AI-Enabled Data Processing Method and Server for Environmental Monitoring

Abstract

The present invention provides a data processing method and a server for AI-based environmental monitoring, which relates to the field of artificial intelligence technology. First, an environmental data set including air quality, noise intensity, and meteorological parameters collected in real time by multiple sensor nodes in a target monitoring area is obtained. Monitoring sub-regions are dynamically divided according to the data spatial distribution characteristics, and data feature extraction rules are assigned to each sub-region. By this means, abnormal fluctuation indicators and region correlation features are extracted to generate a sub-region feature set. A pre-trained dynamic anomaly analysis model is used to perform multi-dimensional correlation analysis on the abnormal fluctuation indicators to determine the types and influence ranges of environmental anomaly events, generate an environmental optimization strategy set, execute the environmental optimization strategy set and send control instructions, and then update the model parameter weights according to the feedback data, so as to realize intelligent processing and analysis of environmental data, effectively cope with environmental anomalies, and improve the level of environmental monitoring and governance.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence technology. Specifically, it relates to a data processing method and a server for AI-based environmental monitoring. Background Art

[0002] With the continuous improvement of people's attention to environmental quality, environmental monitoring plays an increasingly important role in multiple fields such as environmental protection, urban planning, and public health. Environmental monitoring aims to obtain environmental information in real time and accurately, so as to timely discover environmental problems and take corresponding measures.

[0003] Existing methods mostly use simple threshold judgment or pre-set fixed analysis models to identify environmental anomalies. This processing method lacks in-depth exploration of the complex relationships of environmental data and is difficult to accurately distinguish different types of environmental anomaly events and their influence ranges. For example, in the face of complex environmental problems caused by the interaction of multiple environmental factors, traditional methods may not be able to accurately judge the specific roles and contributions of each factor in the anomaly event, resulting in the inability to formulate effective countermeasures.

[0004] Regarding the division of environmental monitoring areas, existing technologies usually adopt a fixed geographical area division method without considering the spatial distribution characteristics of environmental data itself. This fixed division method cannot adapt to the diversity and variability of environmental data in different regions, resulting in the use of unified rules in the process of extracting data features in different regions, and the extracted data features cannot accurately reflect the actual environmental conditions of each region, reducing the accuracy and effectiveness of data analysis.

[0005] In addition, traditional environmental monitoring and processing systems often lack a feedback mechanism for the effect of environmental regulation. After implementing environmental optimization strategies, it is impossible to adjust and optimize the monitoring and analysis models in real time according to the actual regulation effect, making it difficult for the system to continuously and effectively respond to changing environmental problems. Summary of the Invention

[0006] In view of the above-mentioned problems, in combination with the first aspect of the present invention, embodiments of the present invention provide a data processing method for AI-based environmental monitoring, and the method includes:

[0007] Obtain an environmental data set collected in real time by multiple sensor nodes in a target monitoring area, where the environmental data set includes air quality parameters, noise intensity parameters, and meteorological parameters in a time series;

[0008] According to the spatial distribution characteristics of the environmental data set, dynamically divide the target monitoring area into multiple monitoring sub-areas, and assign corresponding data feature extraction rules to each monitoring sub-area;

[0009] Based on the data feature extraction rules, abnormal fluctuation indicators and regional correlation features are extracted from the environmental data of each monitoring sub-region to generate a sub-region feature set;

[0010] Using the pre-trained dynamic anomaly analysis model, a multi-dimensional correlation analysis is performed on the abnormal fluctuation indicators in the sub-region feature set to determine the type and impact range of environmental abnormal events in the target monitoring area, and generate an environmental optimization strategy set;

[0011] At least one environment optimization strategy in the environment optimization strategy set is executed, a control instruction is sent to a target execution device, and a parameter weight of the dynamic anomaly analysis model is updated based on feedback data of the control instruction.

[0012] On the other hand, an embodiment of the present invention further provides a server, including a processor and a machine-readable storage medium, wherein the machine-readable storage medium is connected to the processor, the machine-readable storage medium is used to store programs, instructions or codes, and the processor is used to execute the programs, instructions or codes in the machine-readable storage medium to implement the above method.

[0013] Based on the above aspects, the embodiment of the present application first obtains the environmental data set containing air quality parameters, noise intensity parameters and meteorological parameters collected in real time by multiple sensor nodes in the target monitoring area, and then dynamically divides the monitoring sub-areas according to the spatial distribution characteristics of the environmental data set, and assigns corresponding data feature extraction rules to each sub-area, which can flexibly adapt to the differences in environmental data in different regions, can accurately capture the unique environmental change laws of each sub-area, and improve the accuracy and effectiveness of data feature extraction. Furthermore, based on targeted data feature extraction rules, abnormal fluctuation indicators and regional correlation features are extracted from the environmental data of each monitoring sub-area and a sub-area feature set is generated, which can not only keenly perceive the abnormal fluctuations in environmental data, but also deeply explore the potential correlation between different regions. Then, the pre-trained dynamic anomaly analysis model is used to perform multi-dimensional correlation analysis on the abnormal fluctuation indicators, which can comprehensively consider the impact of multiple factors on environmental anomalies, not only limited to simple causal judgment, but also analyze the causes and development trends of environmental abnormal events from multiple angles and multiple levels, so as to accurately determine the types and scope of environmental abnormal events in the target monitoring area, and the environmental optimization strategy set generated on this basis improves the pertinence and effectiveness of environmental governance and optimization. Finally, at least one environmental optimization strategy in the environmental optimization strategy set is executed, and the parameter weights of the dynamic anomaly analysis model are updated based on the feedback data of the control instructions. The analysis model can be adjusted in real time according to the actual control effect, and the model's analysis and prediction capabilities for environmental anomalies can be continuously improved, so that the entire environmental monitoring and optimization system can continuously adapt to complex and changeable environmental conditions and achieve long-term stable and efficient operation. Brief Description of the Drawings

[0014] Figure 1 It is a schematic execution flow diagram of the data processing method based on AI environmental monitoring provided by an embodiment of the present invention.

[0015] Figure 2 It is a schematic diagram of the server hardware architecture provided by an embodiment of the present invention. Detailed Embodiments

[0016] The present invention will be specifically described below in conjunction with the accompanying drawings of the specification. Figure 1 It is a schematic flow diagram of the data processing method based on AI environmental monitoring provided by an embodiment of the present invention. The data processing method based on AI environmental monitoring will be introduced in detail below.

[0017] Step S110: Obtain an environmental data set collected in real time by multiple sensor nodes in a target monitoring area. The environmental data set includes air quality parameters, noise intensity parameters, and meteorological parameters in a time series.

[0018] In this embodiment, taking a target monitoring area as an industrial park as an example, there are many factory workshops, warehouses, office buildings, and some green areas in the industrial park. In this industrial park, multiple sensor nodes are set up, and each sensor node can be installed at different positions. For example, near the exhaust port of the factory workshop, at the entrance of the warehouse, on the roof of the office building, and in the middle of the green area, etc.

[0019] Each sensor node is used to collect environmental data in real time. Taking air quality parameters as an example, the sensor node can detect the concentration of various pollutants in the air, such as sulfur dioxide (SO2), nitrogen oxides (NOx), particulate matter (PM2.5, PM10), etc., and record the data at regular time intervals, such as once every 5 minutes, to form a time series. In terms of noise intensity parameters, the sensor node can measure the noise decibel value of the surrounding environment, and the meteorological parameters also change with time, reflecting the impact of production activities and traffic flow in the park at different times on the noise environment. The meteorological parameters include temperature, humidity, wind direction, wind speed, etc. For example, wind direction and wind speed will affect the propagation direction and speed of pollutants emitted by factories in the park. Thus, through continuous data collection by these sensor nodes, an environmental data set containing air quality parameters, noise intensity parameters, and meteorological parameters in a time series is obtained.

[0020] Step S120: Dynamically divide the target monitoring area into multiple monitoring sub-areas according to the spatial distribution characteristics of the environmental data set, and assign corresponding data feature extraction rules to each monitoring sub-area.

[0021] In the above industrial park scenario, from the collected environmental data set, first extract the position coordinates and parameter change gradients of each sensor node. For example, for a sensor node near the exhaust outlet of a factory workshop, the change gradient of its air quality parameters may be relatively large, especially during the peak production period when pollutant emissions increase, while the change gradient of air quality parameters for sensor nodes in green areas is relatively small. Thus, a spatial distribution matrix can be constructed based on the above information.

[0022] Then, perform density clustering analysis on the spatial distribution matrix. Identify the parameter mutation boundaries within the industrial park. For example, around a newly put into production highly polluting workshop, the air quality parameters suddenly deteriorate, thus forming a parameter mutation boundary. At the same time, determine the steady-state regions. For example, the air quality parameters in green areas are relatively stable and belong to the steady-state regions.

[0023] According to the geometric shape of the parameter mutation boundary (assuming the mutation boundary around this new workshop is approximately circular because the workshop is a relatively regular building and pollutants spread out centered on the workshop) and the parameter similarity of the steady-state regions (the air quality parameter similarities detected by sensor nodes in green areas are relatively high), generate a dynamic grid segmentation scheme covering the entire industrial park. Merge consecutive grid cells in the dynamic grid segmentation scheme with parameter similarities higher than a preset threshold (assuming this threshold is set at 90%) into the same monitoring sub-region. For example, merge the grid cells of several adjacent green areas into one monitoring sub-region, labeled as "Green Area Monitoring Sub-region". For each monitoring sub-region, label the spatial attribute tags, such as "High Pollution Risk Area around Workshop", "Low Pollution Area in Office Area", etc.

[0024] According to the environmental parameter types corresponding to the spatial attribute tags, match the noise filtering thresholds and feature fusion weights in the data feature extraction rules from the preset rule library. For example, for the "High Pollution Risk Area around Workshop", since the air quality parameters fluctuate greatly here and the noise mainly comes from workshop equipment, the noise filtering threshold is set relatively high to avoid misjudging the noise generated by the normal operation of the workshop as abnormal; while for the "Low Pollution Area in Office Area", the noise filtering threshold can be relatively low because it is relatively quiet here and even small noise changes may need attention. At the same time, according to different environmental parameter types, determine the feature fusion weights. For example, in air quality analysis, for areas with high pollutant concentrations, the weights of certain key pollutants (such as SO2) may be higher.

