Intelligent monitoring method and device for rail transit air conditioning system

Abstract

The invention belongs to the technical field of rail transit intelligent monitoring, and particularly relates to an intelligent monitoring method and device for a rail transit air conditioning system. According to the method, the sensor array with the self-adaptive sampling frequency is used for collecting the multi-modal operation data, then the collected multi-modal operation data is preprocessed, the feature degradation track atlas is established, powerful data support is provided for subsequent fault early warning and diagnosis, and the fault diagnosis accuracy is improved. Historical abnormal events are introduced to dynamically correct the health state evaluation base line, the timeliness and accuracy of the evaluation base line are ensured, in comparative analysis of real-time operation data and the dynamic evaluation base line, a health degree scoring and dynamic threshold mechanism is adopted, quantitative evaluation of the health state of the air conditioner system is achieved, and the evaluation accuracy of the health state of the air conditioner system is improved. According to the method, a decision graph containing fault diagnosis and predictive maintenance suggestions is generated by analyzing the propagation path and time sequence relevance of abnormal parameters in the air conditioning system and combining an equipment topological relation graph, so that the troubleshooting and repairing efficiency is improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of intelligent monitoring of rail transit, and in particular relates to an intelligent monitoring method and device for a rail transit air-conditioning system. Background Art

[0002] With the rapid development of rail transit, the air-conditioning system, as an important component of rail transit vehicles, has become increasingly important. The stability and reliability of its operating status are crucial to ensuring passenger comfort and driving safety. Since trains operate under high-load and changing environmental conditions for a long time, the air-conditioning system is prone to various faults, affecting the normal operation of the train and the riding experience of passengers. Therefore, it is particularly important to ensure the stable operation of the air-conditioning system and to promptly detect and handle potential faults.

[0003] In the existing technology, although there are some methods for monitoring the operating status of air-conditioning systems, most of them rely on manual inspections or regular maintenance, which is not only inefficient but also difficult to detect and handle potential faults in a timely manner. In addition, although there are some sensor-based monitoring systems, these can often only alarm after a fault occurs, lacking predictability and intelligence. Moreover, due to the complexity of the air-conditioning system network, after the fault point is discovered, it is impossible to determine the status of other related equipment in a timely manner, which leads to low efficiency in fault detection and repair. Based on this, this solution proposes an intelligent monitoring method for rail transit air-conditioning systems to solve the above problems. Summary of the Invention

[0004] The purpose of the present invention is to provide an intelligent monitoring method and device for rail transit air-conditioning systems, which can monitor the operating status of the air-conditioning system in real time, improve the efficiency of fault detection and processing, and realize intelligent and predictive maintenance.

[0005] The technical solutions adopted by the present invention are as follows: An intelligent monitoring method for a rail transit air conditioning system, comprising: The distributed sensor array collects multi-modal operating data of the air conditioning system in real time, including temperature parameters, electrical parameters, and mechanical status parameters; Perform dimensionality reduction on multimodal operation data, eliminate redundant information, extract key feature vectors, and establish feature degradation trajectory maps; Build a health status assessment baseline for the air conditioning system based on historical operating data, introduce historical abnormal events to dynamically revise the health status assessment baseline, and output a dynamic assessment baseline; Compare and analyze real-time operating data with dynamic assessment baselines to generate a comprehensive assessment report containing health levels and fault warning information; The propagation path and temporal correlation of abnormal parameters within the air-conditioning system are obtained, and combined with the equipment topology diagram within the air-conditioning system, a decision map containing fault diagnosis and predictive maintenance recommendations is generated.

[0006] In a preferred embodiment, the distributed sensor array adopts an adaptive sampling frequency when collecting multimodal operation data, and dynamically adjusts the sampling interval according to the changing trend of the multimodal operation data. The specific steps are as follows: Presetting an initial sampling frequency for each sensor in the sensor array; Establish a sliding time window and calculate the gradient of the rate of change of multimodal operation data within each sliding time window in real time; Compare the rate of change gradient with a preset corresponding tolerance threshold; When the rate of change gradient exceeds the tolerance threshold, the automatic sampling interval adjustment mechanism is triggered, and the sampling interval of the corresponding sensor is adjusted according to the adjustment result; Among them, after the automatic adjustment mechanism of the sampling interval is triggered, if the gradient of the rate of change in N consecutive sliding time windows still exceeds the tolerance threshold, the sampling frequency will be directly increased to the preset upper limit.

[0007] In a preferred embodiment, when the sampling interval automatic adjustment mechanism is executed, the following steps are included: The difference between the rate of change gradient and the corresponding tolerance threshold is collected and recorded as the adjustment condition parameter; Calculate the adjusted sampling interval of the corresponding sensor based on the adjustment condition parameters, and directly increase the sampling frequency of the corresponding sensor to a preset upper limit when the sampling interval adjustment exceeds the adjustment threshold; Send adjustment signals to similar sensors in the same area to synchronously adjust the sampling frequencies of similar sensors.

[0008] In a preferred embodiment, the steps of performing dimensionality reduction processing on the multimodal operation data, eliminating redundant information, extracting key feature vectors, and establishing a feature degradation trajectory map include: Segment the multimodal operating data through a sliding time window to generate data segments including parameter correlations; The data segments are processed in a standard way, and then a hybrid dimensionality reduction method of principal component analysis and kernel density estimation is used to calculate the principal component scores and kernel density distribution of the data segments and screen the key eigenvectors. The key feature vectors are time-series aligned, and a feature degradation trajectory map is constructed. The feature degradation trend and key inflection points are marked in the degradation trajectory map.

