Hazardous chemical logistics tracking method based on Internet of Things

Abstract

The invention discloses a dangerous chemical logistics tracking method based on the Internet of Things, and the method comprises the steps: carrying out the smoothing of a movement track of a transport vehicle through employing a Kalman filtering algorithm according to a continuous spatial position sequence, and obtaining the smoothed track data; for the smoothed trajectory data, calculating real-time speed and direction parameters of the transport vehicle to obtain a dynamic movement parameter set; adjusting weighted average algorithm parameters fused by the multi-source sensor through the interference influence weight to obtain an optimized spatial position sequence; according to the optimized spatial position sequence, recalculating the moving track and the dynamic moving parameters of the transport vehicle to obtain updated track parameters; and through the updated track parameters, emergency response instructions of dangerous chemical transportation are generated by adopting a real-time data stream processing technology, and an emergency response instruction set is obtained.

Description

Technical Field

[0001] This invention belongs to the field of chemical hazardous materials logistics monitoring technology, and particularly relates to a chemical hazardous materials logistics tracking method based on the Internet of Things. Background Technology

[0002] As a crucial field concerning public safety and environmental protection, the research and application of chemical hazardous materials logistics tracking have irreplaceable value in ensuring social stability and reducing environmental risks. The transportation and storage of chemical hazardous materials involve complex processes; if management is out of control, it can lead to serious safety accidents. Therefore, building an efficient and accurate tracking system has become an urgent need for the industry's development.

[0003] However, current mainstream logistics tracking methods have revealed significant shortcomings when dealing with the specific scenario of hazardous chemicals. Many existing solutions rely too heavily on a single external signal, making them susceptible to environmental interference and resulting in unstable data, especially in complex terrain or indoor environments where positioning accuracy and continuity are difficult to guarantee. Furthermore, existing methods have limited ability to capture the movement of hazardous materials in dynamic environments, failing to fully reflect subtle changes in their spatial location, thus affecting the timeliness and accuracy of emergency response.

[0004] Against this backdrop, the core challenges facing this field are becoming increasingly apparent. The key to tracking the logistics of hazardous chemicals lies in accurately recording spatial locations in changing environments, a problem directly linked to the ability to analyze movement trajectories in real time. Because acquiring spatial location data requires seamless switching between different technologies, such as transitioning from outdoor to indoor positioning environments, the inability to effectively integrate multiple positioning methods leads to data gaps, resulting in deviations in the calculation of key parameters such as direction of movement and speed. This correlation between data gaps and parameter deviations further exacerbates the overall reliability issues of the tracking system. Summary of the Invention

[0005] This invention proposes an Internet of Things-based method for tracking the logistics of hazardous chemicals to address the problems existing in the prior art.

[0006] To achieve the above objectives, the present invention provides a method for tracking the logistics of hazardous chemicals based on the Internet of Things, comprising:

[0007] Spatial location data of chemical hazardous materials transport vehicles are acquired by multi-source sensors, and the spatial location data is integrated by a weighted average algorithm to obtain a continuous spatial location sequence.

[0008] Based on a continuous spatial location sequence, the Kalman filter algorithm is used to smooth the movement trajectory of the transport vehicle to obtain smoothed trajectory data.

[0009] Based on the smoothed trajectory data, the real-time speed and direction parameters of the transport vehicle are calculated to obtain a set of dynamic movement parameters;

[0010] If the speed or direction in the dynamic movement parameter set exceeds the preset threshold, the anomaly detection algorithm is used to determine whether there is an abnormal state in the transportation of hazardous chemicals and to obtain an anomaly status identifier.

[0011] By using anomaly status identifiers and spatial location sequences, the K-nearest neighbor algorithm is used to perform cluster analysis on the abnormal locations of transport vehicles to obtain the distribution of abnormal locations;

[0012] Based on the distribution of abnormal locations, environmental data related to the transportation of hazardous chemicals is obtained, and a decision tree algorithm is used to determine the degree of impact of environmental interference on positioning accuracy, thereby obtaining the interference impact weight.

[0013] By adjusting the weighted average algorithm parameters of multi-source sensor fusion through interference influence weights, an optimized spatial location sequence can be obtained;

[0014] Based on the optimized spatial location sequence, the movement trajectory and dynamic movement parameters of the transport vehicle are recalculated to obtain the updated trajectory parameters;

[0015] Based on the updated trajectory parameters, a set of emergency response instructions for the transportation of hazardous chemicals is generated using real-time data stream processing technology.

[0016] Optionally, obtaining a continuous spatial location sequence includes:

[0017] Spatial location information of chemical hazardous materials transport vehicles is collected by multi-source sensors, and signals from different sources are integrated using fusion technology to obtain a preliminary set of location data.

[0018] Based on the initial location data set, a weighted average algorithm is used to weight the signals from global positioning and inertial navigation to determine the location data set with comprehensive weights.

[0019] The system detects the location data set after weighting. If the deviation of a certain source signal exceeds a preset threshold, the signal is corrected to obtain the corrected location data set.

[0020] By analyzing the stability characteristics of signals from different sources using the corrected location data sets, the reliability level of each signal source is determined, and a reliability ranking result is obtained.

[0021] Based on the reliability ranking results, the priority of each signal source in the fusion process is adjusted to generate an optimized signal fusion strategy;

[0022] An optimized signal fusion strategy is adopted to process the subsequently acquired spatial location information in real time, resulting in continuous and stable location sequence data.

[0023] Optionally, obtaining the smoothed trajectory data includes:

[0024] By processing the spatial location information of the transport vehicles, the Kalman filter algorithm is used to smooth the continuous sequence to obtain smoothed trajectory data.

[0025] Based on the smoothed trajectory data, analyze the abnormal points in the movement trajectory. If a certain position information is detected to deviate from the preset threshold, then the position information is corrected to determine the corrected trajectory data.

[0026] Based on the corrected trajectory data, combined with the vehicle monitoring system, real-time updated location information is obtained to determine whether there is any data loss or delay.

[0027] By analyzing the stability of signal acquisition through real-time updated location information, if the continuity of signal acquisition is lower than a preset threshold, the system switches to a backup signal source to obtain stable location data.

[0028] Based on stable location data, multi-source data information is integrated to generate a comprehensive location information sequence and determine the final movement trajectory.

[0029] Optionally, obtaining the dynamic movement parameter set includes:

[0030] For smooth trajectories, data extraction techniques are used to separate the speed changes and direction adjustments of transport vehicles at different time periods, resulting in a preliminary set of dynamic parameters.

[0031] Based on the initial set of dynamic parameters, the continuity of speed changes is analyzed. If the speed change exceeds the preset threshold, the data segment is processed again to determine the corrected speed information.

[0032] For the corrected speed information, the matching degree between the two is obtained by combining the direction adjustment data. If the matching degree is lower than the preset threshold, the direction information is calibrated to generate calibrated direction parameters.

[0033] By fusing the calibrated direction parameters and the corrected velocity information, a comprehensive set of dynamic parameters is generated.

[0034] Based on the actual set of movement features, and combined with information processing technology, key time and location nodes are extracted to generate the final set of dynamic movement parameters.

[0035] Optionally, obtaining the abnormal status identifier includes:

[0036] For the dynamic movement parameter set in the transportation of hazardous chemicals, data filtering technology is used to separate speed data and direction data to obtain preliminary parameter analysis results.

