divide dynamic timestamps based on the sliding window mechanism. Extract the corresponding historical dynamic environmental factor data at each timestamp to obtain the historical dynamic environmental factor data of each sub - subway - track area at each timestamp;

[0019] Convert the historical dynamic environmental factor data of each sub - subway - track area at each timestamp into corresponding binary feature codes to obtain the hash identification keys corresponding to the historical dynamic environmental factor data of each sub - subway - track area at each timestamp;

[0020] Import the hash identification keys corresponding to the historical dynamic environmental factor data of each sub - subway - track area at each timestamp into the deterioration index hash dictionary respectively;

[0021] In the order of timestamps, calculate the similarity between the hash identification key corresponding to the historical dynamic environmental factor data at the corresponding timestamp and the hash identification keys stored in each key - value pair in the deterioration index hash dictionary in turn, and mark the key - value pair with the highest similarity. Extract the deterioration index of the sub - subway - track area at the corresponding timestamp from the linked - list storage slot of the marked key - value pair; and so on, to obtain the deterioration indices of each sub - subway - track area based on the timestamp order;

[0022] Preset an exponential decay function, and perform timeliness weighting on the deterioration indices of each sub - subway - track area based on the timestamp order according to the exponential decay function, and output the deterioration index evolution curve that decays and aggregates over time for each sub - subway - track area.

[0023] Preferably, evaluate each sub - subway - track area according to the deterioration index evolution curve that decays and aggregates over time for each sub - subway - track area, and obtain the dynamic risk score value of each sub - subway - track area. Specifically:

[0024] Perform sliding segmentation of the time window on the deterioration index evolution curve of each sub - subway - track area, and extract local time - series segments; and calculate the ratio of the standard deviation to the mean of the local time - series segments to obtain the coefficient of variation of each sub - subway - track area; and calculate the first - order derivative and the second - order derivative of the deterioration index evolution curve of each sub - subway - track area;

[0025] Synchronize the timestamps of the coefficient of variation, the first derivative, and the second derivative of the degradation curve in each sub-region through a multi-source data spatio-temporal alignment algorithm to obtain the synchronized coefficient of variation, the synchronized first derivative, and the synchronized second derivative under the same time reference;

[0026] Deploy a multi-dimensional feature fusion channel, and use spatio-temporal convolution kernels to extract the fluctuation frequency domain features of the synchronized coefficient of variation, the trend directionality features of the synchronized first derivative, and the degradation acceleration characteristics of the synchronized second derivative respectively;

[0027] Introduce a multi-head attention mechanism to perform cross-dimensional interaction on the fluctuation frequency domain features of the synchronized coefficient of variation, the trend directionality features of the synchronized first derivative, and the degradation acceleration characteristics of the synchronized second derivative, and generate a fusion feature tensor containing spatio-temporal correlation weights;

[0028] Input the fusion feature tensor into a deep feature fusion network, and use residual connections and a feature pyramid structure to extract multi-scale spatio-temporal correlation patterns, and output a three-dimensional risk index vector containing fluctuation intensity, trend stability, and acceleration probability;

[0029] Map the three-dimensional risk index vector to a preset risk level space, and perform non-linear weighted aggregation on the discrete levels through a radial basis function neural network to generate the dynamic risk score values of each sub-metro track region.

[0030] Preferably, determine the sub-metro track regions that need to be inspected and the sub-metro track regions that do not need to be inspected according to the dynamic risk score values of each sub-metro track region, specifically:

[0031] Obtain the dynamic risk score values of each sub-metro track region, and set a dynamic risk score value threshold; judge whether the dynamic risk score value of each sub-metro track region is greater than the dynamic risk score value threshold;

[0032] If the dynamic risk score value of a certain sub-metro track region is greater than the dynamic risk score value threshold, then determine that this sub-metro track region is a sub-metro track region that needs to be inspected;

[0033] If the dynamic risk score value of a certain sub-metro track region is not greater than the dynamic risk score value threshold, then determine that this sub-metro track region is a sub-metro track region that does not need to be inspected.

[0034] Preferably, obtain the geographical location information of each sub-metro track region that needs to be inspected, and generate the inspection route of the inspection equipment according to the geographical location information of each sub-metro track region that needs to be inspected, specifically:

[0035] Obtain the track distribution map of the subway track to be inspected, and mark the geographical location information of each sub-metro track region that needs to be inspected in the track distribution map;

[0036] Introduce the particle swarm optimization algorithm, construct the initial solution space based on the geographical location information of each subway track area to be inspected, abstract each sub-area as the particle position in the particle swarm optimization algorithm, and initialize the velocity and position distribution of the particle swarm according to the preset moving speed and starting position of the inspection equipment;

[0037] By continuously according to the velocity and position distribution of the current particle swarm, let each particle move and update in the solution space according to the preset rules. After the preset number of iterations, obtain several different moving paths of the particles from the starting position to each subway track area to be inspected;

[0038] Calculate the total path length value of each moving path; take the moving path with the shortest total path length value as the inspection route of the inspection equipment.

[0039] Preferably, perform feature extraction on the subway track feature image to obtain the LBP feature matrix of the subway track area to be inspected, specifically:

[0040] Perform grayscale and illumination equalization preprocessing on the collected subway track image to enhance the local texture contrast; adopt a dynamic radius strategy to generate circular sampling points in the pixel neighborhood;

[0041] Introduce a multi-scale difference comparison mechanism, perform secondary threshold quantization on the gray difference between the central pixel and the circular sampling points, and generate a rotation-invariant binary coding sequence;

[0042] Construct a local texture descriptor through a sliding window, perform direction weighted fusion in combination with the track direction, and eliminate the feature offset caused by mechanical vibration;

[0043] Perform block overlapping scanning on the whole image, integrate multi-level LBP response maps and perform normalization and dimensionality reduction processing to generate an LBP feature matrix with spatial distribution characteristics.

[0044] Preferably, analyze the subway track area to be inspected according to the LBP feature matrix of the subway track area to be inspected, generate an inspection result, and transmit the inspection result to the memory of the inspection equipment for storage, specifically:

[0045] Obtain the LBP feature matrix corresponding to various warning defects of the subway track through the big data network, construct a database, and import the LBP feature matrix corresponding to various warning defects of the subway track into the database;

[0046] When obtaining the LBP feature matrix of a certain subway track area to be inspected, import the LBP feature matrix of the subway track area to be inspected into the database;

[0047] Calculate the matrix coincidence degree between the LBP feature matrix of the subway track area to be inspected and each LBP feature matrix in the database;

[0048] If the matrix coincidence degree between the LBP feature matrix of the subway track area to be inspected and each LBP feature matrix in the database is not greater than the preset coincidence degree threshold, it indicates that there is no warning defect in the subway track area to be inspected, and mark this subway track area to be inspected as a subway track area that does not require maintenance;

[0049] If the matrix coincidence degree between the LBP feature matrix of the subway track area to be inspected and one or more LBP feature matrices in the database is greater than the preset coincidence degree threshold, it indicates that there is a warning defect in the subway track area to be inspected, and mark this subway track area to be inspected as a subway track area that requires maintenance;

[0050] Generate an inspection result based on the subway track areas that require maintenance and those that do not require maintenance, and send the inspection result to the memory of the inspection device for storage.

