Intelligent test method and system for vibration and noise influence of rail transit

Abstract

The application discloses a kind of vibration noise influence intelligent test method and system of rail transit, mainly related to noise test technical field. Including: obtaining target test track, sensor layout is carried out based on the route characteristics of target test track, obtains the noise sensor array of layout completion;Obtain noise monitoring data sequence array;Iterate noise monitoring data sequence array to carry out abnormal monitoring identification to noise sensor array, obtain abnormal noise sensor subarray and normal noise sensor subarray;Utilize abnormal test analysis channel set and normal test analysis network layer respectively to abnormal noise sensor subarray and normal noise sensor subarray carry out online monitoring analysis, obtain target vibration noise test result.The beneficial effects of the present application are that: it solves the technical problem that the existing technology lacks continuous monitoring analysis of the vibration noise influence of rail transit, leading to low reliability of test results, achieving the technical effect of improving test accuracy.

Description

Technical Field

[0001] This invention relates to the field of noise testing technology, and specifically to an intelligent testing method and system for the vibration and noise effects of rail transit. Background Technology

[0002] Currently, the scale of rail transit construction is constantly expanding, and the resulting vibration and noise pollution problems are becoming increasingly serious. Existing rail transit noise monitoring methods mostly rely on manual data collection or periodic spot checks, which cannot effectively achieve real-time online monitoring. Furthermore, the timeliness and accuracy of data collection are insufficient, making it difficult to quickly identify and accurately locate abnormal noise and vibration areas, thus posing challenges to the safe operation and maintenance management of rail transit.

[0003] Existing technologies suffer from a lack of continuous monitoring and analysis of the impact of vibration and noise on rail transit, resulting in low reliability of test results. Summary of the Invention

[0004] This application provides an intelligent testing method and system for the vibration and noise impact of rail transit, which addresses the technical problem of low reliability of test results due to the lack of continuous monitoring and analysis of the vibration and noise impact of rail transit in the prior art.

[0005] In view of the above problems, this application provides an intelligent testing method and system for vibration and noise impact on rail transit.

[0006] The first aspect of this application provides an intelligent testing method for the vibration and noise impact of rail transit, the method comprising:

[0007] Obtain the target test track, and deploy sensors based on the route characteristics of the target test track to obtain a noise sensor array that has been deployed.

[0008] Train passage tests are conducted on the target test track according to the preset test plan, and data from the sensor array is collected simultaneously to obtain a noise monitoring data sequence array.

[0009] The noise monitoring data sequence array is traversed to perform anomaly detection and identification on the noise sensor array, thereby obtaining an abnormal noise sensor subarray and a normal noise sensor subarray, wherein the abnormal noise sensor subarray includes an abnormal monitoring scale subarray;

[0010] An abnormal test analysis channel set is constructed based on the abnormal monitoring scale subarray, and a conventional test analysis network layer is constructed based on the conventional monitoring frequency. The abnormal test analysis channel set and the conventional test analysis network layer are used to perform online monitoring and analysis on the abnormal noise sensor subarray and the conventional noise sensor subarray, respectively, to obtain the target vibration and noise test results.

[0011] A second aspect of this application provides an intelligent testing system for the vibration and noise impact of rail transit, the system comprising:

[0012] The noise sensor array acquisition module is used to acquire the target test track, deploy sensors based on the route characteristics of the target test track, and obtain the deployed noise sensor array.

[0013] The monitoring data sequence array acquisition module is used to conduct train passage tests on the target test track according to a preset test plan, and simultaneously collect data from the sensor array to obtain a noise monitoring data sequence array.

[0014] The sensor subarray acquisition module is used to traverse the noise monitoring data sequence array to perform anomaly monitoring and identification on the noise sensor array, and obtain an abnormal noise sensor subarray and a normal noise sensor subarray, wherein the abnormal noise sensor subarray includes an abnormal monitoring scale subarray;

[0015] The test result acquisition module is used to construct an abnormal test analysis channel set based on the abnormal monitoring scale subarray and a conventional test analysis network layer based on the conventional monitoring frequency. The abnormal test analysis channel set and the conventional test analysis network layer are used to perform online monitoring and analysis on the abnormal noise sensor subarray and the conventional noise sensor subarray, respectively, to obtain the target vibration and noise test results.

