Chemical park environment monitoring method and system based on Internet of Things
Through the Internet of Things-based environmental monitoring method of chemical parks, the problems of unreasonable layout of monitoring points and data lag in air pollution monitoring in chemical parks have been solved, and the high accuracy of air pollution data and rapid response to emission control have been achieved, and the chemical parks have been improved emission control effect.
Patent Information
- Application Number
- CN202411259213.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-10
- Publication Date
- 2024-12-06
- Estimated Expiration
- 2044-09-10
AI Technical Summary
In the existing air pollution monitoring methods in chemical parks, due to the unreasonable layout of monitoring points and the data lag, the accuracy and timeliness of air pollution monitoring and early warning are insufficient, and precise emission control cannot be achieved.
The Internet of Things-based environmental monitoring method of chemical parks is adopted, and through air pollution analysis, emission simulation model generation, monitoring array optimization, edge computing and cloud platform collaborative processing, real-time data prediction and early warning signal generation are realized, and precise emission control is carried out.
It improves the accuracy and scientificity of monitoring points layout, enhances the accuracy of air pollution monitoring data, and improves the accuracy and timeliness of emission control in chemical parks by responding to monitoring abnormalities quickly.
Smart Images

Figure CN118795097B_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of intelligent monitoring technology, and in particular to a chemical park environment monitoring method and system based on the Internet of Things. Background Art
[0002] In the current air pollution monitoring methods in chemical parks, environmental data is usually collected by setting up fixed monitoring points. However, due to the unreasonable layout of monitoring points, it is impossible to fully cover the high-pollution risk areas and key pollution diffusion paths in the park, resulting in insufficient representativeness of the monitoring data. At the same time, this layout method is prone to monitoring blind spots, so that some pollution sources and pollution diffusion processes cannot be captured in time. In addition, the data processing and transmission of traditional monitoring systems often have lags, and real-time data analysis and feedback cannot be achieved.
[0003] In summary, the existing air pollution monitoring methods in chemical parks have technical problems such as insufficient accuracy and timeliness of air pollution monitoring and early warning due to unreasonable layout of monitoring points and lag in monitoring data, which makes it impossible to achieve precise emission control in chemical parks. Summary of the invention
[0004] The purpose of this application is to provide a chemical park environmental monitoring method and system based on the Internet of Things, so as to solve the technical problems in the existing chemical park air pollution monitoring methods, such as the unreasonable layout of monitoring points and the lag of monitoring data, which lead to the lack of accuracy and timeliness of air pollution monitoring and early warning, and make it impossible to achieve precise emission control in chemical parks.
[0005] In view of the above problems, the present application provides a chemical park environment monitoring method and system based on the Internet of Things.
[0006] In the first aspect, the present application provides a chemical park environment monitoring method based on the Internet of Things, which is implemented through a chemical park environment monitoring system based on the Internet of Things, including: calling the park production plan within a predetermined period to perform air pollution analysis and obtain multiple pollution emission information; building an emission simulation space based on the park emission layout information, collecting diffusion interference factors of the target area within the predetermined period, rendering the diffusion interference factors to the emission simulation space, and generating a regional emission simulation model; inputting the multiple pollution emission information into the regional emission simulation model, simulating the gas pollution of the target area, and obtaining multiple simulated pollution distribution atlases, optimizing the layout of the initial monitoring array according to the multiple simulated pollution distribution atlases, and obtaining an optimized monitoring array; performing fixed-point monitoring of the target area through the optimized monitoring array, reducing the dimension of the monitoring data set at the edge layer, and inputting the reduced dimension data set into the cloud platform, wherein the edge layer is embedded in the optimized monitoring array; on the cloud platform, performing regional pollution prediction within a predetermined time window according to the reduced dimension data set, if the predicted pollution distribution does not satisfy the simulated pollution distribution map under the same node, then generating a monitoring early warning signal, and performing emission control on the target chemical park based on the monitoring early warning signal.
[0007] In the second aspect, the present application also provides a chemical park environmental monitoring system based on the Internet of Things, which is used to execute a chemical park environmental monitoring method based on the Internet of Things as described in the first aspect, including: an air pollution analysis module, which is used to call the park production plan within a predetermined period to perform air pollution analysis and obtain multiple pollution emission information; a regional emission simulation model generation module, which is used to build an emission simulation space based on the park emission layout information, collect the diffusion interference factors of the target area within a predetermined period, render the diffusion interference factors to the emission simulation space, and generate a regional emission simulation model; an optimized monitoring array acquisition module, which is used to input the multiple pollution emission information into the regional emission simulation model to perform target area A gas pollution simulation of a domain is performed to obtain multiple simulated pollution distribution atlases, and the initial monitoring array is optimized according to the multiple simulated pollution distribution atlases to obtain an optimized monitoring array; a monitoring data dimensionality reduction module is used to perform fixed-point monitoring of the target area through the optimized monitoring array, reduce the dimensionality of the monitoring data set at the edge layer, and input the reduced dimensionality data set into the cloud platform, wherein the edge layer is embedded in the optimized monitoring array; a park emission control module is used to perform regional pollution prediction within a predetermined time window on the cloud platform according to the reduced dimensionality data set, if the predicted pollution distribution does not satisfy the simulated pollution distribution map under the same node, a monitoring early warning signal is generated, and emission control is performed on the target chemical park based on the monitoring early warning signal.