[0051] It also includes the following steps:

[0052] If a subway track area to be inspected is marked as a subway track area that requires maintenance, obtain the geographical location information of this subway track area that requires maintenance;

[0053] At the same time, control the inspection device to obtain the terrain image information within a preset radius of this subway track area that requires maintenance; and obtain the type of warning defect of this subway track area that requires maintenance;

[0054] Compress and bundle the geographical location information, terrain image information and warning defect type of this subway track area that requires maintenance to generate a subway track maintenance information data packet, and send the subway track maintenance information data packet to the memory of the inspection device for storage.

[0055] The present invention solves the technical defects existing in the background art, and the present invention has the following beneficial effects: constructing a deterioration index evolution curve of the aggregation of each sub-subway track area over time, and evaluating each sub-subway track area according to the deterioration index evolution curve of the aggregation of each sub-subway track area over time; determining the sub-subway track areas that need to be inspected and the sub-subway track areas that do not need to be inspected according to the dynamic risk score values of each sub-subway track area; obtaining the geographical location information of each sub-subway track area that needs to be inspected, and generating an inspection route of the inspection device according to the geographical location information of each sub-subway track area that needs to be inspected; controlling the inspection device to inspect the subway track to be inspected according to the inspection route; when the inspection device moves to the sub-subway track area that needs to be inspected, obtaining a subway track feature image of the sub-subway track area that needs to be inspected through the camera mechanism carried on the inspection device, extracting features from the subway track feature image, and obtaining the LBP feature matrix of the sub-subway track area that needs to be inspected; analyzing the sub-subway track area that needs to be inspected according to the LBP feature matrix of the sub-subway track area that needs to be inspected, and generating an inspection result. The present invention improves the inspection and scheduling capabilities of the inspection device, effectively saves inspection resources, improves inspection efficiency and accuracy, and ensures the safety and normal operation of the subway track. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the following drawings are only some embodiments of the present invention, and for those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0057] Figure 1 It is the first method flow chart of the intelligent inspection device control method;

[0058] Figure 2 It is the second method flow chart of the intelligent inspection device control method;

[0059] Figure 3 It is the third method flow chart of the intelligent inspection device control method. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0060] In order to be able to more clearly understand the above-mentioned objects, features and advantages of the present invention, the present invention will be further described in detail below with reference to the drawings and specific embodiments. It should be noted that, without conflict, the embodiments of the present application and the features in the embodiments can be combined with each other.

[0061] In the following description, many specific details are set forth in order to provide a thorough understanding of the present invention. However, the present invention may be practiced in other ways different from those described herein. Therefore, the scope of protection of the present invention is not limited by the specific embodiments disclosed below.

[0062] As Figure 1 shown, the present invention discloses an intelligent inspection device control method for subway track line inspection, including the following steps:

[0063] S102. Obtain the deterioration index of the subway track to be inspected under the combined action of various environmental factor data, and construct a deterioration index hash dictionary of the subway track to be inspected according to the deterioration index of the subway track to be inspected under the combined action of various environmental factor data;

[0064] S104. Divide the subway track to be inspected into several sub-subway track areas, obtain the historical dynamic environmental factor data sets of each sub-subway track area within a preset time period, and construct a deterioration index evolution curve of each sub-subway track area that decays and aggregates over time according to the historical dynamic environmental factor data sets and in combination with the deterioration index hash dictionary; among them, the historical dynamic environmental factor data sets of each sub-subway track area within a preset time period can be obtained from the monitoring logs of the subway track to be inspected;

[0065] S106. Evaluate each sub-subway track area according to the deterioration index evolution curve of each sub-subway track area that decays and aggregates over time, and obtain the dynamic risk score value of each sub-subway track area; determine the sub-subway track areas that need to be inspected and the sub-subway track areas that do not need to be inspected according to the dynamic risk score values of each sub-subway track area;

[0066] S108. Obtain the geographical location information of each sub-subway track area that needs to be inspected, and generate an inspection route for the inspection device according to the geographical location information of each sub-subway track area that needs to be inspected; control the inspection device to inspect the subway track to be inspected according to the inspection route;

[0067] S110. When the inspection device moves to a sub-subway track area that needs to be inspected, obtain the subway track feature image of the sub-subway track area that needs to be inspected through the camera mounted on the inspection device, perform feature extraction on the subway track feature image, and obtain the LBP feature matrix of the sub-subway track area that needs to be inspected;

[0068] S112. Analyze the sub-subway track area that needs to be inspected according to the LBP feature matrix of the sub-subway track area that needs to be inspected, generate an inspection result, and transmit the inspection result to the memory of the inspection device for storage.

[0069] Preferably, obtain the deterioration index of the subway track to be inspected under the combined action of various environmental factor data, and construct a deterioration index hash dictionary of the subway track to be inspected according to the deterioration index of the subway track to be inspected under the combined action of various environmental factor data, as Figure 2 shown, specifically:

[0070] S202. Obtain the rail material information of the subway track to be inspected, retrieve the deterioration index of the subway track to be inspected under the combined action of various environmental factor data in the big data network; and construct a dynamic hash mapping table;

[0071] Among them, the environmental factors include but are not limited to temperature and humidity, geological deformation, electromagnetic interference, rainfall, salt fog concentration, and traffic load;

[0072] S204. Discretize each environmental factor data combination to obtain the factor discrete value of each environmental factor data combination; and combine each factor discrete value into a binary feature code according to a preset weight to generate a unique hash identification key for each environmental factor data combination;

[0073] S206. Generate a number of key-value pairs in the dynamic hash mapping table according to the hash identification key corresponding to each environmental factor data combination; associate and map the deterioration index under the action of each environmental factor data combination with the corresponding hash identification key to obtain the linked list storage slot mapped by each hash identification key;

[0074] S208. Write the corresponding deterioration index into the linked list storage slot mapped by the corresponding hash identification key in the dynamic hash mapping table through a double hash conflict resolution mechanism to obtain the deterioration index hash dictionary of the subway track to be inspected.

