Dust-slag vehicle linkage supervision system and method based on multi-source data fusion

Abstract

The application relates to the technical field of intelligent traffic Internet of Things, and particularly discloses a dust-muck truck linkage supervision system and method based on multi-source data fusion, which comprises the following steps: acquiring all feasible routes between the starting point and the destination of a muck truck, dividing each feasible route into a plurality of route sections according to a predetermined route interval; collecting the dust amount in the route section, marking the route section as an abnormal section if the dust amount is greater than a preset threshold; otherwise, marking the route section as a normal section; calculating a first score and a second score based on the primary influence score and the secondary influence score of the abnormal section; summing up the first score and the second score of all abnormal sections on a single feasible route to obtain an evaluation score of the feasible route, and taking the feasible route corresponding to the smallest evaluation score as the driving route of the muck truck. The application can reduce the influence of dust on the surrounding area during the driving process of the muck truck to the greatest extent.

Description

Technical Field

[0001] This invention relates to the field of intelligent transportation Internet of Things technology, specifically to a dust-construction truck linkage monitoring system and method based on multi-source data fusion. Background Technology

[0002] Dust is a key contributor to urban air pollution, significantly reducing air quality and impacting residents' respiratory health. In large cities, dense construction sites are a major source of dust: from earthwork excavation and building material stockpiling to material transfer during construction, every step can generate large amounts of dust particles.

[0003] Among these processes, the transportation of construction waste by dump trucks is particularly critical. Some vehicles may spill construction waste due to bumps during the journey, which can easily cause secondary dust pollution on the roads. At the same time, the mud and sand carried by the vehicle tires will also spread on urban roads along the driving trajectory.

[0004] In this situation, if the dump trucks choose inappropriate routes, such as near residential areas, it will further aggravate the dust pollution problem in the area and affect the health of residents. Summary of the Invention

[0005] The purpose of this invention is to provide a dust and construction waste truck linkage monitoring system and method based on multi-source data fusion, and to solve the following technical problems: Improper route selection by dump trucks, such as driving close to residential areas, can further exacerbate dust pollution and affect residents' health.

[0006] The objective of this invention can be achieved through the following technical solutions: The method for joint monitoring of dust and construction waste trucks based on multi-source data fusion includes the following steps: Obtain all feasible routes between the origin and destination of the dump truck. For a single feasible route, divide it into several road segments according to the predetermined distance interval. The system collects the amount of dust generated within a given route. If the amount of dust exceeds a preset threshold, the route is marked as an abnormal route; otherwise, it is marked as a normal route. Based on the location information, wind information and dust volume of the abnormal segment, the impact score of the abnormal segment on the preset type of building is obtained and recorded as the initial impact score. The total value B1 of the initial impact score of a single abnormal segment is calculated and recorded as the first score. The impact score reflects the impact of the dust volume of the abnormal segment on the preset type of building. Obtain the influence score of the abnormal segment on the midpoint of the normal segment, and record it as the secondary influence score. Calculate the total value B2 of the primary influence score of a single abnormal segment, and record the total value B2 as the second score. The evaluation score of a feasible route is obtained by summing the first and second scores of all abnormal segments on a single feasible route, and the feasible route corresponding to the lowest evaluation score is taken as the route for the dump truck.

[0007] As a further aspect of the present invention: obtaining the impact score of the abnormal segment on a preset type of building includes: Draw a circle with a preset radius centered at the midpoint of the abnormal segment, mark the buildings of a preset type inside the circle as target buildings, obtain the distance L between the target building and the midpoint of the abnormal segment, and obtain the wind speed and wind direction at the midpoint of the abnormal segment. Location information includes distance L, and wind information includes wind speed and direction; The dust volume, wind speed, wind direction, and distance L of the abnormal section are input as a real-time sample into a pre-trained artificial intelligence model to obtain the impact score of the abnormal section on the target building.

[0008] As a further aspect of the present invention: the process of obtaining a pre-trained artificial intelligence model includes: Acquire real-time samples from historical monitoring processes, record them as historical samples, and label the impact scores of historical samples; An artificial intelligence model is built based on deep learning, and the model is trained and validated based on labeled historical samples to obtain a pre-trained artificial intelligence model.

