Integrated Monitoring System and Control Method for a Multifunctional Mobile Monitoring Vehicle

Abstract

The present invention relates to the technical field of data processing, and specifically to an integrated monitoring system and a control method for a multi-functional mobile monitoring vehicle. First, according to the data acquisition mobile data, the time period of the monitoring data is divided into the first data mode, the second data mode, and the third data mode time periods; the time period segmentation nodes of all the data acquisition mobile data are obtained, and the time period of the monitoring data is segmented to obtain a segmented data set; the segmented data set is processed to obtain the in-segment data indicators, and further obtain the pollution over-standard duration; the discrete Fourier transform is used to process the in-segment data indicators corresponding to the segmented data to obtain the monitoring pollution data characteristics; the monitoring pollution data characteristics of the historical data are obtained and processed to generate the time series monitoring pollution data characteristics; the time series monitoring pollution data characteristics are processed, and combined with the pollution over-standard duration, the monitoring area indicators are obtained, and the pollution mode is judged according to the monitoring area indicators.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and particularly to an integrated monitoring system and a control method for a multi-functional mobile monitoring vehicle. Background Art

[0002] Traditional environmental monitoring systems mostly rely on decentralized data collection of fixed stations or single mobile devices. The data of each monitoring unit is isolated and lacks a unified analysis framework, resulting in insufficient spatio-temporal correlation of multi-source heterogeneous data (such as VOCs mass spectrometry, inorganic factors, meteorological parameters). Especially in the mobile vehicle scenario, the existing technology is difficult to achieve real-time data fusion and dynamic path optimization of multi-vehicle collaboration, and traditional algorithms cannot effectively process massive mass spectrometry data with second-level response (such as the hybrid mode of TOF rapid monitoring and GC-MS precise analysis), resulting in delays in pollution source tracing and frequent monitoring blind spots. In addition, the functions of historical data comparison and vehicle path playback rely on manual operations and lack an automated analysis model, reducing the efficiency of multi-dimensional pollution feature extraction.

[0003] In the prior art, by constructing a data processing architecture, integrating multi-vehicle mobile monitoring terminals and a cloud collaborative analysis platform, using spatio-temporal alignment algorithms to unify multi-source data streams (such as VOCs components, meteorological parameters, and GPS trajectories), and introducing machine learning models to achieve pollution diffusion simulation and autonomous path planning. Through multi-threaded parallel computing to optimize the real-time analysis of mass spectrometry data (supporting the mixed processing of conventional mass spectrometry, rapid mass spectrometry, and GC-MS data), combined with the automated comparison of historical data playback and factor superposition, the positioning accuracy of pollution sources and the dynamic decision-making response speed under multi-vehicle collaborative monitoring are improved, effectively solving the bottlenecks of data integration and intelligent analysis in complex environmental scenarios.

[0004] Therefore, an integrated monitoring system and a control method for a multi-functional mobile monitoring vehicle are proposed. Summary of the Invention

[0005] The object of the present invention is to provide a comprehensive monitoring system and its control method for a multi-functional mobile monitoring vehicle, so as to improve the pollution monitoring efficiency in traffic jam areas such as intersections. First, according to the data collection mobile data, the time period of the monitoring data is divided into the first data mode, the second data mode, and the third data mode time periods; obtain the time period segmentation nodes of all the data collection mobile data, and segment the time period of the monitoring data to obtain a segmented data set; perform data processing on the segmented data set to obtain in-segment data indicators, and further obtain the pollution over-standard duration; perform discrete Fourier transform on the in-segment data indicators corresponding to the segmented data to obtain the monitoring pollution data characteristics; obtain the monitoring pollution data characteristics of the historical data and perform processing to generate time series monitoring pollution data characteristics; perform processing on the time series monitoring pollution data characteristics, and combine with the pollution over-standard duration to obtain the monitoring area indicators, and judge the pollution mode according to the monitoring area indicators.

[0006] To achieve the above object, the present invention provides the following technical solutions:

[0007] A comprehensive monitoring and control method for a multi-functional mobile monitoring vehicle, including:

[0008] Collect data according to the data collection time period and the data collection route to obtain monitoring data;

[0009] Further, the step of obtaining the data collection time period includes:

[0010] Obtain multi-source data and perform data cleaning and spatio-temporal alignment to obtain multi-source heterogeneous data at intersections;

[0011] Perform clustering analysis on the multi-source heterogeneous data at intersections through a time series clustering analysis algorithm, and then combine with historical environmental data to calculate the Spearman correlation coefficient between traffic flow and pollutants, and screen the data collection time period.