[0025] Step S130, based on the data feature extraction rules, extract the abnormal fluctuation indicators and regional association features from the environmental data of each monitoring sub-region, and generate a sub-region feature set.

[0026] Taking the monitoring sub - area of "high - pollution - risk area around the workshop" as an example, the time - series environmental data of air quality parameters, noise intensity parameters, and meteorological parameters collected by sensor nodes here are first processed by sliding - window normalization to generate a normalized data sequence. For example, the SO2 concentration data collected by each sensor node are normalized within a sliding window (assumed to be 1 hour) so that the data of different sensor nodes are comparable.

[0027] According to the noise - filtering threshold in the data - feature extraction rule, outliers are removed and data interpolation is performed on the normalized data sequence. Suppose the noise - filtering threshold for this monitoring sub - area is set to a relatively high value. When there are short - term high - noise data points (possibly abnormal noise generated by a short - term equipment failure), if the high - noise data point exceeds the normal fluctuation range and does not conform to the noise characteristics during the normal operation of the workshop, it will be removed, and interpolation will be performed based on the data before and after, thus generating a smooth data sequence.

[0028] Calculate the parameter change rate of adjacent time windows in the smooth data sequence. For example, calculate the change rate of SO2 concentration within two adjacent 1 - hour windows. When the parameter change rate exceeds the upper limit of the historical fluctuation range, the corresponding time window is marked as an abnormal - fluctuation interval. If during a certain period, the production process of the workshop changes or equipment fails, resulting in a sudden increase in SO2 emissions, such an abnormal - fluctuation interval will occur.

[0029] Extract the parameter peak value (such as the maximum value of SO2 concentration), duration (how long this abnormally high concentration lasts), and spatial propagation direction (judge the diffusion direction of pollutants based on the wind direction and the changes in data of surrounding sensor nodes) from this abnormal - fluctuation interval to generate abnormal - fluctuation indicators.

[0030] Match the propagation directions of all abnormal - fluctuation indicators in the same monitoring sub - area with those of adjacent sub - areas. For example, the pollutants in the high - pollution - risk area around the workshop may spread to the adjacent office area. By analyzing the propagation directions and change trends of abnormal - fluctuation indicators in the two areas, a cross - regional association path is generated, and its topological structure serves as the regional - association feature. If it is found that the pollutants emitted by the workshop tend to spread to the office area, then this cross - regional association path reflects this potential pollution - propagation relationship.

[0031] Step S140, using the pre - trained dynamic anomaly - analysis model, perform multi - dimensional correlation analysis on the abnormal - fluctuation indicators in the sub - area feature set, determine the types and influence scopes of environmental - anomaly events in the target monitoring area, and generate a set of environmental - optimization strategies.

[0032] In the industrial park scenario, using a pre-trained dynamic anomaly analysis model, the parameter peaks in the anomaly fluctuation indicators obtained from monitoring sub-regions such as the "high-pollution risk area around the workshop" are pattern-matched with the historical environmental event database. Suppose the SO2 concentration peak caused by abnormal emissions in the workshop is very high, similar to the pattern of high-pollution events caused by equipment failures that have occurred in the past. By matching, the potential pollution source type is identified as a workshop equipment failure, along with the corresponding pollution diffusion model (such as a model centered on the workshop and diffusing according to wind direction and speed).

[0033] According to the cross-regional association path in the regional association features, calculate the propagation speed and attenuation coefficient of the pollution diffusion model within the target monitoring area (industrial park). For example, if it is found that pollutants diffuse from the workshop to the office area, calculate the propagation speed as several meters per hour based on the distance between the two areas, wind direction and speed, and the change in pollutant concentration. At the same time, due to the dilution effect of the air, etc., the pollutant concentration will gradually decay, and calculate the attenuation coefficient.

[0034] Combined with the environmental parameter types corresponding to the spatial attribute tags of the monitoring sub-regions, predict the superimposed influence degree of the pollution diffusion model on air quality parameters and meteorological parameters. For the "high-pollution risk area around the workshop" and the "low-pollution area in the office area", considering meteorological parameters such as wind direction and speed, predict the impact of pollutants emitted from the workshop on the air quality in the office area, such as how long it will take for the SO2 concentration in the office area to rise to what level, and the possible minor impacts on meteorological parameters such as temperature and humidity within the park.

[0035] Based on the spatial coordinates of the monitoring sub-regions where the superimposed influence degree exceeds the preset safety threshold (such as the SO2 concentration in the office area exceeding a certain standard), generate an environmental anomaly event type that includes the emergency control priority. If the pollution level in the office area is about to exceed the safety standard, define it as a high-urgency environmental anomaly event, and the event type is the diffusion of workshop pollution to the office area.

[0036] According to the propagation speed and attenuation coefficient of the pollution diffusion model, determine the boundary of the influence range and the predicted value of the duration of the environmental anomaly event. For example, predict which areas of the office area the pollutants will diffuse to in the next few hours, and the possible duration of the entire pollution event.

[0037] According to the pollution source type corresponding to the environmental anomaly event type (workshop equipment failure), select the type of the target execution device and the range of regulation parameters from the equipment control strategy library. For example, for such a workshop pollution event, the target execution device may be the waste gas treatment equipment in the workshop, and the range of regulation parameters may be to increase the power of the treatment equipment, adjust the dosage of the treatment agent, etc. Based on the spatial coordinates of the influence range boundary (the affected area in the office area) and the emergency regulation priority (high urgency), allocate the start time sequence (such as immediately starting the waste gas treatment equipment) and the parameter adjustment gradient (such as how much the power should be increased within a certain period of time) for each target execution device. Combine the predicted duration value and the wind direction and wind speed data in the meteorological parameters to calculate the expected coverage rate of the parameter adjustment gradient of the target execution device within the influence range. If the expected coverage rate is lower than the preset efficiency threshold (for example, the treatment capacity of the waste gas treatment equipment cannot control the pollution in the office area within the safe range within the specified time), add the location deployment suggestions of the auxiliary regulation equipment (such as adding air purification equipment near the office area) and the collaborative working mode (such as how the purification equipment and the waste gas treatment equipment cooperate) to the target execution device. Package the regulation parameter range, start time sequence, and auxiliary regulation equipment deployment suggestions of all target execution devices into an executable instruction set to generate an environmental optimization strategy set.

[0038] Step S150, execute at least one environmental optimization strategy in the environmental optimization strategy set, send a regulation instruction to the target execution device, and update the parameter weights of the dynamic anomaly analysis model based on the feedback data of the regulation instruction.

[0039] Taking the workshop waste gas treatment equipment as an example for the previously generated environmental optimization strategy set. According to the type of the target execution device (waste gas treatment equipment), convert the executable instruction set into an instruction format compatible with the equipment control protocol. For example, convert the instructions for regulating the power and chemical agent dosage of the waste gas treatment equipment into a specific format that the equipment can recognize.

[0040] Send an initialization instruction to the target execution device according to the start time sequence, and receive the status confirmation signal returned by the target execution device. Assume that after sending the instruction to start the waste gas treatment equipment, the equipment returns a signal indicating that it is ready.

[0041] When the status confirmation signal indicates that the target execution device is ready, send regulation instructions in stages according to the parameter adjustment gradient, and monitor the actual parameter values fed back by the target execution device in real time. For example, gradually increase the power of the waste gas treatment equipment according to the set gradient, and at the same time obtain the actual power value, the concentration of pollutants after treatment and other actual parameter values fed back by the equipment in real time.

[0042] Calculate the deviation between the actual parameter value and the parameter target value corresponding to the expected coverage rate. If the deviation exceeds the fault tolerance range, trigger the collaborative instruction of the auxiliary control device. If the actual treatment effect of the waste gas treatment device fails to meet the expectation, that is, the reduction range of the actual pollutant concentration is less than the expectation, then trigger the collaborative working instruction of the air purification device near the office area.

[0043] Record the execution timestamp of the control instruction, the device response delay, and the number of parameter calibrations, and generate a device execution efficiency report. For example, record the time delay from when the waste gas treatment device receives the instruction to when it starts to adjust the power, and the number of times of parameter calibration (such as adjusting the chemical agent dosage) to achieve a better treatment effect, etc.

[0044] Extract the correlation between the number of parameter calibrations and the expected coverage rate from the device execution efficiency report, and construct a strategy execution effect evaluation matrix. Calculate the time prediction error coefficient of the dynamic anomaly analysis model according to the difference between the actual duration and the predicted value of the environmental anomaly event. Assume that the originally predicted pollution event lasted for 3 hours, but actually lasted for 4 hours, and calculate the corresponding error coefficient.

[0045] Input the strategy execution effect evaluation matrix and the time prediction error coefficient into the model weight optimizer to generate the parameter weight adjustment gradient. Use the backpropagation algorithm to iteratively update the weights of the fully connected layer of the dynamic anomaly analysis model until the time prediction error coefficient drops below the convergence threshold. Synchronize the updated model parameter weights to all copies of the dynamic anomaly analysis model corresponding to the monitored sub-regions, so as to be able to more accurately predict and respond to environmental anomaly events in subsequent environmental monitoring and analysis.

[0046] Based on the above steps, in the embodiment of the present application, first, an environmental data set including air quality parameters, noise intensity parameters, and meteorological parameters collected in real time by multiple sensor nodes in the target monitoring area is obtained. Then, according to the spatial distribution characteristics of the environmental data set, the monitoring sub-areas are dynamically divided, and corresponding data feature extraction rules are assigned to each sub-area, which can flexibly adapt to the differences in environmental data in different areas, accurately capture the unique environmental change laws of each sub-area, and improve the accuracy and effectiveness of data feature extraction. Furthermore, based on the targeted data feature extraction rules, the abnormal fluctuation indicators and regional correlation features are extracted from the environmental data of each monitoring sub-area, and a sub-area feature set is generated, which can not only keenly detect the abnormal fluctuations in the environmental data, but also deeply explore the potential correlation relationships between different areas. Then, a multi-dimensional correlation analysis is performed on the abnormal fluctuation indicators by using a pre-trained dynamic anomaly analysis model, which can comprehensively consider the impacts of various factors on environmental anomalies, not limited to simple causal relationship judgments, but analyze the causes and development trends of environmental anomaly events from multiple angles and levels, so as to accurately determine the types and influence ranges of environmental anomaly events in the target monitoring area. On this basis, an environmental optimization strategy set is generated, which improves the pertinence and effectiveness of environmental governance and optimization. Finally, at least one environmental optimization strategy in the environmental optimization strategy set is executed, and the parameter weights of the dynamic anomaly analysis model are updated based on the feedback data of the control instructions, which can adjust the analysis model in real time according to the actual control effects, continuously improve the model's analysis and prediction capabilities for environmental anomalies, so that the entire environmental monitoring and optimization system can continuously adapt to the complex and changeable environmental conditions and achieve long-term stable and efficient operation.

