CN-122015944-A - Rapid positioning and responding system for pollution source of bulk cargo wharf based on meteorological sensing network

CN122015944ACN 122015944 ACN122015944 ACN 122015944ACN-122015944-A

Abstract

The system for rapidly positioning and responding the pollution source of the bulk cargo wharf based on the meteorological sensing network comprises a data acquisition module, a data transmission module, a pollution source positioning analysis module, a response control module and a visual display module. The method integrates the multisource sensing equipment, realizes the comprehensive collection of three-dimensional space data of atmospheric particulates and various pollutant concentration meters, can comprehensively utilize meteorological data and pollutant concentration data by integrating a physical diffusion model and a machine learning algorithm, realizes the quick tracing and probability positioning of a pollution source, overcomes the limitation of a traditional single model in a complex environment, and obviously improves the accuracy, certainty and robustness of positioning.

Inventors

LV CHONGXIAO
JIANG YULONG
JI CHUANHUI
WANG YUMING
WANG KANGNING
YANG JINKAI

Assignees

中交机电工程局有限公司

Dates

Publication Date: 20260512
Application Date: 20251230

Claims (9)

1. Quick positioning and response system of bulk cargo wharf pollution source based on meteorological sensing network, its characterized in that includes: The data acquisition module is used for acquiring meteorological data and pollutant concentration data through a preset meteorological sensor network, wherein the meteorological data at least comprise wind speed, wind direction, temperature, humidity and atmospheric pressure; The data transmission module is used for formatting the meteorological data and the pollutant concentration data into a unified data format according to the preset format requirement and storing the unified data format through a cloud or local database; The pollution source positioning analysis module acquires unified data from the database and performs pollution source positioning and diffusion path simulation in a mode of combining a physical diffusion model with a mechanical learning algorithm; The response control module generates a regulation and control instruction according to the pollution source positioning result and triggers early warning or executes pollution control operation; And the visual display module displays the pollution source position, the diffusion path, the early warning information and the regulation result on an interface in real time.
2. The rapid positioning and responding system of bulk cargo terminal pollution source based on meteorological sensing network according to claim 1, wherein the data acquisition module comprises: The laser scanning radar unit is used for acquiring atmospheric particulate matter data and space-time distribution characteristics thereof; The air quality monitoring unit is used for collecting concentration data of one or more pollutants of PM2.5, PM10, SO2, NOx and O3; The meteorological sensing unit is composed of one or more integrated meteorological stations and is used for acquiring wind speed, wind direction, temperature, humidity and air pressure data by using an ultrasonic anemometer, a temperature sensor, a relative humidity sensor and an atmospheric pressure sensor; And the unmanned aerial vehicle remote sensing unit is used for collecting three-dimensional space data when the pollution abnormality is detected.
3. The rapid positioning and responding system of pollution source of bulk cargo wharf based on meteorological sensing network as claimed in claim 1, wherein said pollution source positioning analysis module comprises: a physical diffusion model unit for simulating a pollutant transmission path based on a Gaussian diffusion model or a reverse diffusion model; the mechanical learning tracing unit is used for carrying out pollution source contribution analysis and probability positioning through a space-time diagram neural network or a Bayesian network; and the dynamic path optimizing unit corrects the pollution diffusion path by utilizing the real-time meteorological data to generate accurate pollution source positioning information.
4. The rapid positioning and responding system of pollution source of bulk cargo wharf based on meteorological sensing network according to claim 3, wherein the pollution source positioning and analyzing module further comprises: establishing a linear relation model based on the pollutant concentration matrix and the source configuration matrix, and calculating a pollution source contribution value through a minimized error matrix; and the diffusion model parameters are corrected by combining the real-time meteorological data and the topographic data, so that the source positioning accuracy is improved.
5. The rapid positioning and responding system of bulk cargo terminal pollution source based on meteorological sensing network according to claim 1, wherein the responding control module comprises: The early warning triggering unit is used for generating a pollution abnormal signal when the concentration of the pollutants exceeds a preset threshold value; the instruction generating unit generates a regulation and control instruction according to the pollution source positioning result, wherein the regulation and control instruction comprises a limit production instruction, a stop production instruction or a watering dust fall instruction; And the feedback execution unit is used for controlling the sprinkler, the spraying device or the unmanned aerial vehicle to carry out dust fall or pollution interception operation.
6. The rapid positioning and responding system of bulk cargo terminal pollution sources based on meteorological sensing network according to claim 1, wherein the visual display module supports at least one of the following display modes: a thermodynamic diagram of the spatial distribution of contaminant concentration; Pollution source position marks and diffusion path animations; Real-time early warning prompt and historical data query interface; And regulating and controlling the instruction execution state and the effect evaluation icon.
7. A method for rapidly positioning and responding a pollution source of a bulk cargo wharf based on a meteorological sensing network is characterized by comprising the following steps: step S1, acquiring meteorological data and pollutant concentration data of a bulk cargo wharf area through a meteorological sensor network; Step S2, formatting the acquired data into a unified format, and transmitting and storing the unified format in a local database through a cloud or a wired network; s3, reading normalized pollution data from a database, performing traceability analysis on pollutants based on a physical model and a machine learning algorithm, and simulating a diffusion path; S4, generating a pollution regulation and control instruction according to a pollution source positioning result, and executing response operation; and S5, visually displaying the positions of the pollution sources, the diffusion paths, the early warning information and the regulation and control results.
