Intelligent dedusting environmental impact assessment method based on big data analysis

By using big data analysis and counterfactual prediction models, the problem of resource mismatch in dust removal equipment in traditional methods has been solved, enabling accurate quantification and interpretable assessment of the environmental impact of dust removal measures, and improving the scientific nature and precision of environmental governance.

CN121328928APending Publication Date: 2026-01-13SHANGHAI LIJIANG INFORMATION TECHNOLOGY CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202511500986.3
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2025-10-21
Publication Date
2026-01-13

AI Technical Summary

Technical Problem

Traditional manual sampling and static models are insufficient to capture the dynamic diffusion patterns of pollution sources, leading to a misallocation of dust removal equipment resources and an inability to scientifically and accurately quantify the environmental improvement contribution of dust removal measures.

Method used

An intelligent dust removal environmental impact assessment method based on big data analysis is adopted. Through multi-source data collection, preprocessing, feature extraction and counterfactual prediction model, an environmental impact assessment model is established to quantify the contribution of dust removal measures and display the assessment results through charts.

Benefits of technology

It enables the scientific and precise quantification of the environmental improvement contribution of dust removal measures, provides a refined and intelligent environmental governance tool, avoids the impact of natural weather changes, and improves the scientificity and accuracy of the assessment.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN121328928A_ABST
    Figure CN121328928A_ABST
Patent Text Reader

Abstract

The invention belongs to the technical field of cup holders, and discloses a big data analysis-based intelligent dust removal environmental impact assessment method, which comprises the following steps of: acquiring multi-source data, and integrating the multi-source data into a unified data model by using a space-time alignment technology to obtain a comprehensive data set of environment and dust removal system operation; performing data preprocessing on the multi-source data in the comprehensive data set, and performing feature extraction to obtain feature data; establishing an environmental impact assessment model based on the feature data; utilizing the trained model to evaluate the influence of the dust removal measures on the environment; the evaluation result is displayed in a chart form; the time sequence diagram shows the change of the environment indexes before and after the dedusting measures are implemented. According to the invention, through big data fusion and an advanced causal inference model, the defect that the influence of natural meteorological change cannot be eliminated by a traditional front-back comparison method is overcome, and the real contribution of dust removal measures to environment improvement can be quantified more scientifically and accurately.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention relates to the field of environmental impact assessment technology, and in particular to an intelligent dust removal environmental impact assessment method based on big data analysis. Background Technology

[0002] Environmental impact assessment is a system for controlling environmental impacts, aiming to reduce pollution caused by project development and maintain human health and ecological balance. Traditional methods relying on manual sampling and static models are difficult to capture the dynamic diffusion patterns of pollution sources (such as the instantaneous migration of PM2.5 under meteorological influences), and experience-based decision-making has limitations; manual scheduling of dust removal equipment often leads to resource misallocation (such as fog cannons operating ineffectively in low-pollution areas). Summary of the Invention

[0003] To overcome the technical defects of existing technologies, this invention provides an intelligent dust removal environmental impact assessment method based on big data analysis. A magnetic attraction mechanism is set up to position the cup and the supporting truncated cone. In this way, when external forces such as wind blow, the cup will not shake or collide with each other, thus achieving better results.

[0004] The technical solution adopted in this invention is: an intelligent dust removal environmental impact assessment method based on big data analysis, comprising the following steps:

[0005] Step 1: Collect multi-source data and use spatiotemporal alignment technology to integrate the multi-source data into a unified data model to obtain a comprehensive dataset of the operation of the environment and dust removal system.

[0006] Step 2: Perform data preprocessing and feature extraction on the multi-source data in the comprehensive dataset to obtain feature data;

[0007] Step 3: Establish an environmental impact assessment model based on feature data; use the trained model to assess the environmental impact of dust removal measures; keep other variables constant, change the operating parameters of the dust removal equipment (such as dust removal efficiency), predict changes in environmental indicators, and calculate the contribution of dust removal measures to environmental indicators.

[0008] Step 4: Present the assessment results in chart form; time series graph shows the changes in environmental indicators before and after the implementation of dust removal measures; spatial distribution map shows the improvement of environmental indicators in different areas; contribution pie chart shows the contribution ratio of each factor to the environmental impact.

