Environmental protection index anomaly detection method, system, equipment and medium

Abstract

The invention relates to an environmental protection index anomaly detection method, system and device and a medium, and the method comprises the steps: obtaining real-time environmental parameter data in a preset monitoring region through a sensor node, and transmitting the real-time environmental parameter data to an edge computing device; preprocessing the real-time environment parameter data by using a preprocessing algorithm to obtain preprocessed environment parameter data, and fusing the multi-dimensional environment data of the sensor nodes to obtain a comprehensive environment index; inputting the preprocessed environment parameter data and the comprehensive environment index into a pre-trained anomaly detection model for analysis to obtain an environment analysis result; when the environment analysis result is that the detected pollutant concentration exceeds a preset threshold value or has an abnormal change trend, triggering an early warning signal, and generating early warning information according to the early warning signal; and sending the early warning information to the user side through a preset response mechanism, and generating an abnormal state report on the cloud platform. The method has the effect of improving the accuracy of environment monitoring.

Description

Technical Field

[0001] The present application relates to the technical field of environmental monitoring, and in particular to a method, system, equipment and medium for detecting anomalies of environmental indicators. Background Art

[0002] At present, with the acceleration of industrialization and the continuous development of social economy, environmental protection has received increasing attention. At present, the common environmental monitoring equipment on the market mostly adopts Internet of Things technology, which collects basic data such as temperature, humidity, PM2.5 concentration, etc. through sensor nodes arranged in different areas, and uploads them to the cloud platform for centralized processing. However, such systems generally have a single algorithm. When pollutants in a complex environment suddenly increase or decrease, a large amount of raw data needs to be processed, and the system often finds it difficult to identify abnormal conditions in a timely and accurate manner.

[0003] The above-mentioned existing technical solutions have the following defects: the existing environmental monitoring technology has a high false alarm rate in complex environments, so there is room for improvement. Summary of the invention

[0004] In order to improve the accuracy of environmental monitoring, the present application provides an environmental indicator anomaly detection method, system, equipment and medium.

[0005] The above-mentioned invention objective of the present application is achieved through the following technical solutions: A method for detecting anomalies of environmental indicators, comprising: Acquire real-time environmental parameter data in a preset monitoring area through sensor nodes, and send the real-time environmental parameter data to an edge computing device, wherein the sensor nodes include a temperature sensor, a humidity sensor, a PM2.5 sensor, and a pollutant sensor; Preprocessing the real-time environmental parameter data using a preprocessing algorithm to obtain preprocessed environmental parameter data, and fusing the multi-dimensional environmental data of the sensor nodes to obtain a comprehensive environmental index; Inputting the preprocessed environmental parameter data and the comprehensive environmental index into a pre-trained anomaly detection model for analysis to obtain an environmental analysis result; When the environmental analysis result detects that the pollutant concentration exceeds a preset threshold or shows an abnormal change trend, a warning signal is triggered, and warning information is generated according to the warning signal, the warning information includes the concentration value of the pollutant, the time and place of the abnormality, and the possible location of the pollution source; The warning information is sent to the user end through a preset response mechanism, and an abnormal status report is generated on the cloud platform.

[0006] By adopting the above technical solution, real-time environmental parameter data in a preset monitoring area is obtained through sensor nodes, and the real-time environmental parameter data is sent to the edge computing device, so that the environmental information of the monitoring area can be obtained in real time, ensuring the timeliness of the environmental data, thereby improving the accuracy of abnormal detection of environmental indicators; by using a preprocessing algorithm to preprocess the real-time environmental parameter data, the preprocessed environmental parameter data is obtained, and the multi-dimensional environmental data of the sensor node is integrated to obtain a comprehensive environmental index, which can eliminate noise and missing information in the data, ensure the integrity and accuracy of the environmental data, and thus improve the reliability of the detection results; by inputting the preprocessed environmental parameter data and the comprehensive environmental index into a pre-trained anomaly detection model for analysis, the environmental analysis results are obtained, which can timely identify abnormal environmental change trends and pollutant concentrations exceeding the standard, thereby improving the ability to identify pollution sources; by triggering early warning signals and generating early warning information, real-time early warning of abnormal conditions can be achieved, helping relevant personnel to respond quickly, thereby improving the response speed of environmental management.

[0007] In a preferred example, the present application can be further configured as follows: the step of acquiring real-time environmental parameter data in a preset monitoring area through a sensor node and sending the real-time environmental parameter data to an edge computing device includes: Monitor the external environment data acquired by the sensor node from the preset monitoring area in real time to obtain the real-time external environment parameter data, and monitor the status of the sensor node in real time, and send a fault alarm to the edge computing device when the sensor node fails; The real-time environmental parameter data is wirelessly transmitted to the edge computing device at a predetermined sampling frequency by adopting an adaptive data transmission protocol, and the real-time environmental parameter is encrypted by adopting transport layer encryption technology during the data transmission process.

[0008] By adopting the above technical solution, real-time external environmental parameter data can be obtained by real-time monitoring of the external environmental data obtained by the sensor nodes from the preset monitoring area, which can ensure the real-time and accuracy of the environmental status information of the monitoring area, thereby improving the comprehensiveness of anomaly detection; the real-time environmental parameter data is wirelessly transmitted to the edge computing device at a predetermined sampling frequency through an adaptive data transmission protocol, and the transport layer encryption technology is used for encryption processing during the data transmission process, which can ensure the security of the data during the transmission process and prevent the data from being tampered with or leaked, thereby improving the security of the system and data integrity.

