An environmental monitoring method and system based on digital twin technology

Abstract

The present application relates to an environmental monitoring method and system based on digital twin technology, and relates to the technical field of environmental monitoring. The system includes a sensor array for obtaining environmental data of the monitoring area; the sensor array includes a temperature and humidity sensor, an oxygen concentration sensor, a combustible gas concentration sensor, a vibration sensor, a noise sensor, and a smoke sensor; an environmental monitoring host for obtaining and processing environmental data sent by each sensor in the sensor array to obtain environmental prediction results; and a back-end monitoring terminal for displaying environmental prediction results, displaying a virtual environment model, and providing a user interaction interface. The technical solution of the present application applies digital twin technology to environmental monitoring, so as to better display environmental monitoring data.

Description

Technical Field

[0001] The present application relates to the technical field of environmental monitoring, and in particular to an environmental monitoring method and system based on digital twin technology. Background Art

[0002] With the development of industry and social progress, environmental protection has become increasingly important. As one of the important tools for environmental protection, environmental monitoring systems are widely used in various scenarios, such as factories, cities, nature reserves, etc. Traditional environmental monitoring systems mainly rely on single-point measurement and simple data analysis, which cannot achieve comprehensive monitoring and prediction of complex environments. With the development of information technology, especially the advancement of the Internet of Things, big data and artificial intelligence technology, environmental monitoring systems are also constantly evolving and upgrading.

[0003] Digital twin technology is an emerging technical means that realizes real-time monitoring and prediction of the actual environment by constructing a virtual mapping of the physical environment.

[0004] How to combine digital twin technology to achieve environmental monitoring is a technical challenge that technicians in this field need to overcome. Summary of the invention

[0005] In order to at least partially solve the above technical problems, the present application provides an environmental monitoring method and system based on digital twin technology.

[0006] In the first aspect, the present application provides an environmental monitoring system based on digital twin technology, which adopts the following technical solution.

[0007] An environmental monitoring system based on digital twin technology, comprising:

[0008] A sensor array is used to obtain environmental data of the monitoring area; the sensor array includes a temperature and humidity sensor, an oxygen concentration sensor, a combustible gas concentration sensor, a vibration sensor, a noise sensor and a smoke sensor;

[0009] The environment monitoring host is used to: acquire and process the environment data sent by each sensor in the sensor array to obtain an environment prediction result;

[0010] The back-end monitoring terminal is used to display the environmental prediction results, present the virtual environment model and provide a user interaction interface;

[0011] Wherein, the environmental monitoring host is configured with a digital twin module, an analysis module and an adaptive optimization module;

[0012] The digital twin module is used to construct a virtual mapping of the actual physical environment based on various environmental data acquired in real time to obtain a virtual environment model that is updated in real time;

[0013] The analysis module is used to perform anomaly detection, risk prediction and trend analysis on the environmental data in the virtual environment model to obtain an environmental prediction result;

[0014] The adaptive optimization module adjusts the environmental monitoring strategy according to the environmental prediction result.

[0015] Optionally, the digital twin module is used to construct a virtual mapping of the actual physical environment based on various environmental data acquired in real time to obtain a real-time updated virtual environment model, including:

[0016] Acquire environmental data sent from each sensor of the sensor array;

[0017] Fuse the environmental data sent by each sensor to improve data consistency;

[0018] Generate a virtual representation of the corresponding actual physical environment based on the fused environmental data; the virtual representation is a digital representation that reflects the distribution of various environmental parameters of the actual physical environment;

[0019] Constructing a real-time virtual environment model including various environmental parameters in the monitoring area according to the virtual representation; the virtual environment model is used to display the spatial distribution of environmental parameters and their changing trends over time;

[0020] The virtual environment model is updated in real time to reflect changes in environmental parameters in the actual physical environment.