[0016] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0017] This application acquires a target test track, deploys sensors based on the track's route characteristics, and obtains a completed noise sensor array. Then, according to a pre-set test plan, train passage tests are conducted on the target test track, simultaneously collecting data from the sensor array to obtain a noise monitoring data sequence array. This array is then traversed to identify anomalies in the noise sensor array, resulting in an abnormal noise sensor subarray and a conventional noise sensor subarray. The abnormal noise sensor subarray includes an abnormal monitoring scale subarray. An abnormal test analysis channel set is constructed based on the abnormal monitoring scale subarray, and a conventional test analysis network layer is built based on the conventional monitoring frequency. The abnormal test analysis channel set and the conventional test analysis network layer are used to perform online monitoring and analysis on the abnormal and conventional noise sensor subarrays, respectively, to obtain the target vibration and noise test results. This achieves the technical effect of intelligent online monitoring of rail transit and improving the reliability of test results. Attached Figure Description

[0018] Appendix Figure 1This is a schematic diagram of a smart testing method for vibration and noise impact on rail transit provided in an embodiment of the present invention.

[0019] Appendix Figure 2 This is a schematic diagram of the structure of an intelligent testing system for vibration and noise impact on rail transit provided in an embodiment of the present invention.

[0020] The labels shown in the attached diagram:

[0021] The module 11 obtains noise sensor array, the module 12 obtains monitoring data sequence array, the module 13 obtains sensor subarray, and the module 14 obtains test results. Detailed Implementation

[0022] The present invention will be further illustrated below with reference to specific embodiments. It should be understood that these embodiments are for illustrative purposes only and are not intended to limit the scope of the invention. Furthermore, it should be understood that after reading the teachings of this invention, those skilled in the art can make various alterations or modifications to the invention, and these equivalent forms also fall within the scope defined by the appended claims. It should be noted that the terms "comprising" and "having" are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or server that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or modules not explicitly listed or inherent to these processes, methods, products, or devices.

[0023] Example 1, as shown in the appendix Figure 1 As shown, this application provides an intelligent testing method for the vibration and noise impact of rail transit, wherein the method includes:

[0024] S1: Obtain the target test track, and deploy sensors based on the route characteristics of the target test track to obtain a noise sensor array that has been deployed.

[0025] In one possible embodiment, the target test track refers to a specific rail transit line that needs to be tested for vibration and noise impact; it can be a complete line or a specific area within the line to be monitored. Route characteristics include track structural features (e.g., rail type, roadbed condition, turnout location, etc.) and surrounding environmental features (e.g., tunnels, viaducts, residential areas, etc.). The noise sensor array refers to a monitoring network formed by combining multiple noise sensors according to a preset spatial layout strategy, used to comprehensively and accurately acquire noise and vibration data along the track.

[0026] In the specific implementation process, the target track is first surveyed on-site or analyzed using historical data to identify and determine the different route characteristics along the line, such as straight sections, curved sections, switch areas, tunnels, and elevated sections, among other track conditions. Then, based on these route characteristics, appropriate noise sensors are selected and their deployment locations are planned. For example, sensor density is appropriately increased in switch and curved areas to improve detection accuracy in these critical areas. Finally, a noise sensor array with reasonable spatial coverage and appropriate density is formed through installation and commissioning. This step effectively improves the accuracy and representativeness of subsequent test data, providing a precise data foundation for subsequent online real-time monitoring and analysis, ensuring that the test results objectively and accurately reflect the impact of actual vibration and noise in rail transit.

[0027] S2: Conduct train passage tests on the target test track according to the preset test plan, and simultaneously collect data from the sensor array to obtain a noise monitoring data sequence array;

[0028] In one possible embodiment, the preset test plan is a specific content of train operation test planned and formulated in advance by those skilled in the art, including the train's operating speed, route, test time, and number of tests. Obtaining the preset test plan lays the groundwork for subsequent actual train operation on the target test track, triggering track vibration and noise under actual operating conditions. Each noise monitoring data point in the noise monitoring data sequence array refers to a data set collected by noise sensors and arranged in chronological order, where the monitoring data generated by each sensor constitutes a separate time series.

[0029] First, based on the pre-defined test plan, the operating conditions of the train on the target test track are clearly defined, including specific operating speeds (e.g., different speed levels such as 60 km / h, 80 km / h, and 120 km / h), the type of test train (e.g., high-speed train, subway train, urban rail train), and the time period and number of tests. Then, when the train passes the target track, the noise sensor array is activated. Each sensor is precisely synchronized using GPS timing technology or a network timing protocol (such as NTP), enabling real-time and continuous acquisition of vibration and noise data during the train's passage, forming a clear temporal noise monitoring data sequence array. This step achieves accurate monitoring of real-world rail transit conditions. The obtained data array accurately reflects the dynamic changes in track vibration and noise under different areas and conditions, providing high-quality basic data for subsequent anomaly identification and online analysis.