[0037] Based on the preliminary parameter analysis results, if the speed data or direction data exceeds the preset threshold, the excess data will be processed by the anomaly detection algorithm to determine the preliminary identification of the abnormal state.

[0038] Based on the initial abnormal status identification, and considering the characteristics of chemical transportation and hazardous materials, relevant environmental data is extracted using information processing technology to obtain contextual information related to the abnormal status.

[0039] If there is an inconsistency between the acquired context information and the abnormal status identifier, the abnormal status identifier is calibrated to obtain a calibrated status identifier.

[0040] For the calibrated status identifier, data fusion technology is used to integrate it with dynamic parameters and other data in the moving set to determine whether there is a persistent abnormal state;

[0041] If a persistent abnormal state is confirmed, a corresponding abnormal warning message will be generated through the information processing stage to determine the final abnormal state identifier.

[0042] Optionally, obtaining the abnormal location distribution includes:

[0043] By using abnormal status identifiers and spatial location sequences, data preprocessing techniques are employed to clean the spatial location data of transport vehicles, resulting in standardized location data sequences.

[0044] Based on the standardized location data sequence, the K-nearest neighbor algorithm is used to cluster the abnormal location data to obtain the preliminary distribution characteristics of the abnormal locations;

[0045] Based on the initial distribution characteristics, if the cluster center of the abnormal location deviates from the preset normal location range, environmental data related to the abnormal location is extracted through information processing technology to obtain environmental context information.

[0046] Based on environmental context information, data fusion technology is used to integrate the preliminary distribution characteristics with environmental data to determine whether the distribution of abnormal locations is persistent, and to obtain the calibrated distribution characteristics.

[0047] If persistent abnormal location distribution is confirmed based on the calibrated distribution characteristics, then abnormal points in the distribution characteristics are marked using anomaly detection technology to obtain an anomaly point set.

[0048] By using the set of anomalies, information processing technology is employed to generate warning information for the anomaly locations, thus determining the final distribution identifiers of the anomalies.

[0049] Optionally, obtaining the interference influence weight includes:

[0050] Based on the abnormal locations and their distribution characteristics, relevant environmental data are obtained from the chemical hazardous materials transportation scenario using information acquisition technology to obtain an environmental data set;

[0051] Based on the environmental data set, environmental interference factors related to positioning accuracy are analyzed using feature extraction techniques to determine the environmental interference feature set;

[0052] Based on the environmental interference feature set, the decision tree algorithm is used to analyze the correlation between interference factors and positioning accuracy, and to determine the weight distribution of interference impact.

[0053] Based on the weighted data of the interference impact, the environmental interference characteristics are correlated and mapped with the distribution of abnormal locations through data integration technology to determine the final range of interference impact.

[0054] To determine the ultimate scope of the interference, data storage technology is used to bind it with the scenario information of chemical hazardous materials transportation to obtain the basic data required for long-term monitoring.

[0055] Optionally, obtaining the optimized spatial location sequence includes:

[0056] To address interference and weighting analysis, raw data from multiple sensors is acquired, and the acquired sensor data is cleaned using data preprocessing techniques to obtain a standardized basic dataset.

[0057] Based on a standardized basic dataset, feature extraction techniques are used to analyze key factors related to interference effects and determine the interference feature set.

[0058] For the set of interference features, the influence of each interference factor on spatial location is calculated by weight analysis to obtain weight distribution data.

[0059] Based on the weight distribution data, adjust the weighted average parameters in the fusion algorithm to generate an optimized algorithm configuration;

[0060] Based on the optimized algorithm configuration, data fusion processing is performed using multi-source sensor data to obtain a preliminary spatial location sequence;

[0061] By smoothing the initial spatial location sequence and optimizing and adjusting the location sequence using time series analysis tools, the final spatial location sequence is obtained.

[0062] If the final spatial location sequence exhibits abnormal fluctuations within certain intervals, the source of the anomaly is determined by comparing the weight distribution data with the historical data of the sensor data, and the abnormal intervals are corrected to determine the final optimized location sequence.

[0063] Optionally, obtaining the updated trajectory parameters includes:

[0064] Continuous location points are obtained from the optimized spatial location sequence, and interpolation techniques are used to complete the location points to generate a smooth trajectory dataset;

[0065] Based on the smooth trajectory dataset, calculate the vehicle's dynamic movement parameters, including speed and acceleration, and generate a set of dynamic parameters;

[0066] If the velocity or acceleration in the dynamic parameter set exceeds the preset threshold, the source of the abnormal parameters is determined by comparing with historical trajectory data, and a corrected dynamic parameter set is obtained.

[0067] Key trajectory features are extracted from the modified dynamic parameter set, and cluster analysis is used to classify the trajectory features to generate a classified trajectory dataset.

[0068] For the classified trajectory dataset, combined with real-time data processing technology, the vehicle's movement trajectory parameters are updated to obtain the final trajectory parameter set;

[0069] Based on the final set of trajectory parameters, time series analysis tools are used to predict the trajectory and generate a predicted trajectory dataset.

[0070] Optionally, the set of emergency response instructions to be generated includes:

[0071] By using real-time data stream processing technology, key transportation path information is extracted from trajectory parameters, and data filtering tools are used to classify the path information to obtain a classified path dataset.

[0072] Based on the classified path dataset and combined with real-time monitoring data, potential risk points in the transportation of hazardous chemicals are analyzed. If the risk points exceed the preset threshold, high-risk paths are selected through data comparison tools to determine the set of high-risk paths.

[0073] For high-risk route sets, vehicle status information is obtained from real-time data, and information integration technology is used to generate targeted emergency response strategies, resulting in a preliminary set of emergency strategies.

[0074] From the initial set of emergency strategies, response mechanisms related to transportation safety are extracted, and logical matching tools are used to prioritize the strategies to determine the highest priority combination of response strategies.

[0075] Based on the highest priority response strategy combination, and in conjunction with the instruction generation module, specific emergency response instructions are generated, resulting in a set of instructions for the transportation of hazardous chemicals.

[0076] By using a set of instructions and employing data transmission technology, instructions are distributed to relevant transportation management systems to obtain the execution status of the distributed instructions and determine the coverage of instruction execution.

[0077] Based on the scope of instruction execution and combined with real-time monitoring data, the support vector machine algorithm is used to evaluate the execution effect and obtain the execution effect analysis results.

[0078] Compared with the prior art, the present invention has the following advantages and technical effects:

[0079] This invention discloses an IoT-based method for tracking chemical hazardous materials logistics. The method acquires vehicle spatial location data through multi-source sensor fusion and integrates GPS, inertial navigation, and indoor positioning signals using a weighted average algorithm. A Kalman filter algorithm is used to smooth the movement trajectory, and real-time speed and direction parameters are calculated. When parameters exceed thresholds, an anomaly detection algorithm is used to determine if an abnormal state exists. The K-nearest neighbor algorithm is combined to perform cluster analysis on abnormal locations, and a decision tree algorithm is used to evaluate the impact of environmental interference on positioning accuracy. The sensor fusion algorithm is optimized based on the interference impact weights, trajectory parameters are updated, and finally, emergency response commands are generated. This invention achieves precise positioning and real-time monitoring of abnormal states for chemical hazardous materials transport vehicles, improving transportation safety and emergency response capabilities. Attached Figure Description

[0080] The accompanying drawings, which form part of this application, are used to provide a further understanding of this application. The illustrative embodiments and descriptions of this application are used to explain this application and do not constitute an undue limitation of this application. In the drawings:

[0081] Figure 1 This is a flowchart of a method according to an embodiment of the present invention. Detailed Implementation

[0082] It should be noted that, unless otherwise specified, the embodiments and features described in this application can be combined with each other. This application will now be described in detail with reference to the accompanying drawings and embodiments.