[0075] Among them, the rail material information of the subway track to be inspected refers to the set of various characteristics and related data of the rails used in the subway track to be inspected. It includes the specific material of the rail, such as what kind of steel or alloy; it also covers the information on the physical properties of the rail, such as hardness, strength, toughness and other indicators; at the same time, it may also involve details such as the chemical composition of the rail.

[0076] The deterioration index is a quantitative index used to measure and evaluate the degree of gradual deterioration of the performance or state of the subway track to be inspected. It comprehensively reflects the decline of the subway track in terms of quality, reliability, safety, etc. relative to the initial state under the influence of various factors, such as environmental conditions (temperature and humidity, geological deformation, etc.), service life, traffic load, etc. The deterioration index can intuitively indicate the current deterioration level of the subway track. The higher the value, the more serious the deterioration degree of the track usually is, and there may be more potential problems or risks.

[0077] It should be noted that first, the rail material information is obtained to accurately retrieve the relevant deterioration indices in the big data network. At the same time, a dynamic hash map is constructed to establish the basic structure for subsequent data processing and storage. Discretizing the environmental factor data combinations and generating unique hash identification keys helps to effectively identify and classify complex environmental factor combinations. By generating key-value pairs and performing associated mapping, the corresponding relationship between the environmental factor data combinations and the deterioration indices is established. The double hash collision resolution mechanism is used to write the deterioration indices into the corresponding storage slots to ensure the accuracy and efficiency of data storage. Through the construction of the deterioration index hash dictionary, the present invention realizes the efficient organization, classification, storage, and association of these complex data, enabling the corresponding deterioration indices to be quickly and accurately found according to the environmental factor combinations, providing a convenient query method for the monitoring, evaluation, and maintenance of subway tracks, and improving the inspection efficiency.

[0078] Preferably, the subway tracks to be inspected are divided into several sub-subway track areas, and the historical dynamic environmental factor data sets of each sub-subway track area within a preset time period are obtained. According to the historical dynamic environmental factor data sets and in combination with the deterioration index hash dictionary, the deterioration index evolution curves of each sub-subway track area decaying and aggregating over time are constructed, as Figure 3 shown, specifically as follows:

[0079] S302. Perform time series feature analysis on the historical dynamic environmental factor data sets of each sub-subway track area within a preset time period, and divide the dynamic timestamps based on the sliding window mechanism. At each timestamp, the corresponding historical dynamic environmental factor data is extracted to obtain the historical dynamic environmental factor data of each sub-subway track area at each timestamp;

[0080] S304. Convert the historical dynamic environmental factor data of each sub-subway track area at each timestamp into the corresponding binary feature codes to obtain the hash identification keys corresponding to the historical dynamic environmental factor data of each sub-subway track area at each timestamp;

[0081] S306. Import the hash identification keys corresponding to the historical dynamic environmental factor data of each sub-subway track area at each timestamp into the deterioration index hash dictionary respectively;

[0082] S308. In the order of timestamps, calculate the similarity between the hash identification keys corresponding to the historical dynamic environmental factor data at the corresponding timestamps and the hash identification keys stored in each key-value pair in the deterioration index hash dictionary in turn, and mark the key-value pair with the highest similarity. Extract the deterioration index of the sub-subway track area at the corresponding timestamp from the linked list storage slot of the marked key-value pair; and so on, to obtain the deterioration indices of each sub-subway track area based on the timestamp order;

[0083] S310. A preset exponential decay function is used to perform timeliness weighting on the deterioration indices of each sub - subway track area based on the time - stamp order, and an evolution curve of the deterioration index aggregated with time decay for each sub - subway track area is output.

[0084] Among them, for the preset exponential decay function, performing timeliness weighting on the deterioration indices of each sub - subway track area based on the time - stamp order by using the exponential decay function and outputting an evolution curve of the deterioration index aggregated with time decay for each sub - subway track area is specifically as follows: The specific form and parameters of the exponential decay function are set by relevant technical personnel. Then, start traversing each sub - subway track area. For each sub - subway track area, obtain the sequence of deterioration indices based on the time - stamp order. Starting from the earliest time - stamp, multiply the deterioration index corresponding to the current time - stamp by the decay weight calculated according to the exponential decay function to obtain the weighted deterioration index value. Then, accumulate this weighted value into a cumulative variable. Then move to the next time - stamp and repeat the calculation and accumulation operations of the deterioration index and the decay weight. During this process, continuously update the cumulative variable. When all time - stamps of this sub - subway track area have been traversed, the value of the cumulative variable obtained is the deterioration index of this sub - subway track area aggregated with time decay. Repeat such operations for all sub - subway track areas, so as to obtain the deterioration index of each sub - subway track area aggregated with time decay. Finally, based on these data, draw an evolution curve of the deterioration index of each sub - subway track area aggregated with time decay to visually display the trend and degree of change of the deterioration index of each sub - subway track area over time.

[0085] It should be noted that by subdividing the subway track to be inspected, it is possible to more accurately analyze the situation of each area. Parse and process the historical dynamic environmental factor data set of each sub - subway track area, extract key features and convert them into hash identification keys, so as to establish a connection with the deterioration index hash dictionary. By calculating the similarity, the corresponding deterioration index is found, realizing the associated acquisition from environmental factor data to the deterioration index. Then, perform weighting processing on the deterioration index to reflect the influence of time factors on deterioration, and finally generate an evolution curve of the deterioration index. Generally speaking, by carefully dividing the subway track into areas, obtaining the historical environmental factor data of each area and conducting in - depth processing, combined with the deterioration index hash dictionary, accurately calculate the deterioration index of each sub - subway track area at different time points, and generate an evolution curve of the deterioration index aggregated with time decay through timeliness weighting, so as to quantify the change trend and degree of deterioration of each sub - subway track area over time.

[0086] Preferably, evaluate each sub - subway track area according to the evolution curve of the deterioration index of each sub - subway track area aggregated with time decay, and obtain the dynamic risk score value of each sub - subway track area, specifically as follows:

[0087] Perform time-window sliding segmentation on the deterioration index evolution curves of each sub-subway track area to extract local time series segments; calculate the ratio of the standard deviation to the mean of the local time series segments to obtain the coefficient of variation of each sub-subway track area; and calculate the first derivative and the second derivative of the deterioration index evolution curves of each sub-subway track area; where the first derivative is the slope of the deterioration index evolution curve and the second derivative is the curvature of the deterioration index evolution curve.