[0030] S3: Traverse the noise monitoring data sequence array to perform anomaly monitoring and identification on the noise sensor array, and obtain an abnormal noise sensor subarray and a normal noise sensor subarray, wherein the abnormal noise sensor subarray includes an abnormal monitoring scale subarray;

[0031] Furthermore, the noise monitoring data sequence array is traversed to perform anomaly detection and identification on the noise sensor array, resulting in an abnormal noise sensor subarray and a conventional noise sensor subarray. The abnormal noise sensor subarray includes an abnormal monitoring scale subarray. In this embodiment, step S3 further includes:

[0032] Arbitrarily extract the first noise monitoring data sequence from the noise monitoring data sequence array;

[0033] Noise monitoring features are extracted from the first noise monitoring data sequence to obtain a first noise monitoring data feature set.

[0034] In the pre-built anomaly detection and identification prototype library, the first noise monitoring data feature set is matched. If the match is successful, the noise sensor corresponding to the first noise monitoring data sequence is marked as abnormal and added to the abnormal noise sensor subarray. The first anomaly monitoring scale obtained by matching is added to the anomaly monitoring scale subarray.

[0035] In one possible embodiment, the anomalous noise sensor subarray refers to the subset of sensors identified as having anomalous data. The conventional noise sensor subarray, on the other hand, is the subset of sensors that did not identify any anomalies. The anomaly monitoring scale subarray represents the interval scale used when monitoring and analyzing the anomalous noise sensor subarray.

[0036] Preferably, a first noise monitoring data sequence is arbitrarily extracted from the noise monitoring data sequence array. Signal processing and analysis algorithms (e.g., Fourier Transform (FFT), Short-Time Fourier Transform (STFT), Wavelet Transform, or statistical feature extraction) are used to extract features from this data sequence, generating a first noise monitoring data feature set, including indicators such as spectral characteristics, energy distribution, peak position, and vibration amplitude variation trend. Optionally, multiple sample noise monitoring data sequences and multiple sample noise monitoring data feature sets are obtained as training data to supervise the training of a framework built on a feedforward neural network until the training converges, obtaining a trained feature recognizer. The feature recognizer is then used to perform feature recognition on the first noise monitoring data sequence to obtain the first noise monitoring data feature set.

[0037] Then, the extracted first noise monitoring data feature set is matched and compared with a pre-constructed anomaly detection and identification prototype library. Specifically, the cosine similarity of the first noise monitoring data feature set with the identification internal nodes in the anomaly detection and identification tree of the anomaly detection and identification prototype library is calculated. If the calculation result is greater than or equal to a preset similarity (pre-set by those skilled in the art), a successful match is achieved, the corresponding noise sensor is determined to be an anomaly, and it is assigned to the anomaly noise sensor subarray. The corresponding anomaly scale is also recorded and added to the anomaly detection scale subarray. If no match is found, the noise sensor is assigned to the conventional noise sensor subarray.

[0038] By utilizing an anomaly monitoring and identification prototype library, abnormal noise data in actual rail transit operation can be quickly identified, enabling timely detection of abnormal operating conditions and potential risks. This provides effective data segmentation and feature support for subsequent online intelligent analysis of abnormal noise, achieving precise and efficient track noise anomaly management.

[0039] Furthermore, step S3 in this embodiment of the application also includes:

[0040] Obtain the set of historical noise monitoring data sequences, the corresponding set of historical noise types, and the set of historical noise response durations;

[0041] Using noise type as an index, the historical noise type set is aggregated into similar categories, and the historical noise monitoring data sequence set and historical noise response duration are mapped and clustered according to the aggregation results to obtain K clustered historical noise monitoring data sequence sets and K clustered historical noise response duration sets, where K is a positive integer;

[0042] The noise monitoring features are extracted by traversing the K clustered historical noise monitoring data sequence sets to obtain K clustered historical noise monitoring feature sets;

[0043] The K clustered historical noise monitoring feature sets are aggregated a second time, and combined with the K clustered historical noise response duration sets, K anomaly monitoring and identification trees are constructed. The K anomaly monitoring and identification trees are then summarized to generate the anomaly monitoring and identification prototype library.

[0044] Furthermore, the K clustered historical noise monitoring feature sets are aggregated a second time, and combined with the K clustered historical noise response duration sets, K anomaly monitoring and identification trees are constructed. Step S3 in this embodiment of the application also includes:

[0045] Extract the first clustered historical noise monitoring feature set from the K clustered historical noise monitoring feature sets;

[0046] M historical noise monitoring features of the first cluster are randomly extracted from the first cluster historical noise monitoring feature set as M initial identification internal nodes;

[0047] The first clustered historical noise monitoring feature set is aggregated using the M initial internal nodes to obtain M aggregated historical noise monitoring feature sets.

