Incineration plant environment data processing method and system based on edge computing

Abstract

The present application provides a kind of incineration plant environmental data processing method and system based on edge computing, first by the environmental perception device of each functional area arrangement of incineration plant collection gas component, temperature distribution and particulate matter concentration etc., then the regionalization feature extraction of environmental perception data is carried out, and the regional environmental feature set of each functional area is generated, then cross-regional sharing and correlation analysis are carried out, the environmental influence correlation between different functional areas is identified, the environmental risk propagation path model is constructed according to the environmental influence correlation, the change trend of each functional area environmental parameter is predicted, finally, according to the change trend of each functional area environmental parameter, incineration equipment control instruction is generated, and is transmitted to corresponding executing mechanism, so that the overall processing of incineration plant environmental data and the intelligent control of equipment are realized, and the operation efficiency and environmental safety are improved.

Description

Technical Field

[0001] This invention relates to the field of edge computing technology, and more specifically, to a method and system for processing environmental data of incineration plants based on edge computing. Background Technology

[0002] In the operation and management of incineration plants, the processing of environmental data and the control of incineration equipment are crucial, directly affecting the plant's operational efficiency, environmental protection effectiveness, and the safety of the surrounding ecological environment. Traditional environmental data processing methods for incineration plants typically involve independently collecting environmental data, such as gas composition, temperature distribution, and particulate matter concentration, from each functional area. Then, only the data from that area is analyzed and processed simply to determine whether the environmental conditions meet standards. This approach has significant limitations. Firstly, complex environmental impact relationships exist between different functional areas. For example, high-temperature gases from the incineration area may affect the temperature and gas composition of subsequent processing areas. Traditional independent processing methods cannot identify these cross-regional environmental impacts, making it difficult to comprehensively and accurately grasp the dynamic changes in the entire incineration plant's environment. Secondly, the lack of in-depth analysis of environmental impact relationships prevents the construction of effective environmental risk propagation path models, thus hindering the accurate prediction of changing trends in environmental parameters across functional areas. Consequently, it is impossible to take targeted measures to adjust the operation of incineration equipment in advance, leaving the incineration plant often in a passive position when responding to environmental changes, increasing environmental risks and operating costs. Summary of the Invention

[0003] In view of the aforementioned problems, and in conjunction with the first aspect of the present invention, embodiments of the present invention provide a method for processing environmental data of an incineration plant based on edge computing, the method comprising: Environmental sensing data is collected by environmental sensing devices deployed in various functional areas of the incineration plant. The environmental sensing data includes gas composition data, temperature distribution data, and particulate matter concentration data. The environmental perception data is processed by regional feature extraction to generate a set of regional environmental features corresponding to each functional area; The set of regional environmental characteristics of each functional area is shared across regions, and the environmental impact correlation between different functional areas is identified through cross-regional correlation analysis. An environmental risk propagation path model is constructed based on the aforementioned environmental impact correlations, and the environmental parameter change trends of each functional area are predicted through the environmental risk propagation path model. Based on the changing trends of the environmental parameters, control commands for the incineration equipment are generated, and these control commands are transmitted to the corresponding incineration equipment actuators.

[0004] In another aspect, embodiments of the present invention also provide an environmental data processing system for incineration plants based on edge computing, including a processor and a machine-readable storage medium connected to the processor. The machine-readable storage medium is used to store programs, instructions, or code, and the processor is used to execute the programs, instructions, or code in the machine-readable storage medium to implement the above-described method.

[0005] Based on the above, this invention, through the deployment of environmental sensing devices in various functional areas of the incineration plant to collect comprehensive environmental sensing data, and performing regional feature extraction processing on the environmental sensing data, can accurately generate regional environmental feature sets corresponding to each functional area, thus clearly presenting the environmental status of each area. By sharing the regional environmental feature sets of each functional area across regions and performing correlation analysis, the environmental impact correlations between different functional areas are successfully identified. Based on these correlations, an environmental risk propagation path model is constructed, which can accurately predict the changing trends of environmental parameters in each functional area. Finally, control commands for the incineration equipment are generated based on these environmental parameter changing trends and transmitted to the actuators, enabling proactive adjustments to the operation of the incineration equipment. This improves the operating efficiency of the incineration plant, effectively reduces environmental risks, and ensures the stable, safe, and environmentally friendly operation of the incineration plant. Attached Figure Description

[0006] Figure 1 This is a schematic diagram of the execution flow of the edge computing-based environmental data processing method for incineration plants provided in an embodiment of the present invention.

[0007] Figure 2 This is a schematic diagram of exemplary hardware and software components of the edge computing-based environmental data processing system for incineration plants provided in an embodiment of the present invention. Detailed Implementation

[0008] The present invention will now be described in detail with reference to the accompanying drawings. Figure 1 This is a flowchart illustrating an edge computing-based environmental data processing method for incineration plants according to an embodiment of the present invention. The following is a detailed description of this edge computing-based environmental data processing method for incineration plants.

[0009] This embodiment provides a method for processing environmental data in an incineration plant based on edge computing. This method uses edge computing technology to collect, process, analyze, and apply environmental data from various functional areas of the incineration plant, achieving precise monitoring of the incineration plant environment and intelligent control of the incineration equipment. The following describes each step of the method in detail.

[0010] Step S110: Collect environmental sensing data through environmental sensing devices deployed in various functional areas of the incineration plant. The environmental sensing data includes gas composition data, temperature distribution data, and particulate matter concentration data.

[0011] Step S111: Divide the functional areas according to the incineration plant's production process. The functional areas include the incineration area, flue gas treatment area, waste storage area, auxiliary equipment area, and office area.

[0012] In this embodiment, functional areas are divided to facilitate more targeted environmental data collection and management. The incineration area is the core of the incineration operation, directly generating large amounts of gas, heat, and particulate matter; the flue gas treatment area is used to purify the flue gas generated from incineration, and its environmental parameters reflect the purification effect; the waste storage area is used to store waste to be incinerated, and the environmental conditions of this waste storage area are related to the stability and safety of the waste; the auxiliary equipment area contains various devices that support the incineration operation, and their operating status affects the surrounding environment; the office area is where personnel work, and a good environmental quality must be ensured. Through the above division, the key environmental monitoring areas for each region can be clearly defined.

[0013] Step S112: Configure area identification codes for each group of environmental sensing devices deployed in each functional area. The environmental sensing device group includes gas sensors, temperature sensors, and particulate matter sensors. The deployment locations of each type of sensor correspond to key environmental monitoring points within the functional area. The area identification code includes a functional area type code and a location coordinate code, which are used to distinguish environmental sensing data from different functional areas.

[0014] For example, in the incineration area, the deployed gas sensors may include sensors for detecting gases such as carbon dioxide, carbon monoxide, and nitrogen oxides; temperature sensors may be distributed at different locations around the incinerator; and particulate matter sensors may be placed near the flue gas outlet. Each sensor belongs to an environmental sensing device group for that area, and each device group is configured with a unique area identification code.

[0015] Functional area type codes can use letter combinations, such as "FS" for incineration area, "YQ" for flue gas treatment area, "FW" for waste storage area, "FZ" for auxiliary equipment area, and "BG" for office area. Location coordinate codes can be set based on the incineration plant's planar coordinate system, for example, using the form "X coordinate value - Y coordinate value," where the X and Y coordinates represent the horizontal and vertical positions of the area in the planar coordinate system, respectively.

[0016] Through the aforementioned area identification coding, once environmental perception data is collected, it is possible to quickly and accurately identify which functional area the data comes from and the specific location of that area, thus avoiding data confusion between different areas.

[0017] Step S113: Obtain the acquisition triggering mechanism of the environmental sensing device group. The acquisition triggering mechanism includes timed triggering and event triggering. Timed triggering starts acquisition at a preset time interval, and event triggering starts acquisition when abnormal fluctuations in environmental parameters are detected.