[0083] It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions, and although a logical order is shown in the flowchart, in some cases the steps shown or described may be executed in a different order than that shown here.

[0084] Example 1

[0085] like Figure 1 As shown, this embodiment provides a method for tracking the logistics of hazardous chemicals based on the Internet of Things, including:

[0086] Spatial location data of chemical hazardous materials transport vehicles are acquired by multi-source sensors, and the spatial location data is integrated by a weighted average algorithm to obtain a continuous spatial location sequence.

[0087] Based on a continuous spatial location sequence, the Kalman filter algorithm is used to smooth the movement trajectory of the transport vehicle to obtain smoothed trajectory data.

[0088] Based on the smoothed trajectory data, the real-time speed and direction parameters of the transport vehicle are calculated to obtain a set of dynamic movement parameters;

[0089] If the speed or direction in the dynamic movement parameter set exceeds the preset threshold, the anomaly detection algorithm is used to determine whether there is an abnormal state in the transportation of hazardous chemicals and to obtain an anomaly status identifier.

[0090] By using anomaly status identifiers and spatial location sequences, the K-nearest neighbor algorithm is used to perform cluster analysis on the abnormal locations of transport vehicles to obtain the distribution of abnormal locations;

[0091] Based on the distribution of abnormal locations, environmental data related to the transportation of hazardous chemicals is obtained, and a decision tree algorithm is used to determine the degree of impact of environmental interference on positioning accuracy, thereby obtaining the interference impact weight.

[0092] By adjusting the weighted average algorithm parameters of multi-source sensor fusion through interference influence weights, an optimized spatial location sequence can be obtained;

[0093] Based on the optimized spatial location sequence, the movement trajectory and dynamic movement parameters of the transport vehicle are recalculated to obtain the updated trajectory parameters;

[0094] Based on the updated trajectory parameters, a set of emergency response instructions for the transportation of hazardous chemicals is generated using real-time data stream processing technology.

[0095] Specifically, the following steps are included:

[0096] S101. Spatial location data of chemical hazardous materials transport vehicles are obtained by multi-source sensor fusion, and GPS and inertial navigation signals are integrated by weighted average algorithm to obtain a continuous spatial location sequence.

[0097] Specifically, spatial location information of chemical hazardous materials transport vehicles is collected through multi-source sensors. Fusion technology is used to integrate signals from different sources to obtain a preliminary location dataset. Based on this preliminary dataset, a weighted average algorithm is applied to weight the signals from Global Positioning (GPS) and Inertial Navigation (INS) to determine a weighted location data set. For this weighted location data set, if a deviation from a certain source signal exceeds a preset threshold, the signal is corrected to obtain a corrected location data set. The stability characteristics of the signals from different sources are analyzed using the corrected location data set to determine the reliability level of each signal source, resulting in a reliability ranking. Based on the reliability ranking, the priority of each signal source in the fusion process is adjusted to generate an optimized signal fusion strategy. Using this optimized signal fusion strategy, subsequent spatial location information is processed in real time to obtain continuous and stable location sequence data.

[0098] For example, the application of multi-source sensors is crucial in the spatial location information acquisition and processing of vehicles transporting hazardous chemicals. Assuming the Global Positioning System (GPS) and Inertial Navigation System (INS) are used as the primary signal sources, data is collected once per second by onboard equipment. GPS provides latitude and longitude information, while INS calculates relative displacement using acceleration and angular velocity. When integrating these signals using fusion technology, it may be discovered that GPS signals are lost in tunnels, and INS suffers from accumulated errors. Therefore, fusion technology is needed to initially integrate the data from both systems, forming a dataset containing multiple sets of latitude and longitude values.

[0099] Specifically, in the application of the weighted average algorithm, weights can be assigned based on the historical accuracy of the signal source. Assuming the weight of the Global Positioning System (GPS) is 0.7 and the weight of the Inertial Navigation System (INS) is 0.3, a weighted average is used to obtain a composite position data set. This method effectively balances the relative strengths and weaknesses of the two signals, improving the accuracy of the position data. A beneficial effect is that even if one signal experiences a temporary failure, the composite data can still maintain a certain level of reliability.

[0100] In this embodiment, if a signal deviation from a certain source is detected to exceed a preset threshold, such as a deviation of more than 50 meters between the location data from the Global Positioning System and the historical trajectory, the signal needs to be corrected. The correction method can be to interpolate and adjust the data by referencing data from other signal sources, ultimately obtaining a corrected set of location data. This correction can significantly reduce the impact of abnormal data on the overall positioning, ensuring data continuity.

[0101] For example, when analyzing the stability characteristics of signals from different sources, reliability can be determined by calculating the amplitude of signal fluctuations. Assuming the fluctuation amplitude of the Global Positioning System (GPS) in an open area is 2 meters, while the fluctuation amplitude of an Inertial Navigation System (INS) reaches 10 meters after long-term operation, the GPS can be judged to have a higher reliability level. Based on this ranking, the priority in the fusion process is adjusted, setting the GPS as the primary reference source. The technical effect of this adjustment is to optimize the accuracy of data processing.

[0102] Specifically, after generating the optimized signal fusion strategy, when processing the subsequently acquired spatial location information in real time, the signal source data with high reliability can be used first, and the system can dynamically switch to the suboptimal signal source when the signal fails.

[0103] For example, in urban high-rise areas, GPS signals may be blocked. In such cases, the weight of the inertial navigation system is automatically increased to ensure the continuity and stability of the location sequence data. The benefit of this real-time processing is that the system can provide stable positioning support regardless of environmental changes.

[0104] In this embodiment, the continuous location sequence data generated by the above method can be further used for vehicle trajectory monitoring and risk warning.

[0105] For example, the system can issue a timely alert when a vehicle deviates from the planned route, ensuring the safety of transporting hazardous chemicals. The overall solution forms a closed-loop processing mechanism, from data collection to strategy optimization, ensuring the continuity and reliability of the technology's effectiveness.

[0106] S102. Based on the continuous spatial position sequence, the Kalman filter algorithm is used to smooth the movement trajectory of the transport vehicle to obtain the smoothed trajectory data.

[0107] Specifically, the spatial location information of the transport vehicle is processed, and a Kalman filter algorithm is used to smooth the continuous sequence to obtain smoothed trajectory data. Based on the smoothed trajectory data, outliers in the movement trajectory are analyzed. If a location deviates from a preset threshold, the location information is corrected to determine the corrected trajectory data. For the corrected trajectory data, in conjunction with the vehicle monitoring system, real-time updated location information is obtained to determine if there are any missing or delayed data. The stability of signal acquisition is analyzed using the real-time updated location information. If the continuity of signal acquisition is lower than a preset threshold, a backup signal source is switched to obtain stable location data. Based on the stable location data, multi-source data information is fused to generate a comprehensive location information sequence, determining the final movement trajectory.

[0108] For example, when processing the spatial location information of vehicles transporting hazardous chemicals, the Kalman filter algorithm can be understood in principle as a dynamic estimation method based on prediction and updating. It gradually optimizes the smoothness of the location data through a comprehensive analysis of historical data and current observations.