[0009] Preferably, the method for multi-source data acquisition involves deploying an Internet of Things (IoT) sensor network to collect real-time data on the energy consumption, dust removal efficiency, and operating status of the dust removal equipment, as well as environmental data such as PM2.5, PM10, temperature, and humidity. In addition, external data sources are accessed, including meteorological data such as wind speed, wind direction, air pressure, and precipitation; geographic information data such as topography and land use type; and socioeconomic data such as population density and industrial distribution.

[0010] Preferably, the method for performing data preprocessing and feature engineering on multi-source data in the comprehensive dataset to obtain feature data involves cleaning missing values, outliers, and duplicate values, and standardizing the data to convert data of different dimensions into a unified standard.

[0011] Preferably, the method for extracting features to obtain feature data is to extract features from preprocessed data, including time features (hour, day, month, season); spatial features (regional division, latitude and longitude); meteorological features (wind speed, wind direction, etc., converted into features that affect pollutant diffusion); dust removal equipment features (dust removal efficiency, operating time, processing air volume); and creating lagged values ​​of all feature variables and target variables from the past T time steps as features to form lagged features.

[0012] Preferably, the method for establishing an environmental impact assessment model based on feature data involves constructing a "counterfactual" prediction model. Specifically, this involves selecting historical data from periods when no strong dust control measures were implemented, or selecting data from a control group area where no dust control measures were implemented, as training data. The model uses meteorological characteristics, pollution source characteristics, and spatiotemporal characteristics, but excludes dust control activity characteristics. Environmental quality indicators are then used as the model output. The model formula is:

[0013] In the above formula , is the model-predicted "counterfactual" PM2.5 concentration at time t for grid i. It is the input feature vector. It is the space of the return tree. It refers to the number of trees;

[0014] The goal of this model is to minimize the loss function. .

[0015] Then, during the evaluation period after the dust control measures were implemented, we compared the "counterfactual" concentrations predicted by the model with the actual observed concentrations to conduct an environmental impact assessment and quantification.

[0016] Preferably, the method for conducting environmental impact assessment and quantification is as follows:

[0017] Absolute concentration reduction (ACA), ,

[0018] in This is the actual observed concentration. The higher this value, the better the dust removal effect.

[0019] Relative Improvement Rate (RIR) ;

[0020] Health Benefit Assessment (HBA) .

[0021] Preferably, contribution decomposition is performed, and the SHAP model interpretation tool is used to analyze the contribution of each feature in the "counterfactual" prediction.

[0022] Preferably, dust removal strategy simulation is performed, and by adjusting the input DII (Dust Intensity Index), the future environmental improvement effect is predicted, providing decision-makers with "if-then" scenario analysis; based on the predicted high-pollution periods and areas, combined with meteorological conditions, the optimal dust removal operation route and timing are recommended.

[0023] For the preferred option, conduct a cost-benefit analysis.

[0024] Benefit-cost analysis formula:

[0025] .

[0026] The beneficial effects of this invention are: 1. This method, through big data fusion and advanced causal inference models, overcomes the shortcomings of the traditional "before-and-after comparison method" in excluding the influence of natural weather changes, and can more scientifically and accurately quantify the true contribution of dust removal measures to environmental improvement. It elevates the assessment from "descriptive" analysis to "interpretive" and "predictive" analysis, providing a powerful tool for the refinement and intelligence of environmental governance. Attached Figure Description

[0027] One or more embodiments are illustrated by way of example with reference numerals in the accompanying drawings. These illustrations do not constitute a limitation on the embodiments. Elements with the same reference numerals in the drawings are denoted as similar elements. Unless otherwise stated, the figures in the drawings are not to be limited by scale.

[0028] Figure 1 This is a schematic diagram of the steps of the present invention. Detailed Implementation

[0029] To make the objectives, technical solutions, and advantages of this invention clearer, the various embodiments of this invention will be described in detail below with reference to the accompanying drawings. However, those skilled in the art will understand that many technical details have been provided in the various embodiments of this invention to facilitate a better understanding of this application. However, the technical solutions claimed in the claims of this application can be implemented even without these technical details and with various variations and modifications based on the following embodiments.