[0018] In this embodiment, the data acquisition triggering mechanism is designed to balance the timeliness and economy of data acquisition. Timed triggering ensures regular data updates, enabling the system to monitor trends in environmental parameters. The preset time interval can be adjusted based on the environmental stability of different functional areas. In areas with rapid environmental changes, such as incineration areas, the time interval can be set shorter; in relatively stable areas, such as office areas, the time interval can be appropriately extended.

[0019] Event triggering is a dynamic response mechanism. When the sensor detects abnormal fluctuations in environmental parameters, such as a sudden and sharp rise in temperature in the incineration area or a concentration of a certain harmful gas in the flue gas treatment area exceeding the normal range, it can immediately start data collection to obtain detailed abnormal data, facilitating timely detection and handling of problems.

[0020] Step S114: Send a data acquisition start signal to the environmental sensing device group so that the environmental sensing device group responds to the data acquisition start signal, starts the sensors to acquire corresponding environmental parameters, and generates raw sensing data containing sensor type identifier, acquisition timestamp and parameter measurement value. The data acquisition start signal includes acquisition duration parameter and data transmission format parameter.

[0021] In this embodiment, the data acquisition start signal is emitted by the edge computing node and transmitted to each environmental sensing device group via wired or wireless communication. Sensor type identifiers are used to distinguish data collected by different types of sensors; for example, "QT" represents a gas sensor, "WD" represents a temperature sensor, and "KL" represents a particulate matter sensor.

[0022] The data acquisition timestamps are accurate to the millisecond level, recording the exact time of data collection to facilitate subsequent time-series analysis. Parameter measurements are the numerical values ​​of specific environmental parameters collected by the sensors, expressed in their corresponding physical units.

[0023] The data collection duration parameter specifies the duration of this data collection, ensuring that sufficient data samples can be obtained; the data transmission format parameter standardizes the data transmission format, such as using JSON format, so that edge computing nodes can correctly parse the received data.

[0024] Step S115: Add area identification codes and data check codes to the original sensing data to generate environmental perception data that conforms to the edge computing network transmission specifications.

[0025] In this embodiment, a region identifier code is added to the raw sensor data, which associates the data with the corresponding functional area, facilitating subsequent regionalization processing. The data checksum is generated using a cyclic redundancy check method, calculated from the raw sensor data, and is used to verify the integrity and accuracy of the data during data transmission.

[0026] When an edge computing node receives environmental sensing data, it can first verify the data checksum. If the verification passes, it indicates that the data was transmitted without errors or loss and can be processed further. If the verification fails, the environmental sensing device group will be required to resend the data. The generated environmental sensing data must conform to the transmission specifications of the edge computing network, including data encapsulation format, transmission rate, and other requirements, to ensure that the data can be transmitted efficiently and stably in the edge computing network.

[0027] Step S120: Perform regional feature extraction processing on the environmental perception data to generate a set of regional environmental features corresponding to each functional area.

[0028] In this embodiment, regional feature extraction is a deep processing of environmental perception data. By extracting the features of environmental data in each functional area, the environmental conditions of each area can be reflected more accurately.

[0029] Step S121: Receive environmental sensing data transmitted from each functional area, perform integrity verification on the environmental sensing data, classify the verified environmental sensing data according to parameter type, divide it into gas composition data sequence, temperature distribution data sequence and particulate matter concentration data sequence, and arrange them in the order of collection timestamp.

[0030] In this embodiment, the edge computing node first receives environmental perception data transmitted from each functional area. Integrity verification is achieved by checking the length, format, and completeness of data fields. For example, it checks whether necessary fields such as area identification code, sensor type identifier, acquisition timestamp, parameter measurement value, and data checksum are included.

[0031] Environmental sensing data, verified for integrity, is categorized according to parameter type. Data collected by gas sensors is classified as a gas composition data sequence, data collected by temperature sensors as a temperature distribution data sequence, and data collected by particulate matter sensors as a particulate matter concentration data sequence.

[0032] Then, each data sequence is arranged according to the chronological order of its collection timestamps, forming an ordered data sequence. This arrangement allows the data to reflect changes over time, facilitating time series analysis. For example, arranging gas composition data sequences in chronological order can accurately show the changes in the concentration of a certain gas at different times.

[0033] Step S122: Perform feature extraction on the gas component data sequence, identify the types of gases contained therein, and determine the concentration change trend characteristics and concentration fluctuation pattern characteristics of each gas type. The concentration change trend characteristics include a continuous upward trend, a continuous downward trend, and a stable fluctuation trend. The concentration fluctuation pattern characteristics include periodic fluctuation patterns and non-periodic fluctuation patterns.

[0034] In this embodiment, feature extraction from a gas composition data sequence first requires identifying the types of gases contained within it. This is achieved by comparing the parameter measurements collected by the gas sensor with the characteristic parameters of known gases. Each gas has its unique physicochemical properties, and the corresponding sensor measurements will also have specific ranges and patterns of variation. For example, the parameter measurements collected by a carbon dioxide sensor fluctuate within a certain range, which can be used to identify carbon dioxide as a gas.

[0035] Step S1221: Extract gas type identifiers from each data record of the gas composition data sequence, and filter the main gas types based on a preset gas priority list, wherein the gas priority list is preset according to the environmental risk level of the incineration plant.

[0036] In this embodiment, each gas composition data record includes a gas type identifier, which is automatically added by the sensor when collecting data to indicate the type of gas collected. The preset gas priority list is based on the environmental risk level of each gas in the incineration plant; the higher the risk level, the higher the priority in the list. For example, toxic and harmful gases such as hydrogen sulfide and chlorine have a higher priority than harmless gases such as carbon dioxide.

[0037] By comparing the gas type identifiers in the data records with a gas priority list, gases with higher priority were selected as the primary gas types. These primary gas types are the focus of subsequent analysis because they have a greater impact on the environment and human safety.

[0038] Step S1222: The gas composition data sequence is split according to the main gas types to obtain the single gas data subsequences corresponding to each main gas type. The single gas data subsequences contain the concentration measurement values ​​of the same gas type at different collection timestamps.

[0039] In this embodiment, the gas composition data sequence is split according to the main gas types, enabling separate analysis for each main gas. For example, after screening and identifying carbon dioxide, carbon monoxide, and nitrogen oxides as the main gas types, the gas composition data sequence is split into carbon dioxide data subsequences, carbon monoxide data subsequences, and nitrogen oxide data subsequences. Each subsequence contains the concentration measurements of the corresponding gas at different acquisition timestamps.

[0040] Step S1223: Arrange the concentration measurements in the single gas data subsequence according to the acquisition timestamp order to generate a gas concentration time series.

[0041] In this embodiment, the concentration measurements in the single-gas data subsequence are arranged in chronological order of the acquisition timestamps to form a gas concentration time series. This arrangement can intuitively show how the gas concentration changes over time; for example, the changes in the concentration of a certain gas in the morning, noon, and evening can all be reflected in the time series.

[0042] Step S1224: Perform trend fitting processing on the gas concentration time series, calculate the overall slope value of the series using the linear regression method, and determine the concentration change trend characteristics based on the sign and absolute value of the slope value. When the slope value is positive and the absolute value is greater than a preset threshold, it is determined to be a continuous upward trend; when the slope value is negative and the absolute value is greater than a preset threshold, it is determined to be a continuous downward trend; and when the absolute value of the slope value is less than a preset threshold, it is determined to be a stable fluctuation trend.

[0043] In this embodiment, trend fitting is used to determine the overall trend of gas concentration change. The linear regression method fits the data points in the gas concentration time series to obtain a straight line, the slope of which reflects the overall direction and degree of concentration change.

[0044] When the slope is positive and its absolute value is greater than the preset threshold, it indicates that the concentration of the gas is showing a continuous upward trend. When the slope is negative and its absolute value is greater than the preset threshold, it indicates that the concentration is showing a continuous downward trend. When the absolute value of the slope is less than the preset threshold, it indicates that the concentration fluctuates within a certain range, with no significant overall change, and represents a stable fluctuation trend.