[0009] In a preferred embodiment, the step of constructing a health status assessment baseline for the air conditioning system based on historical operating data includes: Perform data cleaning and standardization on historical operating data, eliminate abnormal operating condition data, and form a standardized training data set containing temperature, electrical, and mechanical parameters; Based on the train operation mode and real-time environmental parameters, the standardized training dataset is divided into data subsets under multiple working conditions; Cluster analysis is performed on each data subset to extract the health status characteristics under each working condition, and the health status characteristics are time-series aligned to build a health status assessment baseline.

[0010] In a preferred embodiment, the step of dynamically revising the health status assessment baseline by introducing historical abnormal events and outputting the dynamic assessment baseline includes: Extract abnormal feature vectors from historical abnormal events, match them with the feature vectors corresponding to the health status assessment baseline, and output the feature deviation under abnormal events; A time decay function is constructed based on the characteristic deviation, and the impact weights of historical abnormal events are dynamically allocated; The dynamically allocated impact weights are integrated with the historical health status assessment baseline, and the health status assessment baseline is corrected to generate a dynamic assessment baseline.

[0011] In a preferred embodiment, the step of comparing and analyzing the real-time operating data with the dynamic assessment baseline to generate a comprehensive assessment report including health level and fault warning information includes: Standardize the real-time operation data to generate a comparative feature vector with the same dimensions as the dynamic assessment baseline; The similarity between the comparison feature vector and the dynamic assessment baseline is calculated, and the calculation result is output as a health score. The health score is then compared with the preset health assessment threshold to identify the operating status; If the health score is lower than the health assessment threshold, the abnormal warning mechanism is triggered. Based on the difference between the feature vector and the dynamic assessment baseline, the health level is determined, and fault warning information is generated based on the health level, and a comprehensive assessment report is output; If the health score is higher than the health assessment threshold, the system will maintain normal operation, record the current operation data, and continuously monitor health changes; The health assessment threshold is a dynamic threshold that is adaptively adjusted based on fluctuations in real-time operating data and the frequency of historical abnormal events. The specific adjustment process is as follows: Construct a short-term sliding window and calculate the short-term fluctuation coefficient based on the fluctuation range of the real-time running data within the short-term sliding window; Construct a long-term sliding window, count the frequency of historical abnormal events within the long-term sliding window, and calculate the frequency correction factor based on the frequency of historical abnormal events within the long-term sliding window; The short-term fluctuation coefficient and the frequency correction factor are combined and input into the threshold adjustment function to generate a dynamically adjusted health assessment threshold.

[0012] In a preferred embodiment, the step of obtaining the propagation path and temporal correlation of abnormal parameters in the air conditioning system and generating a decision map including fault diagnosis and predictive maintenance recommendations in combination with a topological diagram of equipment in the air conditioning system includes: Based on the spatial distribution of characteristic deviations, the propagation paths of abnormal parameters between devices are analyzed to determine the temporal correlation. Compare the time series differences of multi-source sensor data, identify the propagation delay of abnormal events between different devices, and form a fault propagation chain with time stamps; Combined with the connection density of the device topology diagram, the key devices related to the abnormal parameters and the potential impact range are determined, and the key devices are marked as potential fault sources to generate fault diagnosis results; Based on the time stamps of the fault propagation chain, the fault development trend of equipment within the potential impact range is predicted, preventive maintenance strategies are formulated, and a decision-making map is formed.

[0013] The present invention also provides an intelligent monitoring system for a rail transit air conditioning system, using the above-mentioned intelligent monitoring method for a rail transit air conditioning system, comprising: Parameter acquisition module, used to collect multi-modal operating data of the air conditioning system in real time through a distributed sensor array, including temperature parameters, electrical parameters and mechanical state parameters; Parameter preprocessing module, used to perform dimensionality reduction on multimodal operation data, eliminate redundant information, extract key feature vectors, and establish feature degradation trajectory maps; The baseline optimization module is used to construct the health status assessment baseline of the air conditioning system based on historical operating data, and dynamically revise the health status assessment baseline by introducing historical abnormal events, and output a dynamic assessment baseline; The evaluation module is used to compare and analyze real-time operating data with the dynamic evaluation baseline and generate a comprehensive evaluation report containing health level and fault warning information; The fault diagnosis module is used to obtain the propagation path and timing correlation of abnormal parameters in the air-conditioning system, and combine it with the equipment topology diagram in the air-conditioning system to generate a decision map that includes fault diagnosis and predictive maintenance recommendations.

[0014] And, an intelligent monitoring device for a rail transit air conditioning system, the monitoring device comprising: at least one processor; and a memory communicatively coupled to the at least one processor; The memory stores a computer program that can be executed by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the above-mentioned intelligent monitoring method for rail transit air-conditioning systems.

[0015] The technical effects achieved by the present invention are: The present invention utilizes a sensor array with adaptive sampling frequency and dynamically adjusts the sampling strategy according to the data change trend, which not only ensures the comprehensiveness of the data, but also effectively reduces the data volume and improves the data processing efficiency. By performing dimensionality reduction processing and feature extraction on the collected multimodal operation data, a feature degradation trajectory map is established, which provides strong data support for subsequent fault warning and diagnosis. Historical abnormal events are introduced to dynamically correct the health status assessment baseline, ensuring the timeliness and accuracy of the assessment baseline. In the comparative analysis of real-time operation data and dynamic assessment baseline, health score and dynamic threshold mechanism are adopted to realize the quantitative assessment of the health status of the air-conditioning system, which provides a reference basis for fault warning and decision-making. By analyzing the propagation path and time series correlation of abnormal parameters in the air-conditioning system, combined with the equipment topology relationship diagram, a decision map containing fault diagnosis and predictive maintenance suggestions is generated, which can provide maintenance personnel with intuitive and comprehensive fault information, thereby improving the efficiency of fault detection and repair. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 It is a schematic flow chart of the method of the present invention; Figure 2 It is a schematic diagram of the system modules of the present invention; Figure 3 It is a schematic structural diagram of the monitoring device of the present invention. DETAILED DESCRIPTION

[0017] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0018] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may also be implemented in other ways different from those described herein. Those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0019] Secondly, the term "one embodiment" or "embodiment" herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in a preferred embodiment" appearing in various places throughout this specification does not necessarily refer to the same embodiment, nor does it constitute a separate or selective embodiment that is mutually exclusive of other embodiments.