[0009] In a preferred example, the present application can be further configured as follows: the real-time environmental parameter data is preprocessed by the preprocessing algorithm to obtain the preprocessed environmental parameter data, and the multi-dimensional environmental data of the sensor node is fused to obtain a comprehensive environmental index including: Applying an adaptive filtering algorithm to eliminate noise from the real-time environmental parameter data, and using an interpolation method and a moving average technique to fill in missing data in the real-time environmental parameter data and to smooth the data curve, thereby obtaining the pre-processed environmental parameter data; The principal component analysis technique is used to extract features from the multidimensional data of the sensor nodes, and the comprehensive environmental index is constructed in combination with the fuzzy logic method.

[0010] By adopting the above technical solution, by applying an adaptive filtering algorithm to eliminate noise in real-time environmental parameter data, the noise interference in the environmental data can be effectively removed, the accuracy of the data can be ensured, and the accuracy of subsequent processing can be improved; by filling and smoothing the missing data through interpolation methods and moving average technology, the data missing can be effectively compensated and the continuity of the data can be improved, thereby ensuring the accuracy and consistency of the analysis results; by using principal component analysis technology to extract features from multidimensional data and combining fuzzy logic methods to construct a comprehensive environmental index, key features can be extracted from complex data, redundant information can be reduced, and the efficiency and accuracy of environmental indicator analysis can be improved.

[0011] In a preferred example, the present application may be further configured as follows: before the pre-processed environmental parameter data and the comprehensive environmental index are input into a pre-trained anomaly detection model for analysis and the environmental analysis result is obtained, the environmental index anomaly detection method further includes: Acquire historical environmental data and environmental standard data, normalize and annotate the historical environmental data and the environmental standard data to obtain a training set; Using the training set, forward propagation and back propagation training are performed on an anomaly detection model constructed based on a long short-term memory network and a convolutional neural network to obtain an anomaly detection model after forward propagation and back propagation training; The parameters of the anomaly detection model after forward propagation and back propagation training are optimized by a transfer learning method to obtain the pre-trained anomaly detection model.

[0012] By adopting the above technical solution, by obtaining historical environmental data and environmental standard data, and normalizing and labeling them to obtain a training set, it is possible to ensure that the anomaly detection model has sufficient and standardized data samples, thereby improving the training effect of the model; by using long short-term memory networks and convolutional neural networks for forward and back propagation training, it is possible to effectively capture the temporal dynamic characteristics of complex environmental data, thereby improving the model's ability to detect abnormal events; by using the transfer learning method to optimize the parameters of the anomaly detection model after forward and back propagation training, the model can adapt to different environmental conditions, improve the generalization ability of the model, and thus improve the accuracy of anomaly detection.

[0013] In a preferred example, the present application can be further configured as follows: the pre-processed environmental parameter data and the comprehensive environmental index are input into a pre-trained anomaly detection model for analysis, and the environmental analysis results obtained include: Capturing the temporal dynamic change characteristics of the pre-processed environmental parameter data, analyzing the temporal dynamic change characteristics, and obtaining a prediction scheme for the pollutant diffusion path; Compare and analyze the comprehensive conversion index with the environmental standard data in the pre-trained anomaly detection model to obtain a data comparison and analysis result; The prediction scheme of the pollutant diffusion path and the data comparison analysis results are analyzed in combination with the Bayesian inference method to obtain the environmental analysis results.

[0014] By adopting the above technical solution, by capturing the temporal dynamic change characteristics of the pre-processed environmental parameter data, it is possible to accurately analyze the changing trends of environmental indicators and predict potential pollution diffusion paths, thereby improving the predictability of anomaly detection; by comparing and analyzing the comprehensive environmental index with the environmental standard data in the pre-trained anomaly detection model, it is possible to accurately identify environmental anomalies that exceed the standards, thereby improving the compliance of environmental protection indicators; by combining the Bayesian inference method to conduct a comprehensive analysis of the prediction scheme of the pollutant diffusion path and the data comparison and analysis results, it is possible to enhance the comprehensive understanding of abnormal events, thereby improving the accuracy of decision support.

[0015] In a preferred example, the present application can be further configured as follows: the environmental protection index abnormality detection method further includes: The pre-processed environmental parameter data, the comprehensive environmental index and the environmental analysis results are transmitted to the user terminal for display in the form of a time series diagram, a heat map and a relationship network through Rerun; Dynamically adjust the data collection frequency, anomaly detection threshold and device operation status of the sensor node based on real-time user feedback and changes in the environment in which the sensor node is located; When data is transmitted to the cloud platform, the data transmitted to the cloud platform is encrypted and stored using blockchain technology.

[0016] By adopting the above technical solution, the pre-processed environmental parameter data, comprehensive environmental index and environmental analysis results are transmitted to the user end through Rerun in the form of time series diagrams, heat maps and relationship networks for display, which can intuitively display environmental abnormalities and help users quickly understand abnormal situations, thereby improving the efficiency of abnormal response; by dynamically adjusting the data collection frequency, abnormal detection threshold and equipment operation status based on user feedback and environmental changes of sensor nodes, the flexibility and adaptability of the detection system can be improved, thereby ensuring the accuracy and effectiveness of abnormal detection; by using blockchain technology for encrypted storage when data is transmitted to the cloud platform, the security and non-tamperability of environmental data can be ensured, thereby improving the data management capabilities of the system.

[0017] The second object of the invention is achieved by the following technical solutions: An environmental indicator abnormality detection system, the environmental indicator abnormality detection system comprising: A sensing detection module is used to obtain real-time environmental parameter data in a preset monitoring area through sensor nodes, and send the real-time environmental parameter data to an edge computing device. The sensor nodes include temperature sensors, humidity sensors, PM2.5 sensors, and pollutant sensors. A data preprocessing module is used to preprocess the real-time environmental parameter data using a preprocessing algorithm to obtain preprocessed environmental parameter data, and to fuse the multi-dimensional environmental data of the sensor nodes to obtain a comprehensive environmental index; An analysis module, used for inputting the pre-processed environmental parameter data and the comprehensive environmental index into a pre-trained anomaly detection model for analysis to obtain an environmental analysis result; An early warning module is used to trigger an early warning signal when the environmental analysis result detects that the pollutant concentration exceeds a preset threshold or shows an abnormal change trend, and generate early warning information according to the early warning signal, wherein the early warning information includes the concentration value of the pollutant, the time and place of the abnormal occurrence, and the possible location of the pollution source; The display module is used to send the warning information to the user end through a preset response mechanism and generate an abnormal status report on the cloud platform.

