Method for judging compliance of hazardous chemical vehicles based on big data technology

Abstract

The invention relates to the technical field of vehicle compliance judgment, in particular to a method for judging the compliance of a hazardous chemical substance vehicle based on a big data technology, and the method comprises the steps: obtaining a historical order set, historical vehicle data and historical hazardous chemical substance state data corresponding to a to-be-detected hazardous chemical substance vehicle; and obtaining a historical vehicle trajectory set and a planned vehicle trajectory set corresponding to the historical order set, and matching the historical vehicle trajectory set with the planned vehicle trajectory set to obtain a trajectory matching degree. According to the method, the compliance of the hazardous chemical vehicle is judged by comprehensively using multiple data sources, the comprehensiveness and accuracy of the judgment process are ensured, limitation possibly exists in single-dimension data analysis, compliance judgment can be effectively supplemented and verified in combination with different types of data, and the accuracy of the judgment process is improved. And the historical vehicle track and the planned vehicle track are obtained, and compliance judgment is performed based on the track matching degree, so that whether the hazardous chemical substance vehicle runs according to the set route can be effectively evaluated.

Description

Technical Field

[0001] The present invention relates to the technical field of vehicle compliance judgment, and particularly relates to a method for judging the compliance of hazardous chemical vehicles based on big data technology. Background Art

[0002] The judgment of the compliance of hazardous chemical vehicles refers to the evaluation process of whether the hazardous chemical transport vehicles comply with relevant regulations, standards and safety requirements. The purpose of this judgment process is to ensure that hazardous chemicals will not pose a threat to public safety, the environment or personnel during transportation, and to ensure that the transportation behavior complies with national and local laws and regulations, as well as industry safety specifications.

[0003] Traditional methods usually rely on a single data source, such as only through paper records, manual inspections or a certain type of monitoring equipment for supervision. This method often lacks multi-dimensional data support and cannot comprehensively evaluate the safety of hazardous chemical transportation. For example, simply relying on vehicle inspection forms may not be able to detect abnormal driving behaviors or dangerous routes in a timely manner, and can only rely on historical data to verify compliance, resulting in poor accuracy and comprehensiveness of the judgment; moreover, traditional methods often identify abnormal behaviors based on simple rules and checklists, and it is difficult to detect complex or potential abnormal driving behaviors (such as slight sudden braking, excessive acceleration, etc.), nor can they mine potential risks through big data analysis. This method mainly relies on manual experience and routine inspections, and is prone to ignoring some unnoticed dangerous driving behaviors. Summary of the Invention

[0004] The technical problem to be solved by the present invention is to overcome the above-mentioned shortcomings of the prior art and provide a method for judging the compliance of hazardous chemical vehicles based on big data technology.

[0005] The technical solution adopted to solve the above technical problem is: A method for judging the compliance of hazardous chemical vehicles based on big data technology, including:

[0006] Obtain the historical order set, historical vehicle data and historical hazardous chemical status data corresponding to the hazardous chemical vehicle to be detected;

[0007] Obtain the historical vehicle trajectory set and the planned vehicle trajectory set corresponding to the historical order set, match the historical vehicle trajectory set with the planned vehicle trajectory set to obtain a trajectory matching degree, and judge whether the historical trajectory of the hazardous chemical vehicle to be detected is compliant based on the trajectory matching degree;

[0008] Cluster the historical vehicle data to obtain a set of clustering clusters, and judge whether the driving behavior of the hazardous chemical vehicle to be detected is compliant based on the set of clustering clusters;

[0009] Sort the historical hazardous chemical status data in time series to obtain a historical hazardous chemical status data sequence, and perform curve fitting on the historical hazardous chemical status data sequence to obtain the historical hazardous chemical status curve;

[0010] Compare the historical hazardous chemical status curve with a preset standard hazardous chemical status curve to obtain the deviation of the hazardous chemical status, and determine whether the transportation of hazardous chemicals by the vehicle to be detected is compliant based on the deviation of the hazardous chemical status.

[0011] Preferably, match the historical vehicle trajectory set with the planned vehicle trajectory set to obtain a trajectory matching degree, including:

[0012] Obtain an electronic map of the operation area corresponding to the vehicle to be detected, and divide the electronic map into a plurality of Thiessen polygon grids. Among them, the Thiessen polygon grids include vehicle trajectories and a set of valid paths. Among them, the valid paths include a set of road segments, and the vehicle trajectories are composed of multiple GPS points.