[0109] In one possible implementation, assuming a vehicle is traveling on a highway, the collected location data is updated every second. However, due to signal interference, some data points exhibit jumps in position. By using Kalman filtering, the vehicle's possible position in the next second can be predicted and corrected using the actual collected data, ultimately generating a smooth trajectory curve. This method is particularly suitable for data processing in dynamic environments and can effectively reduce noise interference.

[0110] For example, to detect outliers in trajectory data, a reasonable threshold range can be set to determine deviations in location information. Assuming a vehicle's normal speed is 60 km / h, if the location data suddenly shows a speed of 120 km / h at a certain moment, and there is no significant acceleration trend before or after, it can be identified as an outlier. During correction, the location information at the preceding and following time points can be referenced, and interpolation methods can be used to estimate a reasonable location. This approach ensures the reasonableness of the trajectory from the perspective of data continuity.

[0111] For example, when using a vehicle monitoring system to obtain real-time location updates, the reception interval of data packets can be analyzed to determine if there are any missing or delayed data. Assuming the system is set to receive data every 5 seconds, if an interval exceeds 10 seconds, a delay can be considered. In this case, the system can record the data status for that period and generate an alert for subsequent signal stability analysis. This real-time monitoring method helps to detect problems promptly.

[0112] For example, if the continuity of signal acquisition falls below a preset threshold, such as three consecutive data gaps, the system can automatically switch to a backup signal source, such as switching from the primary GPS signal to an auxiliary inertial navigation signal. In densely populated urban areas with high-rise buildings, where GPS signals frequently interrupt, the backup signal source can estimate the vehicle's position using changes in acceleration and direction, maintaining data continuity. This switching mechanism ensures the stability of location information from the perspective of multi-source data fusion.

[0113] For example, when generating a comprehensive location information sequence, multi-source data can be fused according to weighted allocation. Assuming GPS signals have a weight of 0.8 in open areas and inertial navigation has a weight of 0.2, the weight ratio is dynamically adjusted in areas with signal obstruction. This fusion method, from the perspective of data reliability, ensures the accuracy and completeness of the final movement trajectory. Through these multi-faceted processing methods, a complete technical chain is formed, from data smoothing to anomaly correction, signal switching, and comprehensive fusion, significantly improving the reliability of location data for chemical hazardous materials transport vehicles.

[0114] S103. For the smoothed trajectory data, calculate the real-time speed and direction parameters of the transport vehicle to obtain a set of dynamic movement parameters.

[0115] Specifically, for smooth trajectories, data extraction techniques are used to separate the speed changes and direction adjustments of the transport vehicle at different time periods, obtaining a preliminary set of dynamic parameters. Based on this preliminary set, the continuity of speed changes is analyzed. If the speed change exceeds a preset threshold, the data segment undergoes secondary processing to determine the corrected speed information. For the corrected speed information, the degree of matching between the two is determined by combining it with the direction adjustment data. If the degree of matching is lower than a preset threshold, the direction information is calibrated to generate calibrated direction parameters. The calibrated direction parameters and the corrected speed information are then fused to generate a comprehensive set of dynamic parameters, which is used to determine if it conforms to the actual movement characteristics of the transport vehicle. For the actual movement characteristic set, information processing techniques are used to extract key time and location nodes, generating the final set of dynamic movement parameters.

[0116] For example, when processing smooth trajectory data of transport vehicles, data extraction techniques can be used to separate speed changes and direction adjustment information. For speed changes, the average speed per minute can be calculated by analyzing the vehicle's position data over different time periods. For instance, in a 5-minute trajectory, the vehicle's speed gradually increases from 30 km / h to 50 km / h; this trend can be recorded as preliminary dynamic parameters. Direction adjustment information can be obtained through changes in the angle of the vehicle's position points; for example, a vehicle turning from due north to northeast at an intersection changes its angle by 45 degrees. This information together constitutes a preliminary set of dynamic parameters, laying the foundation for subsequent analysis.

[0117] For example, in analyzing the continuity of speed changes, a preset threshold can be set, such as a speed change rate not exceeding 10 kilometers per hour. If, within a certain trajectory, the speed suddenly jumps from 40 kilometers per hour to 60 kilometers per hour, exceeding the threshold, secondary processing is required. One possible approach is to use data from surrounding time periods for interpolation smoothing. For instance, if the speed was 42 kilometers per hour in the previous minute and 58 kilometers per hour in the next, outliers could be corrected to 50 kilometers per hour. This method helps ensure the reasonableness of speed data and avoids abrupt changes caused by signal interference.

[0118] For example, when analyzing the matching degree between speed information and direction adjustment data, it can be judged by comparing the changing trends of the two at the same time point. If the vehicle's direction changes frequently while its speed increases, for example, if the speed increases from 30 km / h to 50 km / h and the direction changes 3 times within 10 minutes, the matching degree may be lower than a preset threshold. In this case, the direction information can be calibrated, prioritizing the main directional trends in historical trajectories, for example, filtering out frequent small angle changes and retaining only the main turning points. This calibration method can reduce noise interference and make the dynamic parameters more consistent with actual movement characteristics.

[0119] For example, when fusing calibrated direction parameters and corrected speed information, a comprehensive dynamic parameter set can be generated using a weighted approach. Assuming that in a certain trajectory segment, the speed information has a weight of 0.6 and the direction information has a weight of 0.4, the comprehensive parameters can more accurately reflect the vehicle's movement state. When determining whether the parameters conform to actual movement characteristics, vehicle type and road condition information can be considered. For instance, heavy transport vehicles are unlikely to make frequent high-speed turns on urban roads; this assessment helps verify the parameters' rationality.

[0120] For example, when extracting key time and location nodes, focus can be placed on moments of significant change in speed or direction. For instance, if a vehicle's speed suddenly drops from 50 km / h to 10 km / h at a certain point in time, it may indicate congestion or a stop; this time and corresponding location should be marked as key nodes. By integrating these nodes into a final set of dynamic movement parameters through information processing technology, important references can be provided for subsequent trajectory optimization and monitoring. This approach not only improves data analysis efficiency but also provides a more reliable basis for transportation management.

[0121] S104. If the speed or direction in the dynamic movement parameter set exceeds the preset threshold, the abnormality detection algorithm is used to determine whether there is an abnormal state in the transportation of chemical hazardous materials and obtain the abnormal state identifier.

[0122] Specifically, for the dynamic movement parameter set in the transportation of hazardous chemicals, data filtering technology is used to separate speed and direction data to obtain preliminary parameter analysis results. Based on these preliminary results, if the speed or direction data exceeds a preset threshold, an anomaly detection algorithm is used to process the excess data and determine a preliminary abnormal state identifier. This algorithm is implemented using machine learning methods. For the preliminary abnormal state identifier, considering the characteristics of chemical transportation and hazardous materials, relevant environmental data is extracted using information processing technology to obtain contextual information related to the abnormal state. If the obtained contextual information is inconsistent with the abnormal state identifier, the identifier is calibrated to obtain a calibrated identifier. For the calibrated identifier, data fusion technology is used to integrate it with other data from the dynamic parameters and movement set to determine if a persistent abnormal state exists. If a persistent abnormal state is confirmed using the integrated data, a corresponding abnormal warning message is generated through information processing to determine the final abnormal state identifier.