[0012] Further, the step of obtaining the data collection route includes:

[0013] Obtain and process the historical data of the intersections to be monitored to obtain the first passing time through the intersections to be monitored during the data collection time period; according to the number of monitoring vehicles Group the first passing time and the second passing time between adjacent intersections to be monitored to obtain sets, where the sum of the first passing time and the second passing time in each set has an error less than the error threshold with other sets, and further obtain the data collection route.

[0014] According to the data collection mobile data, the time period of the monitoring data is divided into the first data mode, the second data mode, and the third data mode time periods;

[0015] Further, the steps of obtaining the first data pattern, the second data pattern, and the third data pattern time periods include:

[0016] Train an XGBoost model based on historical data, with vehicle speed, acceleration, and engine load as inputs, and output data patterns, including the first data pattern, the second data pattern, and the third data pattern;

[0017] Obtain complete data collection mobile data, which includes vehicle speed, acceleration, engine load, etc.; Using seconds as the time window, calculate feature statistics through a sliding window, input the feature statistics into the trained XGBoost model, and determine the data pattern;

[0018] Divide the first data pattern, the second data pattern, and the third data pattern time periods according to the time where the data pattern is located.

[0019] Obtain the time period segmentation nodes of all data collection mobile data, and segment the time period of the monitoring data to obtain a segmented data set; perform data processing on the segmented data set to obtain in-segment data indicators, judge whether they exceed the standard according to the in-segment data indicators, and further obtain the pollution over-standard duration;

[0020] Further, the steps of obtaining the segmented data set include:

[0021] Obtain the time period segmentation nodes of all the first data pattern, the second data pattern, and the third data pattern time periods, and merge them to obtain a time period segmentation vector; segment all actual data collection time periods according to the time period segmentation vector to obtain a segmented time period set, and obtain a segmented data set according to the monitoring data corresponding to the segmented time periods.

[0022] Further, the steps of obtaining the pollution over-standard duration include:

[0023] Obtain the in-segment data indicators by merging the segmented data sets according to the pollution data; construct an emission threshold according to historical data and environmental indicator requirements. If the in-segment data indicators are greater than the emission threshold, mark the corresponding segmented time period as the first over-standard duration; judge all the segmented data and sum the first over-standard durations to obtain the pollution over-standard duration.

[0024] Process the in-segment data indicators corresponding to the segmented data by discrete Fourier transform to obtain the monitoring pollution data characteristics;

[0025] Obtain the monitoring pollution data characteristics of historical data and process them to generate time series monitoring pollution data characteristics; process the time series monitoring pollution data characteristics and combine them with the pollution over-standard duration to obtain the monitoring area indicators, and judge the pollution pattern of the area to be monitored according to the monitoring area indicators.

[0026] Further, the steps of obtaining the monitoring area indicators include:

[0027] Processing the characteristics of the time-series monitored pollution data, combining with the over-standard duration, to obtain the first monitoring area indicators; processing the first monitoring area indicators of all monitoring vehicles to obtain the monitoring area indicators; if the monitoring area indicators exceed the threshold, the pollution mode is severe pollution.

[0028] The present application also provides an integrated monitoring system for a multi-functional mobile monitoring vehicle, including:

[0029] A data acquisition module, configured to acquire monitoring data according to the data acquisition period and the data acquisition route;

[0030] A data segmentation module, configured to divide the time period of the monitoring data into a first data mode, a second data mode, and a third data mode time period according to the data acquisition mobile data;

[0031] The data calculation module includes a data segmentation unit, a data index calculation unit, and an over-standard duration calculation unit; wherein the data segmentation unit is configured to obtain the time period segmentation nodes of all the data acquisition mobile data and segment the time period of the monitoring data to obtain a segmented data set; the data index calculation unit is configured to perform data processing on the segmented data set to obtain the in-segment data indicators, and the over-standard duration calculation unit is configured to determine whether it exceeds the standard according to the in-segment data indicators, and further obtain the pollution over-standard duration;

[0032] The feature extraction module is configured to process the in-segment data indicators corresponding to the segmented data according to the discrete Fourier transform to obtain the monitored pollution data features; process the monitored pollution data features of the historical data to generate the time-series monitored pollution data features; process the time-series monitored pollution data features and combine with the pollution over-standard duration to obtain the monitoring area indicators;

[0033] A pollution judgment module, configured to judge the pollution mode of the area to be monitored according to the monitoring area indicators.

[0034] Compared with the prior art, the beneficial effects of the present invention are:

[0035] 1. Through the fine division and optimization of the data acquisition period and the acquisition route, not only can the spatio-temporal consistency and high quality of the monitoring data be ensured, but also the most sensitive monitoring period can be screened out; at the same time, the passing time of the intersection to be monitored is calculated using historical data, and the route balance is achieved through grouping, so that the multi-functional mobile monitoring vehicle can efficiently cover and accurately locate the key pollution areas in the entire area; combined with the XGBoost model for pattern recognition of the vehicle's driving data, the data mode is further subdivided, thus providing scientific, accurate, and efficient data support for environmental pollution early warning, traceability analysis, and emergency response.