[0088] It should be noted that time-window sliding segmentation and coefficient of variation calculation quantify local volatility through the standard deviation / mean ratio, eliminate the interference of absolute dimensions, and reveal the relative abnormal amplitude of the deterioration process; the first derivative (slope) captures the linear trend of the deterioration rate, and the second derivative (curvature) identifies the non-linear mutation of the deterioration acceleration. The combination of the two can distinguish progressive deterioration from sudden deterioration.

[0089] Perform timestamp synchronization processing on the coefficient of variation, the first derivative, and the second derivative of the deterioration curve of each sub-area through a multi-source data spatio-temporal alignment algorithm to obtain the synchronized coefficient of variation, the synchronized first derivative, and the synchronized second derivative under the same time reference.

[0090] It should be noted that the spatio-temporal alignment algorithm solves the time-series deviation problem of multi-source heterogeneous data (such as sensor data, inspection records, environmental parameters) through timestamp synchronization processing and establishes a basis for spatio-temporal consistency analysis.

[0091] Deploy a multi-dimensional feature fusion channel, and use spatio-temporal convolution kernels to extract the fluctuating frequency domain features of the synchronized coefficient of variation, the trend directionality features of the synchronized first derivative, and the deterioration acceleration characteristics of the synchronized second derivative respectively.

[0092] Introduce a multi-head attention mechanism to perform cross-dimensional interaction on the fluctuating frequency domain features of the synchronized coefficient of variation, the trend directionality features of the synchronized first derivative, and the deterioration acceleration characteristics of the synchronized second derivative to generate a fusion feature tensor containing spatio-temporal correlation weights.

[0093] It should be noted that the spatio-temporal convolution kernel adopts a three-dimensional convolution structure (time axis + space axis + feature axis) and extracts respectively: fluctuating frequency domain features (identifying periodic anomalies through Fourier transform), trend directionality features (strengthening trend signals through moving average filtering), and deterioration acceleration characteristics (enhancing mutation response through second-order difference). The multi-head attention mechanism establishes a three-dimensional correlation matrix of fluctuation intensity - trend direction - acceleration through parallel multi-group self-attention calculations to capture cross-dimensional coupling effects (such as high-frequency fluctuations triggering acceleration threshold transitions).

[0094] Input the fused feature tensor into the deep feature fusion network, and adopt residual connection and feature pyramid structure to extract multi-scale spatio-temporal correlation patterns, and output a three-dimensional risk index vector including fluctuation intensity, trend stability and acceleration probability;

[0095] It should be noted that the residual connection retains shallow local features (such as single-point anomalies), and the feature pyramid fuses multi-scale spatio-temporal patterns (short-term pulse interference and long-term trend drift) to construct a "micro-meso-macro" risk identification system.

[0096] Map the three-dimensional risk index vector to a preset risk level space, and perform non-linear weighted aggregation on the discrete levels through a radial basis function neural network to generate the dynamic risk score values of each sub-subway track area.

[0097] It should be noted that the radial basis function neural network realizes the non-linear mapping of the risk level space through the Gaussian kernel function, and its local response characteristics can finely distinguish the transition intervals between adjacent risk levels, avoiding the boundary blur caused by traditional linear weighting.

[0098] Generally speaking, through this step, the real-time risk status of different sub-subway track areas can be comprehensively quantified, which not only reflects the instantaneous intensity of deterioration (such as mutation risk), but also reveals its long-term evolution law (such as trend stability), providing a high-precision risk assessment basis for subway track maintenance, supporting the decision-making mode upgrade from "passive response" to "active prediction", effectively reducing the probability of sudden failures and optimizing resource allocation, helping to timely discover high-risk areas and take targeted measures, ensuring the safe and stable operation of the subway track system, and improving the accuracy, comprehensiveness and timeliness of the risk assessment of the subway track area.

[0099] Preferably, determine the sub-subway track areas that need to be inspected and the sub-subway track areas that do not need to be inspected according to the dynamic risk score values of each sub-subway track area, specifically:

[0100] Obtain the dynamic risk score values of each sub-subway track area, and set the dynamic risk score value threshold; judge whether the dynamic risk score value of each sub-subway track area is greater than the dynamic risk score value threshold;

[0101] If the dynamic risk score value of a certain sub-subway track area is greater than the dynamic risk score value threshold, then determine that the sub-subway track area is a sub-subway track area that needs to be inspected;

[0102] If the dynamic risk score value of a certain sub-subway track area is not greater than the dynamic risk score value threshold, then determine that the sub-subway track area is a sub-subway track area that does not need to be inspected.

[0103] It should be noted that by obtaining the dynamic risk score values of each sub - subway track area and comparing them with the set threshold values, the sub - subway track areas can be quickly and accurately divided into two categories: areas that need to be inspected and areas that do not need to be inspected. This can make the inspection work of the subway track more targeted, improve the resource utilization efficiency, and enable timely focus and inspection on the high - risk sub - subway track areas that need to be inspected, which helps to discover potential problems in advance and take corresponding measures.

[0104] Preferably, obtain the geographical location information of each sub - subway track area that needs to be inspected, and generate the inspection route of the inspection equipment according to the geographical location information of each sub - subway track area that needs to be inspected. Specifically:

[0105] Obtain the track distribution map of the subway track to be inspected, and mark the geographical location information of each sub - subway track area that needs to be inspected on the track distribution map;

[0106] Introduce the particle swarm optimization algorithm, construct the initial solution space based on the geographical location information of each sub - subway track area that needs to be inspected, abstract each sub - area as the particle position in the particle swarm optimization algorithm, and initialize the velocity and position distribution of the particle swarm according to the preset moving speed and starting position of the inspection equipment;

[0107] By continuously according to the velocity and position distribution of the current particle swarm, let each particle move and update in the solution space according to the preset rules. After the preset number of iterations, obtain several different moving paths of the particles from the starting position to each sub - subway track area that needs to be inspected;

[0108] Calculate the total path length values of each moving path; take the moving path with the shortest total path length value as the inspection route of the inspection equipment.

[0109] It should be noted that by obtaining the geographical location information of the sub - subway track areas that need to be inspected and marking them on the track distribution map, combined with the particle swarm optimization algorithm, the optimal inspection route can be generated efficiently. It can not only ensure that the inspection equipment covers all sub - subway track areas that need to be inspected, but also improve the inspection efficiency and reduce energy consumption.