[0123] For example, in the processing of dynamic movement parameters in the transportation of hazardous chemicals, data filtering techniques for dynamic movement parameter sets can extract speed and direction data by layering real-time data from transport vehicles. Suppose a transport vehicle records a speed of 60 kilometers per hour over a certain distance, while its direction data shows a continuous deviation from the preset route by 10 degrees. This preliminary parameter analysis provides a foundation for subsequent processing. The core of data filtering technology lies in decomposing complex data streams into analyzable single-dimensional information for targeted processing later.

[0124] For example, anomaly detection algorithms can be used to identify speed or direction data exceeding preset thresholds. Assuming a speed threshold of 80 km / h and a direction deviation threshold of 15 degrees, the system will automatically mark an abnormal state when the vehicle speed reaches 85 km / h or the direction deviation is 20 degrees. The principle behind this algorithm is to quickly locate potential risk points by comparing historical and real-time data, laying the foundation for subsequent analysis.

[0125] For example, in the extraction of environmental data combining chemical transportation and hazardous material characteristics, sensors can be used to acquire information such as temperature and humidity during transportation. Suppose the transported liquid is flammable and the ambient temperature suddenly rises to 35 degrees Celsius, exceeding the safe range. The system will associate this information with an anomaly status indicator to form contextual information. The advantage of this information processing technology is its ability to integrate multi-dimensional data, enhancing the accuracy of anomaly detection.

[0126] For example, if the context information is inconsistent with the anomaly status indicator, the indicator needs to be calibrated. Suppose the speed anomaly indicator shows speeding, but environmental data indicates that the road was on a steep slope; the system will calibrate the speed indicator based on the slope and determine it as normal. This calibration process effectively reduces false alarms and improves system reliability.

[0127] For example, in data fusion technology, integrating calibrated status indicators with dynamic parameters can determine whether a persistent anomaly exists. Suppose a vehicle displays abnormal speed and direction for 30 consecutive minutes, and the environmental data offers no reasonable explanation; the system will then confirm a persistent abnormal state. This integration method helps to comprehensively understand the vehicle's status.

[0128] For example, when generating an anomaly warning message, if the anomaly is confirmed to be persistent, the system will automatically generate a warning and notify the dispatch center. For instance, if the warning message indicates that a vehicle is continuously speeding and directionally unstable on a certain road segment, the dispatch center can intervene promptly. This information processing step enables rapid response to potential risks and ensures transportation safety.

[0129] S105. By using the abnormal status identifier and combining it with the spatial location sequence, the K-nearest neighbor algorithm is used to perform cluster analysis on the abnormal locations of the transport vehicles to obtain the distribution of abnormal locations.

[0130] Specifically, by using anomaly status identifiers and spatial location sequences, data preprocessing techniques are employed to clean the spatial location data of transport vehicles, resulting in standardized location data sequences. Based on these standardized sequences, the K-nearest neighbor algorithm is used to cluster the anomaly location data, yielding preliminary distribution characteristics of the anomaly locations. If the cluster centers of the anomaly locations deviate from the preset normal location range, information processing techniques are used to extract environmental data related to the anomaly locations, obtaining environmental context information. Based on this context information, data fusion techniques are used to integrate the preliminary distribution characteristics with the environmental data to determine the persistence of the anomaly location distribution, resulting in calibrated distribution characteristics. If a persistent anomaly location distribution is confirmed, anomaly point detection techniques are used to mark the anomaly points in the distribution characteristics, obtaining an anomaly point set. Using this anomaly point set, information processing techniques are employed to generate warning information for the anomaly locations, determining the final anomaly location distribution identifier.

[0131] For example, in the scenario of transporting hazardous chemicals, data preprocessing is the primary step in processing abnormal status indicators and spatial location sequences. Data preprocessing mainly involves cleaning the spatial location data of the transport vehicle to remove noise, such as positioning drift caused by signal interference. Suppose a transport vehicle records 100 location points along a certain route, and 5 of these points significantly deviate from the normal trajectory. Cleaning techniques can remove these points, resulting in a standardized location data sequence, laying the foundation for subsequent analysis.

[0132] For example, the K-nearest neighbors algorithm is used to cluster abnormal location data for standardized location data sequences. Its principle is to calculate the distance between location points and group nearby abnormal points into one category. Suppose that in a certain transportation operation, 10 location points are marked as abnormal. The K-nearest neighbors algorithm finds that 8 of these points are concentrated in a specific area, forming a cluster center. This initially reveals the distribution characteristics of the abnormal locations.

[0133] For example, when analyzing preliminary distribution characteristics, if cluster centers are found to deviate from the preset normal location range—for instance, the normal range is a main road, but the cluster centers appear on remote side roads—then information processing techniques are needed to extract environmental data. Environmental data may include road conditions, weather information, etc. For instance, if it is discovered that the side road is under construction, causing vehicles to detour, this contextual information helps explain the cause of the abnormal distribution.

[0134] For example, based on environmental context information, data fusion technology integrates preliminary distribution characteristics with environmental data to determine the persistence of abnormal location distributions. Suppose that fusion analysis reveals vehicles have been circling the same side roads for three consecutive days, then the persistence of the abnormal distribution can be confirmed, yielding calibrated distribution characteristics. This approach can more accurately reflect the actual situation.

[0135] For example, if a persistent abnormal distribution is confirmed after calibration, the anomaly detection technology will further mark the set of anomalies. Suppose that out of 100 location points, 15 points are marked as anomalies; these points are concentrated in the detour path area, forming an anomaly set, which provides a basis for subsequent warnings.

[0136] For example, information processing technology is used to generate abnormal location warning information, determining the final distribution of abnormal locations. Suppose the system generates a warning message based on the set of abnormal points, indicating "The vehicle continues to deviate from the main route, potentially posing a risk," and records the abnormal distribution as "Detour Abnormality" for easy tracking and handling. This method can promptly alert relevant personnel to potential problems. Through the above analysis and examples, from data cleaning to the generation of the final warning information, each step closely revolves around the business needs of transporting hazardous chemicals, ensuring accurate identification and effective management of abnormal locations, while simultaneously improving the safety and controllability of the transportation process.

[0137] S106. Based on the distribution of abnormal locations, obtain environmental data related to the transportation of hazardous chemicals, use a decision tree algorithm to determine the degree of influence of environmental interference on positioning accuracy, and obtain the interference influence weight.

[0138] Specifically, for anomaly locations and their distribution characteristics, relevant environmental data is acquired from the chemical hazardous materials transportation scenario using information acquisition technology, resulting in an environmental data set. Based on this environmental data set, data cleaning techniques are used to remove potential noise or missing values, resulting in a standardized environmental data group. For this standardized environmental data group, feature extraction techniques are used to analyze environmental interference factors related to positioning accuracy, determining an environmental interference feature set. Based on this feature set, a decision tree algorithm is used to analyze the correlation between interference factors and positioning accuracy, determining the weight distribution of interference impact. Based on the weighted data of interference impact, data integration techniques are used to map the environmental interference features to the anomaly location distribution, determining the final interference impact range. For the final interference impact range, data storage techniques are used to bind it to the chemical hazardous materials transportation scenario information, obtaining the basic data required for long-term monitoring.

[0139] For example, in the scenario of transporting hazardous chemicals, the application of information acquisition technology can be carried out from multiple dimensions to analyze the abnormal locations and their distribution characteristics. Information acquisition technology mainly uses sensors and monitoring equipment to acquire environmental data around the transport vehicle, such as temperature, humidity, air pressure, and road conditions. Suppose that during a certain transport operation, the system uses onboard sensors to collect data showing an abnormal temperature rise to 45 degrees Celsius and a sudden drop in humidity to 20% on a certain road section. This data is categorized as an environmental data set, laying the foundation for subsequent analysis.