[0020] See also Figure 1 As shown, the present invention provides an intelligent monitoring method for a rail transit air conditioning system, comprising: S1. Real-time collection of multi-modal operating data of the air conditioning system, including temperature parameters, electrical parameters, and mechanical state parameters, through a distributed sensor array; In step S1, during the real-time data monitoring of the rail transit air-conditioning system, multimodal operating data of the air-conditioning system is first collected in real time through a distributed sensor array. The multimodal operating data covers multiple aspects such as temperature parameters, electrical parameters, and mechanical state parameters, thereby ensuring the comprehensiveness of monitoring and providing an information source for subsequent data analysis and fault diagnosis. The distributed sensor array adopts an adaptive sampling frequency when collecting multimodal operating data, and dynamically adjusts the sampling interval according to the changing trend of the multimodal operating data. The specific steps are as follows: Presetting an initial sampling frequency for each sensor in the sensor array; Establish a sliding time window and calculate the gradient of the rate of change of multimodal operation data within each sliding time window in real time; Compare the rate of change gradient with a preset corresponding tolerance threshold; When the rate of change gradient exceeds the tolerance threshold, the automatic sampling interval adjustment mechanism is triggered, and the sampling interval of the corresponding sensor is adjusted according to the adjustment result; Among them, after the automatic adjustment mechanism of the sampling interval is triggered, if the gradient of the rate of change in N consecutive sliding time windows still exceeds the tolerance threshold, the sampling frequency will be directly increased to the preset upper limit Specifically, when collecting multimodal operation data under the air-conditioning system, an initial sampling frequency is first set for each sensor in the sensor array. The initial sampling frequency is preset according to the normal operation status and monitoring requirements of the air-conditioning system. Then, a sliding time window is established. The size of the sliding time window can be set according to the actual monitoring requirements, which is used for the subsequent calculation of the change rate gradient of the multimodal operation data in each time period. The change rate gradient reflects the speed of change of the data in a short period of time. In this embodiment, it is used as the basis for judging the trend of data changes. The calculated change rate gradient is then compared with the preset tolerance threshold. The tolerance threshold is set according to the normal operation fluctuation range of the air-conditioning system, and is used to judge whether the change of the multimodal operation data is within an acceptable range. When the change rate gradient exceeds the tolerance threshold, it indicates that the corresponding data has an abnormality. If the rate of change is high or low, the sampling interval will fluctuate frequently. At this time, the automatic adjustment mechanism of the sampling interval will be triggered. The automatic adjustment mechanism of the sampling interval will adjust the sampling interval of the corresponding sensor according to the size and duration of the change rate gradient. If the change rate gradient continues to exceed the tolerance threshold, the sampling interval will be gradually reduced to increase the density of data collection, so as to more accurately capture the change trend of the data. On the contrary, if the change rate gradient gradually decreases and returns to within the tolerance threshold, the sampling interval will be gradually increased to reduce the frequency of data collection and save system resources. After the automatic adjustment mechanism of the sampling interval is triggered, if the change rate gradient in N consecutive sliding time windows still exceeds the tolerance threshold, the sampling frequency of the corresponding sensor will be directly increased to the preset upper limit, so as to ensure that in the case of abnormal data fluctuations, the data changes can be quickly captured, and timely and accurate information can be provided for subsequent fault diagnosis and predictive maintenance.

[0021] Secondly, when the sampling interval automatic adjustment mechanism is executed, the following steps are included: The difference between the rate of change gradient and the corresponding tolerance threshold is collected and recorded as the adjustment condition parameter; Calculate the adjusted sampling interval of the corresponding sensor based on the adjustment condition parameters, and directly increase the sampling frequency of the corresponding sensor to a preset upper limit when the sampling interval adjustment exceeds the adjustment threshold; Send adjustment signals to similar sensors in the same area to synchronously adjust the sampling frequencies of similar sensors; The automatic sampling interval adjustment mechanism is automatically activated when abnormal data fluctuations are detected. By collecting the difference between the rate of change gradient and the corresponding tolerance threshold, the degree of data fluctuation can be quantified and recorded as an adjustment condition parameter. Then, based on the adjustment condition parameter, the preset measurement function can be used to calculate the adjusted sampling interval of the corresponding sensor. The expression of the measurement function is: , where Indicates the sampling interval adjustment amount, Indicates the current sampling interval, represents the adjustment coefficient, Represents the difference between the change gradient and the tolerance threshold, Represents the attenuation factor, which is mainly used to control the attenuation rate of the adjustment amplitude. Indicates the duration of abnormal fluctuations. It should be clear that if the adjustment amount of the sampling interval exceeds the preset adjustment threshold, it means that the corresponding data fluctuates greatly and more intensive sampling is required to capture data changes. At this time, the sampling frequency of the corresponding sensor will be directly increased to the preset upper limit. In addition, in order to ensure the consistency and accuracy of monitoring, the automatic adjustment mechanism of the sampling interval will also send adjustment signals to similar sensors in the same area, so that they can adjust the sampling frequency synchronously, thereby avoiding the distortion of monitoring data caused by inconsistent sampling frequencies of different sensors, and providing more comprehensive information support for subsequent data analysis and fault diagnosis.