[0018] By adopting the above technical solution, real-time environmental parameter data in a preset monitoring area is obtained through sensor nodes, and the real-time environmental parameter data is sent to the edge computing device, so that the environmental information of the monitoring area can be obtained in real time, ensuring the timeliness of the environmental data, thereby improving the accuracy of abnormal detection of environmental indicators; by using a preprocessing algorithm to preprocess the real-time environmental parameter data, the preprocessed environmental parameter data is obtained, and the multi-dimensional environmental data of the sensor node is integrated to obtain a comprehensive environmental index, which can eliminate noise and missing information in the data, ensure the integrity and accuracy of the environmental data, and thus improve the reliability of the detection results; by inputting the preprocessed environmental parameter data and the comprehensive environmental index into a pre-trained anomaly detection model for analysis, the environmental analysis results are obtained, which can timely identify abnormal environmental change trends and pollutant concentrations exceeding the standard, thereby improving the ability to identify pollution sources; by triggering early warning signals and generating early warning information, real-time early warning of abnormal conditions can be achieved, helping relevant personnel to respond quickly, thereby improving the response speed of environmental management.

[0019] The third objective of the present application is achieved through the following technical solutions: A computer device comprises a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the above-mentioned method for detecting anomalies of environmental indicators when executing the computer program.

[0020] The fourth objective of the present application is achieved through the following technical solutions: A computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the above-mentioned environmental protection indicator abnormality detection method are implemented.

[0021] In summary, the present application includes at least one of the following beneficial technical effects: 1. Real-time environmental parameter data in the preset monitoring area is obtained through sensor nodes, and the real-time environmental parameter data is sent to the edge computing device, which can obtain the environmental information of the monitoring area in real time, ensure the timeliness of environmental data, and thus improve the accuracy of abnormal detection of environmental indicators; by using the preprocessing algorithm to preprocess the real-time environmental parameter data, the preprocessed environmental parameter data is obtained, and the multi-dimensional environmental data of the sensor nodes is integrated to obtain a comprehensive environmental index, which can eliminate noise and missing information in the data, ensure the integrity and accuracy of the environmental data, and thus improve the reliability of the detection results; 2. By inputting the pre-processed environmental parameter data and comprehensive environmental index into the pre-trained anomaly detection model for analysis, the environmental analysis results are obtained, which can timely identify abnormal environmental change trends and excessive pollutant concentrations, thereby improving the ability to identify pollution sources; by triggering early warning signals and generating early warning information, real-time early warning of abnormal conditions can be achieved, helping relevant personnel to respond quickly, thereby improving the response speed of environmental management. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 This is a flow chart of a method for detecting anomalies of environmental indicators in an embodiment of the present application; Figure 2 This is a flowchart for implementing step S10 in a method for detecting anomalies in environmental indicators in one embodiment of the present application; Figure 3 This is a flowchart for implementing step S20 in a method for detecting an abnormality of an environmental indicator in an embodiment of the present application; Figure 4 This is a flowchart for implementing step S30 in a method for detecting an abnormality of an environmental indicator in an embodiment of the present application; Figure 5 This is a flowchart for implementing step S30 in a method for detecting anomalies in environmental indicators in one embodiment of the present application; Figure 6 This is a flowchart of an implementation method of an environmental indicator abnormality detection method in an embodiment of the present application; Figure 7 This is a principle block diagram of an environmental indicator abnormality detection system in one embodiment of the present application; Figure 8 It is a schematic diagram of a device in an embodiment of the present application. DETAILED DESCRIPTION

[0023] The present application is further described in detail below in conjunction with the accompanying drawings.

[0024] In one embodiment, if Figure 1 As shown, the present application discloses a method for detecting anomalies of environmental indicators, which specifically includes the following steps: S10: Acquire real-time environmental parameter data in a preset monitoring area through sensor nodes, and send the real-time environmental parameter data to an edge computing device. The sensor nodes include temperature sensors, humidity sensors, PM2.5 sensors, and pollutant sensors.

[0025] Specifically, multiple sensor nodes are deployed in the preset monitoring area. These sensor nodes are responsible for real-time monitoring and recording of environmental parameters. The temperature of the environment is measured through temperature sensors, and the moisture content in the environment is monitored through humidity sensors. PM2.5 sensors are used to detect the concentration of fine particulate matter (PM2.5) in the air. Pollutant sensors monitor other types of air pollutants such as sulfur dioxide, carbon monoxide, volatile organic compounds, etc. Sensor nodes send the collected data to edge computing devices through wireless communication technologies such as Wi-Fi, Bluetooth, LoRa, etc. These devices are located near the data source and can quickly process and analyze data, reducing the delay and bandwidth requirements for data transmission to remote servers. Through edge computing, rapid response to environmental data can be achieved, improving the real-time and efficiency of the monitoring system.

[0026] S20: Preprocess the real-time environmental parameter data using a preprocessing algorithm to obtain preprocessed environmental parameter data, and fuse the multi-dimensional environmental data of the sensor nodes to obtain a comprehensive environmental index.

[0027] Specifically, a preprocessing algorithm is applied to process the real-time environmental parameter data collected from the sensor nodes. The preprocessing may include steps such as denoising, missing value processing, data normalization, outlier detection and filtering. The preprocessed data becomes more accurate and reliable and can be used for further analysis. The environmental parameter data from different sensor nodes, such as temperature, humidity, PM2.5, pollutant concentration, etc., are fused. The fusion process may involve data aggregation, feature extraction or multi-sensor data fusion technology. Based on the fused data, a comprehensive environmental index, such as a weighted average, is calculated to reflect the combined impact of multiple environmental parameters.