[0013] Preferably, matching the historical vehicle trajectory set with the planned vehicle trajectory set to obtain a trajectory matching degree further includes:

[0014] Map the historical vehicle trajectory and the planned vehicle trajectory into a plurality of Thiessen polygon grids;

[0015] Calculate the first trajectory similarity between the historical vehicle trajectory and each valid path in the set of valid paths based on a similarity model;

[0016] Calculate the second trajectory similarity between the planned vehicle trajectory and each valid path in the set of valid paths based on a similarity model;

[0017] Calculate the first path matching degree between the historical vehicle trajectory and the set of valid paths based on the first trajectory similarity;

[0018] Use the valid path corresponding to the maximum first path matching degree as the first matching valid path of the historical vehicle trajectory;

[0019] Calculate the second path matching degree between the planned vehicle trajectory and the set of valid paths based on the second trajectory similarity;

[0020] Use the valid path corresponding to the maximum second path matching degree as the second matching valid path of the planned vehicle trajectory;

[0021] Determine whether the road segments in the first matching valid path are the same as those in the second matching valid path. If they are the same, increase the trajectory matching degree between the historical vehicle trajectory and the planned vehicle trajectory by 1. If they are different, increase the trajectory matching degree between the historical vehicle trajectory and the planned vehicle trajectory by 0;

[0022] Repeat the above operations until all historical vehicle trajectories in the historical vehicle trajectory set are processed, or until all planned vehicle trajectories in the planned vehicle trajectory set are processed, to obtain the trajectory matching degree between the historical vehicle trajectory set and the planned vehicle trajectory set.

[0023] Preferably, the expression of the set of valid paths within the Thiessen polygon grid is as follows:

[0024]

[0025] Wherein, represents the set of valid paths within the Thiessen polygon grid, represents the first valid path in the set of valid paths, g m represents the m-th Thiessen polygon grid, and N represents the total number of valid paths within the Thiessen polygon grid;

[0026] The expression of the set of road segments of the valid path is as follows:

[0027]

[0028] Wherein, represents the set of road segments of the valid path, represents the first road segment in the set of road segments, and K represents the total number of road segments of the valid path.

[0029] Preferably, the expression of the similarity model is as follows:

[0030]

[0031] Wherein, d(tr,r) represents the trajectory similarity between the vehicle trajectory and the valid path, n represents the total number of GPS points in the vehicle trajectory and the valid path, and represent the latitude and longitude of the k-th GPS point of the vehicle trajectory, and represent the latitude and longitude of the k-th GPS point of the valid path;

[0032] The calculation formula of the path matching degree is as follows:

[0033]

[0034] Among them, p(tr,r) represents the path matching degree between the planned vehicle trajectory and the effective path, and r i represents the i-th effective path.

[0035] Preferably, clustering is performed on the historical vehicle data to obtain a set of clustering clusters, including:

[0036] Construct an objective function based on the historical vehicle data, where the objective function is used to minimize the sum of the squares of the distances from the points within the cluster to the cluster center. Among them, the objective function is as follows:

[0037]

[0038] Among them, J represents the sum of the squares of the distances from the points within the cluster to the cluster center, and x i represents a data point in the historical vehicle data, and r ik represents whether the data point x i in the historical vehicle data belongs to cluster k, and μ k represents the center of cluster k, and ||x i -μ k || 2 represents the square of the distance from the data point x i in the historical vehicle data to the center of cluster k;

[0039] Allocate the data pairs in the historical vehicle data to different clustering clusters based on the objective function;

[0040] Calculate the silhouette coefficient of the cluster, and compare the silhouette coefficient of the cluster with a preset silhouette coefficient threshold. If the silhouette coefficient of the cluster is greater than the preset silhouette coefficient threshold, it indicates that the historical vehicle data is clustered; otherwise, it indicates that the historical vehicle data is not clustered. The calculation formula of the silhouette coefficient is as follows:

[0041]

[0042] Among them, S(i) represents the silhouette coefficient of the cluster, a(i) represents the average distance from the data point of the data stream in the historical vehicle data to other points within its cluster, and b(i) represents the average distance from the data point of the data stream in the historical vehicle data to the nearest other cluster.

[0043] Preferably, judging whether the historical trajectory of the to-be-detected hazardous chemical vehicle is compliant based on the trajectory matching degree includes:

[0044] Compare the trajectory matching degree with a preset trajectory matching degree threshold;

[0045] If the trajectory matching degree is greater than the preset trajectory matching degree threshold, it is determined that the historical trajectory of the to-be-detected hazardous chemical vehicle is compliant;

[0046] If the trajectory matching degree is not greater than a preset trajectory matching degree threshold, it is determined that the historical trajectory of the hazardous chemical vehicle to be detected is non-compliant.

[0047] Preferably, determining whether the driving behavior of the hazardous chemical vehicle to be detected is compliant based on the set of clustering clusters includes:

[0048] Obtain the abnormal clustering clusters in the set of clustering clusters, and obtain the total number of data points in the abnormal clustering clusters;

[0049] Compare the total number of data points in the abnormal clustering clusters with a preset quantity threshold;

[0050] If the total number of data points in the abnormal clustering clusters is greater than the preset quantity threshold, it is determined that the driving behavior of the hazardous chemical vehicle to be detected is non-compliant;

[0051] If the total number of data points in the abnormal clustering clusters is not greater than the preset quantity threshold, it is determined that the driving behavior of the hazardous chemical vehicle to be detected is compliant.