[0022] S2. Perform dimensionality reduction processing on multimodal operation data, eliminate redundant information, extract key feature vectors, and establish feature degradation trajectory maps; In step S2, after the multimodal operation data is collected, the collected multimodal operation data will be subjected to dimensionality reduction processing to eliminate redundant information in the multimodal operation data. At the same time, key feature vectors will be extracted, and a feature degradation trajectory map will be established on this basis to visually display the changing trend of the air-conditioning system state. The steps of performing dimensionality reduction processing on the multimodal operation data, eliminating redundant information, extracting key feature vectors, and establishing the feature degradation trajectory map include: Segment the multimodal operating data through a sliding time window to generate data segments including parameter correlations; The data segments are processed in a standard way, and then a hybrid dimensionality reduction method of principal component analysis and kernel density estimation is used to calculate the principal component scores and kernel density distribution of the data segments and screen the key eigenvectors. Perform time-series alignment on key feature vectors, then construct a feature degradation trajectory map, and mark the feature degradation trend and key inflection points in the degradation trajectory map; Specifically, when extracting key eigenvectors and establishing corresponding feature degradation trajectory maps, the multimodal operation data will first be segmented through a sliding time window to generate data segments containing parameter correlations. The generation of data segments helps to analyze the changing patterns of the data in more detail, and then the data segments will be subjected to standard processing to unify the dimensions and distribution of the data to improve the accuracy of subsequent dimensionality reduction processing. On the basis of standard processing, a hybrid dimensionality reduction method of principal component analysis and kernel density estimation is used to reduce the dimensionality of the data segments. Principal component analysis is used to extract the main components in the data, namely the key eigenvectors, while kernel density estimation can estimate the distribution characteristics of the data. By calculating the principal components of the data segments, The key feature vectors that have an important impact on the state change of the air-conditioning system can be screened out based on the kernel density distribution. For example, the key feature vectors may include abnormal fluctuations in temperature parameters, sudden changes in electrical parameters, and degradation of mechanical state parameters, etc., which can reflect the potential problems of the air-conditioning system during operation. After extracting the key feature vectors, the screened key feature vectors will be time-series aligned to ensure their consistency in the time series. On this basis, a feature degradation trajectory map will be constructed to visually display the changing trend of the air-conditioning system state. In the degradation trajectory map, the feature degradation trend and key inflection points will be marked to identify the state change law and potential fault points of the air-conditioning system.

[0023] S3. Build a health status assessment baseline for the air conditioning system based on historical operating data, dynamically modify the health status assessment baseline by introducing historical abnormal events, and output a dynamic assessment baseline. In step S3, the health status assessment baseline is used as a reference for assessing the health status of the air conditioning system, and is used to determine the current health status of the air conditioning system. In addition, historical abnormal events are introduced to dynamically revise the health status assessment baseline to ensure the accuracy and real-time performance of the assessment baseline. The dynamic assessment baseline is then output to provide a basis for real-time monitoring. The steps of constructing the health status assessment baseline of the air conditioning system based on historical operating data include: Perform data cleaning and standardization on historical operating data, eliminate abnormal operating condition data, and form a standardized training data set containing temperature, electrical, and mechanical parameters; Based on the train operation mode and real-time environmental parameters, the standardized training dataset is divided into data subsets under multiple working conditions; Perform cluster analysis on each data subset to extract health status characteristics under various working conditions, and perform time series alignment on the health status characteristics to build a health status assessment baseline; Specifically, when constructing the health status assessment baseline, the historical operating data will first be cleaned and standardized, and abnormal operating condition data will be eliminated to ensure the accuracy and consistency of the historical operating data. The processed historical operating data will form a standardized training data set containing temperature, electrical, and mechanical parameters, providing a reliable basis for subsequent analysis. Then, according to the train operation mode and real-time environmental parameters, the standardized training data set will be divided into data subsets under multiple working conditions, so as to analyze the operating status of the air-conditioning system under different working conditions. After that, cluster analysis will be performed on each data subset. Cluster analysis can group similar data points into one category, thereby The health status characteristics under each working condition are extracted. Cluster analysis can be performed using methods such as K-means clustering, hierarchical clustering or density clustering. The specific choice depends on the characteristics of the data and the analysis requirements. This can reflect the key parameters and change trends of the air-conditioning system under normal operating conditions. After extracting the health status characteristics, they will also be time-series aligned to ensure the consistency of the data in the time series, thereby constructing the corresponding health status assessment baseline. Specifically, a curve chart of the health status characteristics changing over time can be drawn, and the upper and lower limit thresholds of the health status assessment baseline can be determined based on the curve chart, which can be used as a reference standard for evaluating the health status of the air-conditioning system.