[0045] Step S1225: Divide the gas concentration time series into multiple consecutive time segments according to a preset time window, calculate the concentration standard deviation of each time segment, and determine the intensity of concentration fluctuation within each time segment based on the magnitude of the concentration standard deviation.

[0046] In this embodiment, the length of the preset time window can be set according to the rate of change of gas concentration. The gas concentration time series is divided according to the preset time window to obtain multiple continuous and non-overlapping time segments. For example, for gases that change rapidly, the time window can be set shorter; for gases that change slowly, the time window can be set longer.

[0047] Calculate the standard deviation of the concentration measurements within each time segment. The larger the standard deviation, the more drastic the fluctuation in gas concentration within that time segment, and the greater the fluctuation intensity. The smaller the standard deviation, the smoother the fluctuation, and the smaller the fluctuation intensity.

[0048] Step S1226: Analyze the variation pattern of concentration fluctuation intensity over time in each time segment, identify the periodic repetition pattern or random change pattern of fluctuation intensity, and determine the characteristics of the concentration fluctuation pattern. The periodic repetition pattern is characterized by the consistent interval between peak fluctuation intensity, while the random change pattern is characterized by the irregular interval between peak fluctuation intensity.

[0049] In this embodiment, the concentration fluctuation pattern characteristics are determined by analyzing the changes in concentration fluctuation intensity over time in each time segment. If the peak values ​​of the fluctuation intensity occur at roughly the same interval, such as a peak occurring at a fixed time each day, it indicates that the gas concentration fluctuation exhibits a periodic repetitive pattern. If the intervals between peak values ​​are irregular, fluctuating in length, it indicates that the fluctuation exhibits a random variation pattern.

[0050] Step S123: Perform feature extraction on the temperature distribution data sequence, analyze the distribution pattern of temperature parameters in the spatial dimension and the variation law in the temporal dimension, and determine the spatial distribution characteristics and temporal variation characteristics of the temperature field. The spatial distribution characteristics of the temperature field include uniform distribution characteristics, gradient distribution characteristics and local clustered distribution characteristics. The temporal variation characteristics of the temperature field include continuous increase characteristics, continuous decrease characteristics and alternating fluctuation characteristics.

[0051] In this embodiment, feature extraction of the temperature distribution data sequence aims to analyze temperature changes from both spatial and temporal dimensions.

[0052] Step S1231: Extract the spatial coordinates of the temperature sensors from the temperature distribution data sequence, associate the temperature measurement values ​​of each sensor with the spatial coordinates, and generate a temperature spatial distribution dataset.

[0053] In this embodiment, each temperature sensor has its fixed spatial coordinates, which are determined and stored during sensor deployment. After extracting the spatial coordinates of each temperature sensor from the temperature distribution data sequence, these coordinates are associated with the corresponding temperature measurement value to form a temperature spatial distribution dataset. For example, if the spatial coordinates of a certain temperature sensor are (X1, Y1) and its corresponding temperature measurement value is T1, then it is recorded in the dataset as (X1, Y1, T1).

[0054] Step S1232: Interpolate the temperature spatial distribution dataset to construct a temperature field spatial distribution grid. Each grid cell of the temperature field spatial distribution grid contains the temperature estimate for the corresponding spatial location.

[0055] In this embodiment, interpolation is used to fill the spatial gaps between temperature sensors and construct a continuous spatial distribution of the temperature field. The interpolation method used can be linear interpolation, polynomial interpolation, etc. Through interpolation, the entire functional area is divided into multiple uniform grid cells, each with a corresponding temperature estimate. These temperature estimates are calculated by interpolation based on the measurements from surrounding temperature sensors.

[0056] Step S1233: Analyze the uniformity of temperature estimates in the spatial distribution grid of the temperature field, calculate the temperature difference coefficient between grid cells, and determine the spatial distribution characteristics of the temperature field based on the magnitude of the temperature difference coefficient. When the temperature difference coefficient is less than a preset uniformity threshold, it is determined to be a uniform distribution characteristic. When the temperature difference coefficient is greater than the preset uniformity threshold and shows a unidirectional gradient change, it is determined to be a gradient distribution characteristic. When the temperature difference coefficient is greater than the preset uniformity threshold and shows a local clustering of high or low values, it is determined to be a local clustering distribution characteristic.

[0057] In this embodiment, the temperature difference coefficient is obtained by calculating the ratio of the standard deviation to the average value of the temperature estimates between grid cells. When the temperature difference coefficient is less than a preset uniformity threshold, it indicates that the temperature difference between grid cells is small and the temperature distribution is relatively uniform, exhibiting uniform distribution characteristics. When the temperature difference coefficient is greater than the preset uniformity threshold, and the temperature estimates gradually increase or decrease along a certain direction, exhibiting a unidirectional gradient change, it is determined to have gradient distribution characteristics. When the temperature difference coefficient is greater than the preset uniformity threshold, and there is a concentrated distribution of high or low temperature values ​​in some local areas, it is determined to have local clustering distribution characteristics.

[0058] Step S1234: Extract temperature measurement values ​​from the same spatial location from the temperature distribution data sequence and arrange them in order of collection timestamps to form a temperature time series.

[0059] In this embodiment, for each spatial location, all temperature measurements at that location are extracted from the temperature distribution data sequence and arranged in chronological order of the acquisition timestamps to form a temperature time series. For example, the temperature measurements at a certain spatial location are T1, T2, T3, etc., and arranged in chronological order to form the temperature time series for that location.

[0060] Step S1235: Perform sliding window analysis on the temperature time series, calculate the average temperature within each sliding window, analyze the direction of change of the average temperature as the sliding window moves, and determine the temperature time change characteristics. When the average temperature continuously increases, it is determined to be a continuous rising characteristic. When the average temperature continuously decreases, it is determined to be a continuous falling characteristic. When the average temperature alternates between increasing and decreasing within a preset range, it is determined to be an alternating fluctuation characteristic.

[0061] In this embodiment, the size of the sliding window can be set according to the rate of temperature change. The sliding window is moved sequentially over a temperature time series, and the average temperature within each window is calculated. By observing the direction of change of the average temperature as the sliding window moves, if the average value continuously increases, it indicates that the temperature at that spatial location has a continuously rising characteristic; if the average value continuously decreases, it indicates a continuously decreasing characteristic; if the average value alternately increases and decreases within a preset range, it indicates an alternating fluctuation characteristic.

[0062] Step S124: Perform feature extraction on the particulate matter concentration data sequence to identify the peak occurrence pattern and diffusion range characteristics of particulate matter concentration, and determine the particulate matter source correlation characteristics and diffusion path characteristics. The particulate matter source correlation characteristics include equipment emission correlation characteristics and regional diffusion correlation characteristics, and the diffusion path characteristics include local aggregation path characteristics and widespread diffusion path characteristics.

[0063] In this embodiment, feature extraction of particulate matter concentration data sequences can help determine the source and diffusion of particulate matter.

[0064] Step S1241: Identify points in the particulate matter concentration data sequence whose concentration values ​​are higher than those of the surrounding concentration values ​​as peak points, and record the collection timestamp and corresponding spatial coordinates of each peak point.

[0065] In this embodiment, peak points are identified by comparing each concentration value in the particulate matter concentration data sequence with concentration values ​​within a certain surrounding range. For example, if a certain concentration value is greater than the concentration values ​​of multiple surrounding points, it is determined as a peak point, and the collection timestamp and spatial coordinates of the peak point are recorded.

[0066] Step S1242: Analyze the collection timestamps of peak points to determine the time interval and frequency of peak occurrences and identify the patterns of peak occurrences.

[0067] In this embodiment, the timing pattern of peak occurrences is analyzed by statistically analyzing the intervals between peak point collection timestamps, such as whether the peaks occur at fixed intervals or randomly. Simultaneously, the frequency of peak occurrences is determined by calculating the number of times the peaks occur per unit time.