[0048] The aggregation verification function is used to perform aggregation authentication on the M aggregated historical noise monitoring feature sets and the M first identification internal nodes. If the authentication is successful, the M initial clustering historical noise monitoring features are used as the M identification internal nodes, and the monitoring scale is marked on the M identification internal nodes based on the corresponding first clustering historical noise response duration set to obtain the constructed first anomaly monitoring and identification tree.

[0049] The K sets of clustered historical noise monitoring features and the K sets of clustered historical noise response durations are analyzed to obtain the K anomaly monitoring and identification trees.

[0050] Furthermore, the aggregation verification function is:

[0051] ;

[0052] in, The degree of aggregation output by the aggregation verification function. Let j be the j-th aggregated historical noise monitoring feature within the i-th aggregated historical noise monitoring feature set of M aggregated historical noise monitoring feature sets. Let i be the first clustered historical noise monitoring feature corresponding to the i-th initially identified internal node in the set of M aggregated historical noise monitoring features. The total number of aggregated historical noise monitoring features in the i-th aggregated historical noise monitoring feature set;

[0053] The aggregation degree is obtained by inputting the M aggregated historical noise monitoring feature sets and the M first identification internal nodes into the aggregation verification function;

[0054] Determine whether the degree of aggregation meets the preset degree of aggregation threshold; if so, the authentication is successful.

[0055] In one possible embodiment, the historical noise monitoring data sequence set refers to the data set obtained from monitoring rail transit noise and vibration over a certain period in the past, with each sequence representing noise data recorded under a specific condition or event. The historical noise type set refers to the categories of different abnormal events corresponding to historical monitoring data, such as track faults, vehicle faults, or other specific anomalies. The historical noise response duration set refers to the time from the onset of signs of an abnormal event to the occurrence of the anomaly. The anomaly monitoring and identification tree is a classification tree constructed based on historical data characteristics and response durations, used for rapid feature comparison and classification during real-time anomaly identification. The anomaly monitoring and identification prototype library is a knowledge base formed by summarizing and organizing all anomaly identification trees, which can be used for matching and identifying real-time abnormal noise data features.

[0056] First, the initial historical dataset is constructed by acquiring a set of historical noise monitoring data sequences and their corresponding historical noise types and abnormal response durations. Then, the data is classified and aggregated using noise type as an index (e.g., switch anomalies, vehicle malfunctions, etc. are aggregated separately). Based on the classification results, the noise monitoring data sequences and their corresponding response durations are mapped and clustered to generate K clustered historical datasets (e.g., K=5 indicates aggregation into 5 different anomaly categories). Subsequently, the feature recognizer obtained above is used to traverse each clustered dataset to extract features, such as extracting indicators like spectral energy, peak frequency, and temporal energy change trends, forming K clustered historical noise monitoring feature sets.

[0057] Furthermore, these feature sets are subjected to secondary aggregation, and combined with the response time characteristics of the corresponding anomalies, K anomaly detection and identification trees are constructed using a tree structure (such as decision trees or random forests). Each tree can efficiently identify data features of specific anomaly types and corresponding response time characteristics. Finally, these K anomaly detection and identification trees are summarized and integrated to form a unified anomaly detection and identification prototype library.

[0058] By establishing a precise and efficient historical data feature recognition model, a stable and rapid feature matching standard is provided for the anomaly identification of real-time noise monitoring data. This achieves the technical effect of improving the accuracy and efficiency of anomaly identification in real-time monitoring and effectively supporting the intelligent management of vibration and noise in rail transit.

[0059] Preferably, the first clustering historical noise monitoring feature set is the first feature set selected from the previously obtained K clustering historical noise monitoring feature sets, representing the feature set of historical monitoring data for a specific category of anomalies. The initial identification internal nodes are feature nodes initially randomly selected as classification criteria or references when constructing the anomaly monitoring identification tree. The aggregated historical noise monitoring feature set is a feature set re-divided by using the initial identification internal nodes as the clustering criterion; each set represents data with similar features. The aggregation validation function is a quantitative evaluation function used to judge the reasonableness and tightness (aggregation degree) of the clustering results, ensuring the effectiveness of the clustering partitioning.

[0060] The similarity between the first clustered historical noise monitoring feature set and the M initial identification internal nodes is calculated using the cosine similarity formula. The first clustered historical noise monitoring features are then assigned to the set corresponding to the initial identification internal node with the highest similarity, resulting in the M aggregated historical noise monitoring feature sets. Furthermore, the aggregation degree of the M aggregated historical noise monitoring feature sets is analyzed and quantified using the aggregation verification function. It is determined whether the aggregation degree meets a preset aggregation degree threshold (the minimum aggregation degree that can pass authentication, pre-set by those skilled in the art). If yes, authentication is successful. If not, the M initial identification internal nodes are reselected.