[0028] S30: The preprocessed environmental parameter data and the comprehensive environmental index are input into a pre-trained anomaly detection model for analysis to obtain an environmental analysis result.

[0029] Specifically, the preprocessed environmental parameter data and the comprehensive environmental index are input into a pre-trained anomaly detection model, which is based on machine learning or deep learning technology, such as convolutional neural network (CNN), long short-term memory network (LSTM) or support vector machine (SVM). The model can analyze the data and identify abnormal patterns or trends in the environment, such as a sudden increase or abnormal change in pollutant concentration. The anomaly detection model outputs analysis results, which indicate whether the current environmental conditions are normal or whether there are any abnormalities, so as to help decision makers understand the environmental conditions and take necessary preventive or response measures.

[0030] S40: When the environmental analysis result detects that the pollutant concentration exceeds the preset threshold or shows an abnormal change trend, a warning signal is triggered, and warning information is generated based on the warning signal. The warning information includes the concentration value of the pollutant, the time and place of the abnormality, and the possible location of the pollution source.

[0031] Specifically, after the anomaly detection model analyzes the pre-processed environmental parameter data and the comprehensive environmental index, if it is found that the pollutant concentration exceeds the preset safety threshold or there is an unexpected increase trend, the system will consider that an environmental anomaly has been detected. Once it is detected that the pollutant concentration exceeds the standard or an abnormal change trend is detected, the system will automatically trigger an early warning signal. This signal is part of the early warning system response mechanism and is used to indicate situations that require immediate attention. Based on the triggered early warning signal, the system will generate detailed early warning information. The early warning information includes the concentration value of the pollutant, the time and place of the anomaly, and the possible location of the pollution source, providing decision makers with a comprehensive perspective to help them understand the specific circumstances of the pollution incident and formulate corresponding response strategies.

[0032] S50: Send the warning information to the user end through the preset response mechanism, and generate an abnormal status report on the cloud platform.

[0033] Specifically, when the environmental monitoring system detects that the pollutant concentration exceeds the safety threshold or shows other abnormal change trends, the system will automatically send warning information to relevant users or managers according to the pre-set response mechanism, such as email, text message, mobile application push notification or automated voice call, etc., to ensure that the information can be quickly conveyed to the user. At the same time, the system will generate a detailed abnormal status report on the cloud platform, recording all relevant warning information and environmental parameter data. The abnormal status report can be used for post-analysis to help environmental management personnel understand the causes, frequency and impacts of the abnormalities, thereby optimizing monitoring strategies and preventive measures.

[0034] By adopting the above technical solution, real-time environmental parameter data in a preset monitoring area is obtained through sensor nodes, and the real-time environmental parameter data is sent to the edge computing device, so that the environmental information of the monitoring area can be obtained in real time, ensuring the timeliness of environmental data, thereby improving the accuracy of abnormal detection of environmental indicators; by using a preprocessing algorithm to preprocess the real-time environmental parameter data, the preprocessed environmental parameter data is obtained, and the multi-dimensional environmental data of the sensor node is integrated to obtain a comprehensive environmental index, which can eliminate noise and missing information in the data, ensure the integrity and accuracy of the environmental data, and thus improve the reliability of the detection results; by inputting the preprocessed environmental parameter data and the comprehensive environmental index into a pre-trained anomaly detection model for analysis, the environmental analysis results are obtained, which can timely identify abnormal environmental change trends and pollutant concentrations exceeding the standard, thereby improving the ability to identify pollution sources; by triggering early warning signals and generating early warning information, real-time early warning of abnormal conditions can be achieved, helping relevant personnel to respond quickly, thereby improving the response speed of environmental management. In one embodiment, if Figure 2 As shown, in step S10, the real-time environmental parameter data in the preset monitoring area is obtained through the sensor node, and the real-time environmental parameter data is sent to the edge computing device, which specifically includes: S11: Monitor the external environment data acquired by the sensor node from the preset monitoring area in real time, obtain real-time external environment parameter data, and monitor the status of the sensor node in real time, and send a fault alarm to the edge computing device when a sensor node fails.

[0035] Specifically, the system continuously collects external environmental data from sensor nodes deployed in the preset monitoring area. The system can grasp the environmental conditions in the monitoring area in real time, including key parameters such as temperature, humidity, and air quality. The system not only collects environmental data, but also continuously monitors the operating status of sensor nodes to ensure that they are working properly. If the system detects that a sensor node has a fault, such as data anomalies, communication interruption, or hardware damage, it will automatically send a fault alarm to the edge computing device. The fault alarm is sent to the edge computing device via wireless networks or other communication methods. These devices are usually responsible for processing and analyzing data from sensor nodes.

[0036] S12: The real-time environmental parameter data is wirelessly transmitted to the edge computing device at a predetermined sampling frequency by adopting an adaptive data transmission protocol, and the real-time environmental parameter is encrypted by adopting transport layer encryption technology during the data transmission process.

[0037] Specifically, the system uses an adaptive data transmission protocol that can dynamically adjust data transmission parameters such as transmission rate and data packet size according to network conditions. Sensor nodes collect environmental parameter data in real time according to a predetermined sampling frequency, that is, the number of times data is collected and sent per second, and transmit it to the edge computing device wirelessly. During the entire process of data transmission from the sensor node to the edge computing device, the data is encrypted using transport layer encryption technology to prevent the data from being intercepted or tampered with during transmission.