[0036] 2. By performing fine segmentation and constructing in-segment data metrics to quantify the pollution status, the spatio-temporal resolution and accuracy of the monitoring data are effectively improved; the data under different data patterns are segmented using a unified time period segmentation vector, and then combined with the historical data and the emission thresholds set by the environmental indicators to determine whether the data of each segment exceeds the standard, thereby accurately calculating the duration of pollution exceeding the standard. This not only provides scientific and reliable data support for pollution early warning response but also further improves the efficiency of overall environmental monitoring.

[0037] 3. By extracting the frequency domain characteristics of the pollution indicators in the segmented data through discrete Fourier transform and combining the time series pollution characteristics constructed from historical data and the actual duration of pollution exceeding the standard, the fluctuations, periodicity, and cumulative effects of pollutant concentrations in the monitoring area can be comprehensively reflected. By integrating the first monitoring area indicators of each monitoring vehicle to generate the final monitoring area indicators and judging the pollution pattern based on this, quantitative assessment and real-time early warning of the regional pollution status can be achieved. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] Figure 1 It is a flowchart of a comprehensive monitoring and control method for a multi-functional mobile monitoring vehicle provided by an embodiment of the present invention;

[0039] Figure 2 It is a flowchart of obtaining a segmented data set provided by an embodiment of the present invention;

[0040] Figure 3 It is a schematic structural diagram of a comprehensive monitoring system for a multi-functional mobile monitoring vehicle provided by an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0041] The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0042] Embodiment 1:

[0043] A certain company introduced a comprehensive monitoring and control method for a multi-functional mobile monitoring vehicle provided by the present invention to monitor the pollution gas emissions at intersections during peak traffic flow periods, aiming to improve the monitoring efficiency and accuracy of pollution gas emissions at intersections during peak traffic flow periods. The process is as Figure 1 shown, and the specific implementation is as follows:

[0044] First, use a clustering algorithm to process and analyze the historical traffic flow data at the intersections in the area to be monitored, obtaining the data collection time period and the data collection route.

[0045] Further, the steps for obtaining the data collection time period include:

[0046] Obtain multi-source data and perform data cleaning and spatio-temporal alignment to obtain multi-source heterogeneous data at the intersections; the multi-source data includes traditional traffic flow data, the trajectory density of online car-hailing, the GPS of logistics vehicles, and social media events, etc.

[0047] Perform clustering analysis on the multi-source heterogeneous data at the intersections through a time series clustering analysis algorithm, and then combine historical environmental data to calculate the Spearman correlation coefficient between traffic flow and tail gas pollutants, and screen the data collection time period.

[0048] Further, the traditional traffic flow data includes data sources such as traffic monitoring cameras, traffic flow sensors, and checkpoint detection devices, recording the historical traffic flow conditions at each intersection; monitor potential traffic anomalies through social platform events (such as accidents, rallies, emergencies, etc.).

[0049] Further, after removing outliers and performing data imputation on the data, map the data from different data sources to the same spatio-temporal coordinate system to ensure that the timestamps and longitude and latitude information of various data are accurately matched, forming complete multi-source heterogeneous data at the intersections.

[0050] Further, perform time series clustering analysis on the multi-source heterogeneous data at the intersections according to the time dimension (such as day, week, month), and extract the traffic flow patterns in each time period; use common time series clustering algorithms, such as K-shape, DTW (Dynamic Time Warping), or DBSCAN, etc., to find time periods with similar traffic flow characteristics.

[0051] Further, obtain historical environmental data, including indicators such as PM2.5, NOx, CO, and VOCs, etc., as a reference for the pollution situation, calculate the Spearman correlation coefficient between the traffic flow and the concentrations of various pollutants in the time periods with similar traffic flow characteristics, and screen out the time periods with strong correlations as the data collection time periods. The formula is:

[0052] ;

[0053] Where represents the Spearman correlation coefficient, and its value range is [-1, 1]. The larger the absolute value, the stronger the correlation. represents the th rank difference between the traffic flow and the pollutant concentration within the Indicates the number of valid data points within the time window. If it is greater than the correlation threshold of 0.6, it indicates the data collection period.

[0054] Table 1. Data collection period

[0055] Time period 1 07:33-08:44 Time period 2 11:15-12:45 Time period 3 13:23-14:11 Time period 4 17:19-18:34

[0056] As shown in Table 1, the data collection periods of the areas to be monitored selected in this embodiment can more comprehensively restore the vehicle flow dynamic characteristics through multi-source heterogeneous data; based on the time series clustering algorithm, it can accurately identify the vehicle flow fluctuation patterns in different time periods, avoid the errors caused by judging only based on the daily or hourly average flow, and improve the scientificity of screening the data collection period; by combining the Spearman correlation coefficient to analyze the strongly correlated periods between the vehicle flow and the pollutant concentration, it can effectively identify the pollution sources and improve the pertinence of pollution monitoring.