[0061] Furthermore, step S3 in this embodiment of the application also includes:

[0062] Based on the one-to-one mapping relationship between historical noise response duration and historical noise monitoring features, the first clustered historical noise response duration set is mapped and aggregated based on the M aggregated historical noise monitoring feature sets to obtain M aggregated historical noise response duration sets.

[0063] The mean of the M aggregated historical noise response duration sets is calculated to obtain M monitoring scales. Based on the M monitoring scales, the M internal nodes are identified to obtain the first anomaly monitoring and identification tree.

[0064] Preferably, based on the established correspondence between historical noise monitoring features and response times, and using the previously obtained M aggregated historical noise monitoring feature sets, the corresponding first clustered historical noise response time sets are clustered together, so that each feature set corresponds to a unique response time set, forming M new aggregated historical noise response time sets. Then, each response time set is iterated and calculated, for example, by taking the average of all response times within each set, to obtain M monitoring scales representing this type of anomaly feature. These scales can be used to represent the average response time of different anomaly situations. Next, using the obtained M monitoring scales as identifiers, they are mapped to M internal nodes in the anomaly monitoring identification tree, forming scale labels. This allows each node in the anomaly identification tree to reflect the quantification degree of the corresponding anomaly feature. Finally, the above steps complete the construction of the first anomaly monitoring identification tree. The important role of this process is to clarify the correspondence between anomaly features and response times, and to intuitively represent the degree of anomaly in a quantitative way, providing accurate and efficient judgment criteria for subsequent real-time anomaly monitoring analysis.

[0065] S4: Construct an abnormal test analysis channel set based on the abnormal monitoring scale subarray, and construct a conventional test analysis network layer based on the conventional monitoring frequency. Use the abnormal test analysis channel set and the conventional test analysis network layer to perform online monitoring and analysis on the abnormal noise sensor subarray and the conventional noise sensor subarray, respectively, to obtain the target vibration and noise test results.

[0066] Furthermore, based on the anomaly monitoring scale subarray, an anomaly test analysis channel set is constructed. In this embodiment, step S4 further includes:

[0067] Multiple sample noise monitoring data sequences are acquired, and anomalies are identified in the multiple sample noise monitoring data sequences based on the anomaly monitoring scale subarray to obtain multiple sample anomaly identification result arrays.

[0068] The multiple sample noise monitoring data sequences are combined with multiple sample anomaly identification result arrays to construct anomaly test analysis channels, thereby obtaining the anomaly test analysis channel set.

[0069] In one possible embodiment, the anomaly test analysis channel set refers to multiple independent data analysis channels specifically designed for analyzing anomalous noise data, constructed based on the anomaly monitoring scale subarray determined in the aforementioned steps. Each channel uses a specific anomaly monitoring scale as its analysis benchmark. The conventional test analysis network layer is constructed based on typical noise monitoring frequencies under normal operating conditions of rail transit, used for real-time analysis of conventional noise data not identified as anomalies. The sample noise monitoring data sequence is a set of multiple typical data sequences extracted from historical or real-time monitoring data, used to construct and validate the anomaly analysis channels. The sample anomaly identification result array represents the array of anomaly determination results formed after anomaly identification of the sample data sequences, with each result corresponding to the anomaly state (e.g., anomaly location, duration, etc.) of a specific data sequence.

[0070] Preferably, multiple noise sample data sequences are first acquired from the historical monitoring data of actual rail transit operation or from recent real-time data collection, such as typical data sequences collected under different track conditions (turnouts, curves, elevated sections). Then, based on the anomaly monitoring scale subarray established in step S3, anomaly identification analysis is performed on each of these sample data sequences to determine whether there are abnormal features in each sample sequence, and the respective anomaly identification result array is output, such as the location of the anomaly, frequency characteristics, vibration amplitude exceeding limits, or duration of the anomaly.

[0071] Furthermore, multiple sample noise monitoring data sequences are combined with multiple sample anomaly identification result arrays to obtain multiple training sample data combinations. These training sample data combinations are then used to supervise the training of a framework built on a feedforward neural network until convergence, resulting in the trained anomaly test analysis channel. By obtaining multiple independent channel combinations constructed for different anomaly types and scale characteristics to form an anomaly test analysis channel set, accurate, efficient, and intelligent analysis of anomalies during real-time monitoring of rail transit vibration and noise is achieved, thereby improving the technical effectiveness of test result reliability.

[0072] Based on the same principle as constructing the anomaly test analysis channel set, the routine test analysis network layer is constructed based on the routine monitoring frequency. This routine test analysis network layer is used to simultaneously monitor and analyze the routine noise sensor subarray.