[0047] In a possible implementation manner, step S120 includes:

[0048] Step S121, extracting the position coordinates and parameter change gradients of each sensor node from the environmental data set, and constructing a spatial distribution matrix.

[0049] In an industrial park, sensor nodes are distributed at various key positions, and each sensor node has its accurate position coordinates. For example, a certain sensor is located near the exhaust port of Workshop A, and its coordinates are (x1, y1, z1), and another sensor located at the entrance of Warehouse B has coordinates (x2, y2, z2), etc. The parameter change gradient reflects the change trend of environmental parameters in space. Taking the sulfur dioxide concentration in air quality parameters as an example, the sulfur dioxide concentration of the sensor near the exhaust port decreases with the increase of distance due to being close to the pollution source, and this rate of change of concentration with distance is the parameter change gradient. By integrating the position coordinates and parameter change gradients of all sensor nodes, a spatial distribution matrix is constructed, which comprehensively reflects the spatial distribution of environmental parameters in the industrial park.

[0050] Step S122: Conduct density clustering analysis on the spatial distribution matrix to identify the parameter mutation boundaries and steady-state regions within the target monitoring area.

[0051] Due to the differences in production activities and environmental conditions in different regions, there are parameter mutation boundaries and steady-state regions. For example, around newly put into use highly polluting workshops, due to the concentrated emission of pollutants, the air quality parameters will mutate. Through density clustering analysis, such regions can be clearly demarcated. The air quality parameters detected by sensors around the workshop are significantly different from those in farther regions, and the edge of this difference forms the parameter mutation boundary. While in green areas, the air quality parameters are relatively stable, and the data detected by each sensor fluctuates less, and this area is identified as a steady-state region.

[0052] Step S123: Generate a dynamic grid segmentation scheme covering the target monitoring area according to the geometric shape of the parameter mutation boundary and the parameter similarity of the steady-state regions.

[0053] Step S124: Merge the continuous grid cells with parameter similarity higher than the preset threshold in the dynamic grid segmentation scheme into the same monitoring sub-area, and label each monitoring sub-area with a spatial attribute label.

[0054] Suppose the preset threshold is 90%. If several adjacent grid cells have a similarity of air quality parameters, noise intensity parameters, and meteorological parameters higher than 90% and these grid cells are continuous, then they are merged into one monitoring sub-area. For example, in the green area, multiple continuous grid cells with high parameter similarity are merged into one monitoring sub-area and labeled as "Green Area Monitoring Sub-area". For the area with heavy pollution and similar parameters around the workshop, after merging, it is labeled as "High Pollution Monitoring Sub-area around Workshop", etc.

[0055] Step S125: Match the noise filtering threshold and feature fusion weight in the data feature extraction rule from the preset rule library according to the environmental parameter type corresponding to the spatial attribute label.

[0056] For example, for the "high-pollution monitoring sub-region around the workshop", since there is more noise generated by production activities here and the air quality parameters fluctuate greatly, a relatively high noise filtering threshold is matched from the preset rule library to avoid misjudging the noise generated by normal production activities as abnormal conditions. At the same time, for air quality analysis, according to the pollution characteristics of this region, the characteristic fusion weights are determined. For example, higher weights are assigned to the detection data of major pollutants such as sulfur dioxide and nitrogen oxides because these pollutants are emitted more during the workshop production process and have a greater impact on the environment. For the "green area monitoring sub-region", since the environment is relatively quiet and the air quality is good, a lower noise filtering threshold is matched, and in the analysis of meteorological parameters, the weights of parameters such as temperature and humidity may be relatively high because the microclimate in the green area is greatly affected by these meteorological parameters. Thus, appropriate data feature extraction rules are assigned to each monitoring sub-region for more accurate analysis and processing of environmental data in the future.

[0057] Among them, step S123 includes:

[0058] Step S1231, extract the spatial coordinate sequence of the boundary points from the parameter mutation boundary, and identify the geometric shape type of the boundary line according to the continuity of the spatial coordinate sequence.

[0059] Step S1232, divide the boundary line into multiple geometric feature segments according to the curvature change characteristics of the geometric shape type, and label the curvature change direction for each geometric feature segment.

[0060] Step S1233, based on the curvature change direction, generate initial grid cells on both sides of the boundary line, and match the side length of the initial grid cells with the parameter similarity of the steady state region to determine the side length adjustment ratio of the initial grid cells.

[0061] Step S1234, expand or contract the initial grid cells according to the side length adjustment ratio to generate boundary grid cells adapted to the geometric shape of the boundary line.

[0062] Step S1235, extract the group of sensor nodes with the highest parameter similarity from the steady state region, and determine the grid cell division density of the steady state region according to the spatial distribution density of the sensor node group.

[0063] Step S1236, based on the grid cell division density, generate uniformly distributed steady state grid cells in the steady state region, and align the side length of the steady state grid cells with the side length of the boundary grid cells to generate transition grid cells.

[0064] Step S1237, according to the side length change trend of the transition grid cells, seamlessly connect the boundary grid cells and the steady-state grid cells to generate a dynamic grid segmentation scheme covering the target monitoring area.

[0065] Suppose there is a parameter mutation boundary around a highly polluted workshop, and the spatial coordinate sequence of its boundary points shows a relatively regular circular feature because the structure of the workshop is relatively regular and pollutants diffuse from the workshop to the surrounding areas. Then, according to the curvature change characteristics of this circular geometric shape type, divide the boundary line into multiple geometric feature segments and label the curvature change direction for each geometric feature segment. For example, on a circular boundary, divide the circumference into several segments according to angles, and label whether the curvature change direction of each segment is inward or outward. Based on the curvature change direction, generate initial grid cells on both sides of the boundary line. The side length of the initial grid cells needs to be matched with the parameter similarity of the steady-state area to determine the side length adjustment ratio. Suppose the parameter similarity of the steady-state area is relatively high, indicating that the environmental parameters in this area are relatively stable. If the side length of the initial grid cells is too large, some local minor changes may be ignored. Therefore, the side length needs to be adjusted according to the parameter situation of the steady-state area. If the sensor nodes in the steady-state area are densely distributed and the parameter similarity is high, it may be necessary to reduce the side length of the initial grid cells. According to the side length adjustment ratio, expand or contract the initial grid cells to generate boundary grid cells adapted to the geometric shape of the boundary line.

[0066] Extract the sensor node group with the highest parameter similarity from the steady-state area, and determine the grid cell division density of the steady-state area according to the spatial distribution density of the sensor node group. In the green area, which is a steady-state area, find the sensor node group with the highest similarity in air quality parameters, noise intensity parameters, and meteorological parameters. If these sensor nodes are densely distributed, it means that the environmental conditions in this area change little within a small range. Then, the grid cell division density of the steady-state area can be appropriately increased, which can not only ensure effective monitoring of the environment but also reduce the calculation amount. Based on this grid cell division density, generate uniformly distributed steady-state grid cells in the steady-state area, and align the side lengths of the steady-state grid cells with those of the boundary grid cells to generate transition grid cells. For example, at the junction of the green area and the area around the workshop, by adjusting the side lengths of the steady-state grid cells and the boundary grid cells, make the two transition smoothly to generate transition grid cells. According to the side length change trend of the transition grid cells, seamlessly connect the boundary grid cells and the steady-state grid cells to generate a dynamic grid segmentation scheme covering the entire industrial park. This dynamic grid segmentation scheme can accurately reflect the environmental characteristics of different areas in the industrial park, taking into account both the complex situation of the mutation area and the relative stability of the steady-state area.

[0067] Step S1238: Map the spatial coordinate sequences of the boundary grid cells, transition grid cells, and steady-state grid cells in the dynamic grid segmentation scheme to the parameter similarity of the parameter mutation boundary and the steady-state region to generate grid cell attribute tags.

[0068] For example, since the boundary grid cells are close to the parameter mutation boundary, their attribute tags may contain information such as "high pollution risk boundary", and at the same time, mark their association degree with the parameter mutation boundary and the internal parameter similarity range. The attribute tags of the transition grid cells reflect their characteristics during the transition in different regions, such as "transition from high pollution area to low pollution area" and the corresponding parameter changes. The attribute tags of the steady-state grid cells reflect the environmental parameter characteristics of the steady-state region where they are located, such as "low pollution steady state in green area", etc.

[0069] Step S1239: According to the grid cell attribute tags, cluster the grid cells in the dynamic grid segmentation scheme according to the parameter similarity to generate multiple grid cell groups, and label each grid cell group with a spatial attribute tag. Based on the spatial attribute tag, match the grid cell groups in the dynamic grid segmentation scheme with the spatial distribution characteristics of the target monitoring area to generate a dynamic grid segmentation scheme covering the target monitoring area.

[0070] In this embodiment, grid cells with similar parameter similarities can be grouped together. For example, all grid cells located in the high pollution risk area around the workshop and with high parameter similarities are grouped together and labeled as "high pollution risk group around the workshop". The grid cells within this group are similar in terms of the change trend of environmental parameters, pollution degree, etc. For other groups, such as the low pollution group in the office area, they are also labeled according to their respective characteristics.