[0030] like Figure 1 As shown in the figure, this embodiment provides an intelligent dust removal environmental impact assessment method based on big data analysis, including the following steps:

[0031] Step 1: Collect multi-source data and use spatiotemporal alignment technology to integrate the multi-source data into a unified data model to obtain a comprehensive dataset of the operation of the environment and dust removal system.

[0032] Step 2: Perform data preprocessing and feature extraction on the multi-source data in the comprehensive dataset to obtain feature data;

[0033] Step 3: Establish an environmental impact assessment model based on feature data; use the trained model to assess the impact of dust removal measures on the environment; keep other variables constant, change the operating parameters of the dust removal equipment (such as dust removal efficiency), and predict changes in environmental indicators; calculate the contribution of dust removal measures to environmental indicators; identify key factors such as dust concentration, humidity, and wind speed by ranking the importance of features in a random forest, and eliminate redundant variables (such as irrelevant temperature fluctuations) to improve the model's R² value to above 0.92.

[0034] Dynamic evaluation phase: Precise attribution under controlled experiments.

[0035] Variable isolation verification: Fix the interference factors such as weather and traffic, and only adjust the dust removal efficiency (e.g., from 80% to 90%). The PM10 concentration reduction curve is output by the model to directly quantify the contribution rate of dust removal measures (e.g., a 10% increase in efficiency can reduce the PM10 peak by 22% ± 3%).

[0036] Multi-scenario simulation: By inputting different combinations of operating parameters for dust removal equipment (such as spray intensity + fan speed), a Monte Carlo simulation is used to generate a heatmap of environmental improvement effects with a 99% confidence interval, assisting in the selection of the optimal solution. This invention, through the technical path of "feature screening → parameter controllable experiment → contribution decoupling," achieves interpretable, quantifiable, and reproducible evaluation of the effects of environmental governance measures, providing a core tool for refined environmental management.

[0037] Step 4: Present the assessment results in chart form; time series graph shows the changes in environmental indicators before and after the implementation of dust removal measures; spatial distribution map shows the improvement of environmental indicators in different areas; contribution pie chart shows the contribution ratio of each factor to the environmental impact.

[0038] The method for acquiring multi-source data involves deploying an Internet of Things (IoT) sensor network to collect real-time data on the energy consumption, dust removal efficiency, and operating status of dust removal equipment, as well as environmental data such as PM2.5, PM10, temperature, and humidity. In addition, external data sources are accessed, including meteorological data such as wind speed, wind direction, air pressure, and precipitation; geographic information data such as topography and land use type; and socioeconomic data such as population density and industrial distribution.

[0039] The method described above for preprocessing and feature engineering multi-source data in a comprehensive dataset to obtain feature data involves data cleaning to remove missing, outlier, and duplicate values, and data standardization to convert data of different dimensions into a unified standard. This avoids model bias caused by missing data, improves the reliability of subsequent feature engineering, eliminates interference from sensor malfunctions and manual data entry errors, and enhances data quality; it also eliminates redundant calculations and reduces model training resource consumption. Multi-source data can be transformed into a high-quality feature dataset, providing a stable and interpretable data foundation for subsequent modeling, and is particularly suitable for the prediction and analysis of complex environmental indicators in dust removal and environmental protection scenarios.

[0040] The method for extracting features and obtaining feature data involves extracting features from preprocessed data, including time features (hour, day, month, season), spatial features (regional division, latitude and longitude), meteorological features (wind speed, wind direction, etc., converted into features that affect pollutant diffusion), and dust removal equipment features (dust removal efficiency, operating time, and processing air volume). The lagged values ​​of all feature variables and the target variable from the past T time steps are created as features to form lagged features.