[0073] In summary, the embodiments of this application have at least the following technical effects:

[0074] This application acquires a target test track, deploys sensors based on the track's route characteristics, and obtains a completed noise sensor array. Then, according to a pre-set test plan, train passage tests are conducted on the target test track, simultaneously collecting data from the sensor array to obtain a noise monitoring data sequence array. This array is then traversed to identify anomalies in the noise sensor array, resulting in an abnormal noise sensor subarray and a conventional noise sensor subarray. The abnormal noise sensor subarray includes an abnormal monitoring scale subarray. An abnormal test analysis channel set is constructed based on the abnormal monitoring scale subarray, and a conventional test analysis network layer is built based on the conventional monitoring frequency. The abnormal test analysis channel set and the conventional test analysis network layer are used to perform online monitoring and analysis on the abnormal and conventional noise sensor subarrays, respectively, to obtain the target vibration and noise test results. This achieves the technical effect of intelligent online monitoring of rail transit and improving the reliability of test results.

[0075] Example 2, based on the same inventive concept as the intelligent testing method for vibration and noise impact on rail transit in the foregoing examples, as shown in the appendix. Figure 2 As shown, this application provides an intelligent testing system for the vibration and noise impact of rail transit. The system and method embodiments in this application are based on the same inventive concept. The system includes:

[0076] The noise sensor array acquisition module 11 is used to acquire the target test track, deploy sensors based on the route characteristics of the target test track, and obtain the deployed noise sensor array.

[0077] The monitoring data sequence array acquisition module 12 is used to conduct train passage tests on the target test track according to a preset test plan, and simultaneously collect data from the sensor array to obtain a noise monitoring data sequence array.

[0078] The sensor subarray acquisition module 13 is used to traverse the noise monitoring data sequence array to perform anomaly monitoring and identification on the noise sensor array, and obtain an abnormal noise sensor subarray and a normal noise sensor subarray, wherein the abnormal noise sensor subarray includes an abnormal monitoring scale subarray;

[0079] The test result acquisition module 14 is used to construct an abnormal test analysis channel set based on the abnormal monitoring scale subarray and construct a conventional test analysis network layer based on the conventional monitoring frequency. The abnormal test analysis channel set and the conventional test analysis network layer are used to perform online monitoring and analysis on the abnormal noise sensor subarray and the conventional noise sensor subarray, respectively, to obtain the target vibration and noise test results.

[0080] Furthermore, the sensor subarray acquisition module 13 is used to perform the following steps:

[0081] Arbitrarily extract the first noise monitoring data sequence from the noise monitoring data sequence array;

[0082] Noise monitoring features are extracted from the first noise monitoring data sequence to obtain a first noise monitoring data feature set.

[0083] In the pre-built anomaly detection and identification prototype library, the first noise monitoring data feature set is matched. If the match is successful, the noise sensor corresponding to the first noise monitoring data sequence is marked as abnormal and added to the abnormal noise sensor subarray. The first anomaly monitoring scale obtained by matching is added to the anomaly monitoring scale subarray.

[0084] Furthermore, the sensor subarray acquisition module 13 is used to perform the following steps:

[0085] Obtain the set of historical noise monitoring data sequences, the corresponding set of historical noise types, and the set of historical noise response durations;

[0086] Using noise type as an index, the historical noise type set is aggregated into similar categories, and the historical noise monitoring data sequence set and historical noise response duration are mapped and clustered according to the aggregation results to obtain K clustered historical noise monitoring data sequence sets and K clustered historical noise response duration sets, where K is a positive integer;

[0087] The noise monitoring features are extracted by traversing the K clustered historical noise monitoring data sequence sets to obtain K clustered historical noise monitoring feature sets;

[0088] The K clustered historical noise monitoring feature sets are aggregated a second time, and combined with the K clustered historical noise response duration sets, K anomaly monitoring and identification trees are constructed. The K anomaly monitoring and identification trees are then summarized to generate the anomaly monitoring and identification prototype library.

[0089] Furthermore, the sensor subarray acquisition module 13 is used to perform the following steps:

[0090] Extract the first clustered historical noise monitoring feature set from the K clustered historical noise monitoring feature sets;

[0091] M historical noise monitoring features of the first cluster are randomly extracted from the first cluster historical noise monitoring feature set as M initial identification internal nodes;

[0092] The first clustered historical noise monitoring feature set is aggregated using the M initial internal nodes to obtain M aggregated historical noise monitoring feature sets.

[0093] The aggregation verification function is used to perform aggregation authentication on the M aggregated historical noise monitoring feature sets and the M first identification internal nodes. If the authentication is successful, the M initial clustering historical noise monitoring features are used as the M identification internal nodes, and the monitoring scale is marked on the M identification internal nodes based on the corresponding first clustering historical noise response duration set to obtain the constructed first anomaly monitoring and identification tree.