[0071] Then, each grid cell group can be corresponding to the actual area in the industrial park to ensure that the grid segmentation scheme can accurately reflect the actual spatial distribution of the industrial park. For example, the grid cells in the high pollution risk group around the workshop cover all areas around the workshop that may be affected by pollution, and the grid cells in the low pollution group in the office area accurately correspond to the office area, etc. In this way, a complete dynamic grid segmentation scheme that conforms to the environmental characteristics of the industrial park is generated.

[0072] In a possible implementation manner, step S130 includes:

[0073] Step S131: Perform sliding window normalization processing on the time-series environmental data of all sensor nodes in the monitoring sub-region to generate a normalized data sequence.

[0074] Taking the "high-pollution monitoring sub-region around the workshop" in the industrial park as an example, various sensor nodes in this sub-region continuously collect environmental data such as air quality parameters, noise intensity parameters, and meteorological parameters. When processing this time-series environmental data, a sliding window normalization processing method is adopted. For example, for the sulfur dioxide concentration data in the air quality parameters, a sliding window time length of one hour is set. Within this one hour, the sulfur dioxide concentration values detected by each sensor node are collected, and through a specific normalization algorithm, these concentration values of different magnitudes are converted into a normalized data sequence, which makes the data of different sensor nodes comparable and convenient for subsequent analysis and processing. Similarly, for the noise intensity parameters and meteorological parameters, such as noise decibel values, temperature, humidity, etc., the same sliding window normalization processing is also adopted to obtain a complete normalized data sequence within the entire monitoring sub-region.

[0075] Step S132, according to the noise filtering threshold in the data feature extraction rule, eliminate outliers and perform data interpolation on the normalized data sequence to generate a smooth data sequence.

[0076] In the "high-pollution monitoring sub-region around the workshop", due to the noise and pollutant emissions generated by the workshop production activities having certain rules, corresponding noise filtering thresholds are set in the data feature extraction rule. For example, for the noise intensity parameter, if the noise decibel value detected at a certain moment suddenly exceeds the noise range that may be generated by normal production activities, and this exceeded value is greater than the noise filtering threshold, this data point is determined as an outlier. In the sulfur dioxide concentration data, if a sudden extremely high value appears and does not conform to the fluctuation range of normal workshop emissions, it is also determined as an outlier based on the noise filtering threshold. For these outliers, elimination processing is performed. Then, data interpolation is performed according to the data before and after to ensure the integrity of the data sequence. Thus, outliers are eliminated and data interpolation is performed on the air quality parameters, noise intensity parameters, and meteorological parameters in the entire normalized data sequence, and finally a smooth data sequence is generated, which can more accurately reflect the true change trend of the environmental parameters in the monitoring sub-region.

[0077] Step S133, calculate the parameter change rate of adjacent time windows in the smooth data sequence, and mark the time window whose parameter change rate exceeds the upper limit of the historical fluctuation range as an abnormal fluctuation interval.

[0078] When processing the smoothed data sequence of the "high-pollution monitoring sub-region around the workshop", for each environmental parameter, calculate the parameter change rate of adjacent time windows. For example, for the sulfur dioxide concentration, compare the concentration change rate within adjacent one-hour sliding windows. Calculate the concentration value of this hour with the concentration value of the previous hour to obtain the change ratio of the concentration. Refer to the fluctuation range of the sulfur dioxide concentration in the historical data of this sub-region. If the parameter change rate of a certain adjacent time window exceeds the upper limit of the historical fluctuation range, mark this time window as an abnormal fluctuation interval. The same method is used for calculating and marking the noise intensity parameter and meteorological parameters. If a production equipment in the workshop fails or the production process is temporarily adjusted, it may cause a sudden change in the pollutant emission rate or a sudden increase in the noise intensity. These situations will be reflected in the parameter change rate and thus be marked as abnormal fluctuation intervals.

[0079] Step S134, extract the parameter peak value, duration, and spatial propagation direction within the abnormal fluctuation interval to generate the abnormal fluctuation index.

[0080] Within the already marked abnormal fluctuation interval, taking the sulfur dioxide concentration as an example, extract the maximum value of the sulfur dioxide concentration within this abnormal fluctuation interval, and this maximum value is the parameter peak value. At the same time, determine the duration of this abnormal fluctuation, that is, the time elapsed from when the concentration starts to exceed the normal fluctuation range to when it returns to the normal range. For the determination of the spatial propagation direction, combine the positions of each sensor node within the sub-region and the concentration change trend to judge. If the sensor near the exhaust port of the workshop first detects an increase in the sulfur dioxide concentration, and then, along with the wind direction, the sensors in the downwind direction also successively detect an increase in the concentration, then it can be determined that the spatial propagation direction of the pollutant is to diffuse from the exhaust port along the wind direction to the downwind direction. For the abnormal fluctuation interval of the noise intensity, similarly extract the maximum value of the noise decibels, duration, and propagation direction (if the noise source has an obvious propagation directionality). These information such as the parameter peak value, duration, and spatial propagation direction constitute the abnormal fluctuation index.

[0081] Step S135, perform similarity matching on the propagation directions of all abnormal fluctuation indexes within the same monitoring sub-region with the abnormal fluctuation indexes of adjacent sub-regions to generate a cross-region association path, and use the topological structure of the cross-region association path as the regional association feature.

[0082] In the industrial park, pollutants in the "high-pollution monitoring sub-region around the workshop" may spread to the adjacent "low-pollution monitoring sub-region of the office area". The similarity between the abnormal fluctuation indicators (such as the propagation direction of pollutants) in the "high-pollution monitoring sub-region around the workshop" and those in the "low-pollution monitoring sub-region of the office area" is matched. If the pollutants emitted by the workshop spread towards the office area along the wind direction, and the trend of pollutant concentration detected in the office area has a certain similarity with the concentration change trend around the workshop in terms of direction and time, then it can be determined that there is a cross-regional association path. The topological structure of this cross-regional association path reflects the propagation relationship of environmental anomalies between different monitoring sub-regions, such as whether it is one-way propagation or multi-way propagation, and whether the propagation path is straight or curved. This topological structure is regarded as the regional association feature.

[0083] In a possible implementation manner, step S140 includes:

[0084] Step S141, using a pre-trained dynamic anomaly analysis model, perform pattern matching between the parameter peaks in the abnormal fluctuation indicators and the historical environmental event database to identify the potential pollution source type and pollution diffusion model.

[0085] For example, for the abnormal fluctuation indicators obtained from sub-regions such as the "high-pollution monitoring sub-region around the workshop", input the parameter peaks (such as the peak concentration of sulfur dioxide) into the pre-trained dynamic anomaly analysis model. This dynamic anomaly analysis model performs pattern matching between the parameter peaks and the data in the historical environmental event database. If the current peak concentration of sulfur dioxide is similar to the peak concentration in the high-pollution event caused by workshop equipment failure in history, then the potential pollution source type can be identified as workshop equipment failure. At the same time, according to the pollutant diffusion situation recorded in historical events, determine the corresponding pollution diffusion model. For example, when the workshop equipment fails, the pollutants spread around the workshop as the center according to a certain wind direction and wind speed, and this diffusion pattern is determined as the pollution diffusion model for the current event.

[0086] Step S142, according to the cross-regional association path in the regional association feature, calculate the propagation speed and attenuation coefficient of the pollution diffusion model in the target monitoring region.

[0087] For example, based on the cross-regional association path from the "high-pollution monitoring sub-region around the workshop" to the "low-pollution monitoring sub-region in the office area" determined previously, the propagation speed of the pollution diffusion model within the industrial park is calculated by combining the distance between the sub-regions, the wind direction and speed, and the variation of pollutant concentration at different locations. For example, the distance from the exhaust port of the workshop to a certain location in the office area is a fixed value. By detecting the time difference in the change of pollutant concentration between these two locations, as well as meteorological parameters such as wind direction and speed, the diffusion speed of pollutants from the workshop to the office area is calculated. At the same time, due to the dilution, sedimentation, etc. of the air, the pollutant concentration will gradually decay, and the decay coefficient is calculated by analyzing the concentration change at different locations.

[0088] Step S143: Combine the environmental parameter types corresponding to the spatial attribute tags of the monitoring sub-region to predict the superimposed influence degree of the pollution diffusion model on air quality parameters and meteorological parameters.

[0089] For example, for the "high-pollution monitoring sub-region around the workshop" and the "low-pollution monitoring sub-region in the office area", consider the environmental parameter types corresponding to their spatial attribute tags. For example, the "high-pollution monitoring sub-region around the workshop" mainly focuses on the impact of high-pollution emissions on the surrounding air quality, while the "low-pollution monitoring sub-region in the office area" pays more attention to the air quality and comfort of the office environment. According to the pollution diffusion model, predict the superimposed influence degree of the pollutants emitted from the workshop on air quality parameters (such as sulfur dioxide, nitrogen oxides, particulate matter concentration, etc.) after spreading to the office area. At the same time, considering the influence of meteorological parameters (such as temperature, humidity, wind direction and speed), for example, wind direction and speed will affect the diffusion direction and speed of pollutants, and temperature and humidity may affect the chemical reaction and sedimentation speed of pollutants, so as to comprehensively predict the superimposed influence degree on air quality parameters and meteorological parameters.

[0090] Step S144: Generate an environmental anomaly event type including an emergency regulation priority based on the spatial coordinates of the monitoring sub-region where the superimposed influence degree exceeds the preset safety threshold.

[0091] If it is predicted that the air quality parameters in the office area exceed the preset safety threshold due to the superimposed influence of the diffusion of pollutants from the workshop, for example, the sulfur dioxide concentration in the office area exceeds the specified health standard, then according to the spatial coordinates of the office area in the industrial park, generate an environmental anomaly event type including an emergency regulation priority. Since the office area is densely populated and has high requirements for air quality, it is defined as an environmental anomaly event with a high emergency level, and the event type is the diffusion of workshop pollution to the office area. If a green area is affected by a certain degree of pollution but is below the safety threshold, it may be defined as an environmental anomaly event with a low emergency level.

[0092] Step S145: Determine the boundary of the influence range and the predicted value of the duration of the environmental anomaly event according to the propagation speed and attenuation coefficient of the pollution diffusion model.