[0038] In one embodiment, if Figure 3 As shown, in step S20, the real-time environmental parameter data is preprocessed using a preprocessing algorithm to obtain preprocessed environmental parameter data, and the multi-dimensional environmental data of the sensor nodes are fused to obtain a comprehensive environmental index, which specifically includes: S21: Apply an adaptive filtering algorithm to eliminate noise from the real-time environmental parameter data, and use an interpolation method and a moving average technique to fill in missing data in the real-time environmental parameter data and smooth the data curve to obtain preprocessed environmental parameter data.

[0039] Specifically, an adaptive filtering algorithm is used to process the real-time environmental parameter data to eliminate noise. The adaptive filter can automatically adjust the filtering parameters according to the characteristics of the signal, effectively remove noise and retain useful signal components. For missing values ​​in the real-time environmental parameter data, an interpolation method is applied to fill them. Interpolation methods such as linear interpolation, polynomial interpolation or nearest neighbor interpolation can estimate reasonable values ​​at the data missing points, and use moving average technology to smooth the environmental parameter data. Moving average technology can effectively reduce random fluctuations in the data, highlight the trends and patterns of the data, and make the data curve smoother, which is convenient for analyzing and identifying the trends of environmental changes. After the above steps, the preprocessed environmental parameter data are obtained. These data have removed noise, filled missing values, and have a smoother curve.

[0040] S22: Use principal component analysis technology to extract features from the multidimensional data of sensor nodes, and combine it with fuzzy logic method to construct a comprehensive environmental index.

[0041] Specifically, through PCA, key features can be extracted from the multidimensional environmental data collected by sensor nodes. These features can represent the main change trends in the original data set. PCA can reduce the dimension of the data while retaining the most important information, making the data easier to analyze and process. The features extracted by PCA can be used as the basis for subsequent analysis. Using the features extracted by PCA and combined with fuzzy logic methods, a comprehensive environmental index can be constructed. This index integrates the influence of multiple environmental parameters and reflects the overall condition of the environment.

[0042] In one embodiment, if Figure 4 As shown, before step S30, that is, before the pre-processed environmental parameter data and the comprehensive environmental index are input into the pre-trained anomaly detection model for analysis to obtain the environmental analysis result, the environmental index anomaly detection method further includes: S301: Acquire historical environmental data and environmental standard data, normalize and label the historical environmental data and environmental standard data, and obtain a training set.

[0043] Specifically, environmental monitoring data over a period of time are collected. These data may include indicators such as temperature, humidity, PM2.5, and pollutant concentrations. At the same time, relevant environmental standard data are collected. These standards are formulated by environmental protection agencies and define safety thresholds for various pollutants. The collected data are normalized, that is, the data is scaled to a small, specified range, usually 0 to 1, or -1 to 1. The normalized data is labeled. The labeling may include classifying the data as "normal" or "abnormal", or marking whether the pollutant concentration exceeds the standard according to environmental standards. The normalized and labeled data are organized into a training set, which will be used to train the machine learning model.

[0044] S302: Perform forward propagation and back propagation training on the anomaly detection model constructed based on the long short-term memory network and the convolutional neural network using the training set to obtain the anomaly detection model after the forward propagation and back propagation training.

[0045] Specifically, a prepared training set, that is, normalized and labeled data, is used to train an anomaly detection model based on a long short-term memory network (LSTM) and a convolutional neural network (CNN). The training set provides the basic data for model learning, so that the model can learn normal and abnormal environmental patterns from historical data. Through forward propagation, the input data is fed into the model, processed by each layer of the network, and finally an output result is generated. Through back propagation, the output result of the model is compared with the actual labeled data, and a loss function such as mean square error or cross entropy loss is calculated. Then, according to the gradient information of the loss function, an optimization algorithm such as gradient descent is used to adjust the weight and bias parameters of the model. Through the iterative process of forward propagation and back propagation, the model gradually learns how to accurately identify abnormal patterns from the input data, and finally an anomaly detection model after forward propagation and back propagation training is obtained.

[0046] S303: Optimize the parameters of the anomaly detection model trained by forward propagation and back propagation by using a transfer learning method to obtain a pre-trained anomaly detection model.

[0047] Specifically, the parameters of the anomaly detection model after forward propagation and back propagation training are optimized through transfer learning methods, the knowledge learned on one task is applied to the training task, and the model is optimized by fine-tuning the parameters of the pre-trained model to make it better adapt to new tasks, such as adjusting the learning rate, optimizer selection, layer freezing and thawing strategies, to obtain a pre-trained anomaly detection model. After transfer learning and parameter optimization, the model can perform anomaly detection tasks more accurately.

[0048] In one embodiment, if Figure 5 As shown, in step S30, the pre-processed environmental parameter data and the comprehensive environmental index are input into the pre-trained anomaly detection model for analysis to obtain the environmental analysis results, which specifically include: S31: Capture the temporal dynamic change characteristics of the pre-processed environmental parameter data, analyze them according to the temporal dynamic change characteristics, and obtain a prediction plan for the pollutant diffusion path.

[0049] Specifically, the system analyzes the pre-processed environmental parameter data and identifies characteristics that change over time, such as seasonal and diurnal changes in pollutant concentrations. It uses the captured temporal dynamic change characteristics combined with mathematical models and algorithms, such as fluid dynamics models and statistical analysis methods, to conduct predictive analysis of pollutant diffusion paths and ultimately obtain a prediction plan for pollutant diffusion paths.

[0050] S32: Compare and analyze the comprehensive conversion index with the environmental standard data in the pre-trained anomaly detection model to obtain a data comparison and analysis result.

[0051] Specifically, the system will compare the comprehensive environmental index collected and integrated by various sensors with the pre-set environmental standard data. The comprehensive environmental index may include data in multiple dimensions such as air quality, temperature, and humidity. Through comparison and analysis, the system can identify whether the current environmental conditions meet the preset environmental standards, thereby determining whether there are any abnormal conditions. After the comparison and analysis, the system will generate a data comparison analysis result. This result shows the difference between the environmental index and the standard data, helping decision makers understand whether the environmental quality meets the expected standards and whether corresponding measures need to be taken.