[0094] The K sets of clustered historical noise monitoring features and the K sets of clustered historical noise response durations are analyzed to obtain the K anomaly monitoring and identification trees.

[0095] Furthermore, the aggregation verification function is:

[0096] ;

[0097] in, The degree of aggregation output by the aggregation verification function. Let j be the j-th aggregated historical noise monitoring feature within the i-th aggregated historical noise monitoring feature set of M aggregated historical noise monitoring feature sets. Let i be the first clustered historical noise monitoring feature corresponding to the i-th initially identified internal node in the set of M aggregated historical noise monitoring features. The total number of aggregated historical noise monitoring features in the i-th aggregated historical noise monitoring feature set;

[0098] The aggregation degree is obtained by inputting the M aggregated historical noise monitoring feature sets and the M first identification internal nodes into the aggregation verification function;

[0099] Determine whether the degree of aggregation meets the preset degree of aggregation threshold; if so, the authentication is successful.

[0100] Furthermore, the sensor subarray acquisition module 13 is used to perform the following steps:

[0101] Based on the one-to-one mapping relationship between historical noise response duration and historical noise monitoring features, the first clustered historical noise response duration set is mapped and aggregated based on the M aggregated historical noise monitoring feature sets to obtain M aggregated historical noise response duration sets.

[0102] The mean of the M aggregated historical noise response duration sets is calculated to obtain M monitoring scales. Based on the M monitoring scales, the M internal nodes are identified to obtain the first anomaly monitoring and identification tree.

[0103] Furthermore, the test result acquisition module 14 is used to perform the following steps:

[0104] Multiple sample noise monitoring data sequences are acquired, and anomalies are identified in the multiple sample noise monitoring data sequences based on the anomaly monitoring scale subarray to obtain multiple sample anomaly identification result arrays.

[0105] The multiple sample noise monitoring data sequences are combined with multiple sample anomaly identification result arrays to construct anomaly test analysis channels, thereby obtaining the anomaly test analysis channel set.

[0106] It should be noted that the order of the embodiments described above is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. Furthermore, the above description focuses on specific embodiments of this specification. Additionally, the processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired results. In some implementations, multitasking and parallel processing are possible or may be advantageous.

[0107] The above description is only a preferred embodiment of this application and is not intended to limit this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

[0108] This specification and accompanying drawings are merely illustrative examples of this application and are intended to cover any and all modifications, variations, combinations, or equivalents within the scope of this application. Clearly, those skilled in the art can make various alterations and modifications to this application without departing from its scope. Therefore, if such modifications and modifications fall within the scope of this application and its equivalents, this application intends to include such modifications and modifications.

Claims

1. A method for intelligently testing the vibration and noise impact of rail transit, characterized in that, The method comprises: acquiring a target test track, arranging sensors based on the route characteristics of the target test track, and obtaining a noise sensor array after arrangement; performing train passing test on the target test track according to a preset test scheme, synchronously collecting data of the sensor array, and obtaining a noise monitoring data sequence array; iterating through the noise monitoring data sequence array to identify abnormalities of the noise sensor array, obtaining an abnormal noise sensor subarray and a normal noise sensor subarray, wherein the abnormal noise sensor subarray comprises an abnormal monitoring scale subarray; constructing an abnormal test analysis channel set based on the abnormal monitoring scale subarray, constructing a normal test analysis network layer based on a normal monitoring frequency, respectively performing online monitoring analysis on the abnormal noise sensor subarray and the normal noise sensor subarray by using the abnormal test analysis channel set and the normal test analysis network layer, and obtaining a target vibration noise test result. 2.The intelligent test method for vibration and noise impact of rail transit according to claim 1, characterized in that, iterating through the noise monitoring data sequence array to identify abnormalities of the noise sensor array, obtaining an abnormal noise sensor subarray and a normal noise sensor subarray, wherein the abnormal noise sensor subarray comprises an abnormal monitoring scale subarray, comprising: arbitrarily extracting a first noise monitoring data sequence from the noise monitoring data sequence array; extracting noise monitoring features of the first noise monitoring data sequence to obtain a first noise monitoring data feature set; matching the first noise monitoring data feature set in a pre-constructed abnormal monitoring identification prototype library, if the matching is successful, identifying the noise sensor corresponding to the first noise monitoring data sequence as abnormal, adding it to the abnormal noise sensor subarray, and adding the first abnormal monitoring scale obtained by matching to the abnormal monitoring scale subarray. 3.The intelligent test method for vibration and noise impact of rail transit according to claim 2, characterized in that, comprising: acquiring a set of historical noise monitoring data sequences, a set of corresponding historical noise types, and a set of historical noise response durations; aggregating the same type in the set of historical noise types with noise type as an index, and mapping and clustering the set of historical noise monitoring data sequences and the set of historical noise response durations according to the same type aggregation result to obtain K sets of clustered historical noise monitoring data sequences and K sets of clustered historical noise response durations, wherein K is a positive integer; iterating through the K sets of clustered historical noise monitoring data sequences to extract noise monitoring features to obtain K sets of clustered historical noise monitoring features; performing secondary aggregation on the K sets of clustered historical noise monitoring features, combining the K sets of clustered historical noise response durations, constructing K abnormal monitoring identification trees, and generating the abnormal monitoring identification prototype library by summarizing the K abnormal monitoring identification trees. 4.The intelligent test method for vibration and noise impact of rail transit according to claim 3, characterized in that, performing secondary aggregation on the K sets of clustered historical noise monitoring features, combining the K sets of clustered historical noise response durations, constructing K abnormal monitoring identification trees, comprising: extracting a first set of clustered historical noise monitoring features from the K sets of clustered historical noise monitoring features; randomly extracting M first clustered historical noise monitoring features from the first set of clustered historical noise monitoring features as M initial identification internal nodes; Aggregating the M initial identified internal nodes to the first clustering historical noise monitoring feature set to obtain M aggregated historical noise monitoring feature sets; Aggregating and authenticating the M aggregated historical noise monitoring feature sets and the M first identified internal nodes by using an aggregation authentication function, if the authentication is passed, taking the M initial clustering historical noise monitoring features as the M identified internal nodes, and identifying the M identified internal nodes based on the corresponding first clustering historical noise response time length set to obtain the first anomaly monitoring identification tree constructed; Analyzing the K clustering historical noise monitoring feature sets and the K clustering historical noise response time length sets to obtain the K anomaly monitoring identification trees.