[0093] Based on the previously calculated propagation speed and attenuation coefficient of the pollution diffusion model, determine the boundary of the diffusion range of pollutants in the industrial park. For example, starting from the workshop, as time goes by, calculate the farthest area that pollutants can spread to within a certain period in the future according to the propagation speed and attenuation coefficient. This area is the boundary of the influence range of the environmental anomaly event. At the same time, predict the duration of the entire environmental anomaly event based on factors such as the pollution diffusion model and meteorological parameters. If the workshop equipment failure continuously generates pollutants, combine factors such as the diffusion speed of pollutants, attenuation coefficient, and the possible time for repairing the workshop equipment to predict the possible duration of this pollution event.

[0094] In a possible implementation manner, step S140 further includes:

[0095] Step S146: Select the type of the target execution device and the range of regulation parameters from the device control strategy library according to the type of pollution source corresponding to the environmental anomaly event type.

[0096] When it is determined that the type of the environmental anomaly event is the pollution in the workshop spreading to the office area, and the corresponding type of pollution source is the workshop equipment failure, screen the target execution device from the device control strategy library. Since the pollution source is the excessive emission of pollutants caused by the workshop equipment failure, the target execution device first considers the waste gas treatment equipment in the workshop. For the waste gas treatment equipment, the range of its regulation parameters needs to be determined according to the specific situation of the workshop equipment failure and the type and concentration of pollutants. For example, if the sulfur dioxide emission exceeds the standard, the range of regulation parameters may involve the dosage of desulfurizer in the waste gas treatment equipment, which may be adjusted from the current X kilograms per hour to Y kilograms per hour. At the same time, the treatment air volume of the equipment may also need to be adjusted, from M cubic meters per hour to N cubic meters per hour. In addition, the regulation of the workshop ventilation equipment may also need to be considered. The range of the wind speed regulation parameter of the ventilation equipment may be adjusted from the current A meters per second to B meters per second to control the diffusion speed of pollutants in the workshop, so as to better cooperate with the waste gas treatment equipment for pollution control.

[0097] Step S147: Allocate the start time sequence and the parameter adjustment gradient for each target execution device based on the spatial coordinates of the influence range boundary and the emergency regulation priority.

[0098] For example, in the scenario where workshop pollution spreads to the office area, the spatial coordinates of the boundary of the affected area determine the areas that need to be key treated, including the office area and the areas on the diffusion path from the workshop to the office area. Since the office area is densely populated and has high air quality requirements, the emergency regulation priorities for waste gas treatment equipment and ventilation equipment are relatively high. For the waste gas treatment equipment, according to the spatial coordinates of the boundary of the affected area, the start time series is determined to be immediate start. In terms of the parameter adjustment gradient, within the first 10 minutes after startup, the dosage of desulfurizer is increased from X kg per hour to (X + ΔX) kg per hour, and the treatment air volume is increased from M cubic meters per hour to (M + ΔM) cubic meters per hour; within the next 20 minutes, the dosage of desulfurizer is further increased to (X + 2ΔX) kg per hour, and the treatment air volume is increased to (M + 2ΔM) cubic meters per hour. For the ventilation equipment, the start time series is also immediate start. Within the first 5 minutes after startup, the wind speed is increased from A meters per second to (A + ΔA) meters per second, and within the next 15 minutes, it is increased to (A + 2ΔA) meters per second. For the air purification equipment (if any) in the area between the workshop and the office area, since its emergency regulation priority is relatively low, the start time series can be set to start 10 minutes after the waste gas treatment equipment starts, and the parameter adjustment gradient is adjusted relatively gently accordingly.

[0099] Step S148, combining the predicted duration value and the wind direction and wind speed data in the meteorological parameters, calculate the expected coverage rate of the parameter adjustment gradient of the target execution device within the affected range.

[0100] For example, it is known that the predicted duration value of the environmental anomaly event that workshop pollution spreads to the office area is T hours, the wind direction in the meteorological parameters is southeast wind, and the wind speed is V meters per second. Based on this information and the parameter adjustment gradient of the target execution device (such as waste gas treatment equipment, ventilation equipment, etc.), calculate the expected coverage rate. Taking the waste gas treatment equipment as an example, according to the adjustment gradient of its treatment air volume, wind direction and wind speed, and the predicted duration value, calculate the proportion of the amount of pollutants that can be treated within T hours in the total amount of pollutants in the entire affected range. Assume that the total amount of pollutants in the affected range is Q kg, and the amount of pollutants treated by the waste gas treatment equipment within T hours according to the adjustment gradient is q kg, then the expected coverage rate is q / Q. For the ventilation equipment, according to the control effect of its wind speed adjustment gradient on pollutant diffusion, calculate the proportion of the area that can effectively prevent pollutant diffusion in the entire affected area within the affected range, which is also taken as part of its expected coverage rate.

[0101] Step S149, when the expected coverage rate is lower than the preset efficiency threshold, add the location deployment suggestions and collaborative working modes of the auxiliary regulation equipment to the target execution device.

[0102] For example, if the calculated expected coverage rates of the waste gas treatment equipment and the ventilation equipment are lower than the preset efficiency threshold, it indicates that it is difficult to effectively control the spread of pollution within the specified time only by these two pieces of equipment. At this time, provide the location deployment suggestions and collaborative working modes of the auxiliary regulation equipment to the target execution equipment. For example, it is recommended to add air purification equipment around the office area, and its location is deployed in the upwind direction of the office area and near the areas with intensive personnel activities. In terms of the collaborative working mode, the air purification equipment works in coordination with the waste gas treatment equipment and the ventilation equipment. The waste gas treatment equipment is responsible for treating the pollutants generated in the workshop, the ventilation equipment controls the diffusion direction and speed of the pollutants in the workshop, and the air purification equipment purifies the pollutants that have spread to the surrounding area of the office area. The operating parameters of the air purification equipment are adjusted according to the operating states of the waste gas treatment equipment and the ventilation equipment. For example, according to the treatment air volume of the waste gas treatment equipment and the wind speed of the ventilation equipment, the purification efficiency of the air purification equipment is adjusted to achieve the best collaborative effect.

[0103] Step S1410, encapsulate the regulation parameter ranges, start time sequences of all target execution equipment, and the deployment suggestions of the auxiliary regulation equipment into an executable instruction set to generate the set of environment optimization strategies.

[0104] For example, integrate the regulation parameter ranges of the waste gas treatment equipment (adjustment ranges of desulfurizer dosage, treatment air volume, etc.), start time sequences (start immediately and subsequent adjustment time points), the regulation parameter ranges of the ventilation equipment (wind speed adjustment range), start time sequences, and the deployment suggestions of the added air purification equipment (location and collaborative working mode), etc. Then, encapsulate this information in a specific format to form an executable instruction set, and this executable instruction set constitutes the set of environment optimization strategies, which covers all the equipment regulation information required to deal with the environmental anomaly event of the pollution in the workshop spreading to the office area.

[0105] In a possible implementation manner, step S150 includes:

[0106] Step S151, convert the executable instruction set into an instruction format compatible with the device control protocol according to the type of the target execution equipment.

[0107] For example, for waste gas treatment equipment, its equipment control protocol may be a specific industrial control protocol. Convert the regulation parameter ranges (such as the dosage of desulfurizer and the treatment air volume) and start time series of the waste gas treatment equipment in the environmental optimization strategy set into an instruction format recognizable by the equipment. For example, convert the adjustment instruction of the desulfurizer dosage into a specific data format stipulated by the equipment control protocol, such as hexadecimal encoding, and arrange it in the instruction sequence specified by the equipment. For ventilation equipment and possibly added air purification equipment, also convert the relevant regulation information in the executable instruction set into the corresponding instruction format according to their respective equipment control protocols.

[0108] Step S152: Send an initialization instruction to the target execution device according to the start time series, and receive the status confirmation signal returned by the target execution device.

[0109] For example, according to the start time series set for the waste gas treatment equipment before, when reaching the time point of immediate start, send an initialization instruction to the waste gas treatment equipment. The initialization instruction contains basic parameters required for equipment startup, such as equipment self-check and initialization of the working mode. After receiving the initialization instruction, the waste gas treatment equipment performs corresponding operations and returns a status confirmation signal after the operations are completed. If the equipment self-check is normal and the initialization of the working mode is successfully set, the returned status confirmation signal indicates that the equipment is ready. Similarly, according to the start time series of the ventilation equipment and the air purification equipment (if deployed), send initialization instructions to them respectively and receive the status confirmation signals returned by them.

[0110] Step S153: When the status confirmation signal indicates that the target execution device is ready, send regulation instructions in stages according to the parameter adjustment gradient, and monitor the actual parameter values fed back by the target execution device in real time.

[0111] For example, after the status confirmation signal returned by the waste gas treatment equipment indicates that it is ready, send regulation instructions in stages according to the previously set parameter adjustment gradient. In the first 10 minutes after startup, send regulation instructions to gradually increase the dosage of desulfurizer and the treatment air volume in stages every 5 minutes. After sending the instructions in each stage, monitor the actual parameter values fed back by the waste gas treatment equipment in real time, such as the actual dosage of desulfurizer and the actual treatment air volume. For the ventilation equipment, after receiving the status confirmation signal of its readiness, send regulation instructions for the wind speed in stages according to the set parameter adjustment gradient, and monitor the actual wind speed value fed back by the ventilation equipment in real time. For the air purification equipment (if deployed), also send regulation instructions in stages according to its parameter adjustment gradient and monitor parameters such as its actual purification efficiency.

[0112] Step S154: Calculate the deviation between the actual parameter value and the parameter target value corresponding to the expected coverage rate. If the deviation exceeds the fault tolerance range, trigger the collaborative instruction for the auxiliary control device.

[0113] For example, for the waste gas treatment equipment, compare the actually monitored actual desulfurizer dosage and the actual treatment air volume with the parameter target values corresponding to the expected coverage rate. For example, it is expected that the desulfurizer dosage should be (X + ΔX) kg at a certain time point, but the actually monitored dosage is (X + ΔX - δ) kg (δ is the deviation value). Calculate the deviation ratio as δ / (X + ΔX). If this deviation ratio exceeds the fault tolerance range, it indicates that the operation effect of the waste gas treatment equipment does not meet the expectation. At this time, trigger the collaborative instruction for the air purification equipment (auxiliary control device). The air purification equipment adjusts its own operation parameters according to the collaborative instruction, such as improving the purification efficiency, to make up for the impact brought by the operation deviation of the waste gas treatment equipment. Similar deviation calculations and collaborative instruction trigger operations are also carried out between the ventilation equipment and the air purification equipment.