[0009] As a further aspect of the present invention: the process of labeling the influence scores of historical samples includes: For a single historical sample, the dust volume at the location of the corresponding target building A is periodically collected and the mean Y1 is calculated; Calculate the average dust emission Y2 at the location of target building A over the past n days; The ratio of Y1 to Y2 is used as the influence score. The larger the influence score, the more severe the impact of the dust in the abnormal section on the preset type of building.

[0010] As a further aspect of the present invention: the process of obtaining the secondary influence score includes: The midpoint of the normal segment is used as the target building to obtain secondary influence scores.

[0011] As a further aspect of the present invention: if two or more evaluation scores are identical and represent the minimum value, then the following steps are performed: The feasible route with the same and smallest corresponding evaluation score is marked as the candidate route. The sum of the first and second scores of a single abnormal segment on the candidate route is used as the ranking score. The ranking scores are sorted in descending order to obtain the score ranking. Generate coordinate point (i, Di), where Di represents the i-th ranked score in the score sorting. Connect two adjacent coordinate points with a straight line to obtain a score line graph. Draw a straight line passing through (0, P) and parallel to the x-axis as a reference line, where P is a preset sorting score threshold, and slowly move the reference line downwards along the y-axis at a preset speed. For a single score line chart, the proportion of the domain of the part above the reference line to the domain of the score line chart is obtained in real time and recorded as the filtering ratio. When the difference between the largest and smallest screening ratios is greater than a preset difference threshold, the candidate curves corresponding to the largest screening ratio are removed until only one candidate route remains, and this route is marked as the driving route of the dump truck.

[0012] As a further aspect of the present invention, it also includes: As the dump truck travels along its route, monitoring videos of the truck are collected in real time by data acquisition devices at preset locations until the truck reaches its destination, at which point the monitoring videos are sent to management personnel for archiving.

[0013] As a further aspect of the present invention: if a first score and / or a second score are greater than a preset score threshold, the corresponding abnormal segment is reported to a preset management personnel, and the corresponding feasible route is not used as a driving route.

[0014] The dust and construction waste truck linkage monitoring system based on multi-source data fusion includes: Road segmentation module: Obtain all feasible routes between the origin and destination of the dump truck. For a single feasible route, divide it into several road segments according to the predetermined distance interval. The system collects the amount of dust generated within a given route. If the amount of dust exceeds a preset threshold, the route is marked as an abnormal route; otherwise, it is marked as a normal route. Impact Determination Module: Based on the location information, wind information and dust volume of the abnormal segment, obtain the impact score of the abnormal segment on the preset type of building, record it as the initial impact score, calculate the total value B1 of the initial impact score of a single abnormal segment, record the total value B1 as the first score, and the impact score reflects the impact of the dust volume of the abnormal segment on the preset type of building. Obtain the influence score of the abnormal segment on the midpoint of the normal segment, and record it as the secondary influence score. Calculate the total value B2 of the primary influence score of a single abnormal segment, and record the total value B2 as the second score. Selection module: The first score and the second score of all abnormal segments on a single feasible route are summed to obtain the evaluation score of the feasible route. The feasible route corresponding to the lowest evaluation score is taken as the driving route of the dump truck.

[0015] The beneficial effects of this invention compared to the prior art are as follows: 1) This invention evaluates the feasible routes of dump trucks in segments and combines dust volume, wind information and building location for comprehensive analysis. It can effectively select driving routes with less overall dust impact, thereby avoiding the aggravation of dust pollution problems in the surrounding area due to improper route selection. 2) By reporting and removing abnormal segments whose evaluation scores exceed the threshold, high-risk routes can be prevented from being selected as actual driving routes in a timely manner, ensuring that the transportation process of dump trucks is more controllable and reducing adverse impacts on the surrounding environment and residents' health. 3) When there are multiple candidate routes with the same evaluation score, this invention further examines the strength of the route's spatial distribution to avoid local environmental degradation, thereby achieving a more favorable selection for the "most areas" and ensuring that the overall operation process has a relatively small environmental impact on most areas of the city. Attached Figure Description

[0016] The invention will now be further described with reference to the accompanying drawings.

