CN111340279A - Method for predicting accumulated content of corrosive chemical pollutants on surface of urban road - Google Patents

Abstract

The invention relates to a method for predicting road surface pollutants, in particular to a method for predicting the cumulative content of corrosive chemical pollutants on urban road surfaces. Model, set the input data as the multi-dimensional data of the study area, and the output data as the cumulative content of corrosive pollutants on the road surface; evaluation and feedback debugging, based on the cumulative content of the corrosive chemical pollutants on the road surface in each study area, the prediction of the construction The model conducts reliability evaluation and feedback debugging; makes predictions. According to the prediction model after reliability evaluation and feedback debugging, the cumulative content of corrosive chemical pollutants on road surfaces in the entire city can be predicted by focusing on corrosive pollution. Based on the analysis of the corrosion process of automobile products by substances, and based on the ambient air data, the cumulative content of corrosive chemical pollutants on the urban road surface is predicted.

Description

A prediction method for the cumulative content of corrosive chemical pollutants on urban road surfaces

technical field

The invention relates to a method for predicting road surface pollutants, in particular to a method for predicting the cumulative content of corrosive chemical pollutants on urban road surfaces.

Background technique

Air pollutants include sulfides, nitrides, chlorides, etc., which can be deposited on the road surface at a faster rate through the natural deposition process, especially during the rainfall process. When the surface humidity is high or it rains, it is easy to form a corrosive mixed medium, which can erode objects containing metal parts such as buildings and automotive products, resulting in a potential corrosion safety hazard.

The pollutants appear as road surface dust under low humidity conditions, and appear as a saturated pollutant dispersion system in the daily watering of roads. In this case, understanding the pollutant accumulation process is of great significance to accurately analyze the road surface pollutants and the corrosion of automotive products by ambient atmospheric conditions. In this context, in-depth investigation of the time distribution of the cumulative content of urban road corrosion pollutants, especially nitrogen oxides, sulfur oxides and other pollutants that have obvious corrosive effects on vehicles, will help to identify vehicle corrosion under urban road environmental conditions. Combined with air quality data, a prediction method for in-depth analysis of the cumulative content of road corrosion pollutants is constructed, so as to assist relevant management departments in the automotive industry to make effective anti-corrosion decisions.

At this stage, the technology on road surface pollutants focuses on the impact of road runoff on water pollution, and the corrosive impact of typical chemical pollutants on automotive products has not been carried out. Due to the obvious differences in the driving environment of automobiles in my country, the corrosion of various environmental factors to automobile products is also different. With the increase in the number of automobiles in my country, the environmental corrosion problem of automobile products is becoming more and more prominent. Urban road corrosion analysis method.

SUMMARY OF THE INVENTION

The purpose of the present invention is to provide a method for predicting the cumulative content of corrosive chemical pollutants on the surface of urban roads, which can analyze the corrosion process of automobile products by focusing on corrosive pollutants, and based on ambient atmospheric data, the corrosiveness of urban road surfaces can be analyzed. Prediction of cumulative levels of chemical pollutants.

The object of the present invention is achieved through the following technical solutions:

A method for predicting the cumulative content of corrosive chemical pollutants on urban road surfaces, comprising the following steps:

S1: Collect data, collect multi-dimensional data in the research area for building a prediction model;

S2: Build a prediction model, set the input data to be the multi-dimensional data of the study area, and the output data to be the cumulative content of corrosive pollutants on the road surface;

S3: Evaluation and feedback debugging, according to the cumulative content of the corrosive chemical pollutants on the road surface in each research area, the reliability evaluation and feedback debugging of the constructed prediction model;

S4: Predict, and predict the cumulative content of corrosive chemical pollutants on the road surface in the entire city according to the prediction model after reliability evaluation and feedback debugging.

As a further optimization of the technical solution, the present invention provides a method for predicting the cumulative content of corrosive chemical pollutants on the surface of urban roads. The step S1 includes the following steps:

S11: target data in the study area;

S12: Source data affecting target data within the study area.

As a further optimization of the technical solution, the present invention is a method for predicting the cumulative content of corrosive chemical pollutants on the surface of urban roads. The target data in the study area is the content data of corrosive pollutants on the road surface, and the factors affecting the target data in the study area are The source data are air pollutants, chemicals, and air quality data in the study area.