[0114] Step S155: Record the execution timestamp, device response delay, and parameter calibration times of the control instruction, and generate a device execution efficiency report.

[0115] For example, during the execution of the entire control instruction, record the execution timestamp of each control instruction. For example, record when the waste gas treatment equipment receives the initialization instruction and when it receives the control instructions at each stage. The same applies to the ventilation equipment and the air purification equipment. At the same time, record the device response delay, that is, the time interval from receiving the instruction to starting to execute the instruction. For the waste gas treatment equipment, if there is a time lag before starting to adjust the desulfurizer dosage after receiving the control instruction for adjusting the desulfurizer dosage, this time interval is the device response delay. In addition, record the parameter calibration times. When the actual parameter value of the waste gas treatment equipment deviates greatly from the target value, parameter calibration may be required, and record the situation of each calibration. Finally, generate a device execution efficiency report based on this recorded information. This report can reflect the actual execution effect of the target execution device during the execution of the environment optimization strategy, providing a basis for subsequent device optimization and strategy adjustment.

[0116] In a possible implementation manner, step S150 further includes:

[0117] Step S156: Extract the correlation between the parameter calibration times and the expected coverage rate from the device execution efficiency report, and construct a strategy execution effect evaluation matrix.

[0118] For example, during the execution of previous equipment, equipment execution efficiency reports were generated for waste gas treatment equipment, ventilation equipment, and potentially existing air purification equipment, etc. Taking the waste gas treatment equipment as an example, the number of parameter calibrations is obtained from the equipment execution efficiency report. For instance, during the entire regulation process, n parameter calibrations were carried out, and the corresponding expected coverage rate was determined. The expected coverage rate reflects the theoretical coverage range of the equipment for pollutant treatment according to the set regulation strategy. If the number of parameter calibrations is large and the expected coverage rate is low, this indicates that there is a significant deviation between the equipment operation and the expectation. Integrate the relationships between the number of parameter calibrations and the expected coverage rate of these different equipment (waste gas treatment equipment, ventilation equipment, air purification equipment, etc.) to construct a strategy execution effect evaluation matrix. The rows of this matrix can represent different equipment, and the columns represent relevant indicators such as the number of parameter calibrations and the expected coverage rate. The elements in the matrix are the corresponding numerical relationships, thus comprehensively reflecting the correlation between the equipment execution situation and the expected effect during the entire strategy execution process.

[0119] Step S157: Calculate the time prediction error coefficient of the dynamic anomaly analysis model according to the difference degree between the actual duration and the predicted value of the environmental anomaly event.

[0120] For example, in the environmental anomaly event where the pollution in the workshop spreads to the office area, the duration of the event was predicted previously, and the predicted value was assumed to be T1 hours. During the actual development of the event, through comprehensive monitoring of the operation of each equipment, the pollutant diffusion situation, and the air quality recovery situation in the office area, etc., it is determined that the actual duration is T2 hours. Calculate the difference degree between the actual duration and the predicted value, that is, (T2 - T1) / T1 (assuming it is expressed in the form of relative error). The value of this difference degree is used as the time prediction error coefficient of the dynamic anomaly analysis model, and this coefficient reflects the accuracy of the model in predicting the duration of environmental anomaly events.

[0121] Step S158: Input the strategy execution effect evaluation matrix and the time prediction error coefficient into the model weight optimizer to generate a parameter weight adjustment gradient.

[0122] For example, the constructed policy execution effect evaluation matrix (which contains various correlation relationships related to the execution efficiency of devices) and the calculated time prediction error coefficient can be input into a dedicated model weight optimizer together. Based on its internal algorithm mechanism, the model weight optimizer processes the input information. For example, it can analyze which devices' execution situations in the policy execution effect evaluation matrix have a greater impact on the overall effect, and combine the time prediction error coefficient to judge the need for weight adjustment due to the deviation of the model in time prediction. According to these comprehensive analysis results, a weight adjustment gradient for each parameter in the dynamic anomaly analysis model is generated. This adjustment gradient clarifies how the weight of each parameter should be adjusted, whether to increase or decrease, and the magnitude of the adjustment, etc.

[0123] Step S159: Use the backpropagation algorithm to iteratively update the weights of the fully connected layer of the dynamic anomaly analysis model until the time prediction error coefficient drops below the convergence threshold.

[0124] For example, in the dynamic anomaly analysis model, the backpropagation algorithm is used to update the weights of the fully connected layer according to the generated parameter weight adjustment gradient. The backpropagation algorithm starts from the output layer of the dynamic anomaly analysis model. According to the difference between the prediction result and the actual result (manifested as the time prediction error coefficient here), it backpropagates the error information and gradually adjusts the weights of the fully connected layer. After each iterative update, the time prediction error coefficient is recalculated to determine whether it drops below the convergence threshold. If not, continue with the iterative update and continuously adjust the weights until the time prediction error coefficient meets the convergence condition. This process ensures that the dynamic anomaly analysis model can more accurately predict the relevant parameters of environmental anomaly events, such as the duration, after being adjusted.

[0125] Step S1510: Synchronize the updated model parameter weights to the dynamic anomaly analysis model copies corresponding to all monitoring sub-regions.

[0126] For example, in a large industrial park, there are multiple monitoring sub-regions, and each monitoring sub-region has a corresponding dynamic anomaly analysis model copy for independently analyzing the environmental data of this region. After the parameter weights of the main dynamic anomaly analysis model are updated, the updated weights are synchronized to the model copies of each monitoring sub-region. For example, the dynamic anomaly analysis model copies of the high-pollution monitoring sub-region around the workshop, the low-pollution monitoring sub-region in the office area, and other sub-regions will all receive the updated model parameter weights. This can ensure that environmental monitoring and analysis within the entire industrial park are based on unified and optimized model parameters, improving the accuracy and consistency of predicting and responding to environmental anomaly events.

[0127] For example, in a possible implementation manner, the method further includes:

[0128] Step S310, when it is monitored that the target execution device fails to respond to the regulation instruction for N consecutive times, start the device fault diagnosis process:

[0129] Taking the waste gas treatment device as an example, if it fails to respond to the regulation instruction for N consecutive times, first obtain its last online time, which is the key information for judging when the device lost contact. At the same time, query its historical maintenance records to understand the past maintenance situation of the device, including the maintenance time, maintenance reasons (such as component replacement, software upgrade, etc.) and the operation situation after maintenance, etc. In addition, obtain the operation status logs of adjacent devices (such as ventilation devices and other production devices related to the waste gas treatment device in the workshop). The operation status log contains the change of various operation parameters of the device over time, such as the change curves of parameters such as current, voltage, and temperature.

[0130] Step S320, obtain the last online time of the target execution device, the historical maintenance records, and the operation status logs of adjacent devices.

[0131] For example, in the operation status log of the adjacent device of the waste gas treatment device, analyze the current fluctuation characteristics. If it is found that the current suddenly increases or decreases and remains abnormal at a certain moment, this may imply problems with the electrical connection between the waste gas treatment device and itself or a circuit fault in the waste gas treatment device. At the same time, observe the temperature change curve. If the device temperature rises abnormally, it may be a fault in the internal heat dissipation components of the device or a hardware problem caused by overload operation. For the communication link, if the communication-related parameters (such as signal strength, communication frequency, etc.) of the adjacent device show abnormal fluctuations during the period when the waste gas treatment device fails to respond to the regulation instruction, combined with the communication protocol and network topology structure, the approximate location of the communication link interruption can be inferred, such as a certain communication node in the workshop or a certain section of the link between the device and the control center.

[0132] Step S330, identify the type of device hardware fault or the location of communication link interruption according to the current fluctuation characteristics and temperature change curve in the operation status log.

[0133] Step S340, if it is a hardware fault, match an available device of the same type from the standby device library and recalculate the impact of the deployment location of the available device on the expected coverage rate.

[0134] For example, when it is determined that the waste gas treatment equipment has a hardware failure, an available device of the same type is searched from the standby device library. Suppose there is a waste gas treatment device of model X available in the standby device library. When deploying the standby device to the workshop, it is necessary to recalculate the impact of its deployment location on the expected coverage rate. Due to factors such as the location of pollution sources, ventilation conditions, and pollutant diffusion paths in the workshop, different deployment locations will affect the treatment effect of the device on pollutants. According to the layout of the workshop, the design of the ventilation system, and the previous analysis of the pollutant diffusion model, calculate the proportion of pollutants that the standby device can handle at different potential deployment locations to the total amount of pollutants generated in the entire workshop, that is, the expected coverage rate. By comparing the expected coverage rates of different locations, select the optimal deployment location.

[0135] Step S350, if it is a communication interruption, switch to the standby communication frequency band to send an encrypted test signal, and locate the link interruption node according to the signal delay time.

[0136] If it is determined that there is a communication interruption, switch the communication of the waste gas treatment equipment to the standby communication frequency band. Send an encrypted test signal on the standby communication frequency band. The encrypted test signal includes the identification information of the device and some special codes for detecting the link status. According to the propagation situation of the encrypted test signal in the link, especially the signal delay time, locate the link interruption node. If the time from sending the test signal to receiving the feedback signal is too long and exceeds the normal signal propagation delay range, combined with the topology of the communication network in the workshop and the location information of each communication node, it can be determined which node or which section of the link has an interruption.

[0137] Step S360, push the device failure type, standby device deployment plan, and communication link repair suggestions to the maintenance terminal.

[0138] Specifically, whether it is determined as a hardware failure or a communication interruption, relevant information can be pushed to the maintenance terminal. If it is a hardware failure, the pushed information includes the failure type of the waste gas treatment equipment (such as circuit failure, heat dissipation component failure, etc.), the selected standby device deployment plan (model of the standby device, deployment location, etc.). If it is a communication interruption, push the location information of the communication link interruption and the corresponding repair suggestions (such as repairing the interruption node, adjusting communication frequency band parameters, etc.). Maintenance personnel can repair or adjust the device in a timely manner according to this information to ensure the normal operation of the environmental monitoring and pollution control equipment in the industrial park.