[0057] Further, the steps of obtaining the data collection route include:

[0058] Obtain and process the historical data of the intersections to be monitored, and obtain the first passing time through the intersections to be monitored during the data collection period; according to the number of monitoring vehicles Group the first passing time and the second passing time between adjacent intersections to be monitored to obtain sets, where the sum of the first passing time and the second passing time in each set has an error less than the error threshold compared with other sets, and further obtain the data collection route.

[0059] Further, according to the historical passing time data of each intersection to be monitored, the average time passing through the intersection to be monitored during the data collection period can be calculated, denoted as the first passing time; according to the historical passing time data between each intersection to be monitored, the average time from the previous intersection to be monitored to the next intersection to be monitored during the data collection period can be calculated, denoted as the second passing time;

[0060] Further, in this embodiment, there are three multi-functional mobile monitoring vehicles. Then, the intersections to be monitored in the area to be monitored are divided into three sets, and the three monitoring vehicles start running at the same time. Calculate the sum of the first passing time and the second passing time in each set, and ensure that the error of the time sum in each set compared with other sets does not exceed the error threshold. For example, the passing time of set 1 is 40 min, the passing time of set 2 is 38 min, and the passing time of set 3 is 41 min, which meets the error threshold of 10 min. Such a division is acceptable.

[0061] By dividing the intersection to be monitored into multiple sets and assigning a monitoring vehicle to each set, it is ensured that each area can be fully monitored during the data collection period; this grouping method can effectively solve the problem that the monitoring vehicle cannot cover all key nodes due to the overly large area and excessive intersections, and improve the integrity and accuracy of pollution data.

[0062] Data collection is carried out according to the data collection period and the data collection route to obtain monitoring data;

[0063] Furthermore, the collected data includes VOCs, etc.;

[0064] According to the data collection mobile data, the time period of the monitoring data is divided into the first data mode, the second data mode, and the third data mode time periods;

[0065] Furthermore, in this embodiment, the data collection mobile data refers to the driving data of the monitoring vehicle, the first data mode refers to the normal driving mode, the second data mode refers to the idling driving mode, and the third data mode refers to the start-stop driving mode.

[0066] Furthermore, the steps of obtaining the first data mode, the second data mode, and the third data mode time periods include:

[0067] Training an XGBoost model based on historical data, with vehicle speed, acceleration, and engine load as inputs, and outputting data modes, including the first data mode, the second data mode, and the third data mode;

[0068] Obtain the complete data collection mobile data, with seconds as the time window, calculate the feature statistics through a sliding window, input the feature statistics into the trained XGBoost model and judge the data mode;

[0069] Divide the first data mode, the second data mode, and the third data mode time periods according to the time when the data mode is located.

[0070] Furthermore, through hardware devices such as the in-vehicle GPS module, vehicle speed sensor, accelerometer, engine control module (ECU), and vibration sensor of the multi-functional mobile monitoring vehicle, the data collection mobile data of the monitoring vehicle can be obtained, including but not limited to vehicle speed, acceleration, engine complexity, and vehicle vibration intensity (such as vibration amplitude and vibration frequency), etc.;

[0071] Further, clean the historical driving data, remove outliers, invalid data, and noise interference, normalize the vehicle speed, acceleration, engine load, and vehicle vibration data to ensure consistent feature dimensions. Slice the time series data into second-level time windows as the training samples for the XGBoost model. The label data defines the data patterns (the first data pattern, the second data pattern, and the third data pattern) corresponding to each time window. After training, obtain the trained XGBoost model;

[0072] Further, divide the complete data acquisition mobile data by seconds as the time window, and adopt a sliding window mechanism to improve the timeliness and accuracy of recognition. Calculate the following feature statistics within each time window, including vehicle speed, acceleration, engine load, and vehicle vibration data, etc.; input the obtained feature statistics into the trained XGBoost model, and the output is the data pattern.

[0073] Further, after processing the complete data acquisition mobile data, divide the actual data acquisition period according to the obtained data pattern, and merge adjacent time periods with the same pattern to obtain the time periods of the first data pattern, the second data pattern, and the third data pattern;

[0074] Further, the actual data acquisition period refers to the actual time spent by the monitoring vehicle during data acquisition, which is not the same as the data acquisition period mentioned above. The data acquisition period mentioned above is just a time window for data acquisition; the duration of the actual data acquisition period has an approximate relationship with the sum of the first and second passing times in the set mentioned above, but it is not exactly the same.