[0110] Preferably, extract the features of the subway track feature image to obtain the LBP feature matrix of the sub - subway track area that needs to be inspected. Specifically:

[0111] Perform gray - scale and illumination equalization pre - processing on the collected subway track image to enhance the local texture contrast; adopt a dynamic radius strategy to generate circular sampling points within the pixel neighborhood;

[0112] Introduce a multi - scale difference comparison mechanism, perform secondary threshold quantization on the gray - scale difference between the central pixel and the circular sampling points, and generate a rotation - invariant binary coding sequence;

[0113] Among them, the specific steps of introducing a multi-scale difference comparison mechanism to perform secondary threshold quantization on the gray difference between the central pixel and the annular sampling points to generate a rotation-invariant binary coding sequence are as follows: First, determine the position of the central pixel and its surrounding annular sampling points, and calculate the gray difference between the central pixel and each annular sampling point; Second, perform a primary threshold quantization on these gray differences to convert them into preliminary binary codes; Then, apply different secondary threshold quantizations for annular sampling points at different scales to further refine the binary codes; Finally, through a rotation-invariant coding rule, ensure that the generated binary sequence is not affected by image rotation, thereby obtaining a stable feature representation.

[0114] Construct a local texture descriptor through a sliding window, perform direction-weighted fusion in combination with the track direction, and eliminate the feature offset caused by mechanical vibration;

[0115] Among them, the specific steps of constructing a local texture descriptor through a sliding window, performing direction-weighted fusion in combination with the track direction, and eliminating the feature offset caused by mechanical vibration are as follows: Define a sliding window of a fixed size, slide it pixel by pixel on the subway track image, and extract the local texture information within the window after each slide. Calculate local texture features (such as gray-level co-occurrence matrix, histogram of gradients, etc.) within each sliding window to generate local texture descriptors; Detect the track direction through image processing techniques (such as edge detection, Hough transform, etc.) to determine the direction information of the track; Assign a direction weight to each local texture descriptor according to the track direction; Fuse the local texture descriptors with direction weights to generate a global texture descriptor.

[0116] Perform block overlapping scanning on the entire image, integrate multi-level LBP response maps and perform normalization and dimensionality reduction processing to generate an LBP feature matrix with spatial distribution characteristics;

[0117] Among them, the specific steps of performing block overlapping scanning on the entire image, integrating multi-level LBP response maps and performing normalization and dimensionality reduction processing to generate an LBP feature matrix with spatial distribution characteristics are as follows: First, divide the entire image into multiple overlapping small blocks, and calculate the LBP feature response for each small block; Then, integrate multi-level LBP response maps at different scales and directions to form a complete feature map; Next, perform normalization processing on the feature map to ensure that the feature values are on a unified scale; Finally, compress the high-dimensional feature map into a low-dimensional feature matrix through dimensionality reduction techniques (such as principal component analysis) to generate an LBP feature matrix with spatial distribution characteristics, thereby retaining key texture information and reducing data redundancy.

[0118] In summary, through the above steps, not only the robustness and stability of the features are enhanced, the interference of factors such as light and mechanical vibration is reduced, but also the accuracy and comprehensiveness of feature extraction are improved, providing high-quality feature descriptions for the intelligent inspection of subway tracks, which helps to more accurately identify and locate track defects and improve the intelligent level of subway track maintenance.

[0119] Preferably, analyze the to-be-inspected sub-subway track area according to the LBP feature matrix of the to-be-inspected sub-subway track area, generate an inspection result, and transmit the inspection result to the memory of the inspection device for storage. Specifically:

[0120] Obtain the LBP feature matrices corresponding to various warning defects of the subway track through the big data network, construct a database, and import the LBP feature matrices corresponding to various warning defects of the subway track into the database; wherein, the warning defects include but are not limited to cracks, wear, deformation, corrosion, looseness, foreign objects, and uneven settlement.

[0121] After obtaining the LBP feature matrix of a certain to-be-inspected sub-subway track area, import the LBP feature matrix of the to-be-inspected sub-subway track area into the database.

[0122] Calculate the matrix coincidence degree between the LBP feature matrix of the to-be-inspected sub-subway track area and each LBP feature matrix in the database.

[0123] If the matrix coincidence degree between the LBP feature matrix of the to-be-inspected sub-subway track area and each LBP feature matrix in the database is not greater than the preset coincidence degree threshold, it means that there are no warning defects in the to-be-inspected sub-subway track area, and mark the to-be-inspected sub-subway track area as an area of subway track that does not require maintenance.

[0124] If the matrix coincidence degree between the LBP feature matrix of the to-be-inspected sub-subway track area and one or more LBP feature matrices in the database is greater than the preset coincidence degree threshold, it means that there are warning defects in the to-be-inspected sub-subway track area, and mark the to-be-inspected sub-subway track area as an area of subway track that requires maintenance.

[0125] Generate an inspection result according to the area of subway track that requires maintenance and the area of subway track that does not require maintenance, and transmit the inspection result to the memory of the inspection device for storage.

[0126] It further includes the following steps:

[0127] If a certain to-be-inspected sub-subway track area is marked as an area of subway track that requires maintenance, obtain the geographical location information of the area of subway track that requires maintenance.

[0128] Meanwhile, control the inspection device to obtain the topographic image information within a preset radius range of the subway track area to be maintained; and obtain the warning defect types of the subway track area to be maintained.

[0129] Among them, if a sub-subway track area to be inspected is marked as a subway track area to be maintained, the specific warning defect types, such as cracks, wear, deformation, corrosion, looseness, foreign objects, or uneven settlement, are determined by the LBP feature matrix of this area and the LBP feature matrix with a coincidence degree greater than the preset coincidence degree threshold obtained by matching in the database.

[0130] Compress and bundle the geographical location information, topographic image information, and warning defect types of the subway track area to be maintained to generate a subway track maintenance information data packet, and transport the subway track maintenance information data packet to the memory of the inspection device for storage.

[0131] It should be noted that the LBP feature matrices corresponding to various warning defects of the subway track are obtained through the big data network, and a database is constructed. The LBP feature matrix of the sub-subway track area to be inspected is imported into the database for feature matching and defect detection. Calculate the coincidence degree between the LBP feature matrix of the sub-subway track area to be inspected and each LBP feature matrix in the database to determine whether there are warning defects. If the coincidence degree between the LBP feature matrix of the sub-subway track area to be inspected and each LBP feature matrix in the database is not greater than the preset threshold, it is considered that there are no warning defects in this area; otherwise, if the coincidence degree is greater than the preset threshold, it is considered that there are warning defects in this area, and corresponding marks are made. Generate inspection results based on the subway track areas that need to be maintained and those that do not need to be maintained, and store the results in the memory of the inspection device.

[0132] If a sub-subway track area to be inspected is marked as a subway track area to be maintained, further obtain the geographical location information, topographic image information, and warning defect types of this area. Compress and bundle the above information to generate a subway track maintenance information data packet, and store it in the memory of the inspection device, so as to provide comprehensive and accurate data support for the maintenance work of the subway track.