[0139] For example, in a possible implementation manner, the method further includes:

[0140] Step S410, within a preset period, count the distribution of environmental abnormal event types and the success rate of policy execution in all monitored sub-regions, and generate a regional security assessment report:

[0141] Step S420: Extract the occurrence frequency, average impact range, and the evolution trend of the associated pollution source types of each environmental anomaly event, calculate the difference in the execution success rates of the same type of environmental optimization strategies in different time periods, and identify the periodic characteristics of external environmental interference factors.

[0142] For example, in each monitoring sub-region of a large industrial park, conduct statistical analysis on different types of environmental anomaly events. Taking the environmental anomaly event of workshop pollution spreading to the office area as an example, count its occurrence frequency within a preset period (such as one month). Calculate the impact range of each event and then obtain the average impact range. At the same time, analyze the evolution trend of the associated pollution source types (such as pollution caused by workshop equipment failures), and observe whether the occurrence frequency of a certain equipment failure increases or decreases over time, or whether the type of failure changes, etc. For the same type of environmental optimization strategy (such as the equipment regulation strategy for workshop pollution spreading to the office area), calculate the difference in its execution success rates in different time periods (such as weekly). For example, the execution success rate in the first week is 80%, and in the second week is 75%, etc. By analyzing the fluctuations of these success rates and combining factors such as the meteorological conditions outside the industrial park (such as the periodic changes in wind direction and wind speed) and the surrounding traffic flow (such as the differences between weekdays and non-working days), identify the periodic characteristics of external environmental interference factors.

[0143] Step S430: According to the evolution trend and periodic characteristics, optimize the parameter similarity determination threshold in the dynamic grid segmentation scheme, and adjust the noise filtering threshold in the data feature extraction rule, so that the noise filtering threshold maintains a dynamic association with the optimized parameter similarity determination threshold.

[0144] Based on the evolution trend of the associated pollution source types of environmental anomaly events and the periodic characteristics of external environmental interference factors obtained from the previous analysis, optimize the parameter similarity determination threshold in the dynamic grid segmentation scheme. For example, if it is found that the pollution spreading pattern in the area around the workshop has changed over time, it may be necessary to adjust the parameter similarity determination threshold to more accurately divide the monitoring sub-regions. At the same time, adjust the noise filtering threshold in the data feature extraction rule. If the periodic changes in external environmental interference factors (such as traffic flow) lead to periodic fluctuations in the noise level, correspondingly adjust the noise filtering threshold to better filter the noise data. And ensure that the noise filtering threshold maintains a dynamic association with the optimized parameter similarity determination threshold. For example, according to the adjustment range of the parameter similarity determination threshold, adjust the noise filtering threshold according to a certain proportional relationship, so that the two can work together during the environmental monitoring and analysis process to improve the accuracy of environmental data processing.

[0145] Step S440: Write the updated dynamic grid segmentation scheme and data feature extraction rules into the version control queue of the preset rule library.

[0146] For example, an optimized dynamic grid segmentation scheme and adjusted data feature extraction rules can be integrated and then written into the version control queue of the preset rule library. In the version control queue, these rules are managed according to the chronological order or the logic of version updates. This can facilitate tracing the historical versions of the rules and ensure that the latest and optimized dynamic grid segmentation scheme and data feature extraction rules are used in the environmental monitoring system of the industrial park, improving the effectiveness and adaptability of the entire environmental monitoring and management system.

[0147] For instance, in a possible implementation manner, the method further includes:

[0148] Step S510: When it is detected that a new sensor node accesses the target monitoring area, execute the node data fusion verification process:

[0149] Step S520: Obtain the device identifier, sampling frequency, and the initial sequence of environmental data collected by the new sensor node.

[0150] In this embodiment, the device identifier is the unique identifier of the new sensor node in the entire monitoring system. For example, it may be a specific code composed of letters and numbers, used to accurately identify the node among numerous devices. The sampling frequency indicates the time interval rule for this sensor node to collect environmental data. For example, environmental data such as air quality parameters, noise intensity parameters, and meteorological parameters are collected every 10 minutes, forming an initial sequence of environmental data. This initial sequence of environmental data contains a series of data collected at the sampling frequency since the sensor node started working.

[0151] Step S530: Perform time alignment processing on the initial sequence and the historical environmental data of adjacent sensor nodes, and calculate the parameter correlation coefficient within the same time window.

[0152] Within the industrial park, there are existing adjacent sensor nodes around the newly added sensor nodes, and these adjacent sensor nodes have their respective historical environmental data records. To conduct data fusion verification, it is necessary to align the initial sequence collected by the newly added sensor nodes with the historical environmental data of the adjacent sensor nodes in terms of time. For example, if the newly added sensor nodes start collecting data at 10 am, and the historical data records of the adjacent sensor nodes cover the entire morning period, then the data after 10 am will be aligned so that they are in the same time frame. Then, for each environmental parameter (such as sulfur dioxide concentration, noise decibel value, temperature, etc.), within the same time window (for example, with 1 hour as a time window), calculate the parameter correlation coefficient, which reflects the degree of association between the data collected by the newly added sensor nodes and the historical data of the adjacent sensor nodes in the corresponding environmental parameters. If within a certain time window, the sulfur dioxide concentration data collected by the newly added sensor nodes shows a similar change trend to the sulfur dioxide concentration data in the historical records of the adjacent sensor nodes, then the parameter correlation coefficient between them will be relatively high; conversely, if the two sets of data have no correlation, the parameter correlation coefficient will be relatively low.

[0153] Step S540, if the parameter correlation coefficient is lower than the node consistency threshold, then start the node calibration mode: send a standard test signal to the newly added sensor node and analyze the error distribution of its response data.

[0154] Suppose the node consistency threshold is set to 0.8. If the calculated parameter correlation coefficient is lower than this threshold, it indicates that the data collected by the newly added sensor nodes has poor consistency with the data of the adjacent sensor nodes and needs to be calibrated. At this time, send a standard test signal to the newly added sensor node. This standard test signal is a known signal with specific environmental parameter values. For example, send an analog standard test signal of sulfur dioxide with a fixed concentration, or a standard test signal of noise with a fixed intensity, etc. After receiving the standard test signal, the newly added sensor node will generate response data, and then analyze the error distribution between these response data and the standard test signal. For the sulfur dioxide concentration response data, if there is a deviation from the concentration value in the standard test signal, analyze the distribution of this deviation at different time points and different concentration ranges, whether the deviation is large at high concentrations, or at low concentrations, or random deviations, etc.

[0155] Step S550, adjust the sampling frequency compensation parameter of the newly added sensor node according to the error distribution, and recalculate the parameter correlation coefficient until the node consistency threshold is reached.

[0156] For example, the sampling frequency compensation parameter of the newly added sensor node can be adjusted according to the error distribution obtained from the previous analysis. If the error distribution shows that the response data deviation is large under the test signal of high-concentration sulfur dioxide, it may be necessary to reduce the sampling frequency compensation parameter to make the sensor collect data more accurately in a high-concentration environment; if the deviation is large at low concentration, it may be necessary to increase the sampling frequency compensation parameter. After adjustment, the data collected by the newly added sensor node is aligned with the historical environmental data of the adjacent sensor node again in terms of time, and the parameter correlation coefficient within the same time window is recalculated. This process is continuously repeated, and the sampling frequency compensation parameter is continuously adjusted and the coefficient is recalculated until the parameter correlation coefficient reaches the node consistency threshold of 0.8.

[0157] Step S560: Add the newly added sensor node that has passed the verification to the spatial distribution matrix and trigger the incremental update calculation of the dynamic grid segmentation scheme.

[0158] For example, when the parameter correlation coefficient of the newly added sensor node reaches the node consistency threshold, it indicates that the data collected by this node has good consistency with the surrounding environmental data, and it can be added to the spatial distribution matrix. The spatial distribution matrix records information such as the position coordinates and parameter change gradients of each sensor node in the industrial park. After the newly added sensor node is added, this spatial distribution matrix is updated. Since the layout of the sensor nodes has changed, it is necessary to trigger the incremental update calculation of the dynamic grid segmentation scheme. In the original dynamic grid segmentation scheme, it was generated based on the previous layout of the sensor nodes. The addition of the newly added sensor node may affect the division of the parameter mutation boundary, the steady-state region, and the layout of the grid cells, etc. Through the incremental update calculation, the spatial distribution characteristics of the environmental data in the industrial park are re-evaluated, the division of the grid cells is adjusted, and the definitions of the parameter mutation boundary and the steady-state region are updated to adapt to the new layout of the sensor nodes and ensure that the entire environmental monitoring system can accurately monitor and analyze the environmental conditions of the industrial park.

[0159] Figure 2 FIG. shows a schematic diagram of exemplary hardware and software components of a server 100 that can implement the ideas of the present application provided by some embodiments of the present application. For example, the processor 120 can be used on the server 100 and is used to execute the functions in the present application.

[0160] The server 100 can be a general-purpose server or a special-purpose server, and both can be used to implement the AI-based environmental monitoring data processing method of the present application. Although only one server is shown in the present application, for convenience, the functions described in the present application can be implemented in a distributed manner on multiple similar platforms to balance the processing load.

[0161] For example, server 100 may include a network port 110 connected to a network, one or more processors 120 for executing program instructions, a communication bus 130, and different forms of storage media 140, such as disks, ROM, or RAM, or any combination thereof. Exemplarily, server 100 may also include program instructions stored in ROM, RAM, or other types of non-transitory storage media, or any combination thereof. The methods of the present application can be implemented according to these program instructions. Server 100 further includes an I / O interface 150 between the computer and other input / output devices.

[0162] For ease of explanation, only one processor is described in server 100. However, it should be noted that server 100 in the present application may also include multiple processors. Therefore, the steps executed by one processor described in the present application may also be executed jointly or separately by multiple processors. For example, if the processor of server 100 executes step A and step B, it should be understood that step A and step B may also be executed jointly by two different processors or separately in one processor. For example, the first processor executes step A, the second processor executes step B, or the first processor and the second processor jointly execute steps A and B.