[0008] One or more technical solutions provided in this application have at least the following technical effects or advantages:
[0009] By analyzing air pollution according to the production plan of the park within a predetermined period, multiple pollution emission information is obtained; on the other hand, an emission simulation space is built based on the emission layout information of the park, and the diffusion interference factors are rendered into the emission simulation space to generate a regional emission simulation model; then, multiple pollution emission information is input into the regional emission simulation model to simulate the gas pollution in the target area, and multiple simulated pollution distribution atlases are obtained; further, the initial monitoring array is optimized according to multiple simulated pollution distribution atlases to obtain an optimized monitoring array; then, fixed-point monitoring of the target area is performed by optimizing the monitoring array, and the monitoring data set is reduced in dimension at the edge layer, and the reduced dimension data set is input into the cloud platform; then, in the cloud platform, regional pollution prediction within a predetermined time window is performed according to the reduced dimension data set. If the predicted pollution distribution does not meet the simulated pollution distribution map under the same node, a monitoring early warning signal is generated; finally, emission control is performed on the target chemical park based on the monitoring early warning signal; the accuracy and scientificity of the monitoring point layout can be improved, and then the accuracy of air pollution monitoring data can be improved. On the other hand, data processing is performed based on cloud-edge collaboration, and pollution distribution prediction is performed according to the monitoring results. Emission control is performed according to the predicted results, which can quickly respond to monitoring anomalies, thereby effectively improving the accuracy and timeliness of emission control in chemical parks.
[0010] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented according to the contents of the specification, and in order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are specifically cited below. It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present application, nor is it intended to limit the scope of the present application. Other features of the present application will become easy to understand through the following description. BRIEF DESCRIPTION OF THE DRAWINGS
[0011] In order to more clearly illustrate the technical solutions in the present application or the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings in the following description are only exemplary, and for ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying any creative work.
[0012] Figure 1 A schematic diagram of a process for chemical park environment monitoring method based on the Internet of Things is provided for this application;
[0013] Figure 2 This is a schematic diagram of a process for collecting diffusion interference factors in a chemical park environment monitoring method based on the Internet of Things in this application;
[0014] Figure 3This application provides a structural schematic diagram of a chemical park environment monitoring system based on the Internet of Things.
[0015] Description of reference numerals:
[0016] Air pollution analysis module 11, regional emission simulation model generation module 12, optimized monitoring array acquisition module 13, monitoring data dimension reduction module 14, park emission control module 15. DETAILED DESCRIPTION
[0017] This application provides a chemical park environmental monitoring method and system based on the Internet of Things, which solves the technical problem that the existing chemical park air pollution monitoring method has insufficient accuracy and timeliness of air pollution monitoring and early warning due to unreasonable monitoring point layout and lag in monitoring data, resulting in the inability to achieve precise emission control of chemical parks. The accuracy and scientificity of the monitoring point layout can be improved, and then the accuracy of air pollution monitoring data can be improved. On the other hand, data processing is performed based on cloud-edge collaboration, and pollution distribution prediction is performed based on monitoring results. Emission control is performed based on the predicted results, which can quickly respond to monitoring anomalies, thereby effectively improving the accuracy and timeliness of emission control in chemical parks.
[0018] Below, the technical solutions in the present application will be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments of the present application. It should be understood that the present application is not limited to the example embodiments described herein. Based on the embodiments of the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present application. It should also be noted that, for the convenience of description, only the parts related to the present application are shown in the accompanying drawings, rather than all of them.
[0019] For example, please refer to the attached Figure 1 The present application provides a chemical park environment monitoring method based on the Internet of Things, which is applied to a chemical park environment monitoring system based on the Internet of Things, and specifically includes the following steps:
[0020] Step 1: Call the park production plan within the predetermined period to perform air pollution analysis and obtain multiple pollution emission information.
[0021] Specifically, first, configure a predetermined period, which is a fixed time period. Those skilled in the art can set it according to actual conditions, such as one week; then call the park production plan within the predetermined period, where the park is a chemical park, that is, collect the production plans of each enterprise in the park within the predetermined period, including production activities, raw materials used, production processes, and expected products. Then, air pollution analysis is performed according to the park production plan, that is, the links and sources that produce air pollution are identified. For the identified pollution sources, their emission factors are determined, including the type of pollutant gas, gas emissions, and emission period. The type of pollutant gas refers to the specific type of pollutant, such as sulfur dioxide (SO2), nitrogen oxides (NOx), volatile organic compounds (VOCs), etc.; gas emissions refer to the total amount of specific pollutant gases emitted into the atmosphere within a certain period of time, usually in kilograms or tons; the emission period refers to the specific time range of pollutant gas emissions, such as 8 am to 4 pm on a certain working day; obtain multiple pollution emission information, where each type of pollutant gas corresponds to a type of pollution emission information. By obtaining multiple pollution emission information, data support is provided for subsequent pollution emission simulation.
[0022] Step 2: Build an emission simulation space based on the emission layout information of the park, collect the diffusion interference factors of the target area within a predetermined period, render the diffusion interference factors to the emission simulation space, and generate a regional emission simulation model.
[0023] Specifically, first, obtain the emission layout information of the park, including the location of the pollution source, emission height, emission rate, emission composition, etc., as well as information such as the park terrain, building distribution, and vegetation coverage; then, in the three-dimensional visualization platform, perform simulation modeling based on the emission layout information of the park, and build an emission simulation space. On the other hand, collect the diffusion interference factors of the target area within a predetermined period. Diffusion interference factors refer to interference factors that affect the diffusion of polluted gases, such as temperature, humidity, wind speed, wind direction, etc. These factors determine the diffusion speed and range of pollutants in the air. Then render the diffusion interference factors to the emission simulation space to generate a regional emission simulation model. First, integrate the collected data of various interference factors into a unified platform or database to ensure data consistency and accessibility; then use the integrated data as input parameters and input them into the emission simulation model, and map these interference factors to the corresponding grids or cells of the simulation space to ensure that each grid cell contains the corresponding interference factor information, and adjust the model parameters such as diffusion coefficient and sedimentation velocity according to the characteristics of the interference factors. By constructing an emission simulation space and rendering the diffusion interference factors into the emission simulation space to generate a regional emission simulation model, it can help to understand and simulate the diffusion behavior of pollutants in the target area.