[0052] Preferably, the calculation formula for the deviation of the hazardous chemical state is as follows:

[0053]

[0054] where M represents the deviation between the historical hazardous chemical state curve and the preset standard hazardous chemical state curve, y i represents the data point of the standard hazardous chemical state curve, represents the data point of the historical hazardous chemical state curve, and B represents the total number of data points.

[0055] Preferably, determining whether the transportation of hazardous chemicals by the hazardous chemical vehicle to be detected is compliant based on the deviation of the hazardous chemical state includes:

[0056] Compare the deviation of the hazardous chemical state with a preset deviation threshold;

[0057] If the deviation of the hazardous chemical state is greater than the preset deviation threshold, it is determined that the transportation of hazardous chemicals by the hazardous chemical vehicle to be detected is compliant;

[0058] If the deviation of the hazardous chemical state is not greater than the preset deviation threshold, it is determined that the transportation of hazardous chemicals by the hazardous chemical vehicle to be detected is non-compliant.

[0059] Preferably, the method further includes:

[0060] Obtain the first qualification certificate of the hazardous chemical vehicle to be detected, obtain the second qualification certificate of the driver of the hazardous chemical vehicle to be detected, and obtain the third qualification certificate of the enterprise to which the hazardous chemical vehicle to be detected belongs;

[0061] Perform character recognition on the first qualification certificate, the second qualification certificate, and the third qualification certificate based on a pre-trained character recognition model to obtain a first qualification text, a second qualification text, and a third qualification text;

[0062] Match the first qualification text, the second qualification text, and the third qualification text with preset qualification keywords to obtain the qualification keywords corresponding to the first qualification text, the second qualification text, and the third qualification text;

[0063] Judge the compliance corresponding to the first qualification certificate, the second qualification certificate, and the third qualification certificate based on the qualification keywords corresponding to the first qualification text, the second qualification text, and the third qualification text.

[0064] The beneficial effects of the present invention are as follows: (1) By comprehensively using multiple data sources to judge the compliance of hazardous chemical vehicles, the present invention ensures the comprehensiveness and accuracy of the judgment process. The analysis of a single dimension of data may have limitations, while combining different types of data can effectively supplement and verify the compliance judgment. By obtaining the historical vehicle trajectory and planning the vehicle trajectory, and judging the compliance based on the trajectory matching degree, it is possible to effectively evaluate whether the hazardous chemical vehicle is driving along the established route. This not only helps to ensure that the vehicle transports according to the specified route and avoids potential violations, but also can identify in advance the paths or driving behaviors that may pose safety hazards; (2) By processing the historical vehicle data through cluster analysis, the present invention can effectively identify abnormal driving behaviors. For example, dangerous driving behaviors (such as sudden braking, speeding, etc.) can be highlighted in the driving data through cluster analysis. By identifying these abnormal behaviors, it is possible to further ensure that the driving behaviors of hazardous chemical vehicles comply with safety standards and prevent safety accidents caused by non-compliant driving behaviors. By fitting the curve of the historical hazardous chemical state data and comparing it with the standard hazardous chemical state curve, it is possible to effectively monitor and identify whether the state of the hazardous chemical during transportation is abnormal. The abnormal state of the hazardous chemical (such as too high temperature, unstable pressure, etc.) is often a potential risk of accidents. Therefore, timely discovering the deviation and taking measures helps to reduce the probability of accidents; (3) Based on big data technology, the present invention can perform real-time or near-real-time dynamic risk assessment. For example, by continuously monitoring the vehicle trajectory, driving behavior, and hazardous chemical state, potential safety hazards can be discovered in a timely manner and early warnings can be issued. This real-time nature improves the supervision efficiency of hazardous chemical vehicles and ensures that quick responses and measures can be taken when problems occur during transportation. Description of the Drawings

[0065] Figure 1 It is a schematic diagram of the step flow of the overall method in an embodiment proposed by the present invention. Detailed Embodiment

[0066] Example 1, as Figure 1 shown, a method for judging the compliance of hazardous chemical vehicles based on big data technology proposed by the present invention includes:

[0067] S1. Obtain the historical order set, historical vehicle data, and historical hazardous chemical status data corresponding to the hazardous chemical vehicle to be detected;

[0068] S2. Obtain the historical vehicle trajectory set and the planned vehicle trajectory set corresponding to the historical order set, match the historical vehicle trajectory set with the planned vehicle trajectory set to obtain a trajectory matching degree, and judge whether the historical trajectory of the hazardous chemical vehicle to be detected is compliant based on the trajectory matching degree;

[0069] S3. Cluster the historical vehicle data to obtain a set of clustering clusters, and judge whether the driving behavior of the hazardous chemical vehicle to be detected is compliant based on the set of clustering clusters;

[0070] S4. Sort the historical hazardous chemical status data in time series to obtain a historical hazardous chemical status data sequence, and perform curve fitting on the historical hazardous chemical status data sequence to obtain a historical hazardous chemical status curve;

[0071] S5. Compare the historical hazardous chemical status curve with a preset standard hazardous chemical status curve to obtain the deviation of the hazardous chemical status, and judge whether the transportation of hazardous chemicals by the hazardous chemical vehicle to be detected is compliant based on the deviation of the hazardous chemical status.