[0140] For example, data cleaning techniques for environmental datasets can improve data quality by filtering noise and filling in missing values.

[0141] In one possible implementation, assuming that some time points are missing in the collected temperature data, the system can interpolate to fill in the gaps by averaging the values ​​of the preceding and following time points, estimating the missing temperature value as 42 degrees Celsius. Meanwhile, for abnormal noise data, such as a sudden temperature jump to 80 degrees Celsius that significantly deviates from the normal range, median filtering can be used to remove it, ensuring a standardized set of environmental data.

[0142] For example, when analyzing environmental interference factors related to positioning accuracy, feature extraction techniques can focus on the impact of specific environmental variables on positioning signals. Suppose that during transportation, the system identifies that high humidity can cause signal attenuation, thus affecting positioning accuracy. By extracting the correlation features between humidity and signal strength, a set of environmental interference features can be determined. For instance, when humidity exceeds 70%, the positioning error may increase to 5 meters, providing a basis for subsequent analysis.

[0143] For example, when using decision tree algorithms to analyze the correlation between interference factors and positioning accuracy, a tree structure can be constructed to determine the weight of each factor. Suppose the system analysis shows that humidity has a weight of 0.6 on positioning accuracy, while temperature has a weight of 0.3, indicating that humidity is the primary interference factor. This weight distribution helps to prioritize high-impact factors and improves the focus of the analysis.

[0144] For example, data integration technology can use spatial overlay analysis to correlate environmental disturbance characteristics with the distribution of abnormal locations. Suppose an abnormal location is concentrated on a certain road segment, and the environmental disturbance characteristics of that segment indicate high humidity. The system can correlate the two to determine that the disturbance's impact range is a 500-meter area surrounding that road segment. This mapping helps to accurately pinpoint the root cause of the problem.

[0145] For example, data storage technology can employ a distributed database storage approach when binding the scope of interference impact with transportation scenario information. Assuming the system binds interference range data with transportation routes, times, and other information, storing it in a cloud database facilitates rapid retrieval during long-term monitoring. This approach ensures data traceability, supporting continuous optimization of transportation safety. Through these multi-dimensional technological applications, a complete technology chain can be formed, from environmental data collection to interference impact analysis and data storage, significantly improving the reliability of abnormal location analysis in the transportation of hazardous chemicals and providing a solid data foundation for safety management.

[0146] S107. Adjust the weighted average algorithm parameters of multi-source sensor fusion by adjusting the interference influence weight to obtain an optimized spatial location sequence.

[0147] Specifically, regarding interference impact and weight analysis, raw data from multiple sensors is acquired. Data preprocessing techniques are used to clean the acquired sensor data, resulting in a standardized base dataset. Based on this standardized base dataset, feature extraction techniques are employed to analyze key factors related to interference impact and determine the interference feature set. For this interference feature set, weight analysis methods are used to calculate the degree of influence of each interference factor on spatial location, obtaining weight distribution data. Based on the weight distribution data, the weighted average parameters in the fusion algorithm are adjusted to generate an optimized algorithm configuration. For the optimized algorithm configuration, data fusion processing is performed using multi-source sensor data to obtain a preliminary spatial location sequence. This preliminary spatial location sequence is smoothed, and time series analysis tools are used to optimize and adjust the position sequence, resulting in the final spatial location sequence. If the final spatial location sequence exhibits abnormal fluctuations in certain intervals, the source of the anomaly is determined by comparing the weight distribution data with historical sensor data, and the abnormal intervals are corrected to determine the final optimized position sequence.

[0148] For example, in the scenario of transporting hazardous chemicals, the business requirements for interference impact and weight analysis can start with the acquisition of multi-source sensor data and gradually delve into the optimization and adjustment of the final location sequence. This involves acquiring raw data from multi-source sensors.

[0149] Understandably, the sensors may include temperature, humidity, air pressure, and vibration sensors, which are deployed at different locations on the transport vehicle to record environmental changes in real time. For example, a temperature sensor collects data once per minute to record abnormally high temperatures that may occur during transportation, while a vibration sensor focuses on the bumps and jolting of the vehicle during travel, with a data frequency that may reach once per second to capture momentary disturbances.

[0150] For example, in data preprocessing techniques, the cleaning process can address outliers or missing values ​​in sensor data. If a temperature sensor has missing data for a certain period, it can be imputed by averaging the values ​​from preceding and following time points. Similarly, if vibration data shows peak values ​​significantly exceeding the normal range, these can be considered noise and removed. This cleaning method helps ensure the standardization of the base dataset for subsequent analysis, avoiding biases caused by data quality issues.

[0151] For example, when analyzing key factors related to interference using feature extraction techniques, features significantly correlated with location deviation can be extracted from environmental variables such as temperature and humidity. Suppose that historical data analysis reveals a significant increase in the positioning signal error rate when humidity exceeds 80%. This feature will be included in the interference feature set as an important basis for subsequent weighting analysis.

[0152] For example, in the weighted analysis method, when calculating the influence of each interference factor on spatial location, historical data comparison can be used to determine that the influence weight of humidity on positioning error is 0.4, while the influence weight of vibration is 0.3. This weight distribution data provides a basis for subsequent algorithm adjustments. When adjusting the weighted average parameter in the fusion algorithm, the proportion of the humidity factor in the algorithm can be increased according to the weight distribution to ensure that the fusion result is closer to the actual environmental impact.

[0153] For example, in data fusion processing, when generating a preliminary spatial location sequence by combining multi-source sensor data, the positioning data from different sensors can be weighted and averaged to obtain a preliminary vehicle location trajectory. Subsequent smoothing processes, such as using time series analysis tools to optimize the location sequence, can reduce jumps caused by interference in a short period, making the trajectory more consistent with the actual driving path.

[0154] For example, if the final spatial location sequence exhibits abnormal fluctuations, such as a sudden deviation from the normal trajectory during a certain segment of the route, the anomaly can be determined by comparing the weighted distribution data with historical sensor records. This can help identify signal interference likely originating from high humidity environments, allowing for correction of the abnormal intervals. This correction process ensures the accuracy of the final location sequence, providing reliable support for real-time monitoring of hazardous chemical transportation.

[0155] For example, in the extended scheme, weather forecast data can be further incorporated as auxiliary input to predict potential high humidity or extreme temperature areas, thereby dynamically adjusting weight parameters in the algorithm configuration. This approach enhances the system's adaptability to environmental changes and ensures the stability of positioning accuracy. Through the above multi-faceted analysis and implementation examples, a complete scheme for interference impact analysis and location sequence optimization has been formed, providing multi-layered protection for transportation safety.

[0156] S108. Based on the optimized spatial location sequence, recalculate the movement trajectory and dynamic movement parameters of the transport vehicle to obtain the updated trajectory parameters.