[0024] Step three: input the multiple pollution emission information into the regional emission simulation model, simulate the gas pollution in the target area, obtain multiple simulated pollution distribution atlases, optimize the layout of the initial monitoring array according to the multiple simulated pollution distribution atlases, and obtain the optimized monitoring array.
[0025] Specifically, the multiple pollution emission information is input into the regional emission simulation model to simulate the gas pollution status of the target area. First, according to the pollution emission information and the characteristics of the target area, the parameters required for the simulation are set, such as the pollutant diffusion coefficient, sedimentation velocity, chemical reaction rate, etc., and the time range and spatial resolution of the simulation are determined to match the time period of the pollution emission information and the geographical scope of the target area. Then, the model is started to run, and the model will simulate the diffusion process of pollutants in the target area according to the input pollution emission information and the set parameters; wherein, during the simulation process, the predetermined period is divided into multiple time intervals, and the air pollution status in multiple time intervals is simulated. After the simulation is completed, the model will output a series of pollution distribution maps, which show the concentration distribution and diffusion range of pollutants in the target area at different time points; multiple simulated pollution distribution atlases are obtained, wherein each type of polluted gas corresponds to a simulated pollution distribution atlas.
[0026] Further identify multiple simulated pollution distribution atlases to determine high-value areas of pollutant concentration, change trends, pollution hotspots, and possible transmission paths; then optimize the layout of the initial monitoring array based on the identification results, for example, increase monitoring points in sub-areas with higher pollution concentrations to ensure coverage of high-risk areas; reduce overly dense monitoring points in sub-areas with lower pollution concentrations to optimize resource utilization; dynamically adjust the location and density of monitoring points according to the concentration and area size of different sub-areas to obtain an optimized monitoring array. By optimizing the layout of the initial monitoring array based on the pollution identification results, the adaptability of the monitoring array to the air pollution status can be improved, ensuring that the monitoring array is more scientific and reasonable, and can more effectively monitor and evaluate the air pollution status of the target area, providing more accurate data support for environmental management and decision-making.
[0027] Step 4: Performing fixed-point monitoring of the target area through the optimized monitoring array, reducing the dimension of the monitoring data set at the edge layer, and inputting the reduced dimension data set into the cloud platform, wherein the edge layer is embedded in the optimized monitoring array.
[0028] Specifically, the optimized monitoring array is used to conduct fixed-point monitoring in the target area to collect the concentration data of air pollutants. The monitoring equipment includes gas sensors, particulate matter monitors, etc., which can record information such as pollutant concentration and pollutant diffusion range in real time or regularly. The monitoring data set is further preprocessed and dimensionally reduced at the edge computing layer of the optimized monitoring array, that is, the original monitoring data is preliminarily processed, such as cleaning, calibration and formatting; at the same time, since the monitoring data is usually high-dimensional, the edge layer needs to perform dimension reduction processing on the data to reduce the amount of data and extract key information to obtain a reduced-dimensional data set. Finally, the reduced-dimensional data set is transmitted to the cloud platform through communication, and data processing is performed by combining edge computing and cloud platform, which helps to reduce data transmission delays, improve the response speed of the system, and reduce the computing burden of the cloud platform.
[0029] Step 5: On the cloud platform, regional pollution prediction is performed within a predetermined time window based on the dimension reduction data set. If the predicted pollution distribution does not satisfy the simulated pollution distribution map under the same node, a monitoring and early warning signal is generated, and emission control is performed on the target chemical park based on the monitoring and early warning signal.
[0030] Specifically, in the cloud platform, historical dimensionality reduction data sequences are obtained, that is, historical monitoring data under multiple monitoring nodes within a predetermined period; further regional pollution prediction within a predetermined time window is performed based on the historical dimensionality reduction data sequences and dimensionality reduction data sets. First, a suitable prediction model is selected according to the prediction target and data characteristics, such as random forest, support vector machine, neural network, etc., and then the selected model is trained using historical data. This step involves parameter tuning and verification of the model to ensure that the model has good prediction performance; then the trained model is used to predict the pollution situation within the predetermined time window to obtain the predicted pollution distribution.
[0031] Further obtain a set of simulated pollution distribution maps at the end node of the predetermined time window, including multiple simulated pollution distribution maps corresponding to multiple types of polluting gases; then perform deviation traversal comparison on the predicted pollution distribution and the simulated pollution distribution map. If the pollution range of the predicted pollution distribution is larger than the pollution range of the simulated pollution distribution map and / or the pollution concentration of the predicted pollution distribution is larger than the pollution concentration of the simulated pollution distribution map in the same area, it indicates that the actual pollution situation is more serious than expected, which may be because the actual emission of the emission source exceeds expectations, or environmental conditions cause faster diffusion and accumulation of pollutants, and then generate a monitoring and early warning signal; finally, according to the monitoring and early warning signal, the target chemical park is subjected to emission control, such as checking the emission sources in the chemical park according to the monitoring and early warning signal, finding the emission sources that cause excessive pollution, and taking measures to reduce emissions. Through the above steps, it is possible to quickly respond to monitoring anomalies and effectively improve the accuracy and timeliness of emission control in chemical parks.