As a further optimization of the technical solution, the present invention provides a method for predicting the cumulative content of corrosive chemical pollutants on the surface of urban roads. The step S11 includes the following steps:

S111: Affect the medium requirements of the target data to fully cover the research area or roughly cover the functional area, and directional or non-stereotyped accumulation of medium mixtures in a small area, preferably, the medium is precipitation/snow, etc.;

S112: Divide the research area or functional area into grid points, where a single or a small amount of single source data affecting target data is located in the grid point, and a single functional area is located within the grid range.

S113: the collection time is the precipitation period, and the preferred sampling time is the monitoring of the depth of the water accumulation since the precipitation;

S114: Collect target data and source data in the research area;

S115: Acquire the total solid mass of each sample and each target data.

As a further optimization of the technical solution, the present invention provides a method for predicting the cumulative content of corrosive chemical pollutants on urban road surfaces, wherein the single functional area is an area capable of producing characteristic corrosive substances.

As a further optimization of the technical solution, the present invention provides a method for predicting the cumulative content of corrosive chemical pollutants on the surface of urban roads. The step S12 includes the following steps:

S121: target data research area division;

S212: Study area data collection affects the source of target data.

As a further optimization of the technical solution, the present invention provides a method for predicting the cumulative content of corrosive chemical pollutants on urban road surfaces. In step S2, Matlab is used to construct the prediction model, and a neural network model algorithm is used.

As a further optimization of the technical solution, the present invention is a method for predicting the cumulative content of corrosive chemical pollutants on the surface of urban roads. The prediction model includes an input layer, one or more hidden layers and an output layer; the input layer The input parameters of the multi-dimensional data in the study area, the number of hidden neurons in the hidden layer is five to ten, and the output layer is the weight of corrosive pollutants in the study area.

As a further optimization of the technical solution, the present invention provides a method for predicting the cumulative content of corrosive chemical pollutants on urban road surfaces. In the step S3, the actual measurement results in different time domains and different airspaces are used for verification and feedback.

As a further optimization of the present technical solution, the present invention provides a method for predicting the cumulative content of corrosive chemical pollutants on the surface of urban roads. In the step S4, a city or climate range is selected in a large range, and the attributes of the functional areas included in the prediction range are the same as The data collection stage is matched, the input data is the environmental data of other research areas in the city, and the output is the content of corrosive pollutants on the road surface.

The beneficial effects of the method for predicting the cumulative content of corrosive chemical pollutants on urban road surfaces of the present invention are:

The present invention is a method for predicting the cumulative content of corrosive chemical pollutants on the surface of urban roads, which can collect road surface water samples during the precipitation period, analyze the information of corrosive pollutants, and establish the correlation between atmospheric pollutants and road pollutants; Based on pollutant data, the concentration of road pollutants in other study areas can be predicted, and then the corrosive environmental conditions can be quantitatively analyzed.

Description of drawings

The present invention will be described in further detail below with reference to the accompanying drawings and specific implementation methods.

Fig. 1 is the flow chart of the method for predicting the cumulative content of corrosive chemical pollutants on the urban road surface of the present invention;

Fig. 2 is the data acquisition flow chart of the present invention;

Fig. 3 is the flow chart of the road surface corrosive pollutant content of the present invention;

Fig. 4 is a flow chart of constructing a prediction model of the present invention.

Detailed ways

The present invention will be further described in detail below in conjunction with the accompanying drawings.

Specific implementation one:

The present embodiment, a method for predicting the cumulative content of corrosive chemical pollutants on the surface of urban roads, will be described below with reference to FIGS. 1-4, including the following steps:

S1: Collect data, collect multi-dimensional data in the study area for building a prediction model; the multi-dimensional data includes the content of corrosive chemical pollutants on roads in the study area, and air quality data around roads; determine the study area The sample collection area and the accumulated content of corrosive chemical pollutants on the road surface of the sample collection area; the surrounding areas of the road include residential areas, commercial areas and industrial areas;

S2: Build a prediction model, set the input data to be the multi-dimensional data of the study area, and the output data to be the cumulative content of corrosive pollutants on the road surface;

S3: Evaluation and feedback debugging, according to the cumulative content of the corrosive chemical pollutants on the road surface in each research area, the reliability evaluation and feedback debugging of the constructed prediction model;

S4: Predict, and predict the cumulative content of corrosive chemical pollutants on the road surface in the entire city according to the prediction model after reliability evaluation and feedback debugging.