[0052] S33: Combine the Bayesian inference method to analyze the prediction scheme of pollutant diffusion path and the data comparison analysis results to obtain environmental analysis results.

[0053] Specifically, using Bayesian inference, combined with the prediction scheme of pollutant diffusion paths, the possibility of different diffusion paths can be evaluated. This method can take into account multiple factors, such as wind speed, wind direction, terrain, etc., and how they affect the spread of pollutants. Through Bayesian inference, combined with the prediction scheme and data comparison results, and using prior knowledge and observational data to calculate the posterior probability distribution of parameters, a comprehensive environmental analysis result can be obtained. This result not only includes the prediction of pollutant diffusion paths, but also includes an assessment of the current environmental conditions and possible future environmental change trends.

[0054] In one embodiment, if Figure 6 As shown, the environmental protection indicator abnormality detection method also includes: S60: The pre-processed environmental parameter data, comprehensive environmental index and environmental analysis results are transmitted to the user end through Rerun in the form of a time series graph, a heat map and a relationship network for display.

[0055] Specifically, the system compares and analyzes the pre-processed environmental parameter data, such as temperature, humidity, PM2.5, etc., with the comprehensive environmental index. Through comparison and analysis, the system can identify the differences between the current environmental conditions and the standards or expectations, and provide a basis for further prediction and decision-making. The system uses the Rerun tool to visualize the analysis results in the form of time series graphs, heat maps and relationship networks. Through Rerun, developers can easily record various types of data, including 2D images, 3D point clouds, text, time series, tensors, etc., and stream these data in real time to the Rerun Viewer for visualization, or save them to files for future use. This can more intuitively display the changing trends and distribution of environmental parameters, helping users better understand and analyze environmental data.

[0056] S70: Based on real-time user feedback and changes in the environment where the sensor node is located, dynamically adjust the data collection frequency, anomaly detection threshold and device operation status of the sensor node.

[0057] Specifically, the system automatically adjusts the sampling frequency of sensor nodes based on real-time environmental parameter data and user feedback. For example, the sampling frequency is increased when the environment changes rapidly to capture rapidly changing environmental characteristics. The system dynamically adjusts the thresholds used for anomaly detection based on environmental changes and user feedback. These thresholds may be related to pollutant concentrations, temperature anomalies, or other key environmental parameters. By adjusting the thresholds, the system can more accurately identify and respond to environmental anomalies and improve the sensitivity and accuracy of anomaly detection. The system monitors the operating status of sensor nodes and makes adjustments based on environmental changes and user needs, such as changing the sensor's operating mode, restarting the sensor, or optimizing its energy consumption. Dynamic adjustment of the device's operating status helps ensure that sensor nodes can operate stably under various environmental conditions, while improving the system's adaptability and reliability.

[0058] S80: When data is transmitted to the cloud platform, the data transmitted to the cloud platform is encrypted and stored using blockchain technology.

[0059] Specifically, when the data of the environmental monitoring system needs to be transmitted to the cloud platform, these data may contain sensitive information, such as pollutant concentrations, environmental quality indicators, etc. In order to protect the security and privacy of these data, blockchain technology is used to encrypt the data. Before the data is transmitted to the cloud, blockchain technology will encrypt the data. This process may involve encryption algorithms such as symmetric encryption, asymmetric encryption or hash functions to ensure that the data is not accessed or tampered with by unauthorized third parties during transmission. The encrypted data is stored in the blocks of the blockchain. Each block contains a certain amount of transaction data and is linked to the previous block through encryption to form a growing chain. The distributed ledger and encryption technology of the blockchain ensure the security and transparency of the data.

[0060] It should be understood that the size of the serial numbers of the steps in the above embodiments does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0061] In one embodiment, a system for detecting anomalies of environmental indicators is provided, and the system for detecting anomalies of environmental indicators corresponds one-to-one to the method for detecting anomalies of environmental indicators in the above embodiment. Figure 7 As shown, the environmental protection index abnormality detection system includes a sensor detection module, a data preprocessing module, an analysis module, an early warning module and a display module. The detailed description of each functional module is as follows: The sensor detection module is used to obtain real-time environmental parameter data in the preset monitoring area through sensor nodes and send the real-time environmental parameter data to the edge computing device. The sensor nodes include temperature sensors, humidity sensors, PM2.5 sensors and pollutant sensors; The data preprocessing module is used to preprocess the real-time environmental parameter data using a preprocessing algorithm to obtain the preprocessed environmental parameter data, and to fuse the multi-dimensional environmental data of the sensor nodes to obtain a comprehensive environmental index; An analysis module is used to input the pre-processed environmental parameter data and the comprehensive environmental index into a pre-trained anomaly detection model for analysis to obtain environmental analysis results; The early warning module is used to trigger an early warning signal when the environmental analysis result detects that the pollutant concentration exceeds a preset threshold or shows an abnormal change trend, and generate early warning information according to the early warning signal. The early warning information includes the concentration value of the pollutant, the time and place of the abnormal occurrence, and the possible location of the pollution source; The display module is used to send the warning information to the user end through a preset response mechanism and generate an abnormal status report on the cloud platform.

[0062] Optionally, the sensor detection module includes: The real-time monitoring submodule is used to monitor the external environment data obtained by the sensor node from the preset monitoring area in real time, obtain the real-time external environment parameter data, and monitor the status of the sensor node in real time, and send a fault alarm to the edge computing device when a sensor node fails; The data transmission submodule is used to wirelessly transmit the real-time environmental parameter data to the edge computing device at a predetermined sampling frequency by adopting an adaptive data transmission protocol, and encrypt the real-time environmental parameters by adopting transport layer encryption technology during the data transmission process.