[0032] The chemical park environment monitoring method based on the Internet of Things is applied to a chemical park environment monitoring system based on the Internet of Things, which can solve the technical problems in the existing chemical park air pollution monitoring method, such as the unreasonable layout of monitoring points and the lag of monitoring data, which lead to the lack of accuracy and timeliness of air pollution monitoring and early warning, and make it impossible to achieve precise emission control of chemical parks. By analyzing air pollution according to the production plan of the park within a predetermined period, multiple pollution emission information is obtained; on the other hand, an emission simulation space is built based on the emission layout information of the park, and the diffusion interference factors are rendered into the emission simulation space to generate a regional emission simulation model; then, multiple pollution emission information is input into the regional emission simulation model to simulate the gas pollution in the target area, and multiple simulated pollution distribution atlases are obtained; further, the initial monitoring array is optimized according to multiple simulated pollution distribution atlases to obtain an optimized monitoring array; then, fixed-point monitoring of the target area is performed by optimizing the monitoring array, and the monitoring data set is reduced in dimension at the edge layer, and the reduced dimension data set is input into the cloud platform; then, in the cloud platform, regional pollution prediction within a predetermined time window is performed according to the reduced dimension data set. If the predicted pollution distribution does not meet the simulated pollution distribution map under the same node, a monitoring early warning signal is generated; finally, emission control is performed on the target chemical park based on the monitoring early warning signal; the accuracy and scientificity of the monitoring point layout can be improved, and then the accuracy of air pollution monitoring data can be improved. On the other hand, data processing is performed based on cloud-edge collaboration, and pollution distribution prediction is performed according to the monitoring results. Emission control is performed according to the predicted results, which can quickly respond to monitoring anomalies, thereby effectively improving the accuracy and timeliness of emission control in chemical parks.
[0033] Further, the diffusion interference factors of the target area within the predetermined period are collected, such as the attached Figure 2 As shown, step 2 of this application includes:
[0034] Acquire pollution emission information, wherein the pollution emission information includes gas type, emission amount and emission period; based on a preset factor space, perform gas diffusion interference analysis on multiple gas types respectively to determine multiple high-frequency interference factor sets, wherein the preset factor space includes at least temperature, humidity, wind speed, wind direction and light; perform intersection operation on the multiple high-frequency interference factor sets to obtain predetermined interference factors, and collect diffusion interference factors of the target area within a predetermined period based on the predetermined interference factors.
[0035] Specifically, first, obtain the pollution emission information, wherein the pollution emission information includes the type of polluting gas, the emission amount and the emission period; then configure the preset interference factor space of gas diffusion, wherein the preset interference factor space includes at least temperature, humidity, wind speed, wind direction and light, and additional interference factors can also be added according to the actual situation. Then, based on the preset interference factor space, perform gas diffusion interference analysis on multiple gas types respectively, including interference intensity, interference frequency, etc.; then, by comparing the simulation results under different scenarios, set the interference factors that have a significant impact on gas diffusion in most scenarios as high-frequency interference factors, and determine multiple high-frequency interference factor sets. Then, perform an intersection operation on the multiple high-frequency interference factor sets, the purpose of which is to find out the high-frequency interference factors common to all gas types, which have a significant impact on the diffusion of multiple gas types, and set the intersection operation results as the predetermined interference factors. Finally, according to the predetermined interference factors, collect the diffusion interference factors of the target area within the predetermined period.
[0036] Further, a plurality of simulated pollution distribution atlases are obtained. Step three of this application includes:
[0037] The predetermined period is divided into a predetermined step size to determine a plurality of predetermined intervals; first pollution emission information is randomly selected from the plurality of pollution emission information, the first pollution emission information is input into the regional emission simulation model, and gas pollution simulation of the target area within the plurality of predetermined intervals is performed to obtain a first simulated pollution distribution atlas, which is added to the plurality of simulated pollution distribution atlases, wherein the simulated pollution distribution atlas includes a gas simulation diffusion range and a gas simulation concentration.
[0038] Specifically, a predetermined step length is configured, and the predetermined step length is a smaller time period, which can be set according to actual conditions, such as setting the predetermined step length to 1 hour; then the predetermined period is divided according to the predetermined step length to determine multiple predetermined intervals. Then, any one of the multiple pollution emission information is randomly selected as the first pollution emission information, and the first pollution emission information is input into the regional emission simulation model to simulate the gas pollution of the target area within multiple predetermined intervals to obtain a first simulated pollution distribution atlas, wherein the simulated pollution distribution map includes the gas simulation diffusion range and the gas simulation concentration; finally, multiple simulated pollution distribution atlases of multiple polluted gas types are obtained using the same method.
[0039] Further, to obtain an optimized monitoring array, step three of this application includes:
[0040] Select a first key distribution map in the first simulated pollution distribution map set, wherein the first key distribution map is the distribution map with the largest gas pollution range and the highest concentration in the first simulated pollution distribution map set; divide the first key distribution map into regions according to the pollution concentration deviation step, and obtain a first distribution sub-region set and a first regional concentration interval set; according to a predetermined layout strategy, optimize the layout of the first monitoring array according to the first distribution sub-region set and the first regional concentration interval set to obtain a first optimized monitoring array, and add it to the optimized monitoring array.
[0041] Specifically, first, select the first key distribution map in the first simulated pollution distribution map set, which is any one of a plurality of simulated pollution distribution map sets, wherein the first key distribution map is the distribution map with the largest gas pollution range and the highest concentration in the first simulated pollution distribution map set. Then, the first key distribution map is divided into regions according to the pollution concentration deviation step, wherein the pollution concentration deviation step can be set according to actual conditions, and the pollution concentration deviation step determines the fineness of the regional division; then, for each grid unit in the first key distribution map, calculate the pollution concentration difference between it and the adjacent grid unit, and divide the first key distribution map into multiple regions according to the set pollution concentration deviation step, wherein each region is composed of adjacent grid units, and the pollution concentration deviation of these grid units is within the set step range, and obtain the first distribution sub-region set and the first regional concentration interval set, wherein the first distribution sub-region and the first regional concentration interval correspond one to one.