[0017] Figure 1 This is a flowchart illustrating the dust-contaminated truck linkage monitoring method based on multi-source data fusion, as proposed in this invention. Detailed Implementation

[0018] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0019] Please see Figure 1 As shown, this invention is a method for joint monitoring of dust and construction waste trucks based on multi-source data fusion, comprising the following steps: Obtain all feasible routes between the origin and destination of the dump truck. For a single feasible route, divide it into several road segments according to the predetermined distance interval. During implementation, the origin and final destination of the dump trucks are first determined, and then all possible driving routes between them are fully acquired. Feasible routes refer to the set of routes that vehicles can successfully complete the transportation task under the premise that road conditions are permissible and traffic rules are followed. After obtaining all feasible routes, any specific route is further refined: starting from the origin, the route is divided into several consecutive road segments according to the pre-set road intervals, so as to facilitate the subsequent collection and comparative analysis of dust conditions in different road segments. The system collects the amount of dust generated within a given route. If the amount of dust exceeds a preset threshold, the route is marked as an abnormal route; otherwise, it is marked as a normal route. Based on the location information, wind information and dust volume of the abnormal segment, the impact score of the abnormal segment on the preset type of building is obtained and recorded as the initial impact score. The total value B1 of the initial impact score of a single abnormal segment is calculated and recorded as the first score. The impact score reflects the impact of the dust volume of the abnormal segment on the preset type of building. In a preferred embodiment of the present invention, obtaining the impact score of anomaly segments on a preset type of building includes: When obtaining the impact score of an abnormal segment on a building of a preset type, first use the midpoint of the abnormal segment as a reference position, and draw a circle with the midpoint as the center and a radius of a preset value around the position; the spread range of dust is usually limited by spatial distance, so it is reasonable to use the range of the circle to define the potentially affected buildings; Buildings located within the circular area and belonging to the preset type will be marked as target buildings. The preset types include, but are not limited to, schools, hospitals, etc., which can be selected according to actual needs. Next, the distance between these target buildings and the midpoint of the anomaly segment is determined. This distance reflects the spatial attenuation process that dust undergoes as it travels from its source to the target location. At the same time, the wind speed and direction at the midpoint of the anomaly segment are collected, because the way and range of dust particles diffuse in the air largely depend on the airflow conditions at the time. Wind speed determines the strength of the diffusion, and wind direction determines the dominant direction of the diffusion. At this point, the location information is represented by the distance from the target building to the midpoint of the anomaly segment, and the wind information is represented by both wind speed and wind direction. By combining the dust volume of the abnormal section, the collected wind speed, wind direction, and distance data, a real-time sample reflecting the impact of dust under specific spatiotemporal conditions is formed. This real-time sample is input into an artificial intelligence model that has been trained using historical monitoring data. The model uses the learned patterns of the relationship between dust diffusion and the impact on buildings to process the input data and output an impact score of the abnormal section on the target building. This score can intuitively reflect the degree to which the building is affected by the dust of the abnormal section under the current conditions.

[0020] In a preferred embodiment, the process of obtaining a pre-trained artificial intelligence model includes: Acquire real-time samples from historical monitoring processes, record them as historical samples, and label the impact scores of historical samples; An artificial intelligence model is built based on deep learning, and the model is trained and validated based on labeled historical samples to obtain a pre-trained artificial intelligence model.

[0021] It is worth noting that historical samples contain the relationships between dust dispersion, meteorological conditions, and environmental characteristics. Labeling these samples can provide data with clear reference significance for subsequent model training. Building and training artificial intelligence models based on these labeled historical samples allows the models to gradually learn and master the patterns of dust impact, and the verification process ensures that they still have reliability and generalization ability under different conditions. The resulting pre-trained artificial intelligence model can quickly give reasonable impact scores when faced with new real-time data, avoiding reliance on human experience judgment, thereby improving the accuracy of identifying environmental risks in abnormal periods.

[0022] It should be noted that the process of labeling the influence scores of historical samples includes: For a single historical sample, the dust volume at the location of the corresponding target building A is periodically collected and the mean Y1 is calculated; Calculate the average dust emission Y2 at the location of target building A over the past n days; The ratio of Y1 to Y2 is used as the influence score. The larger the influence score, the more severe the impact of the dust in the abnormal section on the preset type of building.