[0072] In the present invention, the historical order set refers to the historical order data related to the transportation of hazardous chemicals, which includes the basic information of the transportation task, such as the transportation time, destination, type of goods, quantity, etc.; the historical vehicle data refers to the vehicle driving data related to the transportation of hazardous chemicals in the past, usually including the vehicle speed, acceleration, driving behavior, location, driving route, staying time, etc.; the historical hazardous chemical state data refers to the state data of hazardous chemicals during transportation, such as temperature, humidity, pressure, etc. These data are important indicators for monitoring the safety and stability of the transportation of hazardous chemicals; the historical vehicle trajectory set refers to the actual driving trajectory data of hazardous chemical transportation vehicles in the past, usually recorded by a GPS system or other positioning devices, including the longitude and latitude coordinates of the vehicle at each time point, representing the driving path of the vehicle; the planned vehicle trajectory set refers to the ideal or standard driving path preset according to the transportation task plan and transportation regulations. This trajectory usually takes into account factors such as traffic regulations, route optimization, time requirements, etc.; clustering is an unsupervised learning method aimed at dividing data into different clusters according to the similarity of the data. The goal of clustering is to group similar driving behaviors into the same cluster and separate different driving behaviors; the clustering cluster set refers to the set of each cluster obtained after dividing the historical vehicle data into several clusters through a clustering algorithm. Each cluster represents a typical driving behavior pattern, such as normal driving, sudden acceleration, sudden braking, speeding, etc.; time series sorting refers to sorting the hazardous chemical state data according to the time stamp so that the data is arranged in chronological order. The hazardous chemical state data is usually continuous and can change over time. Therefore, after sorting, it can reflect the dynamic change process of the hazardous chemical state; curve fitting refers to fitting the historical hazardous chemical state data through a mathematical model to obtain a smooth curve reflecting the change trend of the hazardous chemical state over time. Common curve fitting methods include the least squares method, spline interpolation, etc.; the standard hazardous chemical state curve refers to the preset hazardous chemical state change curve based on industry norms, safety standards or transportation regulations. This curve reflects the ideal change trend of the hazardous chemical state (such as temperature, humidity, etc.) under normal transportation conditions; the hazardous chemical state deviation refers to the difference between the historical hazardous chemical state curve and the standard hazardous chemical state curve during the actual transportation process. If the deviation is too large, it may mean that there are abnormal situations during transportation, such as too high temperature, too low humidity, etc., which may pose risks to the hazardous chemicals.

[0073] Embodiment 2. A method for judging the compliance of hazardous chemical vehicles based on big data technology proposed by the present invention. Compared with Embodiment 1, this embodiment further includes: matching the historical vehicle trajectory set with the planned vehicle trajectory set to obtain the trajectory matching degree, including:

[0074] A1. Obtain the electronic map of the operation area corresponding to the hazardous chemical vehicle to be detected, and divide the electronic map into multiple Thiessen polygon grids. Among them, the Thiessen polygon grids include vehicle trajectories and a set of valid paths. The valid paths include a set of road segments, and the vehicle trajectory is composed of multiple GPS points.

[0075] In this embodiment, the electronic map refers to the map displayed on a computer or a mobile device, usually supported by a Geographic Information System (GIS). It can contain roads, geographical locations, Points of Interest (POIs), traffic information, etc., and supports dynamic updates and interactive functions; the Thiessen polygon is a method for dividing the plane based on a set of points, used to divide the space into multiple regions, and all points in each region are closest to the center point (seed point) of the region. Specifically, given a set of points (such as GPS positioning points), the Thiessen polygon will divide the entire map into several polygon regions, each region corresponding to a "seed point", and the distance from all points within each region to this seed point is the shortest; the vehicle trajectory refers to the time series coordinate data recorded by the vehicle during driving through a positioning system such as GPS. Each trajectory point (i.e., GPS point) usually contains a timestamp, longitude and latitude, and possibly information such as speed and direction. The vehicle trajectory set is a trajectory sequence containing multiple GPS points; the valid path refers to the set of roads or road segments that the vehicle can legally and smoothly drive under specific traffic conditions or planning requirements. In the scenario of hazardous chemical transportation, the valid path may be the road segments selected according to factors such as traffic regulations, safety standards, and road conditions. These road segments are considered the most suitable transportation paths. Especially for hazardous material transportation, some road segments may be excluded due to road conditions or risk factors; the set of road segments refers to the set of multiple continuous road sections. Each road segment is usually the connecting part between two points on the map.