[0075] Through the division and judgment of data by the XGBoost model and the introduction of the sliding window, the interference of instantaneous data fluctuations on the judgment results is significantly reduced, and the accuracy of data pattern recognition is improved; the impact of different driving modes on pollutant emissions is significantly different. In the second data pattern time period, the engine is running but the vehicle speed is low, and the exhaust emission efficiency is low, which may become a key period for pollutant accumulation. In the third data pattern time period, there are frequent accelerations and brakes, and the instantaneous pollutant emissions are high. In the first data pattern time period, the engine runs stably and the pollutant emissions are relatively uniform. Precise division of time periods optimizes pollution monitoring and analysis.

[0076] Obtain the time segment division nodes of all data acquisition mobile data, and divide the time period of the monitoring data to obtain a segmented data set; perform data processing on the segmented data set to obtain in-segment data indicators, and judge whether they exceed the standard according to the in-segment data indicators, and further obtain the pollution over-standard duration;

[0077] Further, the process of obtaining the segmented data set is as Figure 2As shown in the figure, it includes:

[0078] Obtain the time period segmentation nodes of the first data mode, second data mode, and third data mode time periods of all monitoring vehicles, and merge them to obtain a time period segmentation vector; segment the actual data collection time periods of all multi-functional mobile monitoring vehicles according to the time period segmentation vector to obtain a set of segmented time periods, and obtain the segmented data set of each monitoring vehicle according to the monitoring data corresponding to the elements in the set of segmented time periods.

[0079] Furthermore, first process the vehicle status of each multi-functional mobile monitoring vehicle (including the first data mode, second data mode, and third data mode), and based on the results obtained from the pre-trained XGBoost model, identify the time nodes of each mode switch. These switch nodes are the key time points for the monitoring vehicle to switch from one mode to another.

[0080] Furthermore, merge and normalize the time period segmentation nodes extracted from each monitoring vehicle to generate a unified "time period segmentation vector". This vector can reflect the mode distribution in different time periods of the entire monitoring system and provide a unified standard for subsequent time period segmentation;

[0081] Furthermore, use the aforementioned obtained time period segmentation vector to segment the actual data collection time periods of all monitoring vehicles; here, the continuous operation time of each monitoring vehicle is cut into multiple continuous sub-time periods according to the segmentation nodes, and each sub-time period corresponds to a segment of pollution data collected by the monitoring vehicle;

[0082] Furthermore, perform the above processing on all the data to obtain the segmented data set of each monitoring vehicle.

[0083] Segment the continuous monitoring data according to the time period segmentation nodes, so that each data segment has a clear time period label, reducing the noise brought by the mixed mode, ensuring that the mode within each segment of data is consistent, thereby improving the data quality and the reliability of subsequent analysis; significantly improving the efficiency of pollution source tracing, real-time monitoring, and monitoring, providing a solid data support and scientific basis for environmental pollution monitoring and control.

[0084] Furthermore, the steps to obtain the pollution over-standard duration include:

[0085] Obtain the segmented data set and construct the in-segment data indicators according to the collected pollution data such as VOCs and other parameters; construct the emission threshold according to the historical data and environmental index requirements. If the in-segment data indicator is greater than the emission threshold, mark the duration of the corresponding segmented time period as the first over-standard duration; judge all the segmented data of the monitoring vehicle and sum up the first over-standard duration to obtain the pollution over-standard duration monitored by the monitoring vehicle.

[0086] Furthermore, the calculation formula for the in-segment data index is as follows:

[0087] ;

[0088] where represents the in-segment data index, represents the number of pollutant types, represents the th type of pollutant's average concentration within the time period, represents the th type of pollutant's emission threshold, which is set according to national or local standards, represents the th type of pollutant's weight coefficient, reflecting its contribution weight to exceeding the standard, represents the duration of this time period. The longer the time, the greater the impact.

[0089] Furthermore, if the in-segment data index is greater than the emission threshold, then mark the time period corresponding to this segmented data as the first exceeded standard duration; judge all the segmented data sets of the monitoring vehicle and sum up the first exceeded standard duration to obtain the pollution exceeded standard duration detected by the monitoring vehicle.

[0090] Using the segmented data set, the pollution data collected by the monitoring vehicle at different time periods can be finely divided to ensure that the patterns within each segment of data are consistent, thus more accurately reflecting the true pollution level within this segment; by calculating the in-segment data index and combining the average concentration, emission standard, and weight of each pollutant, the pollution load in different time periods can be objectively quantified, providing a fine data basis for judging the exceeded standard situation; taking the duration into account indicates that the pollution effect of the same average concentration is greater in a long time period, and this design is closer to the actual environmental impact.