5. The intelligent test method for vibration and noise impact of rail transit according to claim 4, characterized in that, The aggregation authentication function is: ； wherein, is an aggregation degree of the aggregation verification function output, is the jth aggregation history noise monitoring feature in the ith aggregation history noise monitoring feature set in the M aggregation history noise monitoring feature sets, is the first clustering history noise monitoring feature corresponding to the ith initial identification internal node in the ith aggregation history noise monitoring feature set, is the total number of aggregation history noise monitoring features in the ith aggregation history noise monitoring feature set; Inputting the M aggregated historical noise monitoring feature sets and the M first identified internal nodes into the aggregation authentication function to obtain an aggregation degree; If the aggregation degree meets a preset aggregation degree threshold, the authentication is passed. 6.The intelligent test method for vibration and noise impact of rail transit according to claim 5, characterized in that, It includes: Based on the M aggregated historical noise monitoring feature sets, mapping and aggregating the first clustering historical noise response time length set based on the one-to-one mapping relationship between the historical noise response time length and the historical noise monitoring feature to obtain M aggregated historical noise response time length sets; Traversing and calculating the mean of the M aggregated historical noise response time length sets to obtain M monitoring scales, identifying the M identified internal nodes based on the M monitoring scales to obtain the first anomaly monitoring identification tree.

7. The intelligent test method for vibration and noise effects of rail transit according to claim 1, characterized in that, Based on the anomaly monitoring scale subarray, an anomaly test analysis channel set is constructed, including: Obtaining a plurality of sample noise monitoring data sequences, identifying anomalies in the plurality of sample noise monitoring data sequences based on the anomaly monitoring scale subarray to obtain a plurality of sample anomaly identification result arrays; Combining the plurality of sample noise monitoring data sequences with the plurality of sample anomaly identification result arrays to construct an anomaly test analysis channel set.

8. A vibration and noise impact intelligent testing system for rail transit, characterized in that, The system is used to implement the intelligent test method of the vibration and noise impact of rail transit according to any one of claims 1-7, and the system includes: A noise sensor array obtaining module is configured to obtain a target test track, arrange sensors based on the route characteristics of the target test track, and obtain a noise sensor array arranged; A monitoring data sequence array obtaining module is configured to perform train passing tests on the target test track according to a preset test scheme, synchronously collect data of the sensor array, and obtain a noise monitoring data sequence array; A sensor subarray obtaining module is configured to traverse the noise monitoring data sequence array to identify anomalies in the noise sensor array, and obtain an anomaly noise sensor subarray and a normal noise sensor subarray, wherein the anomaly noise sensor subarray includes an anomaly monitoring scale subarray. The test result obtaining module is configured to construct an abnormal test analysis channel set based on the abnormal monitoring scale subarray, construct a normal test analysis network layer based on the normal monitoring frequency, and perform online monitoring analysis on the abnormal noise sensor subarray and the normal noise sensor subarray respectively by using the abnormal test analysis channel set and the normal test analysis network layer, so as to obtain the target vibration noise test result.

Images