[0023] In the process of labeling historical samples, it is necessary to periodically collect dust data at the location of the target building corresponding to each historical sample and calculate the average value over a period of time. This can reflect the actual dust impact level on the target building at the current time. At the same time, it is also necessary to calculate the average dust data of the target building over the past few days. This average value can reflect the normal level of the target building over a longer time scale. By comparing these two sets of results, the degree of deviation of the current sample from the normal can be intuitively reflected, thereby obtaining the impact score. The larger the impact score, the more serious the impact of the abnormal dust on the building is.

[0024] The above method enables the use of a simple indicator to reflect the relative strength of dust impact on buildings at different time periods, providing data with clear reference significance for subsequent model training, so that the model can learn the correspondence between dust volume and environmental impact.

[0025] Obtain the influence score of the abnormal segment on the midpoint of the normal segment, and record it as the secondary influence score. Calculate the total value B2 of the primary influence score of a single abnormal segment, and record the total value B2 as the second score. In another preferred embodiment of the present invention, the process of obtaining the secondary influence score includes: The midpoint of the normal segment is used as the target building to obtain secondary influence scores.

[0026] It should be noted that the method for obtaining the score of the impact of abnormal segments on the target building can be compared with the above method, and will not be elaborated on here; When analyzing abnormal sections, it is necessary to consider not only their direct impact on surrounding buildings, but also their indirect impact on normal road sections, so as to more comprehensively reveal the dust diffusion path and coverage in different spatial locations, and avoid ignoring the risk of being affected simply because a certain section does not exceed the threshold. By summing these impact scores to obtain a second score, the extent of the impact of the abnormal segment on the surrounding normal area can be more intuitively shown, thereby extending the scope of the abnormal segment's influence to the surrounding potentially affected areas and avoiding underestimating the actual risks; when selecting routes, both direct and indirect impacts should be taken into account to ensure a more objective assessment of the environmental situation.

[0027] It should be noted that if the first score and / or the second score are greater than the preset score threshold, the corresponding abnormal segment will be reported to the preset management personnel, and the corresponding feasible route will not be used as the driving route.

[0028] The evaluation score of a feasible route is obtained by summing the first and second scores of all abnormal segments on a single feasible route, and the feasible route corresponding to the lowest evaluation score is taken as the route for the dump truck. If two or more evaluation scores are the same and represent the minimum value, then perform the following steps: The feasible route with the same and smallest corresponding evaluation score is marked as the candidate route. The sum of the first and second scores of a single abnormal segment on the candidate route is used as the ranking score. The ranking scores are sorted in descending order to obtain the score ranking. Generate coordinate point (i, Di), where Di represents the i-th ranked score in the score sorting. Connect two adjacent coordinate points with a straight line to obtain a score line graph. Draw a straight line passing through (0, P) and parallel to the x-axis as a reference line, where P is a preset sorting score threshold, and slowly move the reference line downwards along the y-axis at a preset speed. For a single score line chart, the proportion of the domain of the part above the reference line to the domain of the score line chart is obtained in real time and recorded as the filtering ratio. When the difference between the largest and smallest screening ratios is greater than a preset difference threshold, the candidate curves corresponding to the largest screening ratio are removed until only one candidate route remains, and this route is marked as the driving route of the dump truck.

[0029] Understandably, when multiple feasible routes have the same and minimum evaluation score, further screening is required. First, these routes are marked as candidate routes, and the first and second scores of each abnormal segment are added together to obtain the ranking score. Then, all ranking scores are sorted in descending order. This can more intuitively show the degree of influence of different abnormal segments. Then, by generating coordinate points and drawing score line graphs, the distribution of each ranked score is presented graphically, making it easier to observe the concentration of high scores. A reference line parallel to the horizontal axis is then drawn on the graph and gradually moved downwards along the vertical axis at a preset speed. This process gradually reveals the performance differences of different routes in the high-score range. By calculating the proportion of the line graph covering the area above the reference line, i.e., the screening ratio, the proportion of different routes in the high-influence range can be quantified. When the difference between the maximum and minimum screening ratios exceeds a threshold, it can be determined that some routes account for a large proportion in the high-risk segment, and these routes are thus preferentially eliminated. After this process, only one candidate route was ultimately selected as the route for the dump trucks. The above process not only considered the overall size of the evaluation score, but also further examined the strength of the route's spatial distribution. The aim was to avoid individual road sections causing serious pollution to the surrounding areas, thereby achieving a more environmentally friendly route selection for most areas. This is of great help in reducing the risk of dust spread and protecting the living environment of residents in the final solution.