[0091] The discrete Fourier transform processes the in-segment data index corresponding to the segmented data to obtain the characteristics of the monitored pollution data;

[0092] Obtain the characteristics of the monitored pollution data of the historical data and process them to generate the time-series characteristics of the monitored pollution data; process the time-series characteristics of the monitored pollution data and combine them with the pollution exceeded standard duration to obtain the monitoring area index, and judge the pollution pattern of the area to be monitored according to the monitoring area index.

[0093] Furthermore, the steps to obtain the monitoring area index include:

[0094] Process the time-series characteristics of the monitored pollution data of each multi-functional mobile monitoring vehicle, combine with the exceeded standard duration, to obtain the first monitoring area index; process the first monitoring area indexes of all multi-functional mobile monitoring vehicles to obtain the monitoring area index. If the monitoring area index exceeds the threshold, then the pollution pattern of the area to be monitored is severe pollution.

[0095] Further, the in-segment data metrics within each time period can be regarded as a discrete time signal, which reflects the change and accumulation of pollutant concentration within a certain time. The in-segment data metrics within each time period are transformed from the time domain to the frequency domain using DFT; through this transformation, the spectral information of the signal can be obtained, such as the main frequency components, amplitude, and phase information. These frequency domain features can reveal the potential periodic fluctuations (such as daily and weekly cycles, etc.) and the frequency characteristics of sudden pollution events in the pollution data collected by the monitoring vehicle during the actual data collection period.

[0096] Further, the main frequency components (such as the frequency corresponding to the maximum amplitude), spectral energy distribution (reflecting the overall fluctuation intensity), and phase information (describing the timing characteristics of pollutant concentration fluctuations) are extracted from the spectrum. These features will be used to quantify the pollution pattern within the current monitoring data segment and conduct comparative analysis with the historical data features in the follow-up.

[0097] Further, from the historical data, the frequency domain features of the monitored pollution data are extracted using the same method (i.e., performing discrete Fourier transform on the in-segment data metrics corresponding to the segmented data set); the frequency domain features extracted within each historical time period are arranged in chronological order to form the complete time series of monitored pollution data features.

[0098] Further, based on the above information, the first monitoring area index can be constructed, and this index can be expressed as:

[0099] ;

[0100] where, represents the first monitoring area index, represents the monitored pollution data features (such as main frequency, amplitude, energy distribution, etc.) extracted from the segmented data of the current monitoring vehicle through DFT, represents the time series of monitored pollution data features, represents the pollution over-standard duration corresponding to the monitoring vehicle, and the function can be constructed through weighted average, principal component analysis, or machine learning models (such as multivariate regression) to output a quantitative first monitoring area index, which is convenient for comprehensive evaluation of different regions or time periods.

[0101] Further, the first monitoring area indexes of all multi-functional mobile monitoring vehicles are summed and weighted to obtain the monitoring area index of the area to be monitored. If it exceeds the threshold, the pollution pattern in the area to be monitored is severe pollution.

[0102] Table 2. Regional Monitoring Index

[0103] Time period number Regional monitoring index A023 0.78 A024 0.76 A025 0.77 A026 0.76

[0104] As shown in Table 2, the monitoring area indicators of the area to be monitored in some time periods are shown. It can be seen that none of them exceed the threshold of 0.9. By performing DFT processing on the cumulative indicators of the segmented data, the main frequency components, amplitudes, and phase information can be extracted from the frequency domain. These frequency domain features can reveal the periodic fluctuations and sudden event characteristics within the data segment; incorporating the key indicator of the pollution exceeding the standard duration into the comprehensive evaluation can better reflect the cumulative effect of the pollutant concentration exceeding the standard for a long time; by extracting and quantifying the frequency domain features of the monitoring data through discrete Fourier transform and combining them with the exceeding standard duration and historical data features, the pollution situation in the area can be comprehensively and finely described, thereby improving the accuracy, timeliness, and decision-making support ability of pollution monitoring, and providing a scientific and reliable basis for environmental early warning and governance.

[0105] Through multi-source data cleaning, spatio-temporal alignment, data pattern recognition, fine segmentation, pollution index construction, and frequency domain feature extraction, the full-process automated analysis from data collection to the pollution exceeding the standard duration and monitoring area indicators is realized; this method not only improves the accuracy and spatio-temporal consistency of data collection, but also effectively captures the periodic fluctuations and sudden event characteristics of the pollutant concentration through the fine segmentation processing and DFT frequency domain analysis of the data of each monitoring vehicle; at the same time, combining the pollution exceeding the standard duration, a comprehensive regional pollution index is constructed to realize the accurate judgment and real-time early warning of the pollution pattern in the area to be monitored, providing scientific, comprehensive, and timely decision-making support for environmental management and pollution control.