[0076] In an optional embodiment, matching the historical vehicle trajectory set and the planned vehicle trajectory set to obtain the trajectory matching degree further includes:

[0077] A2. Map the historical vehicle trajectory and the planned vehicle trajectory into multiple Thiessen polygon grids;

[0078] A3. Calculate the first trajectory similarity between the historical vehicle trajectory and each valid path in the set of valid paths based on the similarity model;

[0079] A4. Calculate the second trajectory similarity between the planned vehicle trajectory and each valid path in the set of valid paths based on the similarity model;

[0080] A5. Calculate the first path matching degree between the historical vehicle trajectory and the set of valid paths based on the first trajectory similarity;

[0081] A6. Take the valid path corresponding to the maximum first path matching degree as the first matching valid path of the historical vehicle trajectory;

[0082] A7. Calculate the second path matching degree between the planned vehicle trajectory and the set of valid paths based on the second trajectory similarity;

[0083] A8. Take the valid path corresponding to the maximum second path matching degree as the second matching valid path of the planned vehicle trajectory;

[0084] A9. Determine whether the road segments in the first matching valid path and the road segments in the second matching valid path are the same. If they are the same, add 1 to the trajectory matching degree between the historical vehicle trajectory and the planned vehicle trajectory. If they are not the same, add 0 to the trajectory matching degree between the historical vehicle trajectory and the planned vehicle trajectory;

[0085] A10. Repeat the above operations until all historical vehicle trajectories in the historical vehicle trajectory set are processed, or until all planned vehicle trajectories in the planned vehicle trajectory set are processed, to obtain the trajectory matching degree between the historical vehicle trajectory set and the planned vehicle trajectory set.

[0086] In an alternative embodiment, the expression of the set of valid paths within the Thiessen polygon grid is as follows:

[0087]

[0088] Wherein, represents the set of valid paths within the Thiessen polygon grid, represents the first valid path in the set of valid paths, g m represents the m-th Thiessen polygon grid, and N represents the total number of valid paths within the Thiessen polygon grid;

[0089] The expression of the set of road segments of the valid path is as follows:

[0090]

[0091] Wherein, represents the set of road segments of the valid path, represents the first road segment in the set of road segments, and K represents the total number of road segments of the valid path.

[0092] In an alternative embodiment, the expression of the similarity model is as follows:

[0093]

[0094] Wherein, d(tr,r) represents the trajectory similarity between the vehicle trajectory and the valid path, n represents the total number of GPS points in the vehicle trajectory and the valid path, and represents the latitude and longitude of the k-th GPS point of the vehicle trajectory, and represents the latitude and longitude of the k-th GPS point of the valid path;

[0095] The calculation formula of the path matching degree is as follows:

[0096]

[0097] where p(tr, r) represents the path matching degree between the planned vehicle trajectory and the valid path, and r i represents the i-th valid path.

[0098] In an optional embodiment, historical vehicle data is clustered to obtain a set of clustering clusters, including:

[0099] B1. Construct an objective function based on the historical vehicle data. The objective function is used to minimize the sum of the squared distances from the points within the cluster to the cluster center. Among them, the objective function is as follows:

[0100]

[0101] where J represents the sum of the squared distances from the points within the cluster to the cluster center, x i represents a data point in the historical vehicle data, and r ik represents whether the data point x in the historical vehicle data i belongs to cluster k, and μ k represents the center of cluster k, and ||x i -μ k || 2 represents the squared distance from the data point x in the historical vehicle data i to the center of cluster k;

[0102] B2. Based on the objective function, allocate the data pairs in the historical vehicle data to different clustering clusters;

[0103] B3. Calculate the silhouette coefficient of the cluster, and compare the silhouette coefficient of the cluster with a preset silhouette coefficient threshold. If the silhouette coefficient of the cluster is greater than the preset silhouette coefficient threshold, it indicates that the historical vehicle data is clustered; otherwise, it indicates that the historical vehicle data is not clustered. The calculation formula of the silhouette coefficient is as follows:

[0104]

[0105] where S(i) represents the silhouette coefficient of the cluster, a(i) represents the average distance from the data points of the data stream in the historical vehicle data to other points within its cluster, and b(i) represents the average distance from the data points of the data stream in the historical vehicle data to the nearest other cluster.

[0106] It should be noted that the cluster center is the average position of all points within the cluster, usually the arithmetic mean of the coordinates of all data points within the cluster. In clustering, the center point of the cluster represents the core position of the cluster. Clustering algorithms (such as K-means clustering) usually optimize the clustering effect by continuously updating the cluster center; the objective function determines which cluster each data point belongs to based on the distance from historical vehicle data points to the cluster center, and the data points will be assigned to the cluster center with the smallest distance to it, which is done to optimize the compactness within the cluster; the silhouette coefficient is an important indicator to measure the quality of clustering, and its value ranges from -1 to 1. The larger the value, the better the clustering effect.