[0068] Step S1243: Determine the diffusion range of particulate matter based on the spatial coordinates of the peak point and the distribution of particulate matter concentration in the surrounding area. The diffusion range is defined by the area where the concentration value drops to a certain proportion of the peak point concentration value.

[0069] In this embodiment, the concentration of particulate matter around the peak point is observed. When the concentration drops to a certain percentage of the peak point's concentration, this is used as the boundary to determine the diffusion range of the particulate matter. For example, when the concentration at a certain point in the vicinity drops to a certain percentage of the peak point's concentration, the location of that point is the boundary of the diffusion range. Using this method, the spatial range of particulate matter diffusion from the peak point to the surrounding area can be accurately defined.

[0070] Step S1244: Analyze the spatial coordinates of the particulate matter peak point and their relationship with the location of equipment and functional areas within the incineration plant to determine the source-related characteristics of particulate matter.

[0071] In this embodiment, the spatial coordinates of the peak particulate matter point are compared with the location information of each piece of equipment in the incineration plant and the location range of each functional area. If the peak point is close to a piece of equipment that may generate particulate matter during operation, such as an incinerator or crusher, then the particulate matter is determined to have equipment emission correlation characteristics. If the peak point is located at the boundary of two functional areas, or its diffusion range covers multiple functional areas, and considering the production activities in each area, it is determined that the particulate matter may have diffused from a certain functional area, then it is determined to have regional diffusion correlation characteristics.

[0072] Step S1245: Based on the distribution and changes of particulate matter concentration at different spatial locations, and combined with environmental factors such as wind direction and airflow, determine the characteristics of particulate matter diffusion paths.

[0073] In this embodiment, by analyzing the distribution data of particulate matter concentration at different time points in various spatial locations, the trajectory of the concentration value spreading from the peak point to the surrounding area is observed. If the particulate matter mainly accumulates and spreads within a small area, such as only spreading in a small area around the incinerator, it is determined to be a local accumulation path characteristic; if the particulate matter spreads over a large area, involving multiple functional areas, it is determined to be a widespread diffusion path characteristic. At the same time, by combining wind direction and airflow monitoring data within the incineration plant, the direction and trend of the diffusion path can be determined more accurately.

[0074] Step S125: The features of the gas composition data sequence, the temperature distribution data sequence, and the particulate matter concentration data sequence are associated and integrated according to a preset feature combination rule to generate a regional environmental feature set containing gas feature groups, temperature feature groups, and particulate matter feature groups. The feature combination rule is used to limit the arrangement order and association identification method of the feature groups.

[0075] In this embodiment, the preset feature combination rules specify the order of gas feature groups, temperature feature groups, and particulate matter feature groups in the regional environmental feature set, for example, they are arranged in the order of gas feature groups, temperature feature groups, and particulate matter feature groups.

[0076] The association identification method involves adding association identifiers to mark related features in different feature groups. For example, if the concentration change trend of a certain gas in the gas feature group is related to the temperature change feature of a certain area in the temperature feature group, then the same association identifier is added to both features to indicate their inherent relationship. Through the above association integration, a complete set of regional environmental features is formed for each functional area, which comprehensively reflects the environmental conditions of that area.

[0077] Step S130: Share the set of regional environmental characteristics of each functional area across regions, and identify the environmental impact correlation between different functional areas through cross-regional correlation analysis.

[0078] In this embodiment, cross-regional sharing and correlation analysis are used to break down data barriers between functional areas, explore the environmental impact relationships between different areas, and thus gain a more comprehensive understanding of the overall environmental status of the incineration plant.

[0079] Step S131: Encapsulate the regional environmental feature set of each functional area, add a sending area identifier and a data timeliness tag, wherein the data timeliness tag is used to indicate the effective analysis time of the regional environmental feature set.

[0080] In this embodiment, the encapsulation process packages the set of regional environmental features according to a certain format for transmission and sharing in the edge computing network. The sending region identifier indicates which functional region the set of regional environmental features originates from, and is consistent with the region identifier encoding in step S112.

[0081] The data timeliness label records the generation time and effective analysis duration of the regional environmental feature set. After the effective analysis duration expires, the data of the regional environmental feature set may no longer be applicable to the current analysis, and a new regional environmental feature set needs to be obtained. For example, in areas where the environment changes rapidly, the effective analysis duration of the regional environmental feature set is shorter; in areas with relatively stable environments, the effective analysis duration can be appropriately extended.

[0082] Step S132: The encapsulated set of regional environmental features is sent to a preset shared regional node through the inter-regional transmission channel of the edge computing network. The shared regional node is a node corresponding to other functional regions that have a physical location association with the current functional region.

[0083] In this embodiment, the inter-regional transmission channel of the edge computing network can be a wired Ethernet, wireless LAN, or other communication channel to ensure the stability and security of data transmission. The preset shared area nodes are determined based on the physical location relationship of each functional area. For example, the shared area nodes of the incineration area can be the nodes corresponding to the flue gas treatment area and the auxiliary equipment area, because these areas are physically adjacent to the incineration area and have a closer relationship with the environmental impact.

[0084] By sending the encapsulated set of regional environmental features to these shared regional nodes, the sharing of regional environmental feature data can be achieved.

[0085] Step S133: Receive the regional environmental feature sets from other functional areas, perform timeliness verification on the received regional environmental feature sets, and remove regional environmental feature sets whose data timeliness labels exceed the valid analysis time.

[0086] In this embodiment, when an edge computing node in a certain functional area receives a set of regional environmental features from other functional areas, it first checks the data timeliness label. The generation time in the label is compared with the current time, and the time difference is calculated. If the time difference exceeds the valid analysis time, it indicates that the set of regional environmental features has expired and is removed; if it does not exceed the valid analysis time, the set is retained for subsequent analysis.

[0087] The above processing ensures that the data used for cross-regional correlation analysis is up-to-date and valid, thus improving the accuracy of the analysis results.

[0088] Step S134: Summarize the locally generated regional environmental feature set with the received other regional environmental feature sets to construct a cross-regional feature analysis dataset, which contains regional environmental feature sets of all functional areas within the same time range.

[0089] In this embodiment, the locally generated regional environmental feature set is the environmental feature data of the current functional area itself, while the received environmental feature sets of other areas are environmental feature data from shared area nodes. These data are aggregated to form a cross-regional feature analysis dataset.

[0090] During the aggregation process, it is necessary to ensure that all regional environmental feature sets are within the same time frame to facilitate temporal correlation analysis. If the time frames of environmental feature sets for some regions differ from those for other regions, time alignment is required, and a common time frame must be selected as the time benchmark for analysis.

[0091] Step S135: Perform time synchronization processing on the cross-regional feature analysis dataset, and based on the time-synchronized cross-regional feature analysis dataset, perform feature association rule mining to identify the co-occurrence relationship and causal relationship between the regional environmental feature sets of different functional areas.

[0092] In this embodiment, time synchronization is used to eliminate discrepancies in time recording between different regional environmental feature sets, ensuring that all data is based on a unified time standard. By synchronizing with a time server in the edge computing network, the timestamps of each regional environmental feature set are adjusted to a uniform time format and precision.

[0093] Feature association rule mining analyzes the relationships between environmental features in different regions of a time-synchronized cross-regional feature analysis dataset to identify co-occurrence and causal relationships. Co-occurrence refers to the occurrence of similar or identical changes in environmental features in different regions at the same or similar times; causal relationship refers to the change in environmental features in one region leading to a change in environmental features in another region.

[0094] Step S1351: Extract feature vectors for each functional region from the time-synchronized cross-regional feature analysis dataset. Each feature vector contains all feature parameters from the regional environmental feature set.

[0095] In this embodiment, the regional environmental feature set of each functional area includes a gas feature group, a temperature feature group, and a particulate matter feature group, and each feature group contains multiple feature parameters. These feature parameters are arranged in a certain order to form the feature vector of that functional area.