[0030] It is worth noting that while the dump truck is traveling on its route, the monitoring video of the dump truck is collected in real time by the data collection equipment at the preset location until the dump truck reaches its destination, at which point the monitoring video is sent to the management personnel for archiving.

[0031] The dust and construction waste truck linkage monitoring system based on multi-source data fusion includes: Road segmentation module: Obtain all feasible routes between the origin and destination of the dump truck. For a single feasible route, divide it into several road segments according to the predetermined distance interval. The system collects the amount of dust generated within a given route. If the amount of dust exceeds a preset threshold, the route is marked as an abnormal route; otherwise, it is marked as a normal route. Impact Determination Module: Based on the location information, wind information and dust volume of the abnormal segment, obtain the impact score of the abnormal segment on the preset type of building, record it as the initial impact score, calculate the total value B1 of the initial impact score of a single abnormal segment, record the total value B1 as the first score, and the impact score reflects the impact of the dust volume of the abnormal segment on the preset type of building. Obtain the influence score of the abnormal segment on the midpoint of the normal segment, and record it as the secondary influence score. Calculate the total value B2 of the primary influence score of a single abnormal segment, and record the total value B2 as the second score. Selection module: The first score and the second score of all abnormal segments on a single feasible route are summed to obtain the evaluation score of the feasible route. The feasible route corresponding to the lowest evaluation score is taken as the driving route of the dump truck.

[0032] The foregoing has provided a detailed description of one embodiment of the present invention, but this description is merely a preferred embodiment and should not be construed as limiting the scope of the invention. All equivalent variations and modifications made within the scope of the present invention should still fall within the scope of the present invention.

Claims

1. A method for joint monitoring of dust and construction waste trucks based on multi-source data fusion, characterized in that, Includes the following steps: Obtain all feasible routes between the origin and destination of the dump truck. For a single feasible route, divide it into several road segments according to the predetermined distance interval. The system collects the amount of dust generated within a given route. If the amount of dust exceeds a preset threshold, the route is marked as an abnormal route; otherwise, it is marked as a normal route. Based on the location information, wind information and dust volume of the abnormal segment, the impact score of the abnormal segment on the preset type of building is obtained and recorded as the initial impact score. The total value B1 of the initial impact score of a single abnormal segment is calculated and recorded as the first score. The impact score reflects the impact of the dust volume of the abnormal segment on the preset type of building. Obtain the influence score of the abnormal segment on the midpoint of the normal segment, and record it as the secondary influence score. Calculate the total value B2 of the primary influence score of a single abnormal segment, and record the total value B2 as the second score. The evaluation score of a feasible route is obtained by summing the first and second scores of all abnormal segments on a single feasible route, and the feasible route corresponding to the lowest evaluation score is taken as the route for the dump truck. If two or more evaluation scores are the same and represent the minimum value, then perform the following steps: The feasible route with the same and smallest corresponding evaluation score is marked as the candidate route. The sum of the first and second scores of a single abnormal segment on the candidate route is used as the ranking score. The ranking scores are sorted in descending order to obtain the score ranking. Generate coordinate point (i, Di), where Di represents the i-th ranked score in the score sorting. Connect two adjacent coordinate points with a straight line to obtain a score line graph. Draw a straight line passing through (0, P) and parallel to the x-axis as a reference line, where P is a preset sorting score threshold, and slowly move the reference line downwards along the y-axis at a preset speed. For a single score line chart, the proportion of the domain of the part above the reference line to the domain of the score line chart is obtained in real time and recorded as the filtering ratio. When the difference between the largest and smallest screening ratios is greater than a preset difference threshold, the candidate curves corresponding to the largest screening ratio are removed until only one candidate route remains, and this route is marked as the driving route of the dump truck.