[0042] Configure a predetermined deployment strategy, which can be set according to actual conditions, such as the deployment strategy needs to meet the requirements of fully covering high-pollution risk areas and key pollution diffusion paths; increase monitoring points in sub-areas with higher pollution concentrations to ensure coverage of high-risk areas; reduce overly dense monitoring points in sub-areas with lower pollution concentrations to optimize resource utilization; dynamically adjust the location and density of monitoring points according to the concentration and area size of different sub-areas. Then, according to the predetermined deployment strategy, the first monitoring array is deployed and optimized according to the first distribution sub-area set and the first regional concentration interval set. The first monitoring array is the initial monitoring array, such as adding monitoring points in high pollution concentration areas or areas with significant changes in pollution concentration, removing monitoring points in low pollution concentration areas or those that contribute less to the overall monitoring network, adjusting the location of monitoring points, ensuring that monitoring points can better reflect the characteristics of pollution distribution, etc., to obtain the first optimized monitoring array, which is added to the optimized monitoring array. By optimizing the initial monitoring array according to the pollution simulation results, an optimized monitoring array that can accurately monitor the distribution of pollutants can be obtained, thereby improving the effectiveness and efficiency of air pollution monitoring.
[0043] Further, the monitoring data set is reduced in dimension at the edge layer. Step 4 of this application includes:
[0044] The optimized monitoring array is used to perform fixed-point monitoring of the target area to obtain a real-time monitoring data set. At the edge layer, a deviation mapping comparison is performed on the real-time monitoring data set and the historical monitoring data set of the previous node, and the real-time monitoring data with a deviation greater than a predetermined deviation threshold is retained to obtain the reduced dimension data set.
[0045] Specifically, the air state of the target area is monitored at a fixed point through the optimized monitoring array, such as collecting data every 1 minute to obtain a real-time monitoring data set. Further, at the edge computing layer of the optimized monitoring array, a deviation mapping comparison is performed on the real-time monitoring data set and the historical monitoring data set of the previous adjacent monitoring node. The deviation mapping comparison refers to comparing the difference in monitoring values between real-time data and historical data at the same location and time point. At the same time, a predetermined deviation threshold is set according to the actual situation. The predetermined deviation threshold is a standard for judging whether the deviation between real-time data and historical data is significant. Then, the real-time monitoring data is screened according to the predetermined deviation threshold, and the real-time monitoring data with a deviation greater than the predetermined deviation threshold is retained as reduced-dimensional data to obtain a reduced-dimensional data set. By reducing the dimensionality of the monitoring data set at the edge computing layer, it can help reduce data transmission delays, improve the response speed of the system, and reduce the computing burden of the cloud platform.
[0046] Further, according to the dimension reduction data set, regional pollution prediction within a predetermined time window is performed, and step five of the present application includes:
[0047] Obtain a predetermined time window, wherein the predetermined time window is configured based on a data acquisition delay and a data transmission delay; randomly select first dimensionality reduction data of a first gas type from the dimensionality reduction data set, and obtain a first historical dimensionality reduction data sequence of the first gas type; input the first dimensionality reduction data and the first historical dimensionality reduction data sequence into a first pollution prediction channel, output first predicted pollution data, generate a first predicted pollution distribution according to the first predicted pollution data, and add the first predicted pollution distribution to the predicted pollution distribution.
[0048] Specifically, first, a predetermined time window is configured according to data collection delay and data transmission delay. Data collection delay refers to the time difference between the actual occurrence of a pollution event and the start of data collection by the data collection device. Data transmission delay includes the transmission time of a wired or wireless network, as well as the time for data processing and storage. Then, the first dimensionality reduction data of the first gas type is randomly selected from the dimensionality reduction data set, and the first historical dimensionality reduction data sequence of the first gas type is obtained. Then, a first pollution prediction channel is constructed based on machine learning, and the first dimensionality reduction data and the first historical dimensionality reduction data sequence are input into the first pollution prediction channel for prediction, and the first predicted pollution data is obtained, including the predicted pollution gas concentration and the predicted gas diffusion range; finally, a first predicted pollution distribution is generated according to the first predicted pollution data and added to the predicted pollution distribution.
[0049] Further, to construct the first pollution prediction channel, the present application also includes the following steps:
[0050] Taking the first gas type as a retrieval constraint, a sample prediction data set is obtained, wherein the sample prediction data includes sample monitoring data, a sample historical monitoring data sequence and sample predicted pollution data; the sample prediction data set is divided into N equal parts, and supervised training is performed on N prediction operators respectively to obtain N pollution prediction units and N prediction accuracies that meet predetermined convergence conditions, wherein N is an integer greater than or equal to 2; and the first pollution prediction channel is constructed based on the N pollution prediction units and N prediction accuracies.
[0051] Specifically, the first gas type is used as a search constraint condition, and big data technology is used to perform sample retrieval to obtain a sample prediction data set, wherein the sample prediction data includes sample monitoring data, sample historical monitoring data sequence and sample prediction pollution data. Then, N prediction operators are configured, wherein the prediction operator type can be configured according to actual needs, such as support vector machine, random decision forest, BP neural network, etc.; the sample prediction data set is further divided into N parts to obtain N data sets; then, the N data sets are used to supervise the N prediction operators respectively, such as combining the gradient descent algorithm and the loss function training model, to obtain N pollution prediction units and N prediction accuracies that meet the predetermined convergence conditions, and the prediction accuracy is the prediction accuracy of the pollution prediction unit. Based on the principle of ensemble learning, the first pollution prediction channel is constructed according to the fusion of the N pollution prediction units and the N prediction accuracies, wherein the output of the first pollution prediction channel is the weighted calculation result of the N pollution prediction units. By configuring multiple prediction operators and fusion-constructing the first pollution prediction channel based on the principle of ensemble learning, the prediction error of a single operator can be reduced, and the accuracy and rationality of pollution prediction can be improved.