[0096] For example, the gas feature group of a certain functional area includes parameters such as the concentration change trend of gas A and the concentration fluctuation pattern of gas B; the temperature feature group includes parameters such as the spatial distribution characteristics of the temperature field and the temperature change characteristics over time; and the particulate matter feature group includes parameters such as the source correlation characteristics of particulate matter and the diffusion path characteristics. Arranging these parameters in sequence constitutes the feature vector of that area.

[0097] Step S1352: Calculate the similarity between feature vectors of any two functional regions and generate a feature similarity matrix, wherein the elements of the feature similarity matrix represent the degree of similarity between the feature vectors of the corresponding two functional regions.

[0098] In this embodiment, the method for calculating the similarity of feature vectors can be cosine similarity, Pearson correlation coefficient, etc. For any two feature vectors of functional regions, their similarity value is obtained through the corresponding calculation method. The larger the similarity value, the more similar the two feature vectors are, that is, the more similar the environmental feature changes of the two regions are.

[0099] Arrange the pairwise similarity values ​​of all functional regions in matrix form to form a feature similarity matrix. The rows and columns of the matrix correspond to different functional regions, and the elements in the matrix are the similarity values ​​of the feature vectors of the corresponding two functional regions.

[0100] Step S1353: Based on the feature similarity matrix, filter functional region pairs with similarity values ​​greater than a preset threshold to determine potential combinations of functional regions with related relationships.

[0101] In this embodiment, the preset threshold is set based on the actual environmental conditions and analysis requirements of the incineration plant. By comparing the elements in the feature similarity matrix with the preset threshold, functional region pairs with similarity values ​​greater than the preset threshold are selected.

[0102] These functional regions are identified as potentially related combinations of functional regions due to the high similarity of their feature vectors, indicating that their environmental feature changes are similar. Therefore, they will be used as the objects of further in-depth analysis.

[0103] Step S1354: Perform time series cross-analysis on the set of regional environmental features of the functional area combination to identify the order of changes in feature parameters and determine the leading and lagging regions of feature changes.

[0104] In this embodiment, time series cross-analysis identifies the order in which feature parameters change by comparing the changes in the set of regional environmental features of two regions in a functional region combination over time. For example, if a feature parameter changes first in functional region A and then a similar change occurs in functional region B, then functional region A is the leading region and functional region B is the lagging region.

[0105] The above analysis allows us to determine which region changes first and which changes later during the process of environmental characteristic changes.

[0106] Step S1355: Calculate the probability that the characteristics of the lagging region will change accordingly after the characteristics of the leading region change, and calculate the correlation confidence parameter, which represents the probability that the characteristics of the lagging region will change due to the characteristics of the leading region.

[0107] In this embodiment, historical data is statistically analyzed to record the number of times the lagging region characteristics change within a corresponding time period after a change in the leading region characteristics, as well as the total number of changes in the leading region characteristics. The association confidence parameter is calculated as the ratio of the number of corresponding changes to the total number of changes.

[0108] The larger the value of this parameter, the greater the likelihood that changes in the characteristics of the leading region will lead to changes in the characteristics of the lagging region, that is, the higher the degree of correlation between the two.

[0109] Step S1356: Based on the order of feature changes and the associated confidence parameter, distinguish between co-occurrence relationship and causal relationship. The co-occurrence relationship is characterized by no obvious order of feature changes in the two regions. The causal relationship is characterized by the feature changes in the leading region preceding the feature changes in the lagging region and the associated confidence parameter being greater than a preset confidence threshold.

[0110] In this embodiment, if the feature changes of the two regions do not have a clear temporal order, that is, they change almost simultaneously, it indicates that there is a co-occurrence relationship between them; if the feature change of the leading region is significantly earlier than the feature change of the lagging region, and the calculated association confidence parameter is greater than the preset confidence threshold, it indicates that there is a causal relationship between the two, that is, the feature change of the leading region is one of the reasons for the feature change of the lagging region.

[0111] Step S136: Based on the feature association rule mining results, determine the direction and intensity of the environmental impact, and generate environmental impact association relationships that include the source area identifier, the affected area identifier, the impact path description, and the impact degree parameter.

[0112] In this embodiment, the direction of environmental impact is determined based on the analysis results of co-occurrence and causal relationships. For causal relationships, the direction of impact is from the leading region to the lagging region, that is, the leading region is the source region of impact, and the lagging region is the affected region; for co-occurrence relationships, although a clear causal direction cannot be determined, it can be considered that there is a mutual influence relationship between them.

[0113] The intensity of the impact is determined by a combination of factors, including the correlation confidence parameter and the magnitude of feature changes, while the degree of impact parameter is a quantitative representation of the impact intensity. The impact path description is a textual description of the path by which the environmental impact propagates from the source region to the affected region, generated by combining information such as diffusion path characteristics.

[0114] By integrating the source area identifier, affected area identifier, impact path description, and impact degree parameters, an environmental impact correlation is formed, which comprehensively describes the environmental impact between different functional areas.

[0115] Step S140: Construct an environmental risk propagation path model based on the environmental impact correlation, and predict the environmental parameter change trends of each functional area through the environmental risk propagation path model.

[0116] In this embodiment, the environmental risk propagation path model is constructed to simulate and predict the propagation of environmental risks between functional areas.

[0117] Step S141: Based on the source area identifier and affected area identifier in the environmental impact association, construct an environmental impact directed graph, where the nodes of the environmental impact directed graph represent functional areas and the directed edges represent the direction of environmental impact.

[0118] In this embodiment, the directed environmental impact graph is a graphical representation where each node represents a functional area, and the node's identifier matches the area identifier code of the functional area. Directed edges point from nodes in the source region to nodes in the affected region, visually representing the direction of the environmental impact.

[0119] For example, if there is an impact from the incineration area to the flue gas treatment area in the environmental impact correlation, then in the directed graph of environmental impact, there is a directed edge from the node representing the incineration area to the node representing the flue gas treatment area.

[0120] Step S142: Add an influence intensity parameter to the directed edges of the directed graph of environmental impact, the influence intensity parameter being determined based on the degree of influence parameter in the environmental impact association relationship.

[0121] In this embodiment, the impact intensity parameter is a quantitative representation of the degree of environmental impact, and its value is derived from the impact degree parameter in the environmental impact correlation. For example, the impact degree parameter can be converted into the corresponding impact intensity parameter value according to a certain mapping rule; the greater the impact degree, the larger the impact intensity parameter value.

[0122] By adding an influence strength parameter to directed edges, the strength of environmental influences in different directions can be represented more accurately, making directed graphs of environmental influences more valuable for quantitative analysis.

[0123] Step S143: Collect historical environmental parameter change data, which includes the environmental parameter measurement values ​​of each functional area at different time periods and the corresponding environmental impact correlations.

[0124] In this embodiment, historical environmental parameter change data is collected from the incineration plant's database. This data records the environmental parameter measurements of each functional area over a period of time, as well as the environmental impact correlations that exist within the corresponding time period.

[0125] The purpose of collecting this data is to calibrate and validate the environmental risk propagation path model, thereby improving the model's predictive accuracy. The historical data should have a sufficiently long time span to cover various scenarios such as different seasons and different production conditions.

[0126] Step S144: Use the historical environmental parameter change data to calibrate the influence intensity parameter of the directed graph of environmental impact and adjust the weight values ​​of the directed edges.

[0127] In this embodiment, historical environmental parameter change data is input into the directed environmental impact graph. By comparing the environmental impact results predicted by the model with the actual environmental impact results in the historical data, the deviation between the two is calculated. Based on the magnitude and direction of the deviation, the influence intensity parameter of the directed edge is adjusted, i.e., the weight value is adjusted.

[0128] After multiple iterative adjustments, the impact intensity parameters were calibrated until the deviation between the model's predictions and historical data was within an acceptable range. The calibrated impact intensity parameters better reflect the actual environmental impact.