[0063] Optionally, the data preprocessing module includes: The filtering submodule is used to apply an adaptive filtering algorithm to eliminate noise from the real-time environmental parameter data, and to fill in the missing data in the real-time environmental parameter data and smooth the data curve using an interpolation method and a moving average technique to obtain the pre-processed environmental parameter data; The feature extraction submodule is used to extract features from the multi-dimensional data of sensor nodes using principal component analysis technology, and to construct a comprehensive environmental index in combination with fuzzy logic methods.

[0064] Optionally, the environmental protection indicator anomaly detection system further includes: A training set acquisition module is used to acquire historical environmental data and environmental standard data, normalize and annotate the historical environmental data and environmental standard data, and obtain a training set; A model training module is used to perform forward propagation and back propagation training on anomaly detection models built based on long short-term memory networks and convolutional neural networks using training sets, thereby obtaining anomaly detection models after forward propagation and back propagation training; The parameter optimization module is used to optimize the parameters of the anomaly detection model after forward propagation and back propagation training through the transfer learning method to obtain a pre-trained anomaly detection model.

[0065] The display module is used to transmit the pre-processed environmental parameter data, comprehensive environmental index and environmental analysis results to the user end for display in the form of time series graph, heat map and relationship network through Rerun; Dynamic adjustment module, used to dynamically adjust the data collection frequency, anomaly detection threshold and device operation status of sensor nodes based on real-time user feedback and changes in the environment where the sensor nodes are located; The encryption storage module is used to encrypt and store the data transmitted to the cloud platform through blockchain technology when the data is transmitted to the cloud platform.

[0066] Optional analysis modules include: The capture submodule is used to capture the temporal dynamic change characteristics of the pre-processed environmental parameter data, analyze the temporal dynamic change characteristics, and obtain a prediction scheme for the pollutant diffusion path; The comparison and analysis submodule is used to compare and analyze the comprehensive conversion index with the environmental standard data in the pre-trained anomaly detection model to obtain data comparison and analysis results; The inference submodule is used to analyze the prediction scheme of pollutant diffusion path and the data comparison analysis results in combination with the Bayesian inference method to obtain environmental analysis results.

[0067] For the specific definition of an environmental indicator anomaly detection system, please refer to the definition of an environmental indicator anomaly detection method mentioned above, which will not be repeated here. Each module in the above-mentioned environmental indicator anomaly detection system can be implemented in whole or in part by software, hardware and a combination thereof. The above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above modules.

[0068] In one embodiment, a computer device is provided. The computer device may be a server, and its internal structure diagram may be as follows: Figure 8As shown. The computer device includes a processor, a memory, a network interface and a database connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a method for detecting an abnormality of an environmental indicator is implemented.

[0069] In one embodiment, a computer device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the following steps when executing the computer program: The sensor nodes are used to obtain real-time environmental parameter data in the preset monitoring area and send the real-time environmental parameter data to the edge computing device. The sensor nodes include temperature sensors, humidity sensors, PM2.5 sensors and pollutant sensors. The real-time environmental parameter data is preprocessed by using a preprocessing algorithm to obtain preprocessed environmental parameter data, and the multi-dimensional environmental data of the sensor nodes are fused to obtain a comprehensive environmental index; The pre-processed environmental parameter data and comprehensive environmental index are input into the pre-trained anomaly detection model for analysis to obtain the environmental analysis results; When the environmental analysis result detects that the pollutant concentration exceeds the preset threshold or shows an abnormal change trend, an early warning signal is triggered, and early warning information is generated based on the early warning signal. The early warning information includes the concentration value of the pollutant, the time and location of the abnormality, and the possible location of the pollution source; The warning information is sent to the user end through the preset response mechanism, and an abnormal status report is generated on the cloud platform.

[0070] In one embodiment, a computer readable storage medium is provided, on which a computer program is stored, and when the computer program is executed by a processor, the following steps are implemented: The sensor nodes are used to obtain real-time environmental parameter data in the preset monitoring area and send the real-time environmental parameter data to the edge computing device. The sensor nodes include temperature sensors, humidity sensors, PM2.5 sensors and pollutant sensors. The real-time environmental parameter data is preprocessed by using a preprocessing algorithm to obtain preprocessed environmental parameter data, and the multi-dimensional environmental data of the sensor nodes are fused to obtain a comprehensive environmental index; The pre-processed environmental parameter data and comprehensive environmental index are input into the pre-trained anomaly detection model for analysis to obtain the environmental analysis results; When the environmental analysis result detects that the pollutant concentration exceeds the preset threshold or shows an abnormal change trend, an early warning signal is triggered, and early warning information is generated based on the early warning signal. The early warning information includes the concentration value of the pollutant, the time and location of the abnormality, and the possible location of the pollution source; The warning information is sent to the user end through the preset response mechanism, and an abnormal status report is generated on the cloud platform.

[0071] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0072] Those skilled in the art will clearly understand that for the sake of convenience and brevity of description, only the division of the above-mentioned functional units and modules is used as an example. In actual applications, the above-mentioned functions can be distributed and completed by different functional units and modules as needed, that is, the internal structure of the system can be divided into different functional units or modules to complete all or part of the functions described above.

[0073] The embodiments described above are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, a person skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application, and should all be included in the protection scope of the present application.

Claims

1. A method for detecting anomalies of environmental indicators, characterized in that: The environmental protection indicator abnormality detection method comprises: Acquire real-time environmental parameter data in a preset monitoring area through sensor nodes, and send the real-time environmental parameter data to an edge computing device, wherein the sensor nodes include a temperature sensor, a humidity sensor, a PM2.5 sensor, and a pollutant sensor; Preprocessing the real-time environmental parameter data using a preprocessing algorithm to obtain preprocessed environmental parameter data, and fusing the multi-dimensional environmental data of the sensor nodes to obtain a comprehensive environmental index; Inputting the preprocessed environmental parameter data and the comprehensive environmental index into a pre-trained anomaly detection model for analysis to obtain an environmental analysis result; When the environmental analysis result detects that the pollutant concentration exceeds a preset threshold or shows an abnormal change trend, a warning signal is triggered, and warning information is generated according to the warning signal, the warning information includes the concentration value of the pollutant, the time and place of the abnormality, and the possible location of the pollution source; The warning information is sent to the user end through a preset response mechanism, and an abnormal status report is generated on the cloud platform.