[0106] Example Two:

[0107] This application also provides a comprehensive monitoring system for a multi-functional mobile monitoring vehicle. The system structure is as Figure 3 shown, and the specific implementation method is as follows:

[0108] A data collection module, used to collect data according to the data collection time period and data collection route to obtain monitoring data;

[0109] Further, the steps of obtaining the data collection time period include:

[0110] Obtain multi-source data and perform data cleaning and spatio-temporal alignment to obtain multi-source heterogeneous data at intersections;

[0111] Perform clustering analysis on the multi-source heterogeneous data at intersections through a time series clustering analysis algorithm, and then combine historical environmental data to calculate the Spearman correlation coefficient between traffic flow and pollutants to screen the data collection time period.

[0112] Further, the steps of obtaining the data collection route include:

[0113] Obtain historical data of intersections to be monitored and process it to obtain the first passing time of passing through the intersections to be monitored during the data collection period; according to the number of monitoring vehicles Group the first passing time and the second passing time between adjacent intersections to be monitored to obtain sets, where the sum of the first passing time and the second passing time in each set has an error less than the error threshold compared to other sets, and further obtain the data collection route.

[0114] A data segmentation module, used to divide the time period of the monitoring data into the first data mode, the second data mode, and the third data mode time periods according to the data collection mobile data;

[0115] Furthermore, the steps of obtaining the first data mode, the second data mode, and the third data mode time periods include:

[0116] Train an XGBoost model based on historical data, with vehicle speed, acceleration, and engine load as inputs, and output data modes, including the first data mode, the second data mode, and the third data mode;

[0117] Obtain the complete data collection mobile data, with seconds as the time window, calculate the feature statistics through a sliding window, input the feature statistics into the trained XGBoost model and judge the data mode;

[0118] Divide the first data mode, the second data mode, and the third data mode time periods according to the time when the data mode is located.

[0119] The data calculation module includes a data segmentation unit, a data index calculation unit, and an over-standard duration calculation unit; among them, the data segmentation unit is used to obtain the time period segmentation nodes of all data collection mobile data and segment the time period of the monitoring data to obtain a segmented data set; the data index calculation unit is used to process the segmented data set to obtain in-segment data indexes, and the over-standard duration calculation unit is used to judge whether it is over-standard according to the in-segment data indexes, and further obtain the pollution over-standard duration;

[0120] Furthermore, the steps of obtaining the segmented data set include:

[0121] Obtain the time period segmentation nodes of all the first data mode, the second data mode, and the third data mode time periods, and merge them to obtain a time period segmentation vector; segment all the actual data collection time periods according to the time period segmentation vector to obtain a segmented time period set, and obtain a segmented data set according to the monitoring data corresponding to the segmented time periods.

[0122] Furthermore, the steps of obtaining the pollution over-standard duration include:

[0123] Obtain segmented data sets and construct in-segment data metrics based on contaminated data; construct emission thresholds according to historical data and environmental metric requirements. If the in-segment data metrics are greater than the emission thresholds, mark the corresponding segmented time periods as the first over-standard duration; judge all segmented data and sum up the first over-standard duration to obtain the pollution over-standard duration. Table 3 shows the pollution over-standard durations counted by four multi-functional mobile monitoring vehicles during this monitoring process.

[0124] Table 3. Pollution over-standard duration

[0125] Vehicle number Actual data collection time period Duration of pollution exceeding the standard JC001 2024.8.18 13:02-14:23 32 minutes 24 seconds JC002 2024.8.18 13:01-14:17 31 minutes 15 seconds JC003 2024.8.18 13:00-14:20 30 minutes 49 seconds JC004 2024.8.18 13:04-14:19 31 minutes 25 seconds

[0126] The feature extraction module is used to process the in-segment data metrics corresponding to the segmented data according to the discrete Fourier transform to obtain the characteristics of the monitored pollution data; obtain the characteristics of the monitored pollution data of the historical data for processing to generate the time series characteristics of the monitored pollution data; process the time series characteristics of the monitored pollution data and combine with the pollution over-standard duration to obtain the monitored area metrics.

[0127] Furthermore, the steps of obtaining the monitored area metrics include:

[0128] Process the time series characteristics of the monitored pollution data and combine with the over-standard duration to obtain the first monitored area metrics; process the first monitored area metrics of all monitoring vehicles to obtain the monitored area metrics.

[0129] The pollution judgment module is used to judge the pollution mode of the area to be monitored according to the monitored area metrics. If the monitored area metrics exceed the threshold, the pollution mode is severe pollution.