[0107] In an optional embodiment, determining whether the historical trajectory of a hazardous chemical vehicle to be detected is compliant based on the trajectory matching degree includes:

[0108] C1. Comparing the trajectory matching degree with a preset trajectory matching degree threshold;

[0109] C2. If the trajectory matching degree is greater than the preset trajectory matching degree threshold, it is determined that the historical trajectory of the hazardous chemical vehicle to be detected is compliant;

[0110] C3. If the trajectory matching degree is not greater than the preset trajectory matching degree threshold, it is determined that the historical trajectory of the hazardous chemical vehicle to be detected is non-compliant.

[0111] In an optional embodiment, determining whether the driving behavior of a hazardous chemical vehicle to be detected is compliant based on the clustering cluster set includes:

[0112] D1. Obtaining the abnormal clustering clusters in the clustering cluster set and obtaining the total number of data points in the abnormal clustering clusters;

[0113] D2. Comparing the total number of data points in the abnormal clustering clusters with a preset quantity threshold;

[0114] D3. If the total number of data points in the abnormal clustering clusters is greater than the preset quantity threshold, it is determined that the driving behavior of the hazardous chemical vehicle to be detected is non-compliant;

[0115] D4. If the total number of data points in the abnormal clustering clusters is not greater than the preset quantity threshold, it is determined that the driving behavior of the hazardous chemical vehicle to be detected is compliant.

[0116] It should be noted that the abnormal clustering clusters refer to those clusters that have significantly different characteristics compared to other clusters. For example, some clustering clusters may contain data points far from other clusters, and these data points are usually caused by certain abnormal behaviors (such as sudden dangerous driving behaviors). Such clusters may show abnormal distributions in terms of characteristics and are therefore considered "abnormal" clusters.

[0117] In an optional embodiment, the calculation formula for the deviation of the hazardous chemical state is as follows:

[0118]

[0119] Among them, M represents the deviation between the historical hazardous chemical state curve and the preset standard hazardous chemical state curve, y i represents the data points of the standard hazardous chemical state curve, represents the data points of the historical hazardous chemical state curve, and B represents the total number of data points.

[0120] In an alternative embodiment, determining whether the transportation of hazardous chemicals by the hazardous chemical vehicle to be detected is compliant based on the deviation of the hazardous chemical state includes:

[0121] E1. Comparing the deviation of the hazardous chemical state with a preset deviation threshold;

[0122] E2. If the deviation of the hazardous chemical state is greater than the preset deviation threshold, it is determined that the transportation of hazardous chemicals by the hazardous chemical vehicle to be detected is compliant;

[0123] E3. If the deviation of the hazardous chemical state is not greater than the preset deviation threshold, it is determined that the transportation of hazardous chemicals by the hazardous chemical vehicle to be detected is non-compliant.

[0124] In an alternative embodiment, the method further includes:

[0125] F1. Obtaining the first qualification certificate of the hazardous chemical vehicle to be detected, obtaining the second qualification certificate of the driver of the hazardous chemical vehicle to be detected, and obtaining the third qualification certificate of the enterprise to which the hazardous chemical vehicle to be detected belongs;

[0126] F2. Performing character recognition on the first qualification certificate, the second qualification certificate, and the third qualification certificate based on a pre-trained character recognition model to obtain a first qualification text, a second qualification text, and a third qualification text;

[0127] F3. Matching the first qualification text, the second qualification text, and the third qualification text with preset qualification keywords to obtain the qualification keywords corresponding to the first qualification text, the second qualification text, and the third qualification text;

[0128] F4. Determining the compliance corresponding to the first qualification certificate, the second qualification certificate, and the third qualification certificate based on the qualification keywords corresponding to the first qualification text, the second qualification text, and the third qualification text.

[0129] It should be noted that qualification certificates refer to legal certification documents related to certain professional activities, which are used to prove that an individual, enterprise or institution meets specific qualification requirements; the first qualification certificate may be a legal and compliant certificate related to the vehicle itself, such as a vehicle registration certificate, a hazardous chemicals transportation license, etc.; the second qualification certificate refers to the certificate related to the driver, such as a hazardous chemicals transportation driver's license, a vocational qualification certificate, etc.; the third qualification certificate may be a qualification certificate related to the enterprise itself, such as a legal business license, a work safety license, etc. for a hazardous chemicals transportation enterprise; a text recognition model, also known as OCR, is a technology that automatically recognizes and extracts text in an image through computer vision technology. In this scenario, through a pre-trained OCR model, the system can extract the text content from the qualification certificate and convert it into text data for subsequent processing; qualification keywords refer to the key information used to identify the legality and compliance of the qualification certificate; for example, some certificates will clearly mention words such as "the hazardous chemicals transportation license is valid" or "the driver qualification is qualified", etc. These are preset qualification keywords used to match the text to determine the validity and compliance of the certificate; compliance judgment is to analyze the text extracted from the certificate and determine whether these certificates meet the relevant regulations according to the preset standards or laws and regulations; for example, checking whether the driver holds a valid hazardous chemicals transportation certificate, whether the enterprise has legal transportation qualifications, etc.; the purpose of this step is to ensure that all qualification certificates meet the legal requirements and avoid the situation of illegal transportation of hazardous chemicals.