[0129] Step S145: Construct an environmental risk propagation path model based on the calibrated directed environmental impact graph. The environmental risk propagation path model includes impact propagation rules and path probability calculation methods. The impact propagation rules specify the way environmental impacts are transmitted between functional areas, and the path probability calculation methods are used to determine the probability of occurrence of different propagation paths.

[0130] In this embodiment, the influence propagation rules specify how environmental influences are transmitted between functional areas. For example, influences can only be transmitted along the direction of directed edges, and the intensity of influences decreases as the transmission distance increases.

[0131] The path probability calculation method analyzes all possible paths from the source region to the affected region, combining the influence strength parameters of each directed edge with the propagation frequency in historical data to calculate the probability of each path occurring. A higher probability indicates a greater likelihood of environmental impact propagation through that path in reality.

[0132] These rules and methods enable the construction of an environmental risk propagation path model that can simulate the process of environmental risk propagation.

[0133] Step S146: Input the set of regional environmental characteristics of each functional area into the environmental risk propagation path model to simulate the propagation process of environmental impacts between functional areas.

[0134] In this embodiment, the current set of regional environmental features for each functional area contains the latest environmental feature data. This data is input into the environmental risk propagation path model. The model simulates the process of environmental impacts propagating from the source area to the affected area based on the impact propagation rules and path probability calculation methods.

[0135] During the simulation, the model considers factors such as the influence intensity parameters of each directed edge and the probability of path occurrence to calculate the propagation and intensity changes of environmental impacts between different functional areas.

[0136] Step S147: Based on the simulation results, predict the possible changes in environmental parameters of each functional area within a future preset time period, analyze the direction, magnitude and rate of change, and generate an environmental parameter change trend that includes parameter change direction identifier, change magnitude parameter and change rate parameter.

[0137] In this embodiment, the preset time period can be set according to actual needs, such as the next hour or the next day. Based on the simulation results of the environmental risk propagation path model, the possible changes in environmental parameters of each functional area within the preset time period are analyzed.

[0138] The direction of parameter change is indicated by "+" to suggest that the parameter may increase, and "-" to suggest that the parameter may decrease. The magnitude of change parameter represents the ratio of the change in parameter value to the current parameter value, reflecting the degree of change. The rate of change parameter represents the amount of change in parameter value per unit time, reflecting how quickly the change occurs.

[0139] These parameters are used to determine the changing trends of environmental parameters in each functional area.

[0140] Step S150: Generate control instructions for the incineration equipment based on the changing trend of the environmental parameters, and transmit the control instructions for the incineration equipment to the corresponding incineration equipment actuator.

[0141] In this embodiment, generating control commands based on the changing trends of environmental parameters and transmitting them to the actuator is a key step in realizing intelligent control of the incineration equipment. This enables timely adjustment of the equipment's operating status according to environmental changes, ensuring the safe and stable operation of the incineration plant.

[0142] Step S151: Extract the predicted values ​​of environmental parameters for each functional area from the trend of environmental parameter changes, compare the predicted values ​​of environmental parameters with the preset parameter safety threshold range, and identify abnormal predicted values ​​that exceed the parameter safety threshold range.

[0143] In this embodiment, the preset safety threshold ranges for parameters are established based on relevant environmental protection standards, incineration plant design specifications, and actual operating experience. Different environmental parameters correspond to different safety threshold ranges. For example, the preset safety threshold range for the temperature parameter in the incineration zone is determined based on the type of incinerated material and the requirements of the incineration process; the safety threshold range for the concentration of harmful gases in the flue gas treatment zone strictly follows the national emission standards.

[0144] After extracting the predicted environmental parameters for each functional area from the trends in environmental parameter changes, each predicted value is compared one by one with the corresponding safety threshold range. If a predicted value is higher than the upper limit of the range or lower than the lower limit, it is identified as an abnormal predicted value. These abnormal predicted values ​​indicate that the corresponding environmental parameters may exceed the safe range in the future, requiring attention and appropriate control measures.

[0145] Step S152: Determine the functional area identifier and parameter type identifier corresponding to the abnormal predicted value, trace the environmental impact correlation that led to the abnormal predicted value, and locate the functional area of ​​the impact source.

[0146] In this embodiment, after identifying an abnormal predicted value, the corresponding functional area identifier is first determined based on the changing trend of the environmental parameter to which the predicted value belongs, i.e., which functional area the abnormal predicted value originates from. Simultaneously, the parameter type identifier is determined to clarify which environmental parameter is abnormal, such as temperature, concentration of a certain type of gas, or particulate matter concentration.

[0147] Then, combining the previously identified environmental impact correlations, the causes of the abnormal predicted value are traced. By analyzing the environmental impact correlations of the functional area where the abnormal predicted value is located, other functional areas that may have an impact are identified, thereby pinpointing the source functional area. For example, if the predicted concentration of a certain harmful gas in the flue gas treatment area is abnormal, tracing the environmental impact correlations may reveal that the anomaly is likely due to excessively high incineration temperature in the incineration area; in this case, the incineration area is the source functional area.

[0148] Step S153: Based on the magnitude of the abnormal predicted value and the importance of the functional area of ​​the source of influence, determine the control priority parameter, which is used to indicate the execution order of the control commands of the incineration equipment.

[0149] In this embodiment, the excess range of the abnormal prediction value refers to the difference between the abnormal prediction value and the boundary value of the parameter safety threshold range. The larger the difference, the greater the excess range, and the greater the potential threat to the environment and production safety.

[0150] The importance of the source functional area is determined by its role in the incineration plant's production process. Core production areas, such as the incineration area and flue gas treatment area, are more important than auxiliary equipment areas and office areas.

[0151] The control priority parameter is determined by comprehensively considering the magnitude of the abnormal predicted value and the importance of the affected functional area. The greater the magnitude of the abnormal value and the higher the importance of the affected functional area, the higher the control priority parameter, and the corresponding incineration equipment control commands need to be executed first; conversely, the lower the control priority parameter, the later the command execution order.

[0152] Step S154: Query the preset control strategy library, and select the corresponding equipment control scheme according to the parameter type identifier and control priority parameter of the abnormal prediction value. The equipment control scheme includes target control parameters and parameter adjustment methods.

[0153] In this embodiment, a preset control strategy library stores equipment control schemes for different parameter types and control priorities. Each equipment control scheme is developed based on historical experience and experimental data to guide the adjustment of the incineration equipment.

[0154] Once the parameter type identifier and control priority parameter of the abnormal prediction value are determined, a matching query is performed in the control strategy library to select the equipment control scheme that matches these two identifiers. For example, if the parameter type identifier is the temperature of the incineration zone and the control priority parameter is high, the queried equipment control scheme may be to reduce the fuel supply to the incinerator to lower the incineration temperature, where the target control parameter is the adjusted fuel supply and the parameter adjustment method is to gradually reduce the supply.

[0155] Step S155: Generate incineration equipment control instructions according to the equipment control scheme. The incineration equipment control instructions include equipment identifier, parameter type identifier, target parameter value and execution time parameter.

[0156] In this embodiment, the device identifier is used to specify the specific incineration equipment that needs to execute control commands, such as an incinerator with a specific number, an induced draft fan, a dust collector, etc. The parameter type identifier is consistent with the previously determined one, indicating the type of environmental parameter that needs to be adjusted.

[0157] The target parameter value is the specific numerical value of the target control parameter specified in the equipment control scheme, that is, the parameter value that the equipment needs to achieve after adjustment. The execution time parameter specifies the execution time of the control command, such as immediate execution or execution at a specific point in time.

[0158] By integrating this information, complete control instructions for the incineration equipment are generated, ensuring that the equipment can accurately execute adjustment actions according to the instructions.

[0159] Step S156: Determine the target transmission path corresponding to the control command of the incineration equipment. The target transmission path is determined based on the functional area identifier corresponding to the abnormal prediction value and the functional area identifier of the influencing source.