Specific implementation two:

The present embodiment will be described below with reference to FIGS. 1-4 , and the present embodiment will further describe the first embodiment. The step S1 includes the following steps:

S11: The content of corrosive pollutants on the road surface;

S12: Air quality data in the study area.

Specific implementation three:

The present embodiment will be described below with reference to FIGS. 1-4. This embodiment will further describe the second embodiment. The target data in the study area is the content data of corrosive pollutants on the road surface, and the source data affecting the target data in the study area is the study area. Regional air pollutants, chemicals, air quality data.

Specific implementation four:

The present embodiment will be described below with reference to FIGS. 1-4 , and the third embodiment will be further described in this embodiment. Step S11 includes the following steps:

S111: Affect the medium requirements of the target data to fully cover the research area or roughly cover the functional area, and directional or non-stereotyped accumulation of medium mixtures in a small range, preferably, the medium is precipitation/snow, etc., the cumulative precipitation of a single field is not Less than 1.3mm, not more than 2.5mm, road surface water area is not more than 2m, not less than 1m ² , the depth of water accumulation is not less than 2mm, not more than 10mm;

S112: Divide the research area or functional area into grid points, and the grid points are single or a small amount of single source data that affects the target data. Preferably, the periphery of the road is divided into square grids with a radius of 200m. is a single functional area;

S113: the collection time is the precipitation period, and the preferred sampling time is the monitoring of the depth of the water accumulation since the precipitation;

S114: Collect target data and source data in the study area. Preferably, the target data is a road surface water sample in the area. There are more than 5 road surface water areas in each study area, and no less than 1 sample is collected from a single water accumulation area. The collection volume is not less than 5ml, and the total number of collected samples is not less than 5; preferably, the source data period is from the generation of the medium in the previous stage to the generation of the medium in this stage.

S115: Obtain the total solid mass of each sample and each target data, preferably, obtain the concentration of corrosive pollutants, and obtain the cumulative content of the corrosive chemical pollutants on the road surface.

Specific implementation five:

The present embodiment will be described below with reference to FIGS. 1-4 . This embodiment further describes the embodiment in four steps. The single functional area is an area capable of producing characteristic corrosive substances, such as a residential area, an industrial area, and a commercial area.

Specific implementation six:

This embodiment is described below with reference to FIGS. 1-4. This embodiment further describes the second embodiment in five steps. The step S12 includes the following steps:

S121: Divide the target data research area, divide the research area according to source data or other impact target data, preferably, divide the city according to latitude and longitude or administrative area, and divide the city into no less than 2000 research areas, or each not less than 3 study areas in each administrative region;

S212: Data collection in the study area affects the source of the target data. Preferably, the data includes data such as temperature, humidity, sulfide, nitride, CO, PM _2.5 , PM ₁₀ , AQI, etc. The data dimension is preferably hourly and daily average.

Specific implementation seven:

The present embodiment will be described below with reference to FIGS. 1-4. This embodiment will further describe the sixth embodiment. In the step S2, the prediction model is constructed by using Matlab, and a neural network model algorithm is used; Matlab is called MatrixLaboratory, which is introduced by MathWorks. Commercial mathematical software for high-level technical computing languages and interactive environments for algorithm development, data visualization, data analysis, and numerical computing.

Eighth specific implementation:

This embodiment is described below with reference to FIGS. 1-4. This embodiment further describes Embodiment 7. The prediction model includes an input layer, one or more hidden layers and an output layer; the input parameters of the input layer are: For multidimensional data in the study area, the number of hidden neurons in the hidden layer is preferably five to ten, and the output layer is the weight of corrosive pollutants in the study area.

Specific implementation nine:

The present embodiment will be described below with reference to FIGS. 1-4 , and the present embodiment will further describe the eighth embodiment. In step S3 , verification and feedback are performed through actual measurement results in different time domains and different air domains.

Specific implementation ten:

This embodiment is described below with reference to FIGS. 1-4. This embodiment further describes the ninth embodiment. In the step S4, a city or climate range is selected in a large range, and the attributes of the functional area included in the prediction range are matched with the data collection stage. , the input data is the environmental data of other research areas in the city, and the output is the content of corrosive pollutants on the road surface.