[0130] The embodiments of the present invention have been described in detail above with reference to the accompanying drawings. However, the present invention is not limited thereto, and various changes can be made without departing from the spirit of the present invention within the scope of knowledge possessed by those skilled in the art.

Claims

1. A method for judging the compliance of hazardous chemical vehicles based on big data technology, characterized in that, Including: Obtain the historical order set, historical vehicle data, and historical hazardous chemical status data corresponding to the hazardous chemical vehicle to be detected; Obtain the historical vehicle trajectory set and the planned vehicle trajectory set corresponding to the historical order set, match the historical vehicle trajectory set with the planned vehicle trajectory set to obtain a trajectory matching degree, and judge whether the historical trajectory of the hazardous chemical vehicle to be detected is compliant based on the trajectory matching degree; Cluster the historical vehicle data to obtain a cluster set, and judge whether the driving behavior of the hazardous chemical vehicle to be detected is compliant based on the cluster set; Sort the historical hazardous chemical status data in time series to obtain a historical hazardous chemical status data sequence, and perform curve fitting on the historical hazardous chemical status data sequence to obtain the historical hazardous chemical status curve; Compare the historical hazardous chemical status curve with a preset standard hazardous chemical status curve to obtain the deviation of the hazardous chemical status, and judge whether the transportation of hazardous chemicals by the hazardous chemical vehicle to be detected is compliant based on the deviation of the hazardous chemical status.

2. The method for judging the compliance of hazardous chemical vehicles based on big data technology according to claim 1, characterized in that, Matching the historical vehicle trajectory set with the planned vehicle trajectory set to obtain a trajectory matching degree, including: Obtain the electronic map of the operation area corresponding to the hazardous chemical vehicle to be detected, divide the electronic map into multiple Thiessen polygon grids, where the Thiessen polygon grids include vehicle trajectories and a set of valid paths, where the valid paths include a set of road segments, the vehicle trajectory is composed of multiple GPS points, and the expression of the set of valid paths in the Thiessen polygon grid is as follows: Among them, represents the set of valid paths within the Thiessen polygon grid, represents the first valid path in the set of valid paths, g m represents the m-th Thiessen polygon grid, and N represents the total number of valid paths within the Thiessen polygon grid; The expression of the road segment set of the valid path is as follows: Among them, represents the set of road segments of the valid path, represents the first road segment in the set of road segments, and K represents the total number of road segments of the valid path.

3. The method for judging the compliance of hazardous chemical vehicles based on big data technology according to claim 2, wherein, Matching the historical vehicle trajectory set with the planned vehicle trajectory set to obtain a trajectory matching degree, further including: Map the historical vehicle trajectory and the planned vehicle trajectory into multiple Thiessen polygon grids; Calculate the first trajectory similarity between the historical vehicle trajectory and each valid path in the set of valid paths based on the similarity model; Calculate the second trajectory similarity between the planned vehicle trajectory and each valid path in the set of valid paths based on the similarity model; Calculate the first path matching degree between the historical vehicle trajectory and the set of valid paths based on the first trajectory similarity; Take the valid path corresponding to the maximum first path matching degree as the first matching valid path of the historical vehicle trajectory; Calculate the second path matching degree between the planned vehicle trajectory and the set of valid paths based on the second trajectory similarity; Take the valid path corresponding to the maximum second path matching degree as the second matching valid path of the planned vehicle trajectory; Judge whether the road segments in the first matching valid path are the same as the road segments in the second matching valid path. If they are the same, add 1 to the trajectory matching degree between the historical vehicle trajectory and the planned vehicle trajectory. If they are not the same, add 0 to the trajectory matching degree between the historical vehicle trajectory and the planned vehicle trajectory; Repeat the above operations until all the historical vehicle trajectories in the historical vehicle trajectory set are processed, or until all the planned vehicle trajectories in the planned vehicle trajectory set are processed, so as to obtain the trajectory matching degree between the historical vehicle trajectory set and the planned vehicle trajectory set.

4. A method for judging the compliance of hazardous chemical vehicles based on big data technology according to claim 1, characterized in that, The method further includes: Obtaining a first qualification certificate of the hazardous chemical vehicle to be detected, obtaining a second qualification certificate of the driver of the hazardous chemical vehicle to be detected, and obtaining a third qualification certificate of the enterprise to which the hazardous chemical vehicle to be detected belongs; Performing character recognition on the first qualification certificate, the second qualification certificate, and the third qualification certificate based on a pre-trained character recognition model to obtain a first qualification text, a second qualification text, and a third qualification text; Matching the first qualification text, the second qualification text, and the third qualification text with preset qualification keywords to obtain the qualification keywords corresponding to the first qualification text, the second qualification text, and the third qualification text; Judging the compliance of the first qualification certificate, the second qualification certificate, and the third qualification certificate based on the qualification keywords corresponding to the first qualification text, the second qualification text, and the third qualification text.