[0041] The method for building an environmental impact assessment model based on feature data involves constructing a "counterfactual" prediction model. Specifically, this involves selecting historical data from periods without strong dust control measures, or data from a control group area without dust control measures, as training data. The model uses meteorological characteristics, pollution source characteristics, and spatiotemporal characteristics, but excludes dust control activity characteristics. Environmental quality indicators are then used as the model output. This avoids "data leakage" and ensures unbiased assessment. The core advantage is that this is the most important reason. If "dust control activity characteristics" (such as "spray water volume") are included in the model training, the model will learn to rely on this strong feature to predict pollutant concentrations. It will discover that "as long as the spray is on, the concentration is low." When using this model to predict the future, it essentially repeats what we already know: "Because we sprayed water, the concentration should be low." This completely fails to assess the effectiveness of the measures, creating a logical loop. By excluding these features, we force the model to learn the generation and diffusion patterns of pollutants only from natural and production factors. Thus, when we input the natural and production conditions after the implementation of measures into the model, the "counterfactual" prediction it provides is a truly independent and reliable frame of reference.

[0042] In the real world, changes in environmental quality are the result of multiple factors working together. For example, it's possible that after measures are implemented, wind speeds are high, leading to favorable natural dispersion conditions. Without constructing a counterfactual model, we might overestimate the effectiveness of dust control measures. This model, by fixing (or "controlling") the confounding variables of meteorology and production activities, perfectly isolates the influence of the single variable "dust control measures," achieving causal inference of the effects.

[0043] The final result can be presented very intuitively as follows: "Look, under exactly the same weather and production intensity, without our dust removal system, the concentration should have reached the red alert level (counterfactual prediction), but in reality, through our efforts, we controlled it at a good level (actual observed value). The essence of constructing this 'counterfactual' prediction model that excludes the characteristics of dust removal activities is to create a scientifically credible 'parallel universe'—in this universe, all natural conditions and production activities remain the same, except that dust removal measures are not implemented. This is achieved by comparing the pollution level of this 'parallel universe' with the measured levels in our real world."

[0044] The model formula is:

[0045] In the above formula , is the model-predicted "counterfactual" PM2.5 concentration at time t for grid i. It is the input feature vector. It is the space of the return tree. It refers to the number of trees;

[0046] The goal of this model is to minimize the loss function. ;

[0047] Then, during the evaluation period after the dust control measures were implemented, we compared the "counterfactual" concentrations predicted by the model with the actual observed concentrations to conduct an environmental impact assessment and quantification.

[0048] The method described above for conducting environmental impact assessment and quantification is...

[0049] Absolute concentration reduction (ACA), ,

[0050] in This is the actual observed concentration. The higher this value, the better the dust removal effect.

[0051] Relative Improvement Rate (RIR) ;

[0052] Health Benefit Assessment (HBA) .

[0053] We decompose the contribution and use the SHAP model interpretation tool to analyze the contribution of each feature in the counterfactual prediction.

[0054] The system simulates dust removal strategies and predicts future environmental improvement effects by adjusting the input DII (Dust Intensity Index), providing decision-makers with "if-then" scenario analysis. Based on the predicted high-pollution periods and areas, combined with meteorological conditions, it recommends the optimal dust removal operation routes and timing.

[0055] By adjusting the DII (Dust Intensity Index), a quantitative indicator that integrates information such as spray volume, number of fog cannons, and operating time, the model can instantly predict the environmental improvement effect under different levels of investment. "If" we increase the DII from the current 60% to 80% (meaning an increase of 20% in energy and resource costs), "then" the model predicts that the average PM10 concentration in the park will decrease by another 15%, and the duration of heavy pollution will be reduced by 4 hours per month. "If" we only maintain the DII at 60%, but temporarily increase it to 90% during high pollution warning periods, "then" the model predicts that the total cost will only increase by 5%, but a 70% peak reduction effect can be achieved, avoiding events exceeding standards. Advantages: This allows decision-makers to clearly see the cost-benefit ratio of different strategies before making actual investments, much like simulating a strategic scenario, thus enabling them to make optimal budget allocations and strategic plans. It allows for cost-benefit analysis.

[0056] Benefit-cost analysis formula: .

[0057] Furthermore, any content not described in detail in this specification is existing technology known to those skilled in the art.

[0058] Those skilled in the art will understand that the above embodiments are specific examples of implementing the present invention, and in practical applications, various changes in form and detail may be made without departing from the spirit and scope of the present invention.