[0133] This method can not only quickly and accurately identify the subway track areas that need to be maintained, but also further obtain detailed maintenance information (such as geographical location, topography, and defect types), generate a maintenance information data packet and store it in the inspection device, thereby providing comprehensive and accurate data support for the maintenance work of the subway track, improving the efficiency and accuracy of the inspection work, and ensuring the safety and normal operation of the subway track.

[0134] In this embodiment, when the inspection device moves to the subway track area to be inspected, the subway track feature image of the area to be inspected is obtained through the camera mounted on the inspection device, specifically as follows:

[0135] The linear acceleration and angular velocity data of the inspection device in the three-dimensional space are collected in real time by the triaxial acceleration sensor and the six-degree-of-freedom inertial measurement unit. Combining the pitch, roll, and yaw angle dynamic parameters obtained by the gyroscope, the motion data of the inspection device is obtained;

[0136] Using the multi-modal data fusion algorithm, the motion data and the displacement of the visual feature points of the camera are fused by Kalman filter, a composite feature matrix containing the device motion vector and the environmental vibration spectrum is constructed, and the motion trajectory of the preset frame image is dynamically trajectory-fitted through the sliding window mechanism to identify the non-autonomous motion component signal caused by mechanical vibration;

[0137] Using the frequency domain decomposition technology, the non-autonomous motion component signal is decomposed into three characteristic sub-bands: the device active motion fundamental frequency, the mechanical resonance frequency band, and the environmental interference harmonic through the fast Fourier transform and combined with the composite feature matrix;

[0138] Based on the image optical flow field analysis, an inter-frame motion compensation model is established. The affine transformation matrix of adjacent frames is calculated through feature point matching, and the characteristic sub-band and the affine transformation matrix are imported into the inter-frame motion compensation model for parameter compensation. After compensation, the displacement compensation parameter difference is obtained;

[0139] If the displacement compensation parameter difference is greater than the preset deviation threshold, a vibration recognition mechanism assisted by deep learning is introduced, and the vibration spectrum characteristics are input into the pre-trained convolutional neural network for pattern classification, and the filter parameter weights are dynamically adjusted; until the displacement compensation parameter difference is less than the preset difference threshold, the compensated image sequence is output;

[0140] Using the multi-exposure fusion technology, the compensated image sequence is processed by spatio-temporal domain fusion, the non-local mean denoising algorithm is used to eliminate the residual blur, and the redundant frames are automatically removed through adaptive threshold detection, and the subway track feature image sequence with spatial continuity and temporal consistency is output.

[0141] It should be noted that before taking a photo, the linear acceleration and angular velocity data of the inspection device in the three-dimensional space are collected, and combined with the dynamic parameters of the pitch, roll and yaw angles obtained by the gyroscope, the motion data of the inspection device are obtained. The dynamic trajectory fitting of the motion trajectory of the preset frame image is carried out through the sliding window mechanism, and the involuntary motion component signal caused by mechanical vibration is identified. The sliding window mechanism is used to process the dynamically changing data to identify and separate the involuntary motion signal caused by mechanical vibration. The frequency domain decomposition technology is used to decompose the involuntary motion component signal into three characteristic sub-bands: the fundamental frequency of the device's active motion, the mechanical resonance frequency band and the environmental interference harmonics through the fast Fourier transform. Through frequency domain analysis, the complex vibration signal is decomposed into different frequency components for subsequent processing. Based on the analysis of the image optical flow field, an inter-frame motion compensation model is established. The affine transformation matrix of adjacent frames is calculated through feature point matching, and the feature sub-band and the affine transformation matrix are imported into the inter-frame motion compensation model for parameter compensation.

[0142] If the difference value of the displacement compensation parameter is greater than the preset deviation value threshold, a vibration recognition mechanism assisted by deep learning is introduced. The vibration spectrum characteristics are input into the pre-trained convolutional neural network for pattern classification, and the filter parameter weights are dynamically adjusted. When the quality of the compensated image does not meet the requirements, the deep learning method is used to further optimize the processing of the vibration signal. The compensated image sequence is subjected to spatio-temporal domain fusion processing, the non-local mean denoising algorithm is used to eliminate the residual blur, and the redundant frames are automatically removed through adaptive threshold detection. Through multi-exposure fusion and denoising, a high-quality, spatially continuous and temporally consistent subway track feature image sequence is finally generated. Through this method, the motion data of the inspection device in the subway track area to be inspected can be accurately obtained, the involuntary motion caused by mechanical vibration and the like can be analyzed and compensated, and finally a subway track feature image sequence with high definition, spatial continuity and temporal consistency is output, effectively improving the accuracy and reliability of obtaining the subway track feature image, reducing the influence of interference factors, providing high-quality image data support for the inspection and analysis of the subway track, and improving the reliability of the inspection results obtained by the inspection device.

[0143] In this embodiment, the inspection device control method further includes the following steps:

[0144] The flight attitude data, environmental parameters and dynamic change values of the device weight of the inspection device are collected in real time through multi-source sensors, including three-dimensional acceleration, angular velocity, GPS position information, barometer altitude data, wind speed and direction sensor readings, air density sensor data and the remaining power information provided by the battery management system;

[0145] Fuse the flight attitude data, environmental parameters, and dynamic change values of the equipment weight, correct the GPS position error using lidar point cloud data, and combine the pose data of the inertial navigation system to calculate the current three-dimensional space coordinates of the inspection equipment;

[0146] Directly incorporate the readings of the wind speed and direction sensor and the data of the air density sensor into the environmental state matrix, and import the remaining power information provided by the battery management system as a key parameter into the environmental state matrix, so as to construct a multi-dimensional environmental state matrix including the current three-dimensional space coordinates of the equipment, wind speed and direction, air density, and remaining battery power;

[0147] Establish a non-linear energy consumption function model with lift coefficient, drag coefficient, and motor efficiency as variables in advance, calculate the power loss curve at different flight speeds in real time through the non-linear energy consumption function model, and obtain the theoretical minimum energy consumption cruise speed interval;

[0148] Introduce an adaptive weight allocation mechanism, and dynamically adjust the priority weights of the three control variables of flight altitude, heading angle deflection rate, and hover duration according to the theoretical minimum energy consumption cruise speed interval and the wind speed fluctuation, track geometric complexity, and remaining power threshold in the multi-dimensional environmental state matrix, and generate a flight parameter combination matching the real-time working conditions;

[0149] Use the genetic algorithm, with the global minimum energy consumption as the objective function, combine the kinematic constraints of the inspection equipment, the obstacle avoidance safety distance limit, and the flight parameter combination matching the real-time working conditions, perform segmented dynamic replanning on the preset inspection path, and generate an energy-optimal smooth flight attitude by iteratively calculating the optimal acceleration vector and pitch angle increment of each waypoint;

[0150] Control the inspection equipment to perform smooth flight according to the energy-optimal smooth flight attitude, ensure that it always operates within the theoretical minimum energy consumption interval throughout the inspection process, so as to achieve the lowest consumption of flight energy.