5. The method for judging the compliance of hazardous chemical vehicles based on big data technology according to claim 4, wherein, The expression of the similarity model is as follows: where d(t r , r) represents the trajectory similarity between the vehicle trajectory and the valid path, n represents the total number of GPS points in the vehicle trajectory and the valid path, and represent the latitude and longitude of the k-th GPS point of the vehicle trajectory, and represent the latitude and longitude of the k-th GPS point of the valid path; The calculation formula of the path matching degree is as follows: Among them, p(tr, r) represents the path matching degree between the planned vehicle trajectory and the effective path, and r i represents the i-th effective path.

6. The method for judging the compliance of hazardous chemical vehicles based on big data technology according to claim 1, characterized in that, Clustering the historical vehicle data to obtain a set of clustering clusters, including: Constructing an objective function based on the historical vehicle data, where the objective function is used to minimize the sum of the squared distances from the points within the cluster to the cluster center, and the objective function is as follows: Among them, J represents the sum of the squares of the distances from the points within the cluster to the cluster center, and x i represents a data point in the historical vehicle data, and r ik represents whether the data point x i in the historical vehicle data belongs to cluster k, and μ k represents the center of cluster k, and ||x i - μ k || 2 represents the square of the distance from the data point x i in the historical vehicle data to the center of cluster k; Allocating the data pairs in the historical vehicle data to different clustering clusters based on the objective function; Calculating the silhouette coefficient of the cluster, comparing the silhouette coefficient of the cluster with a preset silhouette coefficient threshold. If the silhouette coefficient of the cluster is greater than the preset silhouette coefficient threshold, it indicates that the historical vehicle data is clustered; otherwise, it indicates that the historical vehicle data is not clustered. The calculation formula of the silhouette coefficient is as follows: Where S(i) represents the silhouette coefficient of the cluster, a(i) represents the average distance from the data points of the data stream in the historical vehicle data to other points within its cluster, and b(i) represents the average distance from the data points of the data stream in the historical vehicle data to the nearest other cluster.

7. A method for judging the compliance of hazardous chemical transportation vehicles based on big data technology according to claim 5, characterized in that, Judging whether the historical trajectory of the hazardous chemical vehicle to be detected is compliant based on the trajectory matching degree, including: Comparing the trajectory matching degree with a preset trajectory matching degree threshold; If the trajectory matching degree is greater than the preset trajectory matching degree threshold, it is judged that the historical trajectory of the hazardous chemical vehicle to be detected is compliant; If the trajectory matching degree is not greater than the preset trajectory matching degree threshold, it is judged that the historical trajectory of the hazardous chemical vehicle to be detected is non-compliant.

8. A method for judging the compliance of hazardous chemical vehicles based on big data technology according to claim 6, characterized in that, Judging whether the driving behavior of the hazardous chemical vehicle to be detected is compliant based on the set of clustering clusters, including: Obtaining the abnormal clustering clusters in the set of clustering clusters and obtaining the total number of data points in the abnormal clustering clusters; Comparing the total number of data points in the abnormal clustering clusters with a preset quantity threshold; If the total number of data points in the abnormal clustering cluster is greater than a preset quantity threshold, it is determined that the driving behavior of the hazardous chemical vehicle to be detected is non-compliant; If the total number of data points in the abnormal clustering cluster is not greater than a preset quantity threshold, it is determined that the driving behavior of the hazardous chemical vehicle to be detected is compliant.

9. A method for judging the compliance of hazardous chemical vehicles based on big data technology according to claim 1, characterized in that The calculation formula for the deviation of the hazardous chemical state is as follows: Among them, M represents the deviation between the historical hazardous chemical state curve and the preset standard hazardous chemical state curve, and y i represents the data point of the standard hazardous chemical state curve, represents the data point of the historical hazardous chemical state curve, and B represents the total number of data points.

10. A method for judging the compliance of hazardous chemical vehicles based on big data technology according to claim 9, characterized in that, Judging whether the transportation of hazardous chemicals of the hazardous chemical vehicle to be detected is compliant based on the deviation of the hazardous chemical state includes: Comparing the deviation of the hazardous chemical state with a preset deviation threshold; If the deviation of the hazardous chemical state is greater than the preset deviation threshold, it is determined that the transportation of hazardous chemicals of the hazardous chemical vehicle to be detected is compliant; If the deviation of the hazardous chemical state is not greater than the preset deviation threshold, it is determined that the transportation of hazardous chemicals of the hazardous chemical vehicle to be detected is non-compliant.

Images