[0160] In this embodiment, determining the target transmission path requires considering the location of the functional region corresponding to the anomaly prediction value and the functional region of the influencing source. If the functional region of the influencing source is consistent with the region where the device that needs to execute control commands is located, the target transmission path can directly point from the edge computing node to the communication interface of the region where the device is located.

[0161] If the area affected by the source function differs from the area where the equipment is located, an optimal path needs to be selected based on the incineration plant's network topology. This path should run from the edge computing node through intermediate transmission nodes to the communication interface in the area where the equipment is located. The selection of the optimal path typically considers factors such as transmission delay and stability to ensure that control commands can be transmitted quickly and accurately.

[0162] Step S157: The control command of the incineration equipment is transmitted to the device communication interface of the target functional area through the command transmission channel of the edge computing network. The device communication interface is connected to the actuator of the incineration equipment so that the actuator of the incineration equipment receives the control command of the incineration equipment, parses the command content and performs the corresponding parameter adjustment action.

[0163] In this embodiment, the command transmission channel of the edge computing network adopts a reliable communication protocol, such as TCP / IP, to ensure that the control commands of the incineration equipment are not lost or damaged during transmission.

[0164] Once the control command is transmitted to the device communication interface of the target functional area, the device communication interface passes the command to the incineration equipment actuator. The actuator parses the command content, clarifies the equipment to be adjusted, parameter types, target parameter values, and execution time, and then performs the corresponding parameter adjustment actions according to the parsing results.

[0165] For example, after receiving a control command to reduce the fuel supply to the incinerator, the actuator can adjust the opening of the fuel supply valve to gradually reduce the fuel supply until the target parameter value is reached, thereby controlling the incinerator temperature and mitigating the potential risks brought about by abnormal predicted values.

[0166] Figure 2 Schematic diagrams are shown of exemplary hardware and software components of an edge computing-based incinerator environmental data processing system 100 that can implement the inventive ideas provided in some embodiments of the present invention. For example, a processor 120 may be used in the edge computing-based incinerator environmental data processing system 100 and to perform the functions in the present invention.

[0167] For example, the edge computing-based incinerator environmental data processing system 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 various forms of storage media 140, such as a disk, ROM, or RAM, or any combination thereof. Exemplarily, the edge computing-based incinerator environmental data processing system 100 may also include program instructions stored in ROM, RAM, or other types of non-transitory storage media, or any combination thereof. The method of the present invention can be implemented according to these program instructions. The edge computing-based incinerator environmental data processing system 100 also includes an I / O interface 150 between the computer and other input / output devices.

[0168] Furthermore, this embodiment of the invention also provides a readable storage medium, wherein computer-executable instructions are preset in the readable storage medium, and when the processor executes the computer-executable instructions, the above-mentioned edge computing-based incineration plant environmental data processing method is implemented.

[0169] It should be noted that, in order to simplify the description of the present invention and thus help to understand one or more embodiments of the invention, multiple features may sometimes be grouped into one embodiment, drawing or description thereof in the foregoing description of the embodiments of the present invention.

Claims

1. A method for processing environmental data of an incineration plant based on edge computing, characterized in that, The method includes: Environmental sensing data is collected by environmental sensing devices deployed in various functional areas of the incineration plant. The environmental sensing data includes gas composition data, temperature distribution data, and particulate matter concentration data. The environmental perception data is processed by regional feature extraction to generate a set of regional environmental features corresponding to each functional area; The regional environmental characteristics of each functional area are encapsulated, and sending area identifiers and data timeliness tags are added. The encapsulated set of regional environmental features is sent to a preset shared regional node through the inter-regional transmission channel of the edge computing network. The shared regional node is a node corresponding to other functional regions that have a physical location association with the current functional region. Receive regional environmental feature sets from other functional areas, perform timeliness verification on the received regional environmental feature sets, and remove regional environmental feature sets whose data timeliness labels exceed the valid analysis time. The locally generated set of regional environmental features is combined with the received set of environmental features from other regions to construct a cross-regional feature analysis dataset. Time synchronization processing is performed on the cross-regional feature analysis dataset, and feature association rule mining is performed based on the time-synchronized cross-regional feature analysis dataset to identify the co-occurrence relationship and causal relationship between the regional environmental feature sets of different functional areas; Based on the results of feature association rule mining, the direction and intensity of environmental impact are determined, and environmental impact associations are generated, including the source area identifier, the affected area identifier, the impact path description, and the impact degree parameter. Based on the aforementioned environmental impact relationships, a directed graph of environmental impacts is constructed, and an impact intensity parameter is added to the directed edges of the directed graph of environmental impacts. Collect historical environmental parameter change data, and use the historical environmental parameter change data to calibrate the influence intensity parameter of the directed graph of environmental impact, and adjust the weight value of the directed edge; An environmental risk propagation path model is constructed based on the calibrated directed graph of environmental impacts. The environmental risk propagation path model includes impact propagation rules and path probability calculation methods. The set of regional environmental characteristics of each functional area is input into the environmental risk propagation path model to simulate the propagation process of environmental impacts between functional areas. Based on the simulation results, predict the possible changes in environmental parameters of each functional area within a preset time period in the future, and generate environmental parameter change trends including parameter change direction indicators, change magnitude parameters, and change rate parameters. Based on the changing trends of the environmental parameters, control commands for the incineration equipment are generated, and these control commands are transmitted to the corresponding incineration equipment actuators.

2. The edge computing-based environmental data processing method for incineration plants according to claim 1, characterized in that, The environmental sensing data collected through environmental sensing devices deployed in various functional areas of the incineration plant includes: The incineration plant is divided into functional areas according to its production process. These functional areas include an incineration area, a flue gas treatment area, a waste storage area, an auxiliary equipment area, and an office area. Each group of environmental sensing devices deployed within a functional area is configured with a regional identification code. The environmental sensing device group includes gas sensors, temperature sensors, and particulate matter sensors. The deployment locations of each type of sensor correspond to key environmental monitoring points within the functional area. The regional identification code includes a functional area type code and a location coordinate code, which are used to distinguish environmental sensing data from different functional areas. The acquisition triggering mechanism of the environmental sensing device group is obtained. The acquisition triggering mechanism includes timed triggering and event triggering. Timed triggering starts acquisition at a preset time interval, and event triggering starts acquisition when abnormal fluctuations in environmental parameters are detected. A data acquisition start signal is sent to the environmental sensing device group so that the environmental sensing device group responds to the data acquisition start signal, starts the sensors to acquire corresponding environmental parameters, and generates raw sensing data containing sensor type identifier, acquisition timestamp and parameter measurement value. The data acquisition start signal includes acquisition duration parameter and data transmission format parameter. Add region identification codes and data check codes to the original sensor data to generate environmental perception data that conforms to the edge computing network transmission specifications.

3. The edge computing-based environmental data processing method for incineration plants according to claim 1, characterized in that, The process of performing regional feature extraction on the environmental perception data to generate a set of regional environmental features corresponding to each functional area includes: Receive environmental sensing data transmitted from each functional area, perform integrity verification on the environmental sensing data, classify the environmental sensing data that has passed the verification according to parameter type, divide it into gas composition data sequence, temperature distribution data sequence and particulate matter concentration data sequence, and arrange them in order of collection timestamp; Feature extraction is performed on the gas component data sequence to identify the types of gases contained therein, and to determine the concentration change trend characteristics and concentration fluctuation pattern characteristics of each gas type. The concentration change trend characteristics include a continuous upward trend, a continuous downward trend, and a stable fluctuation trend. The concentration fluctuation pattern characteristics include a periodic fluctuation pattern and a non-periodic fluctuation pattern. Feature extraction is performed on the temperature distribution data sequence to analyze the spatial distribution pattern and temporal variation of temperature parameters, and to determine the spatial distribution characteristics and temporal variation characteristics of the temperature field. The spatial distribution characteristics of the temperature field include uniform distribution characteristics, gradient distribution characteristics, and local clustering distribution characteristics. The temporal variation characteristics of the temperature field include continuous increase characteristics, continuous decrease characteristics, and alternating fluctuation characteristics. Feature extraction is performed on the particulate matter concentration data sequence to identify the peak occurrence pattern and diffusion range characteristics of particulate matter concentration, and to determine the particulate matter source correlation characteristics and diffusion path characteristics. The particulate matter source correlation characteristics include equipment emission correlation characteristics and regional diffusion correlation characteristics, and the diffusion path characteristics include local aggregation path characteristics and widespread diffusion path characteristics. The features of the gas composition data sequence, the temperature distribution data sequence, and the particulate matter concentration data sequence are associated and integrated according to a preset feature combination rule to generate a regional environmental feature set containing gas feature groups, temperature feature groups, and particulate matter feature groups. The feature combination rule is used to limit the arrangement order and association identification method of the feature groups.