Divide a city into 2700 grid points according to latitude and longitude, and collect pollutant data on 30-50 grid points. In the grid point, select a 200*200㎡ area near the road, a certain precipitation, and the precipitation in the area of grid point A is 4mm. Since the precipitation, the road surface has accumulated water. When the accumulated water area on both sides of the road is about ^2m2 , and the accumulated water depth is about 1mm, the accumulated water samples are collected and their composition is detected. The data structure is as follows:

A grid point pollutant information data

The neural network model is used, the input layer is the air pollutant data, and the output layer is the water pollutant information. 80% of the grid data is selected to establish a prediction model, and 20% of them are selected for debugging and learning. The prediction model error is less than 10%. Use this to predict other areas of a city.

Of course, the above description does not limit the present invention, and the present invention is not limited to the above examples. Changes, modifications, additions or substitutions made by those of ordinary skill in the art within the essential scope of the present invention also belong to the present invention. protected range.

Claims (10)

1. A prediction method for the accumulated content of corrosive chemical pollutants on the surface of an urban road is characterized by comprising the following steps: the method comprises the following steps:

s1: collecting data, namely collecting multidimensional data used for constructing a prediction model in a research area;

s2: constructing a prediction model, setting input data as multi-dimensional data of a research area, and outputting data as the accumulated content of the corrosive pollutants on the road surface;

s3: evaluating and feedback debugging, namely evaluating the reliability and performing feedback debugging on the constructed prediction model according to the accumulated content of the corrosive chemical pollutants on the road surface in each research area;

s4: and predicting, namely predicting the accumulated content of the corrosive chemical pollutants on the road surface in the whole city range according to the prediction model after reliability evaluation and feedback debugging.

2. The method for predicting the cumulative content of corrosive chemical pollutants on the surface of urban roads according to claim 1, wherein the method comprises the following steps: step S1 includes the following steps:

s11: target data within the study area;

s12: source data within the study area that affects the target data.

3. The method for predicting the cumulative content of corrosive chemical pollutants on the surface of urban roads according to claim 2, wherein the method comprises the following steps: the target data in the research area is data of corrosive pollutant content on the surface of the road, and the source data influencing the target data in the research area is data of atmospheric pollutants, chemicals and air quality in the research area.

4. The method for predicting the cumulative content of corrosive chemical pollutants on the surface of urban roads according to claim 2, wherein the method comprises the following steps: step S11 includes the following steps:

s111: (ii) media requirements affecting target data to cover the study area or substantially the functional area in general and to produce a media mixture accumulation within a small range, oriented or amorphous;

s112: and carrying out lattice point division on the research area or the functional area, wherein single or a small amount of single source data influencing target data are arranged in the lattice points, and a single functional area is arranged in the grid range.

S113: the collection time is the precipitation period, and the preferable sampling time is the monitoring of the depth of the accumulated water from the beginning of precipitation;

s114: collecting target data and source data in a research area;

s115: and acquiring the total solid mass and each target data of each sample.

5. The method for predicting the cumulative content of corrosive chemical pollutants on the surface of urban roads according to claim 4, wherein the method comprises the following steps: the single functional region is a region capable of producing a characteristic corrosive substance.

6. The method for predicting the cumulative content of corrosive chemical pollutants on the surface of urban roads according to claim 2, wherein the method comprises the following steps: step S12 includes the following steps:

s121: dividing a research area of the target data, namely dividing the research area according to source data or other influence target data;

s212: the study area data is used for acquiring environmental data.

7. The method for predicting the cumulative content of corrosive chemical pollutants on urban road surfaces according to any one of claims 1 to 6, wherein: in step S2, the prediction model is constructed by Matlab, and a neural network model algorithm is used.

8. The method for predicting the cumulative content of corrosive chemical pollutants on the surface of urban roads according to claim 7, wherein the method comprises the following steps: the prediction model comprises an input layer, one or more hidden layers and an output layer; the input parameters of the input layer are multidimensional data in the research area, the number of hidden neurons of the hidden layer is five to ten, and the output layer is the weight of corrosive pollutants in the research area.

9. The method for predicting the cumulative content of corrosive chemical pollutants on urban road surfaces according to any one of claims 1, 2, 3, 4, 5, 6 and 8, wherein the method comprises the following steps: in step S3, verification and feedback are performed according to the actual measurement results of different time domains and different spatial domains.

10. The method for predicting the cumulative content of corrosive chemical pollutants on urban road surfaces according to any one of claims 1, 2, 3, 4, 5, 6 and 8, wherein the method comprises the following steps: in step S4, a city or climate range is selected in a large range, the attribute of a functional area contained in the prediction range is matched with the data acquisition stage, the input data is environmental data of other research areas of the city, and the output is the content of corrosive pollutants on the surface of the road.

Images