8. The method for rapidly positioning and responding to the pollution source of the bulk cargo wharf based on the meteorological sensing network according to claim 7, wherein the step S2 specifically comprises the following steps: step P1, data standardization: Uniformly converting multi-source data from a laser scanning radar, an air quality monitoring unit, a meteorological sensing unit and an unmanned aerial vehicle remote sensing unit into a preset standardized data format, wherein the standardized data format comprises a data acquisition time stamp, equipment geographic coordinates, a sensor type, a numerical value and a unit field; Step P2, data quality control: Preprocessing standardized data, wherein the preprocessing comprises the steps of identifying and removing abnormal values by adopting a Z-score method or a quarter bit distance method, filling the missing data by adopting time sequence interpolation or correlation analysis based on meteorological elements, smoothing and filtering the data to reduce random noise interference; step P3, data storage and indexing: And establishing a joint index taking the timestamp and the space coordinate as cores in the database so as to support the high-efficiency space-time query of a follow-up traceability analysis module.
9. The method for rapidly positioning and responding to the pollution source of the bulk cargo wharf based on the meteorological sensing network according to claim 8, wherein the step S3 specifically comprises the following steps: t1, data reading and feature extraction, namely acquiring formatted pollutant detection data and relevant meteorological parameters from a database, wherein the specific steps are as follows: Inputting data, namely, pollutant concentration time sequence data, wind speed, wind direction, temperature, air pressure, wharf topography, building layout and other static data of each monitoring point; Extracting statistical characteristics of pollutant concentration, such as mean value, peak value and gradient change; converting wind speed and wind direction into wind vectors (u, v); Carrying out spatial association on monitoring points and potential pollution sources, such as loading and unloading machines, storage yards and conveyor belts, by combining Geographic Information System (GIS) data; and T2, simulating physical diffusion, namely simulating the diffusion behavior of the pollutant by using a physical law, and providing physical constraint and priori knowledge for a machine learning model, wherein the method specifically comprises the following steps of: The core model adopts a Gaussian plume model or a Gaussian smoke mass model, and for a continuous point source, the Gaussian plume model has the following formula: wherein The method comprises the steps of measuring the concentration of pollutants at the position, which is x meters away from an origin, y meters away from a cross wind direction and z meters away from the height, of the downwind direction, wherein Q is the discharge speed of a pollution source, namely the source intensity, u is the average wind speed, and He is the effective discharge height of the pollution source; , the diffusion parameters in the horizontal and vertical directions are functions of the atmospheric stability and the downwind direction distance x; concentration at known downstream multiple monitoring points In the case of (2) the most likely source location is back-calculated by least squares or optimization algorithms And source strength Q; And T3, analyzing the contribution degree of the pollution source and positioning the probability based on machine learning, namely learning a pollution propagation space-time mode in a complex environment by using a data-driven method, and making up the defects of a pure physical model when dealing with non-ideal conditions, such as a complex wind field and building turbulence: and carrying out pollution source contribution analysis and probability positioning by adopting a space-time diagram neural network STGNN or a Bayesian network: When the space-time diagram neural network STGNN is adopted to construct a diagram structure, each monitoring sensor is used as a node of the diagram, and the space-time diagram neural network STGNN captures the spatial dependence and the time evolution rule of the node characteristics at the same time, so that the most probable pollution source upstream of which node is deduced and the source contribution degree is given; The position and the intensity of a pollution source are regarded as random variables when Bayesian reasoning is adopted, and the posterior probability is calculated based on a Bayesian formula: wherein Is posterior probability, i.e. the probability of the pollution source under the condition of known observation data is wanted; the probability of obtaining current observation data under the condition of known pollution sources is represented as a likelihood function; Is a priori probability representing a priori knowledge of the probability of a source of contamination before there is no data; the evidence is edge likelihood and represents the average probability of observed data under all possible pollution sources; T4, model fusion and dynamic path optimization, wherein the results of the physical model and the machine learning model are fused, and dynamic correction is carried out by utilizing real-time data to obtain a final positioning result and a diffusion path; Establishing a linear system to quantify contributions of a plurality of potential pollution sources to each monitoring point; Wherein C is a vector of m×1 representing the pollutant concentration measurement values of m monitoring points, S is a vector of n×1 representing the emission intensity of n potential pollutant sources, i.e. the source contribution value to be solved, A is a source configuration matrix of m×n, matrix elements The concentration contribution coefficient of the jth pollution source to the ith monitoring point is represented, and the concentration contribution coefficient is usually calculated by a physical diffusion model and comprises wind speed, wind direction, distance and atmospheric stability information, wherein E is an m multiplied by 1 error vector; solving the source contribution S by minimizing the error matrix E, using the least squares method: The solution is as follows: And then dynamically updating parameters in the source configuration matrix A by utilizing the real-time meteorological data and the topographic data, thereby realizing the real-time correction of the diffusion path and the original positioning result and improving the positioning accuracy.