[0130] Although the embodiments of the present invention have been shown and described, for those of ordinary skill in the art, it can be understood that various changes, modifications, substitutions and variations can be made to these embodiments without departing from the principles and spirit of the present invention. The scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. An integrated monitoring and control method for a multi-functional mobile monitoring vehicle, characterized in that, Including: Collect monitoring data according to the data collection period and data collection route; Divide the time period of the monitoring data into the first data mode, the second data mode, and the third data mode time periods according to the data collection mobile data; Obtain the time period segmentation nodes of all the data collection mobile data, and segment the time period of the monitoring data to obtain a segmented data set; perform data processing on the segmented data set to obtain in-segment data indicators, determine whether they exceed the standard according to the in-segment data indicators, and further obtain the pollution over-standard duration; Process the in-segment data indicators corresponding to the segmented data by discrete Fourier transform to obtain the monitoring pollution data characteristics; Obtain the monitoring pollution data characteristics of historical data and process them to generate time series monitoring pollution data characteristics; Process the time series monitoring pollution data characteristics and combine them with the pollution over-standard duration to obtain the monitoring area indicators, and determine the pollution mode of the area to be monitored according to the monitoring area indicators; Obtain multi-source data and perform data cleaning and spatio-temporal alignment to obtain multi-source heterogeneous data at intersections; Perform cluster analysis on the multi-source heterogeneous data at intersections through a time series clustering analysis algorithm, and then combine historical environmental data to calculate the Spearman correlation coefficient between traffic flow and pollutants, and screen the data collection period; Train an XGBoost model based on historical data, with vehicle speed, acceleration, and engine load as inputs, and output data modes including the first data mode, the second data mode, and the third data mode; Obtain complete data acquisition mobile data to Take seconds as the time window, calculate the feature statistic through a sliding window, input the feature statistic into the trained XGBoost model and judge the data pattern; Divide the first data mode, the second data mode, and the third data mode time periods according to the time where the data mode is located.

2. The integrated monitoring and control method of a multifunctional mobile monitoring vehicle according to claim 1, characterized in that, The steps to obtain the data collection route include: Obtain historical data of the intersection to be monitored and process it to obtain the first passing time of passing through the intersection to be monitored during the data collection period; according to the number of monitoring vehicles Group the first passing time and the second passing time between adjacent intersections to be monitored to obtain a set of collections, where the sum of the first passing time and the second passing time in each collection has an error less than the error threshold compared with other collections, and further obtain the data collection route.

3. The comprehensive monitoring and control method of a multifunctional mobile monitoring vehicle according to claim 1, characterized in that The steps to obtain the segmented data set include: Obtain the time period segmentation nodes of all the first data mode, the second data mode, and the third data mode time periods, and merge them to obtain a time period segmentation vector; segment all the actual data collection periods according to the time period segmentation vector to obtain a set of segmented time periods, and obtain the segmented data set according to the monitoring data corresponding to the segmented time periods.

4. The integrated monitoring and control method of a multifunctional mobile monitoring vehicle according to claim 1, characterized in that, The steps to obtain the pollution over-standard duration include: Obtain the segmented data set and construct in-segment data indicators according to the pollution data; construct an emission threshold according to historical data and environmental indicator requirements. If the in-segment data indicator is greater than the emission threshold, mark the corresponding segmented time period as the first over-standard duration; judge all the segmented data and sum the first over-standard durations to obtain the pollution over-standard duration.

5. The integrated monitoring and control method of a multifunctional mobile monitoring vehicle according to claim 1, characterized in that, The steps to obtain the monitoring area indicators include: Process the time series monitoring pollution data characteristics, combine with the over-standard duration, to obtain the first monitoring area indicators; process the first monitoring area indicators of all the monitoring vehicles to obtain the monitoring area indicators; if the monitoring area indicators exceed the threshold, the pollution mode is severe pollution.

6. An integrated monitoring system for a multi-functional mobile monitoring vehicle, which executes the method according to claim 1, characterized in that, Including: A data collection module, used to collect monitoring data according to the data collection period and data collection route; A data segmentation module, used to divide the time period of the monitoring data into the first data mode, the second data mode, and the third data mode time periods according to the data collection mobile data; The data calculation module includes a data segmentation unit, a data index calculation unit, and an over-standard duration calculation unit; among them, the data segmentation unit is used to obtain the time period segmentation nodes for collecting all mobile data, and segment the time period of the monitoring data to obtain a segmented data set; the data index calculation unit is used to process the segmented data set to obtain the in-segment data index; the over-standard duration calculation unit is used to determine whether it exceeds the standard according to the in-segment data index, and further obtain the pollution over-standard duration. The feature extraction module is used to process the in-segment data index corresponding to the segmented data according to the discrete Fourier transform to obtain the monitoring pollution data features; process the monitoring pollution data features of the historical data to generate the time series monitoring pollution data features. Process the time series monitoring pollution data features and combine them with the pollution over-standard duration to obtain the monitoring area index. The pollution judgment module is used to judge the pollution mode of the area to be monitored according to the monitoring area index.

Images