[0024] Secondly, historical abnormal events are introduced to dynamically revise the health status assessment baseline. The steps to output the dynamic assessment baseline include: Extract abnormal feature vectors from historical abnormal events, match them with the feature vectors corresponding to the health status assessment baseline, and output the feature deviation under abnormal events; A time decay function is constructed based on the characteristic deviation, and the impact weights of historical abnormal events are dynamically allocated; The dynamically allocated impact weights are integrated with the historical health status assessment baseline, and the health status assessment baseline is corrected to generate a dynamic assessment baseline. In this embodiment, in order to improve the accuracy of the health status assessment baseline, this embodiment introduces historical abnormal events to dynamically correct it. First, an abnormal feature vector is extracted from the historical abnormal event. The abnormal feature vector represents the key parameter state of the air-conditioning system when an abnormality occurred in the past. Then, the abnormal feature vector is matched with the feature vector corresponding to the health status assessment baseline. The degree of deviation between the two can be used to output the characteristic deviation under the abnormal event. The magnitude of the characteristic deviation directly reflects the degree of influence of the abnormal event on the health status of the air-conditioning system. In order to determine the impact of historical abnormal events on the health status assessment baseline, a time decay function is constructed based on the characteristic deviation to simulate the process in which the impact of abnormal events gradually weakens over time. Based on this, the impact weight of historical abnormal events is dynamically allocated, where the expression of the time decay function is: , where Indicates the impact weight of historical abnormal events, represents the initial weight, represents the decay rate, Indicates the current timestamp, It represents the timestamp of the occurrence of historical abnormal events. The historical abnormal events closer to the current time will be given a greater impact weight, while the historical events farther away will have a smaller impact and their corresponding impact weight will also be smaller. After obtaining the dynamically allocated impact weight, it will be fused with the historical health status assessment baseline. The fusion process is also the process of correcting the health status assessment baseline. By comprehensively considering the impact of historical abnormal events, a dynamic assessment baseline that is closer to the actual health status of the current air-conditioning system can be generated, thereby providing a more reliable reference basis for subsequent real-time monitoring and fault diagnosis.

[0025] S4. Compare and analyze real-time operating data with the dynamic assessment baseline to generate a comprehensive assessment report containing health level and fault warning information; In step S4, when performing real-time operating status monitoring of the equipment under the air-conditioning system, the real-time operating data is compared and analyzed with the dynamic evaluation baseline, so that abnormal conditions in the operation of the air-conditioning system can be discovered in a timely manner, and a comprehensive evaluation report including health levels and fault warning information is generated to provide decision support for maintenance personnel. The step of comparing and analyzing the real-time operating data with the dynamic evaluation baseline to generate a comprehensive evaluation report including health levels and fault warning information includes: Standardize the real-time operation data to generate a comparative feature vector with the same dimensions as the dynamic assessment baseline; The similarity between the comparison feature vector and the dynamic assessment baseline is calculated, and the calculation result is output as a health score. The health score is then compared with the preset health assessment threshold to identify the operating status; If the health score is lower than the health assessment threshold, the abnormal warning mechanism is triggered. Based on the difference between the feature vector and the dynamic assessment baseline, the health level is determined, and fault warning information is generated based on the health level, and a comprehensive assessment report is output; If the health score is higher than the health assessment threshold, the system will maintain normal operation, record the current operation data, and continuously monitor health changes; The health assessment threshold is a dynamic threshold that is adaptively adjusted based on fluctuations in real-time operating data and the frequency of historical abnormal events. The specific adjustment process is as follows: Construct a short-term sliding window and calculate the short-term fluctuation coefficient based on the fluctuation range of the real-time running data within the short-term sliding window; Construct a long-term sliding window, count the occurrence frequency of historical abnormal events within the long-term sliding window, and calculate the frequency correction factor based on the occurrence frequency of historical abnormal events within the long-term sliding window; Combine the short-term fluctuation coefficient and the frequency correction factor, input them into the threshold adjustment function, and generate a dynamically adjusted health assessment threshold; Specifically, when evaluating real-time operation data, the real-time operation data will first be standardized to ensure the comparability and accuracy of the data. The standardization process includes operations such as data dimension unification, outlier removal, and data smoothing to generate a comparison feature vector with the same dimension as the dynamic evaluation baseline. The comparison feature vector can fully reflect the real-time operation status of the air-conditioning system and provide a basis for subsequent similarity calculations. The comparison feature vector and the dynamic evaluation baseline will then be similarity calculated. The similarity calculation can specifically use methods such as cosine similarity, Euclidean distance, or Manhattan distance. The specific selection depends on the characteristics of the data and analysis requirements. The result of the similarity calculation can quantify the degree of difference between the real-time operation data and the dynamic evaluation baseline, and the calculation result will be output as a health score. The level of the health score directly reflects the health status of the air-conditioning system. The health score will then be compared with the preset health assessment threshold to identify the operating status of the air-conditioning system. The health assessment threshold is set according to the normal operating status of the air-conditioning system and historical abnormal events, and is used to distinguish between normal and abnormal states. When the health score is lower than the health assessment threshold, it indicates that the operating status of the air-conditioning system is abnormal. At this time, the abnormal warning mechanism will be triggered. The abnormal warning mechanism can respond quickly and send warning information to maintenance personnel to remind them to pay attention to the operating status of the air-conditioning system. At the same time, the health level will be determined based on the difference between the compared feature vector and the dynamic assessment baseline. The health level can reflect the severity of the abnormality of the air-conditioning system. Different health levels correspond to different maintenance strategies. Fault warning information is generated according to the health level. The fault warning information includes a specific description of the abnormality of the air-conditioning system, possible causes of the fault, and recommended maintenance measures. It is integrated into a comprehensive assessment report and output to the management end for feedback to maintenance personnel so that they can understand the health status of the air-conditioning system in a timely manner and take corresponding maintenance measures to avoid the occurrence or expansion of faults. If the health score is higher than the health assessment threshold, it indicates that the operating status of the air-conditioning system is normal. At this time, the normal operating state will be maintained, and the current operating data will be recorded for subsequent analysis and evaluation. The health changes will be continuously monitored to ensure the stable operation of the air-conditioning system. It should be noted that when determining the health assessment threshold, the fluctuation of real-time operation data and the frequency of historical abnormal events will be comprehensively considered to improve the accuracy and adaptability of the health assessment. Specifically, a short-term sliding window will be constructed, and the short-term fluctuation coefficient will be calculated based on the fluctuation amplitude of the real-time operation data in the short-term sliding window. The short-term fluctuation coefficient can reflect the stability of the air-conditioning system's short-term operating state. When the real-time operation data fluctuates greatly, the short-term fluctuation coefficient will also increase accordingly, indicating that the air-conditioning system may be in an unstable state. At the same time, a long-term sliding window will be constructed, and the frequency of historical abnormal events in the long-term sliding window will be counted. The frequency correction factor is calculated based on the frequency of historical abnormal events in the long-term sliding window. The frequency correction factor can reflect the abnormal condition of the air-conditioning system during long-term operation. When the frequency of historical abnormal events is high, the frequency correction factor will also increase accordingly, indicating that the air-conditioning system has potential failure risks. Finally, the short-term fluctuation coefficient and the frequency correction factor will be combined and input into the threshold adjustment function to generate a dynamically adjusted health assessment threshold. The expression of the threshold adjustment function is: , where Indicates the health assessment threshold after dynamic adjustment, Indicates the initial health assessment threshold, and Represent the weight coefficients of short-term volatility coefficient and frequency correction factor respectively, Represents the short-term fluctuation coefficient (the ratio of the standard deviation of the real-time running data in the short-term sliding window to the standard deviation of the corresponding dynamic evaluation baseline). It represents the frequency correction factor (the ratio of the number of abnormalities within a long-term sliding window to the total number of sampling times). The dynamically adjusted health assessment threshold can more accurately reflect the current health status of the air-conditioning system, providing a more reliable basis for subsequent real-time monitoring and fault warning.