2. The method for detecting anomalies of environmental indicators according to claim 1, characterized in that: The step of acquiring real-time environmental parameter data in a preset monitoring area through a sensor node and sending the real-time environmental parameter data to an edge computing device includes: Monitor the external environment data acquired by the sensor node from the preset monitoring area in real time to obtain the real-time external environment parameter data, and monitor the status of the sensor node in real time, and send a fault alarm to the edge computing device when the sensor node fails; The real-time environmental parameter data is wirelessly transmitted to the edge computing device at a predetermined sampling frequency by adopting an adaptive data transmission protocol, and the real-time environmental parameter is encrypted by adopting transport layer encryption technology during the data transmission process.

3. The method for detecting anomalies of environmental indicators according to claim 1, characterized in that: The preprocessing algorithm is used to preprocess the real-time environmental parameter data to obtain the preprocessed environmental parameter data, and the multi-dimensional environmental data of the sensor node is fused to obtain the comprehensive environmental index, including: Applying an adaptive filtering algorithm to eliminate noise from the real-time environmental parameter data, and using an interpolation method and a moving average technique to fill in missing data in the real-time environmental parameter data and to smooth the data curve, thereby obtaining the pre-processed environmental parameter data; The principal component analysis technique is used to extract features from the multidimensional data of the sensor nodes, and the comprehensive environmental index is constructed in combination with the fuzzy logic method.

4. The method for detecting anomalies of environmental indicators according to claim 1, characterized in that: Before the pre-processed environmental parameter data and the comprehensive environmental index are input into a pre-trained anomaly detection model for analysis to obtain an environmental analysis result, the environmental index anomaly detection method further includes: Acquire historical environmental data and environmental standard data, normalize and annotate the historical environmental data and the environmental standard data to obtain a training set; Using the training set, forward propagation and back propagation training are performed on an anomaly detection model constructed based on a long short-term memory network and a convolutional neural network to obtain an anomaly detection model after forward propagation and back propagation training; The parameters of the anomaly detection model after forward propagation and back propagation training are optimized by a transfer learning method to obtain the pre-trained anomaly detection model.

5. The method for detecting anomalies of environmental indicators according to claim 1, characterized in that: The pre-processed environmental parameter data and the comprehensive environmental index are input into a pre-trained anomaly detection model for analysis to obtain an environmental analysis result, which includes: Capturing the temporal dynamic change characteristics of the pre-processed environmental parameter data, analyzing the temporal dynamic change characteristics, and obtaining a prediction scheme for the pollutant diffusion path; Compare and analyze the comprehensive conversion index with the environmental standard data in the pre-trained anomaly detection model to obtain a data comparison and analysis result; The prediction scheme of the pollutant diffusion path and the data comparison analysis results are analyzed in combination with the Bayesian inference method to obtain the environmental analysis results.

6. The method for detecting anomalies of environmental indicators according to claim 1, characterized in that: The environmental protection indicator abnormality detection method also includes: The pre-processed environmental parameter data, the comprehensive environmental index and the environmental analysis results are transmitted to the user terminal for display in the form of a time series diagram, a heat map and a relationship network through Rerun; Dynamically adjust the data collection frequency, anomaly detection threshold and device operation status of the sensor node based on real-time user feedback and changes in the environment in which the sensor node is located; When data is transmitted to the cloud platform, the data transmitted to the cloud platform is encrypted and stored using blockchain technology.

7. An environmental index abnormality detection system, characterized in that: The environmental protection index abnormality detection system comprises: A sensing detection module is used to obtain real-time environmental parameter data in a preset monitoring area through sensor nodes, and send the real-time environmental parameter data to an edge computing device. The sensor nodes include temperature sensors, humidity sensors, PM2.5 sensors, and pollutant sensors. A data preprocessing module is used to preprocess the real-time environmental parameter data using a preprocessing algorithm to obtain preprocessed environmental parameter data, and to fuse the multi-dimensional environmental data of the sensor nodes to obtain a comprehensive environmental index; An analysis module, used for inputting the pre-processed environmental parameter data and the comprehensive environmental index into a pre-trained anomaly detection model for analysis to obtain an environmental analysis result; An early warning module is used to trigger an early warning signal when the environmental analysis result detects that the pollutant concentration exceeds a preset threshold or shows an abnormal change trend, and generate early warning information according to the early warning signal, wherein the early warning information includes the concentration value of the pollutant, the time and place of the abnormal occurrence, and the possible location of the pollution source; The display module is used to send the warning information to the user end through a preset response mechanism and generate an abnormal status report on the cloud platform.

8. The environmental protection index abnormality detection system according to claim 7, characterized in that: The environmental protection index abnormality detection system also includes: A data display module is used to transmit the pre-processed environmental parameter data, the comprehensive environmental index and the environmental analysis results to the user terminal for display in the form of a time series diagram, a heat map and a relationship network through Rerun; An adjustment module, for dynamically adjusting the data collection frequency, anomaly detection threshold and device operation status of the sensor node based on real-time user feedback and changes in the environment in which the sensor node is located; The encryption module is used to encrypt and store the data transmitted to the cloud platform through blockchain technology when the data is transmitted to the cloud platform.

9. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the steps of the environmental indicator abnormality detection method as described in any one of claims 1 to 6 are implemented.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the steps of the method for detecting anomalies of environmental indicators as described in any one of claims 1 to 6 are implemented.

Images