4. The edge computing-based environmental data processing method for incineration plants according to claim 3, characterized in that, The step of performing feature extraction on the gas component data sequence to identify the types of gases contained therein and determine the concentration change trend characteristics and concentration fluctuation pattern characteristics of each gas includes: Gas type identifiers are extracted from each data record of the gas composition data sequence, and major gas types are filtered based on a preset gas priority list, which is preset according to the environmental risk level of the incineration plant. The gas composition data sequence is split according to the main gas type to obtain the single gas data subsequence corresponding to each main gas type. The single gas data subsequence contains the concentration measurement values ​​of the same gas type at different collection timestamps. The concentration measurements in a single gas data subsequence are arranged in order of acquisition timestamps to generate a gas concentration time series. The gas concentration time series is subjected to trend fitting processing. The overall slope value of the series is calculated by linear regression. The concentration change trend characteristics are determined according to the sign and absolute value of the slope value. When the slope value is positive and the absolute value is greater than a preset threshold, it is determined to be a continuous upward trend. When the slope value is negative and the absolute value is greater than a preset threshold, it is determined to be a continuous downward trend. When the absolute value of the slope value is less than a preset threshold, it is determined to be a stable fluctuation trend. The gas concentration time series is divided into multiple consecutive time segments according to a preset time window. The concentration standard deviation of each time segment is calculated, and the concentration fluctuation intensity within each time segment is determined based on the magnitude of the concentration standard deviation. The variation pattern of concentration fluctuation intensity over time in each time segment is analyzed to identify periodic repetition patterns or random variation patterns of fluctuation intensity, and to determine the characteristics of concentration fluctuation patterns. The periodic repetition pattern is characterized by consistent peak intervals of fluctuation intensity, while the random variation pattern is characterized by irregular peak intervals of fluctuation intensity.

5. The edge computing-based environmental data processing method for incineration plants according to claim 3, characterized in that, The step of performing feature extraction on the temperature distribution data sequence, analyzing the spatial distribution pattern and temporal variation law of temperature parameters, and determining the spatial distribution characteristics and temporal variation characteristics of the temperature field includes: The spatial coordinates of temperature sensors are extracted from the temperature distribution data sequence, and the temperature measurement values ​​of each sensor are associated with the spatial coordinates to generate a temperature spatial distribution dataset. The temperature spatial distribution dataset is interpolated to construct a temperature field spatial distribution grid, and each grid cell of the temperature field spatial distribution grid contains the temperature estimate value of the corresponding spatial location. The uniformity of temperature estimates in the spatial distribution grid of the temperature field is analyzed, the temperature difference coefficient between grid cells is calculated, and the spatial distribution characteristics of the temperature field are determined according to the magnitude of the temperature difference coefficient. When the temperature difference coefficient is less than a preset uniformity threshold, it is determined to be a uniform distribution characteristic. When the temperature difference coefficient is greater than the preset uniformity threshold and shows a unidirectional gradient change, it is determined to be a gradient distribution characteristic. When the temperature difference coefficient is greater than the preset uniformity threshold and shows a local clustering of high or low values, it is determined to be a local clustering distribution characteristic. Temperature measurements from the same spatial location are extracted from the temperature distribution data sequence and arranged in order of collection timestamps to form a temperature time series. A sliding window analysis is performed on the temperature time series to calculate the average temperature within each sliding window. The direction of change of the average temperature as the sliding window moves is analyzed to determine the temperature time change characteristics. When the average temperature continuously increases, it is determined to be a continuous rising characteristic. When the average temperature continuously decreases, it is determined to be a continuous falling characteristic. When the average temperature alternates between increasing and decreasing within a preset range, it is determined to be an alternating fluctuation characteristic.

6. The edge computing-based environmental data processing method for incineration plants according to claim 1, characterized in that, Based on the time-synchronized cross-regional feature analysis dataset, feature association rule mining is performed to identify co-occurrence and causal relationships among regional environmental feature sets of different functional areas, including: Feature vectors for each functional region are extracted from the cross-regional feature analysis dataset after time synchronization. Each feature vector contains all feature parameters in the set of regional environmental features. Calculate the similarity between feature vectors of any two functional regions and generate a feature similarity matrix, wherein the elements of the feature similarity matrix represent the degree of similarity between the feature vectors of the corresponding two functional regions. Based on the feature similarity matrix, pairs of functional regions with similarity values ​​greater than a preset threshold are selected to determine combinations of functional regions that may have a correlation. A time-series cross-analysis is performed on the set of regional environmental features of the functional area combination to identify the order of changes in feature parameters and determine the leading and lagging regions of feature changes. The probability that the characteristics of the lagging region will change accordingly after the characteristics of the leading region change is statistically analyzed, and the correlation confidence parameter is calculated. The correlation confidence parameter represents the probability that the characteristics of the lagging region will change due to the characteristics of the leading region. Based on the order of feature changes and the associated confidence parameter, co-occurrence relationships and causal relationships are distinguished. The co-occurrence relationship is characterized by no obvious order of feature changes in the two regions, while the causal relationship is characterized by the feature changes in the leading region preceding the feature changes in the lagging region and the associated confidence parameter being greater than a preset confidence threshold.

7. The edge computing-based environmental data processing method for incineration plants according to claim 1, characterized in that, The step of generating control commands for the incineration equipment based on the changing trends of the environmental parameters, and transmitting the control commands to the corresponding incineration equipment actuators, includes: The predicted values ​​of environmental parameters for each functional area are extracted from the trend of environmental parameter changes. The predicted values ​​of environmental parameters are compared with the preset parameter safety threshold range to identify abnormal predicted values ​​that exceed the parameter safety threshold range. Determine the functional area identifier and parameter type identifier corresponding to the abnormal predicted value, trace the environmental impact correlation that led to the abnormal predicted value, and locate the functional area of ​​the impact source. Based on the magnitude of the abnormal predicted value and the importance of the functional area of ​​the affected source, a control priority parameter is determined, which is used to indicate the execution order of the control commands of the incineration equipment. The system queries the preset control strategy library and selects the corresponding equipment control scheme based on the parameter type identifier and control priority parameter of the abnormal prediction value. The equipment control scheme includes the target control parameters and parameter adjustment methods. Based on the equipment control scheme, an incineration equipment control instruction is generated, which includes an equipment identifier, a parameter type identifier, a target parameter value, and an execution time parameter. Determine the target transmission path corresponding to the control command of the incineration equipment. The target transmission path is determined based on the functional area identifier corresponding to the abnormal prediction value and the functional area identifier of the impact source. The control commands of the incineration equipment are transmitted to the device communication interface of the target functional area through the command transmission channel of the edge computing network. The device communication interface is connected to the actuator of the incineration equipment so that the actuator receives the control commands of the incineration equipment, parses the command content and performs the corresponding parameter adjustment actions.

8. An environmental data processing system for an incineration plant based on edge computing, characterized in that, The device includes a processor and a memory, the memory and the processor being connected. The memory is used to store programs, instructions or code, and the processor is used to execute the programs, instructions or code in the memory to implement the edge computing-based environmental data processing method for incineration plants as described in any one of claims 1-7.

Images