Claims

1. A smart dust removal environmental impact assessment method based on big data analysis, characterized in that, Includes the following steps: Step 1: Collect multi-source data and use spatiotemporal alignment technology to integrate the multi-source data into a unified data model to obtain a comprehensive dataset of the operation of the environment and dust removal system. Step 2: Perform data preprocessing and feature extraction on the multi-source data in the comprehensive dataset to obtain feature data; Step 3: Establish an environmental impact assessment model based on feature data; use the trained model to assess the environmental impact of dust removal measures; keep other variables constant, change the operating parameters of the dust removal equipment (such as dust removal efficiency), predict changes in environmental indicators, and calculate the contribution of dust removal measures to environmental indicators. Step 4: Present the assessment results in chart form; time series graph shows the changes in environmental indicators before and after the implementation of dust removal measures; spatial distribution map shows the improvement of environmental indicators in different areas; contribution pie chart shows the contribution ratio of each factor to the environmental impact.

2. The intelligent dust removal environmental impact assessment method based on big data analysis according to claim 1, characterized in that: The method for acquiring multi-source data involves deploying an Internet of Things (IoT) sensor network to collect real-time data on the energy consumption, dust removal efficiency, and operating status of dust removal equipment, as well as environmental data such as PM2.5, PM10, temperature, and humidity. In addition, external data sources are accessed, including meteorological data such as wind speed, wind direction, air pressure, and precipitation; geographic information data such as topography and land use type; and socioeconomic data such as population density and industrial distribution.

3. The intelligent dust removal environmental impact assessment method based on big data analysis according to claim 1, characterized in that: The method described above for preprocessing and feature engineering multi-source data in a comprehensive dataset to obtain feature data involves cleaning missing values, outliers, and duplicate values, and standardizing the data to convert data of different dimensions into a unified standard.

4. The intelligent dust removal environmental impact assessment method based on big data analysis according to claim 3, characterized in that: The method for extracting features and obtaining feature data is to extract features from preprocessed data, including time features: Hour, day, month, season; Spatial characteristics: regional division, latitude and longitude; meteorological characteristics: wind speed, wind direction, etc., which are converted into characteristics that affect the diffusion of pollutants; dust removal equipment characteristics: dust removal efficiency, operating time, and processing air volume; create lagged features by using the lagged values ​​of all feature variables and target variables from the past T time steps as features.

5. The intelligent dust removal environmental impact assessment method based on big data analysis according to claim 1, characterized in that: The method for establishing an environmental impact assessment model based on feature data involves constructing a "counterfactual" prediction model. Specifically, this involves selecting historical data from periods without strong dust control measures, or data from a control group area without dust control measures, as training data. The model uses meteorological characteristics, pollution source characteristics, and spatiotemporal characteristics, but excludes dust control activity characteristics. Environmental quality indicators are then used as the model output. The model formula is as follows: In the above formula , is the model-predicted "counterfactual" PM2.5 concentration at time t for grid i. It is the input feature vector. It is the space of the return tree. It refers to the number of trees; The goal of this model is to minimize the loss function. ; Then, during the evaluation period after the dust control measures were implemented, we compared the "counterfactual" concentrations predicted by the model with the actual observed concentrations to conduct an environmental impact assessment and quantification.

6. The intelligent dust removal environmental impact assessment method based on big data analysis according to claim 5, characterized in that: The method described above for conducting environmental impact assessment and quantification is... Absolute concentration reduction (ACA), , in This is the actual observed concentration. The higher this value, the better the dust removal effect. Relative Improvement Rate (RIR) ; Health Benefit Assessment (HBA) .

7. The intelligent dust removal environmental impact assessment method based on big data analysis according to claim 5, characterized in that: We decompose the contribution and use the SHAP model interpretation tool to analyze the contribution of each feature in the "counterfactual" prediction.

8. The intelligent dust removal environmental impact assessment method based on big data analysis according to claim 5, characterized in that: The system simulates dust removal strategies and predicts future environmental improvement effects by adjusting the input DII (Dust Intensity Index), providing decision-makers with "if-then" scenario analysis. Based on the predicted high-pollution periods and areas, combined with meteorological conditions, it recommends the optimal dust removal operation routes and timing.

9. The intelligent dust removal environmental impact assessment method based on big data analysis according to claim 5, characterized in that: Conduct a cost-benefit analysis. Benefit-cost analysis formula: 。