[0052] Further, to generate a monitoring warning signal, step five of this application includes:
[0053] A first predicted pollution distribution of a first gas type is randomly selected in the predicted pollution distribution, and a node simulated pollution distribution map of the first gas type under the end node of a predetermined time window is obtained; a deviation traversal comparison is performed on the first predicted pollution distribution and the node simulated pollution distribution map; if the pollution range of the first predicted pollution distribution is larger than the pollution range of the node simulated pollution distribution map and / or the pollution concentration of the first predicted pollution distribution is larger than the pollution concentration of the node simulated pollution distribution map in the same area, a first warning signal is generated and added to the monitoring warning signal.
[0054] Specifically, first, a first predicted pollution distribution of a first gas type is randomly selected in the predicted pollution distribution, and a node simulated pollution distribution map of the first gas type at the end node of a predetermined time window is obtained; then, a deviation traversal comparison is performed on the first predicted pollution distribution and the node simulated pollution distribution map. If the pollution range of the first predicted pollution distribution is larger than the pollution range of the node simulated pollution distribution map and / or the pollution concentration of the first predicted pollution distribution is larger than the pollution concentration of the node simulated pollution distribution map in the same area, a first warning signal is generated, wherein the first warning signal includes the polluted gas type and the warning type, such as a concentration warning or a diffusion range warning, and finally the first warning signal is added to the monitoring warning signal. The above method can solve the technical problems existing in the existing chemical park air pollution monitoring methods, such as the unreasonable layout of monitoring points and the lag of monitoring data, which lead to the insufficient accuracy and timeliness of air pollution monitoring and early warning, and the inability to achieve precise emission control in chemical parks. It can improve the accuracy and scientificity of the monitoring point layout, and thus improve the accuracy of air pollution monitoring data. On the other hand, data processing is carried out based on cloud-edge collaboration, and pollution distribution is predicted based on the monitoring results. Emission control is carried out based on the prediction results, which can quickly respond to monitoring anomalies, thereby effectively improving the accuracy and timeliness of emission control in chemical parks.
[0055] In summary, the chemical park environment monitoring method based on the Internet of Things provided in this application has the following technical effects:
[0056] By analyzing air pollution according to the production plan of the park within a predetermined period, multiple pollution emission information is obtained; on the other hand, an emission simulation space is built based on the emission layout information of the park, and the diffusion interference factors are rendered into the emission simulation space to generate a regional emission simulation model; then, multiple pollution emission information is input into the regional emission simulation model to simulate the gas pollution in the target area, and multiple simulated pollution distribution atlases are obtained; further, the initial monitoring array is optimized according to multiple simulated pollution distribution atlases to obtain an optimized monitoring array; then, fixed-point monitoring of the target area is performed by optimizing the monitoring array, and the monitoring data set is reduced in dimension at the edge layer, and the reduced dimension data set is input into the cloud platform; then, in the cloud platform, regional pollution prediction within a predetermined time window is performed according to the reduced dimension data set. If the predicted pollution distribution does not meet the simulated pollution distribution map under the same node, a monitoring early warning signal is generated; finally, emission control is performed on the target chemical park based on the monitoring early warning signal; the accuracy and scientificity of the monitoring point layout can be improved, and then the accuracy of air pollution monitoring data can be improved. On the other hand, data processing is performed based on cloud-edge collaboration, and pollution distribution prediction is performed according to the monitoring results. Emission control is performed according to the predicted results, which can quickly respond to monitoring anomalies, thereby effectively improving the accuracy and timeliness of emission control in chemical parks.
[0057] Embodiment 2, based on the same invention concept as the chemical park environment monitoring method based on the Internet of Things in the previous embodiment, this application also provides a chemical park environment monitoring system based on the Internet of Things, please refer to the attached Figure 3 ,include:
[0058] The air pollution analysis module 11 is used to call the park production plan within a predetermined period to perform air pollution analysis and obtain multiple pollution emission information; the regional emission simulation model generation module 12 is used to build an emission simulation space based on the park emission layout information, collect the diffusion interference factors of the target area within the predetermined period, render the diffusion interference factors to the emission simulation space, and generate a regional emission simulation model; the optimized monitoring array acquisition module 13 is used to input the multiple pollution emission information into the regional emission simulation model, simulate the gas pollution of the target area, obtain multiple simulated pollution distribution atlases, optimize the layout of the initial monitoring array according to the multiple simulated pollution distribution atlases, and obtain the optimized monitoring array; the monitoring data dimension reduction module 14 is used to perform fixed-point monitoring of the target area through the optimized monitoring array, reduce the dimension of the monitoring data set at the edge layer, and input the reduced dimension data set into the cloud platform, wherein the edge layer is embedded in the optimized monitoring array; the park emission control module 15 is used to predict the regional pollution within a predetermined time window on the cloud platform according to the reduced dimension data set. If the predicted pollution distribution does not meet the simulated pollution distribution map under the same node, a monitoring warning signal is generated, and emission control is performed on the target chemical park based on the monitoring warning signal.