2. The method for joint monitoring of dust and construction waste trucks based on multi-source data fusion according to claim 1, characterized in that, The impact score of the abnormal segment on the preset type of building includes: Draw a circle with a preset radius centered at the midpoint of the abnormal segment, mark the buildings of a preset type inside the circle as target buildings, obtain the distance L between the target building and the midpoint of the abnormal segment, and obtain the wind speed and wind direction at the midpoint of the abnormal segment. Location information includes distance L, and wind information includes wind speed and direction; The dust volume, wind speed, wind direction, and distance L of the abnormal section are input as a real-time sample into a pre-trained artificial intelligence model to obtain the impact score of the abnormal section on the target building.

3. The method for joint monitoring of dust and construction waste trucks based on multi-source data fusion according to claim 2, characterized in that, The process of obtaining a pre-trained artificial intelligence model includes: Acquire real-time samples from historical monitoring processes, record them as historical samples, and label the impact scores of historical samples; An artificial intelligence model is built based on deep learning, and the model is trained and validated based on labeled historical samples to obtain a pre-trained artificial intelligence model.

4. The method for joint monitoring of dust and construction waste trucks based on multi-source data fusion according to claim 3, characterized in that, The process of labeling the impact scores of historical samples includes: For a single historical sample, the dust volume at the location of the corresponding target building A is periodically collected and the mean Y1 is calculated; Calculate the average dust emission Y2 at the location of target building A over the past n days; The ratio of Y1 to Y2 is used as the influence score. The larger the influence score, the more severe the impact of the dust in the abnormal section on the preset type of building.

5. The method for joint monitoring of dust and construction waste trucks based on multi-source data fusion according to claim 4, characterized in that, The process of obtaining the secondary influence score includes: The midpoint of the normal segment is used as the target building to obtain secondary influence scores.

6. The method for joint monitoring of dust and construction waste trucks based on multi-source data fusion according to claim 1, characterized in that, Also includes: As the dump truck travels along its route, monitoring videos of the truck are collected in real time by data acquisition devices at preset locations until the truck reaches its destination, at which point the monitoring videos are sent to management personnel for archiving.

7. The method for joint monitoring of dust and construction waste trucks based on multi-source data fusion according to claim 1, characterized in that, If a first score and / or a second score are greater than a preset score threshold, the corresponding abnormal segment will be reported to the preset management personnel, and the corresponding feasible route will not be used as the driving route.

8. A multi-source data fusion-based dust and construction waste truck joint monitoring system, including: Road segmentation module: Obtain all feasible routes between the origin and destination of the dump truck. For a single feasible route, divide it into several road segments according to the predetermined distance interval. The system collects the amount of dust generated within a given route. If the amount of dust exceeds a preset threshold, the route is marked as an abnormal route; otherwise, it is marked as a normal route. Impact Determination Module: Based on the location information, wind information and dust volume of the abnormal segment, obtain the impact score of the abnormal segment on the preset type of building, record it as the initial impact score, calculate the total value B1 of the initial impact score of a single abnormal segment, record the total value B1 as the first score, and the impact score reflects the impact of the dust volume of the abnormal segment on the preset type of building. Obtain the influence score of the abnormal segment on the midpoint of the normal segment, and record it as the secondary influence score. Calculate the total value B2 of the primary influence score of a single abnormal segment, and record the total value B2 as the second score. Selection module: The first score and the second score of all abnormal segments on a single feasible route are summed to obtain the evaluation score of the feasible route. The feasible route corresponding to the lowest evaluation score is taken as the driving route of the dump truck. If two or more evaluation scores are the same and represent the minimum value, then perform the following steps: The feasible route with the same and smallest corresponding evaluation score is marked as the candidate route. The sum of the first and second scores of a single abnormal segment on the candidate route is used as the ranking score. The ranking scores are sorted in descending order to obtain the score ranking. Generate coordinate point (i, Di), where Di represents the i-th ranked score in the score sorting. Connect two adjacent coordinate points with a straight line to obtain a score line graph. Draw a straight line passing through (0, P) and parallel to the x-axis as a reference line, where P is a preset sorting score threshold, and slowly move the reference line downwards along the y-axis at a preset speed. For a single score line chart, the proportion of the domain of the part above the reference line to the domain of the score line chart is obtained in real time and recorded as the filtering ratio. When the difference between the largest and smallest screening ratios is greater than a preset difference threshold, the candidate curves corresponding to the largest screening ratio are removed until only one candidate route remains, and this route is marked as the driving route of the dump truck.

Images