[0026] S5. Obtain the propagation path and temporal correlation of abnormal parameters within the air conditioning system, and generate a decision map containing fault diagnosis and predictive maintenance recommendations based on the equipment topology diagram within the air conditioning system; In step S5, when abnormal parameters are present during the evaluation of the air-conditioning system, the propagation path and temporal correlation of the abnormal parameters within the air-conditioning system are first determined. Then, combined with the equipment topology diagram within the air-conditioning system, a decision map including fault diagnosis and predictive maintenance suggestions can be generated, thereby providing maintenance personnel with targeted maintenance strategies to ensure the stable operation of the air-conditioning system. The steps of obtaining the propagation path and temporal correlation of the abnormal parameters within the air-conditioning system and combining the equipment topology diagram within the air-conditioning system to generate a decision map including fault diagnosis and predictive maintenance suggestions include: Based on the spatial distribution of characteristic deviations, the propagation paths of abnormal parameters between devices are analyzed to determine the temporal correlation. Compare the time series differences of multi-source sensor data, identify the propagation delay of abnormal events between different devices, and form a fault propagation chain with time stamps; Combined with the connection density of the device topology diagram, the key devices related to the abnormal parameters and the potential impact range are determined, and the key devices are marked as potential fault sources to generate fault diagnosis results; Based on the time stamps of the fault propagation chain, the fault development trend of equipment within the potential impact range is predicted, and preventive maintenance strategies are formulated to form a decision-making map; Specifically, after the abnormal parameters are output, the propagation path of the abnormal parameters between the various devices in the air-conditioning system and their temporal correlation will be analyzed first based on the spatial distribution of the characteristic deviation. Then, the time series differences of the multi-source sensor data will be compared, mainly to identify the possible delays when the abnormal event propagates between different devices. Based on this, a fault propagation chain with time marks can be constructed to facilitate the determination of the development trajectory of the fault in the air-conditioning system. Then, combined with the equipment topology relationship diagram in the air-conditioning system, the key equipment and potential impact range related to the abnormal parameters will be further determined. In this process, the key equipment will be marked as potential fault sources, and the corresponding fault diagnosis results will be generated. Finally, based on the time mark of the fault propagation chain, the fault development trend of the equipment within the potential impact range will be predicted, so that targeted preventive maintenance strategies can be formulated, thereby forming a decision map that includes fault diagnosis and predictive maintenance recommendations, providing maintenance personnel with intuitive maintenance guidance, and ensuring the stable operation of the air-conditioning system.

[0027] See also Figure 2 , an intelligent monitoring system for a rail transit air conditioning system, using the above-mentioned intelligent monitoring method for a rail transit air conditioning system, comprising: Parameter acquisition module, used to collect multi-modal operating data of the air conditioning system in real time through a distributed sensor array, including temperature parameters, electrical parameters and mechanical state parameters; Parameter preprocessing module, used to perform dimensionality reduction on multimodal operation data, eliminate redundant information, extract key feature vectors, and establish feature degradation trajectory maps; The baseline optimization module is used to construct the health status assessment baseline of the air conditioning system based on historical operating data, and dynamically revise the health status assessment baseline by introducing historical abnormal events, and output a dynamic assessment baseline; The evaluation module is used to compare and analyze real-time operating data with the dynamic evaluation baseline and generate a comprehensive evaluation report containing health level and fault warning information; The fault diagnosis module is used to obtain the propagation path and timing correlation of abnormal parameters in the air-conditioning system, and combine it with the equipment topology diagram in the air-conditioning system to generate a decision map that includes fault diagnosis and predictive maintenance recommendations.