[0157] Specifically, continuous location points are obtained from the optimized spatial location sequence, and interpolation techniques are used to complete the location points, generating a smooth trajectory dataset. Based on the smooth trajectory dataset, the vehicle's dynamic movement parameters, including speed and acceleration, are calculated, generating a dynamic parameter set. If the speed or acceleration in the dynamic parameter set exceeds a preset threshold, the source of the abnormal parameters is determined by comparing with historical trajectory data, resulting in a corrected dynamic parameter set. Key trajectory features are extracted from the corrected dynamic parameter set, and cluster analysis is used to classify the trajectory features, generating a categorized trajectory dataset. For the categorized trajectory dataset, real-time data processing techniques are used to update the vehicle's movement trajectory parameters, obtaining the final trajectory parameter set. Based on the final trajectory parameter set, time series analysis tools are used to predict the trajectory, generating a predicted trajectory dataset.

[0158] S109. Using the updated trajectory parameters, emergency response instructions for the transportation of hazardous chemicals are generated using real-time data stream processing technology, and a set of emergency response instructions is obtained.

[0159] Specifically, real-time data stream processing technology is used to extract key transportation route information from trajectory parameters. Data filtering tools are then used to classify the route information, resulting in a categorized route dataset. Based on this dataset and real-time monitoring data, potential risks in the transportation of hazardous chemicals are analyzed. If these risks exceed a preset threshold, high-risk routes are identified using data comparison tools, forming a high-risk route set. For this set, vehicle status information from real-time data is obtained, and information integration technology is used to generate targeted emergency response strategies, resulting in a preliminary emergency strategy set. From this set, response mechanisms related to transportation safety are extracted, and logical matching tools are used to prioritize the strategies, identifying the highest-priority response strategy combination. Based on this highest-priority combination, and in conjunction with an instruction generation module, specific emergency response instructions are generated, resulting in an instruction set for the transportation of hazardous chemicals. Using this instruction set, data transmission technology distributes the instructions to relevant transportation management systems, and the execution status of the distributed instructions is obtained to determine the coverage of instruction execution. Finally, based on the coverage of instruction execution and real-time monitoring data, a support vector machine algorithm is used to evaluate the execution effectiveness, yielding the execution effectiveness analysis results.

[0160] For example, in the business scenario of transporting hazardous chemicals, real-time data stream processing technology can be used to extract key transportation route information from trajectory parameters.

[0161] Specifically, by streaming vehicle location data, the system can quickly identify major road segments and nodes that vehicles pass through, such as frequently congested areas or accident-prone areas on a highway. For example, if a transport vehicle's average speed suddenly drops to 10 km / h on a certain route, the system will mark that route as a potential bottleneck and record it.

[0162] For example, when classifying route information using data filtering tools, rules can be set to categorize routes into three types: highways, urban roads, and rural paths. Suppose a chemical hazardous materials transport vehicle primarily travels on urban roads; the system will analyze historical data on traffic density and accident rates for this type of route to generate a categorized route dataset. This classification aids in subsequent risk assessment.

[0163] For example, when analyzing potential risk points by combining real-time monitoring data, the system can obtain weather data and traffic conditions in real time. Suppose a route is affected by heavy rain, reducing visibility to below 50 meters, and the risk value exceeds a preset threshold of 80%. The system will use data comparison tools to filter out this route and categorize it into a high-risk route set. This method can promptly identify problematic routes.

[0164] For example, to obtain vehicle status information for a set of high-risk routes, onboard sensors can be used to acquire vehicle load and tire pressure data. If a vehicle's load exceeds a specified value by 10%, the system will use information integration technology to generate emergency response strategies, such as suggesting slowing down or taking an alternate route, thus initially forming a set of emergency strategies.

[0165] For example, when extracting and prioritizing safety-related response mechanisms from the initial emergency strategy set, a logical matching tool can be used to set collision avoidance strategies as the highest priority. Suppose a strategy suggests that vehicles travel at a speed limit of 30 km / h; the system will prioritize this strategy to ensure transportation safety.

[0166] For example, when generating specific emergency response instructions, the instruction generation module can refine speed limit instructions into specific operational steps, such as starting to slow down 5 kilometers in advance on a certain road section. This refined instruction can improve the accuracy of execution.

[0167] For example, after an instruction is distributed to the transportation management system, the system uses data transmission technology to confirm the coverage of the instruction. Suppose an instruction covers 90% of the target vehicles, the system will record the execution status to ensure nothing is missed.

[0168] For example, when evaluating the performance of a vehicle using a support vector machine (SVM) algorithm, the system analyzes real-time monitoring data to determine whether the vehicle adjusts its speed as instructed. If a vehicle successfully decelerates, the system marks this performance as positive. This evaluation provides a basis for subsequent optimization.

[0169] The above are merely preferred embodiments of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A method for tracking the logistics of hazardous chemicals based on the Internet of Things, characterized in that, Includes the following steps: Spatial location data of chemical hazardous materials transport vehicles are acquired by multi-source sensors, and the spatial location data is integrated by a weighted average algorithm to obtain a continuous spatial location sequence. Based on a continuous spatial location sequence, the Kalman filter algorithm is used to smooth the movement trajectory of the transport vehicle to obtain smoothed trajectory data. Based on the smoothed trajectory data, the real-time speed and direction parameters of the transport vehicle are calculated to obtain a set of dynamic movement parameters; If the speed or direction in the dynamic movement parameter set exceeds the preset threshold, the anomaly detection algorithm is used to determine whether there is an abnormal state in the transportation of hazardous chemicals and to obtain an anomaly status identifier. By using anomaly status identifiers and spatial location sequences, the K-nearest neighbor algorithm is used to perform cluster analysis on the abnormal locations of transport vehicles to obtain the distribution of abnormal locations; Based on the distribution of abnormal locations, environmental data related to the transportation of hazardous chemicals is obtained, and a decision tree algorithm is used to determine the degree of impact of environmental interference on positioning accuracy, thereby obtaining the interference impact weight. By adjusting the weighted average algorithm parameters of multi-source sensor fusion through interference influence weights, an optimized spatial location sequence can be obtained; Based on the optimized spatial location sequence, the movement trajectory and dynamic movement parameters of the transport vehicle are recalculated to obtain the updated trajectory parameters; Based on the updated trajectory parameters, a set of emergency response instructions for the transportation of hazardous chemicals is generated using real-time data stream processing technology.

2. The method according to claim 1, characterized in that, The process of obtaining a continuous spatial location sequence includes: Spatial location information of chemical hazardous materials transport vehicles is collected by multi-source sensors, and signals from different sources are integrated using fusion technology to obtain a preliminary set of location data. Based on the initial location data set, a weighted average algorithm is used to weight the signals from global positioning and inertial navigation to determine the location data set with comprehensive weights. The system detects the location data set after weighting. If the deviation of a certain source signal exceeds a preset threshold, the signal is corrected to obtain the corrected location data set. By analyzing the stability characteristics of signals from different sources using the corrected location data sets, the reliability level of each signal source is determined, and a reliability ranking result is obtained. Based on the reliability ranking results, the priority of each signal source in the fusion process is adjusted to generate an optimized signal fusion strategy; An optimized signal fusion strategy is adopted to process the subsequently acquired spatial location information in real time, resulting in continuous and stable location sequence data.

3. The method according to claim 1, characterized in that, The acquisition of smoothed trajectory data includes: By processing the spatial location information of the transport vehicles, the Kalman filter algorithm is used to smooth the continuous sequence to obtain smoothed trajectory data. Based on the smoothed trajectory data, analyze the abnormal points in the movement trajectory. If a certain position information is detected to deviate from the preset threshold, then the position information is corrected to determine the corrected trajectory data. Based on the corrected trajectory data, combined with the vehicle monitoring system, real-time updated location information is obtained to determine whether there is any data loss or delay. By analyzing the stability of signal acquisition through real-time updated location information, if the continuity of signal acquisition is lower than a preset threshold, the system switches to a backup signal source to obtain stable location data. Based on stable location data, multi-source data information is integrated to generate a comprehensive location information sequence and determine the final movement trajectory.