[0059] Furthermore, the chemical park environment monitoring system based on the Internet of Things is also used for:
[0060] Acquire pollution emission information, wherein the pollution emission information includes gas type, emission amount and emission period; based on a preset factor space, perform gas diffusion interference analysis on multiple gas types respectively to determine multiple high-frequency interference factor sets, wherein the preset factor space includes at least temperature, humidity, wind speed, wind direction and light; perform intersection operation on the multiple high-frequency interference factor sets to obtain predetermined interference factors, and collect diffusion interference factors of the target area within a predetermined period based on the predetermined interference factors.
[0061] Furthermore, the chemical park environment monitoring system based on the Internet of Things is also used for:
[0062] The predetermined period is divided into a predetermined step size to determine a plurality of predetermined intervals; first pollution emission information is randomly selected from the plurality of pollution emission information, the first pollution emission information is input into the regional emission simulation model, and gas pollution simulation of the target area within the plurality of predetermined intervals is performed to obtain a first simulated pollution distribution atlas, which is added to the plurality of simulated pollution distribution atlases, wherein the simulated pollution distribution atlas includes a gas simulation diffusion range and a gas simulation concentration.
[0063] Furthermore, the chemical park environment monitoring system based on the Internet of Things is also used for:
[0064] Select a first key distribution map in the first simulated pollution distribution map set, wherein the first key distribution map is the distribution map with the largest gas pollution range and the highest concentration in the first simulated pollution distribution map set; divide the first key distribution map into regions according to the pollution concentration deviation step, and obtain a first distribution sub-region set and a first regional concentration interval set; according to a predetermined layout strategy, optimize the layout of the first monitoring array according to the first distribution sub-region set and the first regional concentration interval set to obtain a first optimized monitoring array, and add it to the optimized monitoring array.
[0065] Furthermore, the chemical park environment monitoring system based on the Internet of Things is also used for:
[0066] The optimized monitoring array is used to perform fixed-point monitoring of the target area to obtain a real-time monitoring data set. At the edge layer, a deviation mapping comparison is performed on the real-time monitoring data set and the historical monitoring data set of the previous node, and the real-time monitoring data with a deviation greater than a predetermined deviation threshold is retained to obtain the reduced dimension data set.
[0067] Furthermore, the chemical park environment monitoring system based on the Internet of Things is also used for:
[0068] Obtain a predetermined time window, wherein the predetermined time window is configured based on a data acquisition delay and a data transmission delay; randomly select first dimensionality reduction data of a first gas type from the dimensionality reduction data set, and obtain a first historical dimensionality reduction data sequence of the first gas type; input the first dimensionality reduction data and the first historical dimensionality reduction data sequence into a first pollution prediction channel, output first predicted pollution data, generate a first predicted pollution distribution according to the first predicted pollution data, and add the first predicted pollution distribution to the predicted pollution distribution.
[0069] Furthermore, the chemical park environment monitoring system based on the Internet of Things is also used for:
[0070] Taking the first gas type as a retrieval constraint, a sample prediction data set is obtained, wherein the sample prediction data includes sample monitoring data, a sample historical monitoring data sequence and sample predicted pollution data; the sample prediction data set is divided into N equal parts, and supervised training is performed on N prediction operators respectively to obtain N pollution prediction units and N prediction accuracies that meet predetermined convergence conditions, wherein N is an integer greater than or equal to 2; and the first pollution prediction channel is constructed based on the N pollution prediction units and N prediction accuracies.
[0071] Furthermore, the chemical park environment monitoring system based on the Internet of Things is also used for:
[0072] A first predicted pollution distribution of a first gas type is randomly selected in the predicted pollution distribution, and a node simulated pollution distribution map of the first gas type under the end node of a predetermined time window is obtained; a deviation traversal comparison is performed on the first predicted pollution distribution and the node simulated pollution distribution map; if the pollution range of the first predicted pollution distribution is larger than the pollution range of the node simulated pollution distribution map and / or the pollution concentration of the first predicted pollution distribution is larger than the pollution concentration of the node simulated pollution distribution map in the same area, a first warning signal is generated and added to the monitoring warning signal.
[0073] Each embodiment in this specification is described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The chemical park environment monitoring method based on the Internet of Things and the specific examples in the aforementioned embodiment 1 are also applicable to a chemical park environment monitoring system based on the Internet of Things in this embodiment. Through the aforementioned detailed description of a chemical park environment monitoring method based on the Internet of Things, those skilled in the art can clearly know a chemical park environment monitoring system based on the Internet of Things in this embodiment, so for the sake of brevity of the specification, it will not be described in detail here. For the system disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method part description.
[0074] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.
[0075] Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application belong to the scope of the present application and its equivalent technology, the present application is also intended to include these modifications and variations.