[0028] In the above, the parameter acquisition module is responsible for acquiring various operating data of the air-conditioning system in real time, which is specifically collected through the sensor array distributed in the air-conditioning system to ensure the comprehensiveness and accuracy of the data. The parameter preprocessing module processes these raw data, removes redundant information through dimensionality reduction technology, and extracts feature vectors that are key to the health status assessment of the air-conditioning system. Based on this, a feature degradation trajectory map is established. The feature degradation trajectory map can intuitively show the performance trend of the air-conditioning system over time, and provide a corresponding reference for subsequent health status assessment and fault diagnosis. The baseline optimization module constructs an initial health status assessment baseline based on historical operating data, which reflects the key parameter range of the air-conditioning system under normal operating conditions. However, considering that the operating status of the air-conditioning system may be affected by various factors and change, historical abnormal events are introduced to evaluate the health status. The evaluation baseline is dynamically revised. By comprehensively considering the impact of historical abnormal events, a dynamic evaluation baseline that is closer to the actual health status of the current air-conditioning system can be generated, providing a more accurate reference for subsequent real-time monitoring and fault diagnosis. The evaluation module is responsible for comparing and analyzing the real-time operation data with the dynamic evaluation baseline. By comparing the differences between the real-time operation data and the dynamic evaluation baseline, it can promptly discover abnormal conditions in the operation of the air-conditioning system and generate a comprehensive evaluation report containing health levels and fault warning information, so as to comprehensively reflect the current health status of the air-conditioning system and provide decision support for maintenance personnel. The fault diagnosis module integrates the propagation path of abnormal parameters, time series correlation and equipment topology relationship diagram within the air-conditioning system to generate a decision map containing fault diagnosis and predictive maintenance suggestions, providing maintenance personnel with targeted maintenance strategies to ensure the stable operation of the air-conditioning system.

[0029] See also Figure 3 , an intelligent monitoring device for a rail transit air conditioning system, the monitoring device comprising: at least one processor; and a memory communicatively coupled to the at least one processor; The memory stores a computer program that can be executed by at least one processor, and the computer program is executed by at least one processor so that the at least one processor can execute the above-mentioned intelligent monitoring method for rail transit air-conditioning systems.

[0030] The processor of the above-mentioned monitoring device can be a high-performance central processing unit (CPU) or a graphics processing unit (GPU), etc., and the memory can include random access memory (RAM), read-only memory (ROM), solid-state drive (SSD) or mechanical hard disk and other storage devices. In addition, the monitoring device can also include an arithmetic unit, input device, output device and network interface, etc. The arithmetic unit can be a logic operation unit, which is used to perform various arithmetic and logical operations and assist the processor to complete complex data processing tasks; the input device can include a keyboard, a mouse, a touch screen, etc., which is used to receive user input instructions and data; the output device can include a display, a printer, etc., which is used to display processing results and output reports; the network interface is used to realize network communication between the monitoring device and other systems or devices for data exchange and remote monitoring.

[0031] It should be noted that, in this document, the terms "comprises," "includes," or any other variations thereof are intended to encompass non-exclusive inclusion, such that a process, apparatus, article, or method comprising a series of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such process, apparatus, article, or method. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of other identical elements in the process, apparatus, article, or method comprising the element.

[0032] The foregoing is merely a preferred embodiment of the present invention. It should be noted that those skilled in the art may make various improvements and modifications without departing from the principles of the present invention, and such improvements and modifications are also within the scope of protection of the present invention. Structures, devices, and operating methods not specifically described or explained herein shall, unless otherwise specified or limited, be implemented in accordance with conventional means in the art.

Claims

1. An intelligent monitoring method for a rail transit air conditioning system, characterized by: include: The distributed sensor array collects multi-modal operating data of the air conditioning system in real time, including temperature parameters, electrical parameters, and mechanical status parameters; Perform dimensionality reduction on multimodal operation data, eliminate redundant information, extract key feature vectors, and establish feature degradation trajectory maps; Build a health status assessment baseline for the air conditioning system based on historical operating data, introduce historical abnormal events to dynamically revise the health status assessment baseline, and output a dynamic assessment baseline; Compare and analyze real-time operating data with dynamic assessment baselines to generate a comprehensive assessment report containing health levels and fault warning information; The propagation path and temporal correlation of abnormal parameters within the air-conditioning system are obtained, and combined with the equipment topology diagram within the air-conditioning system, a decision map containing fault diagnosis and predictive maintenance recommendations is generated.

2. The intelligent monitoring method for rail transit air conditioning system according to claim 1, characterized in that: The distributed sensor array uses an adaptive sampling frequency when collecting multimodal operation data, and dynamically adjusts the sampling interval according to the changing trend of the multimodal operation data. The specific steps are as follows: Presetting an initial sampling frequency for each sensor in the sensor array; Establish a sliding time window and calculate the gradient of the rate of change of multimodal operation data within each sliding time window in real time; Compare the rate of change gradient with a preset corresponding tolerance threshold; When the rate of change gradient exceeds the tolerance threshold, the automatic sampling interval adjustment mechanism is triggered, and the sampling interval of the corresponding sensor is adjusted according to the adjustment result; Among them, after the automatic adjustment mechanism of the sampling interval is triggered, if the gradient of the rate of change in N consecutive sliding time windows still exceeds the tolerance threshold, the sampling frequency will be directly increased to the preset upper limit.

3. The intelligent monitoring method for rail transit air conditioning system according to claim 1, characterized in that: When the sampling interval automatic adjustment mechanism is executed, the following steps are included: The difference between the rate of change gradient and the corresponding tolerance threshold is collected and recorded as the adjustment condition parameter; Calculate the adjusted sampling interval of the corresponding sensor based on the adjustment condition parameters, and directly increase the sampling frequency of the corresponding sensor to a preset upper limit when the sampling interval adjustment exceeds the adjustment threshold; Send adjustment signals to similar sensors in the same area to synchronously adjust the sampling frequencies of similar sensors.