4. The method according to claim 1, characterized in that, The obtained dynamic movement parameter set includes: For smooth trajectories, data extraction techniques are used to separate the speed changes and direction adjustments of transport vehicles at different time periods, resulting in a preliminary set of dynamic parameters. Based on the initial set of dynamic parameters, the continuity of speed changes is analyzed. If the speed change exceeds the preset threshold, the data segment is processed again to determine the corrected speed information. For the corrected speed information, the matching degree between the two is obtained by combining the direction adjustment data. If the matching degree is lower than the preset threshold, the direction information is calibrated to generate calibrated direction parameters. By fusing the calibrated direction parameters and the corrected velocity information, a comprehensive set of dynamic parameters is generated. Based on the actual set of movement features, and combined with information processing technology, key time and location nodes are extracted to generate the final set of dynamic movement parameters.

5. The method according to claim 1, characterized in that, The acquisition of the abnormal status identifier includes: For the dynamic movement parameter set in the transportation of hazardous chemicals, data filtering technology is used to separate speed data and direction data to obtain preliminary parameter analysis results. Based on the preliminary parameter analysis results, if the speed data or direction data exceeds the preset threshold, the excess data will be processed by the anomaly detection algorithm to determine the preliminary identification of the abnormal state. Based on the initial abnormal status identification, and considering the characteristics of chemical transportation and hazardous materials, relevant environmental data is extracted using information processing technology to obtain contextual information related to the abnormal status. If there is an inconsistency between the acquired context information and the abnormal status identifier, the abnormal status identifier is calibrated to obtain a calibrated status identifier. For the calibrated status identifier, data fusion technology is used to integrate it with dynamic parameters and other data in the moving set to determine whether there is a persistent abnormal state; If a persistent abnormal state is confirmed, a corresponding abnormal warning message will be generated through the information processing stage to determine the final abnormal state identifier.

6. The method according to claim 1, characterized in that, The obtained abnormal location distribution includes: By using abnormal status identifiers and spatial location sequences, data preprocessing techniques are employed to clean the spatial location data of transport vehicles, resulting in standardized location data sequences. Based on the standardized location data sequence, the K-nearest neighbor algorithm is used to cluster the abnormal location data to obtain the preliminary distribution characteristics of the abnormal locations; Based on the initial distribution characteristics, if the cluster center of the abnormal location deviates from the preset normal location range, environmental data related to the abnormal location is extracted through information processing technology to obtain environmental context information. Based on environmental context information, data fusion technology is used to integrate the preliminary distribution characteristics with environmental data to determine whether the distribution of abnormal locations is persistent, and to obtain the calibrated distribution characteristics. If persistent abnormal location distribution is confirmed based on the calibrated distribution characteristics, then abnormal points in the distribution characteristics are marked using anomaly detection technology to obtain an anomaly point set. By using the set of anomalies, information processing technology is employed to generate warning information for the anomaly locations, thus determining the final distribution identifiers of the anomalies.

7. The method according to claim 1, characterized in that, The obtained interference impact weight includes: Based on the abnormal locations and their distribution characteristics, relevant environmental data are obtained from the chemical hazardous materials transportation scenario using information acquisition technology to obtain an environmental data set; Based on the environmental data set, environmental interference factors related to positioning accuracy are analyzed using feature extraction techniques to determine the environmental interference feature set; Based on the environmental interference feature set, the decision tree algorithm is used to analyze the correlation between interference factors and positioning accuracy, and to determine the weight distribution of interference impact. Based on the weighted data of the interference impact, the environmental interference characteristics are correlated and mapped with the distribution of abnormal locations through data integration technology to determine the final range of interference impact. To determine the ultimate scope of the interference, data storage technology is used to bind it with the scenario information of chemical hazardous materials transportation to obtain the basic data required for long-term monitoring.

8. The method according to claim 1, characterized in that, The process of obtaining the optimized spatial location sequence includes: To address interference and weighting analysis, raw data from multiple sensors is acquired, and the acquired sensor data is cleaned using data preprocessing techniques to obtain a standardized basic dataset. Based on a standardized basic dataset, feature extraction techniques are used to analyze key factors related to interference effects and determine the interference feature set. For the set of interference features, the influence of each interference factor on spatial location is calculated by weight analysis to obtain weight distribution data. Based on the weight distribution data, adjust the weighted average parameters in the fusion algorithm to generate an optimized algorithm configuration; Based on the optimized algorithm configuration, data fusion processing is performed using multi-source sensor data to obtain a preliminary spatial location sequence; By smoothing the initial spatial location sequence and optimizing and adjusting the location sequence using time series analysis tools, the final spatial location sequence is obtained. If the final spatial location sequence exhibits abnormal fluctuations within certain intervals, the source of the anomaly is determined by comparing the weight distribution data with the historical data of the sensor data, and the abnormal intervals are corrected to determine the final optimized location sequence.

9. The method according to claim 1, characterized in that, The updated trajectory parameters obtained include: Continuous location points are obtained from the optimized spatial location sequence, and interpolation techniques are used to complete the location points to generate a smooth trajectory dataset; Based on the smooth trajectory dataset, calculate the vehicle's dynamic movement parameters, including speed and acceleration, and generate a set of dynamic parameters; If the velocity or acceleration in the dynamic parameter set exceeds the preset threshold, the source of the abnormal parameters is determined by comparing with historical trajectory data, and a corrected dynamic parameter set is obtained. Key trajectory features are extracted from the modified dynamic parameter set, and cluster analysis is used to classify the trajectory features to generate a classified trajectory dataset. For the classified trajectory dataset, combined with real-time data processing technology, the vehicle's movement trajectory parameters are updated to obtain the final trajectory parameter set; Based on the final set of trajectory parameters, time series analysis tools are used to predict the trajectory and generate a predicted trajectory dataset.

10. The method according to claim 1, characterized in that, The set of emergency response instructions generated includes: By using real-time data stream processing technology, key transportation path information is extracted from trajectory parameters, and data filtering tools are used to classify the path information to obtain a classified path dataset. Based on the classified path dataset and combined with real-time monitoring data, potential risk points in the transportation of hazardous chemicals are analyzed. If the risk points exceed the preset threshold, high-risk paths are selected through data comparison tools to determine the set of high-risk paths. For high-risk route sets, vehicle status information is obtained from real-time data, and information integration technology is used to generate targeted emergency response strategies, resulting in a preliminary set of emergency strategies. From the initial set of emergency strategies, response mechanisms related to transportation safety are extracted, and logical matching tools are used to prioritize the strategies to determine the highest priority combination of response strategies. Based on the highest priority response strategy combination, and in conjunction with the instruction generation module, specific emergency response instructions are generated, resulting in a set of instructions for the transportation of hazardous chemicals. By using a set of instructions and employing data transmission technology, instructions are distributed to relevant transportation management systems to obtain the execution status of the distributed instructions and determine the coverage of instruction execution. Based on the scope of instruction execution and combined with real-time monitoring data, the support vector machine algorithm is used to evaluate the execution effect and obtain the execution effect analysis results.

Images