Description

Rapid positioning and responding system for pollution source of bulk cargo wharf based on meteorological sensing network Technical Field The invention relates to the technical field of pollution positioning, in particular to a bulk cargo wharf pollution source rapid positioning and response system based on a meteorological sensing network. Background Bulk cargo wharf can produce a large amount of dust in the processes of loading, unloading, transporting and piling up materials such as coal, ore, and the like, meanwhile, the transportation in the port can also produce various gaseous pollutants, serious threat is formed to the surrounding environment of the port and the health of staff, limited ground fixed monitoring stations are commonly adopted for monitoring port pollution, and the remote sensing data of meteorological satellites are combined to carry out macroscopic analysis by constructing a pollution distribution map. For pollution source tracing, a numerical simulation method based on a Gaussian diffusion model or a Lagrange particle model is generally adopted, and reverse deduction is carried out by combining meteorological data. However, these conventional methods have obvious limitations, such as combining the patent application CN117092300a with an air quality monitor through a laser radar, analyzing the pollutant transmission diffusion, chemical reaction and characteristic source spectrum, mainly focusing on monitoring and post analysis, being very deficient in terms of quick response and active control of pollution, and focusing on a single monitoring and tracing link, failing to form a complete closed-loop automatic response system for monitoring-tracing-early warning-control, so that the tracing result cannot be quickly converted into a practical and effective pollution-reducing measure. Disclosure of Invention The invention aims to solve the defects of the prior art and provides a rapid positioning and responding system for a pollution source of a bulk cargo wharf based on a meteorological sensing network. The invention adopts the following technical scheme to realize the aim: bulk cargo wharf pollution source rapid positioning and response system based on meteorological sensing network includes: The data acquisition module is used for acquiring meteorological data and pollutant concentration data through a preset meteorological sensor network, wherein the meteorological data at least comprise wind speed, wind direction, temperature, humidity and atmospheric pressure; The data transmission module is used for formatting the meteorological data and the pollutant concentration data into a unified data format according to the preset format requirement and storing the unified data format through a cloud or local database; The pollution source positioning analysis module acquires unified data from the database and performs pollution source positioning and diffusion path simulation in a mode of combining a physical diffusion model with a mechanical learning algorithm; The response control module generates a regulation and control instruction according to the pollution source positioning result and triggers early warning or executes pollution control operation; And the visual display module displays the pollution source position, the diffusion path, the early warning information and the regulation result on an interface in real time. The data acquisition module comprises: The laser scanning radar unit is used for acquiring atmospheric particulate matter data and space-time distribution characteristics thereof; The air quality monitoring unit is used for collecting concentration data of one or more pollutants of PM2.5, PM10, SO2, NOx and O3; The meteorological sensing unit is composed of one or more integrated meteorological stations and is used for acquiring wind speed, wind direction, temperature, humidity and air pressure data by using an ultrasonic anemometer, a temperature sensor, a relative humidity sensor and an atmospheric pressure sensor; And the unmanned aerial vehicle remote sensing unit is used for collecting three-dimensional space data when the pollution abnormality is detected. The pollution source positioning analysis module comprises: a physical diffusion model unit for simulating a pollutant transmission path based on a Gaussian diffusion model or a reverse diffusion model; the mechanical learning tracing unit is used for carrying out pollution source contribution analysis and probability positioning through a space-time diagram neural network or a Bayesian network; and the dynamic path optimizing unit corrects the pollution diffusion path by utilizing the real-time meteorological data to generate accurate pollution source positioning information. The pollution source positioning analysis module further comprises: establishing a linear relation model based on the pollutant concentration matrix and the source configuration matrix, and calculating a pollution source contribution value