[0151] In summary, this method ensures that the inspection equipment is always in the optimal energy consumption state throughout the inspection process by dynamically adjusting the flight parameters and attitude, thereby improving the energy consumption efficiency during the inspection process, reducing energy consumption, extending the service life of the equipment, and improving the overall efficiency of the inspection work, maximizing the single inspection distance of the inspection equipment, and improving resource utilization.

[0152] The above is only the specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present invention can easily think of changes or substitutions, which should all be covered within the protection scope of the present invention.

Claims

1. A method for controlling intelligent inspection equipment for subway track line inspection, characterized in that: The following steps are involved: Obtaining a degradation index of the subway track to be inspected under the combined effects of various environmental factor data, and constructing a degradation index hash dictionary of the subway track to be inspected according to the degradation index; The subway track to be inspected is divided into several sub-subway track areas, and a historical dynamic environmental factor data set of each sub-subway track area within a preset time period is obtained. According to the historical dynamic environmental factor data set and in combination with a degradation index hash dictionary, a degradation index evolution curve of each sub-subway track area attenuated and aggregated over time is constructed; Evaluate each sub-subway track area according to the degradation index evolution curve to obtain a dynamic risk score value of each sub-subway track area; determine the sub-subway track area that needs to be inspected and the sub-subway track area that does not need to be inspected according to the dynamic risk score value; Generate inspection routes for inspection equipment based on geographical location information of each sub-subway track area that needs to be inspected; Controlling the inspection equipment to inspect the subway track to be inspected according to the inspection route; When the inspection device moves to the sub-subway track area to be inspected, the subway track feature image of the sub-subway track area to be inspected is obtained through the camera mechanism carried by the inspection device, and the feature of the subway track feature image is extracted to obtain the LBP feature matrix of the sub-subway track area to be inspected; the sub-subway track area to be inspected is analyzed according to the LBP feature matrix, and the inspection result is generated, and the inspection result is transmitted to the memory of the inspection device for storage; Among them, the methods of controlling the inspection equipment also include: The multi-source sensors are used to collect the flight attitude data, environmental parameters and dynamic weight change values ​​of the inspection equipment in real time, including GPS location, wind speed and direction, air density and remaining power information provided by the battery management system; a multi-dimensional environmental state matrix is ​​constructed that includes the current three-dimensional spatial coordinates of the inspection equipment, wind speed and direction, air density and remaining battery power; Real-time calculation of power loss curves at different flight speeds to obtain the theoretical minimum energy consumption cruising speed range; An adaptive weight allocation mechanism is introduced to dynamically adjust the priority weights of multiple flight control variables according to the theoretical minimum energy consumption cruising speed range and the wind speed fluctuation, orbit geometry complexity and remaining power threshold corresponding to the multi-dimensional environmental state matrix, so as to generate a flight parameter combination matching the real-time working condition; Taking the global minimum energy consumption as the objective function, combined with kinematic constraints, obstacle avoidance safety distance limits and flight parameter combinations matching the real-time working conditions, the preset inspection path is dynamically replanned in sections, and the optimal acceleration vector and pitch angle increment of each waypoint are iteratively calculated to generate a smooth flight attitude with optimal energy. The inspection equipment is controlled to perform smooth flight according to the energy-optimal smooth flight posture.

2. The intelligent inspection equipment control method for subway track line inspection according to claim 1 is characterized in that: Obtain the degradation index of the subway track to be inspected under the combined effect of various environmental factor data, and construct a degradation index hash dictionary of the subway track to be inspected according to the degradation index, specifically: Acquire rail material information of the subway track to be inspected, and retrieve the degradation index of the subway track to be inspected under the combined effect of various environmental factor data in the big data network according to the rail material information; And build a dynamic hash map; Discretize each environmental factor data combination to obtain the factor discrete value of each environmental factor data combination; and combine each factor discrete value into a binary feature code according to a preset weight to generate a unique hash identification key for each environmental factor data combination; Generate a number of key-value pairs in the dynamic hash mapping table according to the hash identification key corresponding to each environmental factor data combination; associate the degradation index under the action of each environmental factor data combination with the corresponding hash identification key to perform key-value pair mapping to obtain the linked list storage slot mapped by each hash identification key; The corresponding degradation indexes are written into the linked list storage slots mapped by the corresponding hash identification keys in the dynamic hash mapping table through the double hash conflict resolution mechanism, and the degradation index hash dictionary of the subway track to be inspected is obtained.

3. The intelligent inspection equipment control method for subway track line inspection according to claim 1 is characterized in that: The subway track to be inspected is divided into several sub-subway track areas, and a historical dynamic environmental factor dataset of each sub-subway track area within a preset time period is obtained. According to the historical dynamic environmental factor dataset and in combination with the degradation index hash dictionary, a degradation index evolution curve of each sub-subway track area attenuated and aggregated over time is constructed, specifically: Perform time series feature analysis on the historical dynamic environmental factor data set of each sub-subway track area within a preset time period, divide the dynamic timestamps based on the sliding window mechanism, extract the corresponding historical dynamic environmental factor data at each timestamp, and obtain the historical dynamic environmental factor data of each sub-subway track area at each timestamp; Convert the historical dynamic environmental factor data of each sub-subway track area at each time stamp into a corresponding binary feature code to obtain a hash identification key corresponding to the historical dynamic environmental factor data of each sub-subway track area at each time stamp; Importing the hash identification key corresponding to the historical dynamic environmental factor data of each sub-subway track area at each time stamp into the degradation index hash dictionary; In order of timestamps, the similarities between the hash identification key corresponding to the historical dynamic environmental factor data of the corresponding timestamp and the hash identification key stored in each key-value pair in the degradation index hash dictionary are calculated in sequence, and the key-value pair with the highest similarity is marked, and the degradation index of the sub-subway track area at the corresponding timestamp is extracted from the linked list storage slot of the marked key-value pair; By analogy, the degradation index of each sub-subway track area based on the timestamp order is obtained; An exponential decay function is preset, and based on the exponential decay function, the degradation index of each sub-subway track area based on the timestamp order is time-weighted, and an evolution curve of the degradation index of each sub-subway track area that decays and aggregates over time is output.