[0163] In addition, an embodiment of the present invention further provides a readable storage medium, in which computer-executable instructions are preset. When the processor executes the computer-executable instructions, the data processing method based on AI environment monitoring as described above is implemented.

[0164] It should be noted that, in order to simplify the description of the present invention disclosure and thus help the understanding of one or more embodiments of the invention, in the foregoing description of the embodiments of the present invention, sometimes multiple features are merged into one embodiment, drawing, or description thereof.

Claims

1. A data processing method for AI-based environmental monitoring, characterized in that, The method includes: Obtaining a set of environmental data collected in real time by multiple sensor nodes in a target monitoring area, where the set of environmental data includes air quality parameters, noise intensity parameters, and meteorological parameters in a time series; According to the spatial distribution characteristics of the set of environmental data, dynamically dividing the target monitoring area into multiple monitoring sub-areas, and assigning corresponding data feature extraction rules to each monitoring sub-area; Based on the data feature extraction rules, extracting abnormal fluctuation indicators and regional association features from the environmental data of each monitoring sub-area to generate a set of sub-area features; Using a pre-trained dynamic anomaly analysis model to perform multi-dimensional correlation analysis on the abnormal fluctuation indicators in the set of sub-area features, determining the types and influence scopes of environmental anomaly events in the target monitoring area, and generating a set of environmental optimization strategies; Executing at least one environmental optimization strategy in the set of environmental optimization strategies, sending a control instruction to a target execution device, and updating the parameter weights of the dynamic anomaly analysis model based on the feedback data of the control instruction; The step of, according to the spatial distribution characteristics of the set of environmental data, dynamically dividing the target monitoring area into multiple monitoring sub-areas, and assigning corresponding data feature extraction rules to each monitoring sub-area, includes: Extracting the position coordinates and parameter change gradients of each sensor node from the set of environmental data to construct a spatial distribution matrix; Performing density clustering analysis on the spatial distribution matrix to identify parameter mutation boundaries and steady-state regions in the target monitoring area; Generating a dynamic grid segmentation scheme covering the target monitoring area according to the geometric shape of the parameter mutation boundary and the parameter similarity of the steady-state regions; Merging consecutive grid cells with parameter similarity higher than a preset threshold in the dynamic grid segmentation scheme into the same monitoring sub-area, and labeling spatial attribute tags for each monitoring sub-area; According to the environmental parameter types corresponding to the spatial attribute tags, matching the noise filtering thresholds and feature fusion weights in the data feature extraction rules from a preset rule library; The step of, using a pre-trained dynamic anomaly analysis model to perform multi-dimensional correlation analysis on the abnormal fluctuation indicators in the set of sub-area features, determining the types and influence scopes of environmental anomaly events in the target monitoring area, includes: Using a pre-trained dynamic anomaly analysis model to perform pattern matching on the parameter peaks in the abnormal fluctuation indicators with a historical environmental event database to identify potential pollution source types and pollution diffusion models; According to the cross-regional association paths in the regional association features, calculating the propagation speed and attenuation coefficient of the pollution diffusion model in the target monitoring area; Combining the environmental parameter types corresponding to the spatial attribute tags of the monitoring sub-area, predicting the superimposed influence degree of the pollution diffusion model on air quality parameters and meteorological parameters; Based on the spatial coordinates of the monitoring sub-areas where the superimposed influence degree exceeds a preset safety threshold, generating an environmental anomaly event type including an emergency control priority; According to the propagation speed and attenuation coefficient of the pollution diffusion model, determining the boundary of the influence scope and the predicted value of the duration of the environmental anomaly event.

2. The data processing method based on AI environmental monitoring according to claim 1, wherein Generating a dynamic grid segmentation scheme covering the target monitoring area according to the geometry of the parameter mutation boundary and the parameter similarity of the steady state area, includes: Extracting the spatial coordinate sequence of the boundary points from the parameter mutation boundary, and identifying the geometric shape type of the boundary line according to the continuity of the spatial coordinate sequence; Dividing the boundary line into multiple geometric feature segments according to the curvature change characteristics of the geometric shape type, and labeling the curvature change direction for each geometric feature segment; Generating initial grid cells on both sides of the boundary line based on the curvature change direction, and matching the side length of the initial grid cells with the parameter similarity of the steady state area to determine the side length adjustment ratio of the initial grid cells; Expanding or contracting the initial grid cells according to the side length adjustment ratio to generate boundary grid cells adapted to the geometry of the boundary line; Extracting the sensor node group with the highest parameter similarity from the steady state area, and determining the grid cell division density of the steady state area according to the spatial distribution density of the sensor node group; Generating uniformly distributed steady state grid cells in the steady state area based on the grid cell division density, and aligning the side length of the steady state grid cells with the side length of the boundary grid cells to generate transition grid cells; Seamlessly connecting the boundary grid cells and the steady state grid cells according to the side length change trend of the transition grid cells to generate a dynamic grid segmentation scheme covering the target monitoring area; Mapping the spatial coordinate sequences of the boundary grid cells, transition grid cells and steady state grid cells in the dynamic grid segmentation scheme to the parameter mutation boundary and the parameter similarity of the steady state area to generate grid cell attribute labels; Clustering the grid cells in the dynamic grid segmentation scheme according to the parameter similarity according to the grid cell attribute labels to generate multiple grid cell groups, and labeling spatial attribute labels for each grid cell group; Based on the spatial attribute labels, matching the grid cell groups in the dynamic grid segmentation scheme with the spatial distribution characteristics of the target monitoring area to generate a dynamic grid segmentation scheme covering the target monitoring area.

3. The data processing method based on AI environmental monitoring according to claim 1, wherein, Extracting the abnormal fluctuation index and regional association characteristics from the environmental data of each monitoring sub-area based on the data feature extraction rule, includes: Performing sliding window normalization processing on the time series environmental data of all sensor nodes in the monitoring sub-area to generate a normalized data sequence; Removing abnormal points and performing data interpolation on the normalized data sequence according to the noise filtering threshold in the data feature extraction rule to generate a smoothed data sequence; Calculating the parameter change rate of adjacent time windows in the smoothed data sequence, and marking the time windows with the parameter change rate exceeding the upper limit of the historical fluctuation range as abnormal fluctuation intervals; Extracting the parameter peak value, duration and spatial propagation direction within the abnormal fluctuation interval to generate the abnormal fluctuation index; Match the propagation directions of all abnormal fluctuation indicators within the same monitoring sub-region with the abnormal fluctuation indicators of adjacent sub-regions to generate cross-region association paths, and use the topological structure of the cross-region association paths as regional association features.

4. The data processing method based on AI environmental monitoring according to claim 1, characterized in that The generation of the set of environment optimization strategies includes: Select the type of target execution device and the range of regulation parameters from the device control strategy library according to the type of pollution source corresponding to the environmental abnormal event type; Based on the spatial coordinates of the influence range boundary and the emergency regulation priority, allocate the start time sequence and parameter adjustment gradient for each target execution device; Combine the predicted duration value and the wind direction and wind speed data in the meteorological parameters to calculate the expected coverage rate of the parameter adjustment gradient of the target execution device within the influence range; When the expected coverage rate is lower than the preset efficiency threshold, add the location deployment suggestions and collaborative working modes of auxiliary regulation devices to the target execution device; Package the regulation parameter ranges, start time sequences, and auxiliary regulation device deployment suggestions of all target execution devices into an executable instruction set to generate the set of environment optimization strategies.

5. The data processing method based on AI environmental monitoring according to claim 4, characterized in that, Executing at least one of the environment optimization strategies in the set of environment optimization strategies and sending regulation instructions to the target execution device includes: Convert the executable instruction set into an instruction format compatible with the device control protocol according to the type of the target execution device; Send an initialization instruction to the target execution device according to the start time sequence and receive the status confirmation signal returned by the target execution device; When the status confirmation signal indicates that the target execution device is ready, send regulation instructions in stages according to the parameter adjustment gradient and monitor the actual parameter values fed back by the target execution device in real time; Calculate the deviation between the actual parameter value and the parameter target value corresponding to the expected coverage rate. If the deviation exceeds the tolerance range, trigger the collaborative instruction of the auxiliary regulation device; Record the execution timestamp, device response delay, and parameter calibration times of the regulation instruction to generate a device execution efficiency report.

6. The data processing method based on AI environmental monitoring according to claim 5, wherein, Updating the parameter weights of the dynamic anomaly analysis model based on the feedback data of the regulation instruction includes: Extract the correlation between the parameter calibration times and the expected coverage rate from the device execution efficiency report to construct a strategy execution effect evaluation matrix; Calculate the time prediction error coefficient of the dynamic anomaly analysis model according to the difference between the actual duration and the predicted value of the environmental abnormal event; Input the strategy execution effect evaluation matrix and the time prediction error coefficient into the model weight optimizer to generate a parameter weight adjustment gradient; Use the backpropagation algorithm to iteratively update the weights of the fully connected layer of the dynamic anomaly analysis model until the time prediction error coefficient drops below the convergence threshold; Synchronize the updated model parameter weights to the dynamic anomaly analysis model copies corresponding to all monitoring sub-regions.

7. The data processing method based on AI environmental monitoring according to claim 1, wherein The method further includes: When it is detected that the target execution device fails to respond to the regulation instruction continuously for N times, start the device fault diagnosis process: Obtain the last online time, historical maintenance records of the target execution device, and the operation status logs of adjacent devices; Identify the types of device hardware failures or the locations of communication link interruptions based on the current fluctuation characteristics and temperature change curves in the operating status log; If it is a hardware failure, match available devices of the same type from the standby device library, and recalculate the impact of the deployment locations of the available devices on the expected coverage rate; If it is a communication interruption, switch to the standby communication frequency band to send encrypted test signals, and locate the link interruption node based on the signal delay time; Push the device failure type, standby device deployment plan, and communication link repair suggestions to the maintenance terminal.

8. A server, characterized in that, The server includes a processor and a memory. The memory is connected to the processor. The memory is used to store programs, instructions, or codes, and the processor is used to execute the programs, instructions, or codes in the memory to implement the data processing method based on AI environment monitoring described in any one of claims 1-7 above.

Images