4. The intelligent monitoring method for rail transit air conditioning system according to claim 1, characterized in that: The steps of performing dimensionality reduction processing on the multimodal operation data, eliminating redundant information, extracting key feature vectors, and establishing a feature degradation trajectory map include: Segment the multimodal operating data through a sliding time window to generate data segments including parameter correlations; The data segments are processed in a standard way, and then a hybrid dimensionality reduction method of principal component analysis and kernel density estimation is used to calculate the principal component scores and kernel density distribution of the data segments and screen the key eigenvectors. The key feature vectors are time-series aligned, and a feature degradation trajectory map is constructed. The feature degradation trend and key inflection points are marked in the degradation trajectory map.

5. The intelligent monitoring method for rail transit air conditioning system according to claim 1, characterized in that: The step of constructing a health status assessment baseline for the air conditioning system based on historical operating data includes: Perform data cleaning and standardization on historical operating data, eliminate abnormal operating condition data, and form a standardized training data set containing temperature, electrical, and mechanical parameters; Based on the train operation mode and real-time environmental parameters, the standardized training dataset is divided into data subsets under multiple working conditions; Cluster analysis is performed on each data subset to extract the health status characteristics under each working condition, and the health status characteristics are time-series aligned to build a health status assessment baseline.

6. The intelligent monitoring method for rail transit air conditioning system according to claim 1, characterized in that: The step of introducing historical abnormal events to dynamically correct the health status assessment baseline and outputting the dynamic assessment baseline includes: Extract abnormal feature vectors from historical abnormal events, match them with the feature vectors corresponding to the health status assessment baseline, and output the feature deviation under abnormal events; A time decay function is constructed based on the characteristic deviation, and the impact weights of historical abnormal events are dynamically allocated; The dynamically allocated impact weights are integrated with the historical health status assessment baseline, and the health status assessment baseline is corrected to generate a dynamic assessment baseline.

7. The intelligent monitoring method for rail transit air conditioning system according to claim 1, characterized in that: The step of comparing and analyzing the real-time operating data with the dynamic assessment baseline to generate a comprehensive assessment report including health level and fault warning information includes: Standardize the real-time operation data to generate a comparative feature vector with the same dimensions as the dynamic assessment baseline; The similarity between the comparison feature vector and the dynamic assessment baseline is calculated, and the calculation result is output as a health score. The health score is then compared with the preset health assessment threshold to identify the operating status; If the health score is lower than the health assessment threshold, the abnormal warning mechanism is triggered. Based on the difference between the feature vector and the dynamic assessment baseline, the health level is determined, and fault warning information is generated based on the health level, and a comprehensive assessment report is output; If the health score is higher than the health assessment threshold, the system will maintain normal operation, record the current operation data, and continuously monitor health changes; The health assessment threshold is a dynamic threshold that is adaptively adjusted based on fluctuations in real-time operating data and the frequency of historical abnormal events. The specific adjustment process is as follows: Construct a short-term sliding window and calculate the short-term fluctuation coefficient based on the fluctuation range of the real-time running data within the short-term sliding window; Construct a long-term sliding window, count the frequency of historical abnormal events within the long-term sliding window, and calculate the frequency correction factor based on the frequency of historical abnormal events within the long-term sliding window; The short-term fluctuation coefficient and the frequency correction factor are combined and input into the threshold adjustment function to generate a dynamically adjusted health assessment threshold.

8. The intelligent monitoring method for rail transit air conditioning system according to claim 6, characterized in that: The step of obtaining the propagation path and time sequence correlation of the abnormal parameters in the air conditioning system and generating a decision map including fault diagnosis and predictive maintenance suggestions in combination with the equipment topology diagram in the air conditioning system includes: Based on the spatial distribution of characteristic deviations, the propagation paths of abnormal parameters between devices are analyzed to determine the temporal correlation. Compare the time series differences of multi-source sensor data, identify the propagation delay of abnormal events between different devices, and form a fault propagation chain with time stamps; Combined with the connection density of the device topology diagram, the key devices related to the abnormal parameters and the potential impact range are determined, and the key devices are marked as potential fault sources to generate fault diagnosis results; Based on the time stamps of the fault propagation chain, the fault development trend of equipment within the potential impact range is predicted, preventive maintenance strategies are formulated, and a decision-making map is formed.

9. An intelligent monitoring system for rail transit air conditioning systems, characterized by: The intelligent monitoring method for rail transit air conditioning systems according to any one of claims 1 to 8 comprises: Parameter acquisition module, used to collect multi-modal operating data of the air conditioning system in real time through a distributed sensor array, including temperature parameters, electrical parameters and mechanical state parameters; Parameter preprocessing module, used to perform dimensionality reduction on multimodal operation data, eliminate redundant information, extract key feature vectors, and establish feature degradation trajectory maps; The baseline optimization module is used to construct the health status assessment baseline of the air conditioning system based on historical operating data, and dynamically revise the health status assessment baseline by introducing historical abnormal events, and output a dynamic assessment baseline; The evaluation module is used to compare and analyze real-time operating data with the dynamic evaluation baseline and generate a comprehensive evaluation report containing health level and fault warning information; The fault diagnosis module is used to obtain the propagation path and timing correlation of abnormal parameters in the air-conditioning system, and combine it with the equipment topology diagram in the air-conditioning system to generate a decision map that includes fault diagnosis and predictive maintenance recommendations.

10. An intelligent monitoring device for rail transit air conditioning systems, characterized by: The monitoring device comprises: at least one processor; and a memory communicatively coupled to the at least one processor; The memory stores a computer program that can be executed by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the intelligent monitoring method for rail transit air-conditioning systems according to any one of claims 1 to 8.

Images