Claims
1. A chemical park environment monitoring method based on the Internet of Things, characterized in that: include: Call the park production plan within the scheduled period to analyze air pollution and obtain multiple pollution emission information; Building an emission simulation space based on the emission layout information of the park, collecting diffusion interference factors of the target area within a predetermined period, rendering the diffusion interference factors to the emission simulation space, and generating a regional emission simulation model; Inputting the plurality of pollution emission information into the regional emission simulation model, simulating gas pollution in the target area, obtaining a plurality of simulated pollution distribution atlases, optimizing the layout of the initial monitoring array according to the plurality of simulated pollution distribution atlases, and obtaining an optimized monitoring array; Performing fixed-point monitoring of a target area through the optimized monitoring array, reducing the dimension of a monitoring data set at an edge layer, and inputting the reduced-dimensional data set into a cloud platform, wherein the edge layer is embedded in the optimized monitoring array; On the cloud platform, regional pollution prediction within a predetermined time window is performed based on the dimension reduction data set. If the predicted pollution distribution does not satisfy the simulated pollution distribution map under the same node, a monitoring and early warning signal is generated, and emission control is performed on the target chemical park based on the monitoring and early warning signal; Among them, the dimension reduction of the monitoring data set is carried out at the edge layer, including: Performing fixed-point monitoring of the target area through the optimized monitoring array to obtain a real-time monitoring data set; At the edge layer, performing deviation mapping comparison on the real-time monitoring data set and the historical monitoring data set of the previous node, retaining the real-time monitoring data with a deviation greater than a predetermined deviation threshold, and obtaining the dimension reduction data set; Performing regional pollution prediction within a predetermined time window based on the dimension reduction data set includes: Obtaining a predetermined time window, wherein the predetermined time window is configured based on a data collection delay and a data transmission delay; Randomly selecting first dimensionality reduction data of a first gas type from the dimensionality reduction data set, and obtaining a first historical dimensionality reduction data sequence of the first gas type; Inputting the first dimensionality reduction data and the first historical dimensionality reduction data sequence into a first pollution prediction channel, outputting first predicted pollution data, generating a first predicted pollution distribution according to the first predicted pollution data, and adding the first predicted pollution distribution to the predicted pollution distribution; Construct the first pollution prediction channel, including: Taking the first gas type as a search constraint, obtaining a sample prediction data set, wherein the sample prediction data includes sample monitoring data, a sample historical monitoring data sequence, and sample prediction pollution data; Dividing the sample prediction data set into N equal parts, and performing supervised training on N prediction operators respectively, to obtain N pollution prediction units and N prediction accuracies that meet a predetermined convergence condition, wherein N is an integer greater than or equal to 2; The first pollution prediction channel is constructed based on the N pollution prediction units and the N prediction accuracies.
2. The method for monitoring the environment of a chemical park based on the Internet of Things according to claim 1, characterized in that: Collect the diffusion interference factors of the target area within the predetermined period, including: Obtaining pollution emission information, wherein the pollution emission information includes gas type, emission amount and emission period; Based on a preset factor space, gas diffusion interference analysis is performed on multiple gas types respectively to determine multiple high-frequency interference factor sets, wherein the preset factor space includes at least temperature, humidity, wind speed, wind direction and light; An intersection operation is performed on the multiple high-frequency interference factor sets to obtain a predetermined interference factor, and diffuse interference factors of the target area within a predetermined period are collected based on the predetermined interference factor.
3. The method for monitoring the chemical park environment based on the Internet of Things according to claim 1, characterized in that: Multiple simulated pollution distribution atlases are obtained, including: Dividing the predetermined period according to a predetermined step length to determine a plurality of predetermined intervals; A first pollution emission information is randomly selected from the multiple pollution emission information, and the first pollution emission information is input into the regional emission simulation model to perform gas pollution simulation on the target area within multiple predetermined intervals to obtain a first simulated pollution distribution atlas, which is added to the multiple simulated pollution distribution atlases, wherein the simulated pollution distribution map includes a gas simulation diffusion range and a gas simulation concentration.
4. The method for chemical park environment monitoring based on Internet of Things according to claim 3, characterized in that: Get an optimized monitoring array, including: Selecting a first key distribution map in the first simulated pollution distribution map set, wherein the first key distribution map is a distribution map with the largest gas pollution range and the highest concentration in the first simulated pollution distribution map set; Dividing the first key distribution map into regions according to the pollution concentration deviation step length to obtain a first distribution sub-region set and a first regional concentration interval set; According to a predetermined layout strategy, the first monitoring array is optimized according to the first distribution sub-region set and the first regional concentration interval set to obtain a first optimized monitoring array, which is added to the optimized monitoring array.
5. The method for monitoring the chemical park environment based on the Internet of Things according to claim 1, characterized in that: Generate monitoring and early warning signals, including: Randomly selecting a first predicted pollution distribution of a first gas type from the predicted pollution distribution, and obtaining a node simulation pollution distribution diagram of the first gas type at an end node of a predetermined time window; A deviation traversal comparison is performed on the first predicted pollution distribution and the node simulated pollution distribution map. If the pollution range of the first predicted pollution distribution is larger than the pollution range of the node simulated pollution distribution map and / or the pollution concentration of the first predicted pollution distribution is greater than the pollution concentration of the node simulated pollution distribution map in the same area, a first warning signal is generated and added to the monitoring warning signal.
6. The chemical park environment monitoring system based on the Internet of Things is characterized by: The steps for implementing the chemical park environment monitoring method based on the Internet of Things as described in any one of claims 1 to 5 include: The air pollution analysis module is used to call the park production plan within a predetermined period to perform air pollution analysis and obtain multiple pollution emission information; A regional emission simulation model generation module is used to build an emission simulation space based on the park emission layout information, collect diffusion interference factors of the target area within a predetermined period, render the diffusion interference factors to the emission simulation space, and generate a regional emission simulation model; An optimized monitoring array acquisition module is used to input the plurality of pollution emission information into the regional emission simulation model, perform gas pollution simulation in the target area, obtain a plurality of simulated pollution distribution atlases, optimize the layout of the initial monitoring array according to the plurality of simulated pollution distribution atlases, and obtain an optimized monitoring array; A monitoring data dimension reduction module, used for performing fixed-point monitoring of a target area through the optimized monitoring array, reducing the dimension of a monitoring data set at an edge layer, and inputting the reduced dimension data set into a cloud platform, wherein the edge layer is embedded in the optimized monitoring array; The park emission control module is used to predict regional pollution within a predetermined time window on the cloud platform based on the reduced dimension data set. If the predicted pollution distribution does not satisfy the simulated pollution distribution map under the same node, a monitoring and early warning signal is generated, and emission control is performed on the target chemical park based on the monitoring and early warning signal.