4. The intelligent inspection equipment control method for subway track line inspection according to claim 1 is characterized in that: Each sub-subway track area is evaluated according to the degradation index evolution curve to obtain a dynamic risk score value of each sub-subway track area, specifically: Perform time window sliding segmentation on the degradation index evolution curve of each sub-subway track area to extract local time series segments; calculate the ratio of the standard deviation to the mean of the local time series segments to obtain the coefficient of variation of each sub-subway track area; and calculate the first-order derivative and the second-order derivative of the degradation index evolution curve of each sub-subway track area; The coefficient of variation, first-order derivative and second-order derivative of the degradation curve of each sub-region are synchronized by time stamp through the multi-source data spatiotemporal alignment algorithm to obtain the synchronized coefficient of variation, synchronized first-order derivative and synchronized second-order derivative under the same time reference; Deploy a multi-dimensional feature fusion channel and use spatiotemporal convolution kernels to extract the fluctuation frequency domain characteristics of the synchronized coefficient of variation, the trend directional characteristics of the synchronized first-order derivative, and the degradation acceleration characteristics of the synchronized second-order derivative. A multi-head attention mechanism is introduced to perform cross-dimensional interaction on the fluctuation frequency domain characteristics of the synchronized coefficient of variation, the trend directional characteristics of the synchronized first-order derivative, and the degradation acceleration characteristics of the synchronized second-order derivative, generating a fused feature tensor containing spatiotemporal correlation weights. The fused feature tensor is input into the deep feature fusion network, and the multi-scale spatiotemporal correlation pattern is extracted using residual connection and feature pyramid structure, and a three-dimensional risk indicator vector including volatility intensity, trend stability and acceleration probability is output; The three-dimensional risk index vector is mapped to a preset risk level space, and the discrete levels are nonlinearly weighted aggregated through a radial basis function neural network to generate a dynamic risk score value for each sub-subway track area.

5. The intelligent inspection equipment control method for subway track line inspection according to claim 1 is characterized in that: The sub-subway track areas that need to be inspected and the sub-subway track areas that do not need to be inspected are determined according to the dynamic risk score values, specifically: Obtaining a dynamic risk score value of each sub-subway track area and setting a dynamic risk score value threshold; determining whether the dynamic risk score value of each sub-subway track area is greater than the dynamic risk score value threshold; If the dynamic risk score value of a certain sub-subway track area is greater than the dynamic risk score value threshold, the sub-subway track area is determined as a sub-subway track area that needs to be inspected; If the dynamic risk score value of a certain sub-subway track area is not greater than the dynamic risk score value threshold, the sub-subway track area is determined as a sub-subway track area that does not require inspection.

6. The intelligent inspection equipment control method for subway track line inspection according to claim 1 is characterized in that: The inspection route of the inspection equipment is generated according to the geographical location information of each sub-subway track area to be inspected, specifically: Obtain a track distribution map of the subway track to be inspected, and mark the geographical location information of each sub-subway track area to be inspected in the track distribution map; The particle swarm optimization algorithm is introduced to construct the initial solution space based on the geographical location information of each subway track area to be inspected. Each sub-area is abstracted as a particle position in the particle swarm optimization algorithm, and the speed and position distribution of the particle swarm are initialized according to the preset moving speed and starting position of the inspection equipment. By continuously distributing the speed and position of the current particle group, each particle is moved and updated in the solution space according to the preset rules. After a preset number of iterations, several different moving paths of particles from the starting position to each sub-subway track area to be inspected are obtained; Calculate the total path length value of each moving path; The moving path with the shortest total path length value is used as the inspection route of the inspection equipment.

7. The intelligent inspection equipment control method for subway track line inspection according to claim 1 is characterized in that: Feature extraction is performed on the subway track feature image to obtain the LBP feature matrix of the sub-subway track area to be inspected, specifically: The collected subway track images are preprocessed by grayscale conversion and illumination balance to enhance local texture contrast; a dynamic radius strategy is used to generate annular sampling points within the pixel neighborhood; A multi-scale differential comparison mechanism is introduced to perform secondary threshold quantization on the grayscale difference between the central pixel and the ring sampling point to generate a rotationally invariant binary code sequence. The local texture descriptor is constructed through a sliding window, and the direction is weighted and fused in combination with the track direction to eliminate the feature offset caused by mechanical vibration. The whole image is scanned in blocks and overlapped, and the multi-level LBP response images are integrated and normalized dimension reduction is performed to generate an LBP feature matrix with spatial distribution characteristics.

8. The intelligent inspection equipment control method for subway track line inspection according to claim 1 is characterized in that: The sub-subway track area to be inspected is analyzed according to the LBP feature matrix, an inspection result is generated, and the inspection result is transmitted to the memory of the inspection device for storage, specifically: Obtaining LBP feature matrices corresponding to various early warning defects of subway tracks through a big data network, building a database, and importing the LBP feature matrices corresponding to various early warning defects of subway tracks into the database; After obtaining the LBP feature matrix of a certain subway track area that needs to be inspected, importing the LBP feature matrix of the subway track area that needs to be inspected into the database; Calculate the matrix overlap between the LBP feature matrix of the sub-subway track area to be inspected and each LBP feature matrix in the database; If the matrix overlap between the LBP feature matrix of the sub-subway track area to be inspected and each LBP feature matrix in the database is not greater than the preset overlap threshold, it means that there is no warning defect in the sub-subway track area to be inspected, and the sub-subway track area to be inspected is marked as a subway track area that does not require maintenance; If the matrix overlap between the LBP feature matrix of the sub-subway track area to be inspected and one or more LBP feature matrices in the database is greater than a preset overlap threshold, it means that there is a warning defect in the sub-subway track area to be inspected, and the sub-subway track area to be inspected is marked as a subway track area that needs maintenance; An inspection result is generated according to the subway track area that needs maintenance and the subway track area that does not need maintenance, and the inspection result is transmitted to a memory of the inspection device for storage.

9. The intelligent inspection equipment control method for subway track line inspection according to claim 8, characterized in that: The following steps are also included: If a certain sub-subway track area requiring inspection is marked as a subway track area requiring maintenance, obtaining geographic location information of the subway track area requiring maintenance; At the same time, the inspection equipment is controlled to obtain the topographic image information of the subway track area that needs maintenance within a preset radius; and the warning defect type of the subway track area that needs maintenance; The geographical location information, topographic image information and warning defect type of the subway track area requiring maintenance are compressed and bundled to generate a subway track maintenance information data packet, and the subway track maintenance information data packet is transmitted to the memory of the inspection equipment for storage.

Images