[0021] Optionally, the analysis module performs anomaly detection on the environmental data in the virtual environment model, including:

[0022] The analysis module calculates the importance score of each environmental data feature; the importance score is calculated based on the mutual information;

[0023] The analysis module selects the top N features with the highest importance scores as a feature set for anomaly detection;

[0024] The analysis module obtains a baseline model based on the environmental data during historical normal operation, compares the real-time environmental data with the baseline module, and calculates the distance between the real-time environmental data point and the normal state environmental data point using the Euclidean distance calculation method to obtain the difference between the two;

[0025] The analysis module calculates an anomaly score for each environmental data point based on the difference and in combination with the feature importance scores in the feature set;

[0026] When the abnormality score is greater than a preset value, it is determined that the environmental data point is abnormal.

[0027] Optionally, the analysis module performs risk prediction and trend analysis on the environmental data in the virtual environment model, including:

[0028] The analysis module is based on the historical environmental data, and analyzes the change trend of the historical environmental data to obtain the analysis results of the historical environmental data;

[0029] The analysis module builds a risk prediction model based on the analysis results of historical environmental data;

[0030] The analysis module inputs the real-time environmental data into the risk prediction model to obtain the prediction results of the change trend of the environmental parameters in the future period of time output by the risk prediction model;

[0031] The analysis module evaluates possible future risks based on the forecast results and classifies the risks into different levels;

[0032] The analysis module generates a risk prediction report including prediction results and risk levels;

[0033] The analysis results of the historical environmental data obtained by analyzing the changing trend of the historical environmental data include:

[0034] Use time series analysis methods to capture the changing patterns of environmental parameters over time;

[0035] Analyze the conditions and patterns of historical abnormal events and identify the combination of environmental parameters that may lead to abnormalities;

[0036] The analysis module builds a risk prediction model based on the analysis results of historical environmental data, including:

[0037] A generator network is constructed, wherein the generator network is used to predict future prediction results based on historical data; the generator network includes: multiple convolutional layers for extracting features from historical data, an upsampling layer for increasing the time step to predict future time points, a residual block for alleviating gradient disappearance and enhancing feature learning capabilities, and a jump connection module for maintaining information transfer of low-level features during upsampling; wherein the loss function of the generator network is: ;in, A function that measures the difference between the future environment parameters predicted by the generator network and the actual environment parameters; ; is the predicted value of the generator network for the i-th time point, is the actual value of i time points, and n is the total number of time points; Used to characterize the environment data generated by the generator network After passing through the discriminator network D, the probability that the discriminator network marks it as real data; ; When D(G(Y)) is close to 1, that is, when the discriminator believes that the generated data G(Y) is real data, the loss tends to zero; The degree of match between the generated data G(Y) and the feature importance score W; ; represents the importance score of the i-th feature; Used to measure and ensure the similarity between the interdependencies between features in the generated data G(Y) and the real data; , and is the weight coefficient;

[0038] Constructing a discriminator network, the discriminator network is used to evaluate whether the prediction result of the generator network is consistent with the actual risk status; the discriminator network includes: a plurality of convolutional layers for extracting features from the prediction result and a fully connected layer for evaluating whether the risk prediction is consistent with the actual risk status;

[0039] Constructing a loss function module, the loss function module is used to quantify the difference between the generated risk prediction and the actual risk status; the loss function module includes: a content loss module for measuring the difference between the generated risk prediction and the actual risk status and an adversarial loss module for encouraging the generator to generate predictions that are closer to the actual risk status;

[0040] The initial model is obtained based on the constructed generator network, discriminator network and loss function module;

[0041] Use historical data to train the initial model and use a validation set to evaluate the prediction performance of the initial model and update the generator network and the discriminator network during the validation process;

[0042] When the confidence of the initial model is greater than the preset value, the risk prediction model is obtained.

[0043] Optionally, the adaptive optimization module adjusts the environmental monitoring strategy according to the environmental prediction result, including:

[0044] The adaptive optimization module evaluates the safety level of the current monitored environment state based on the received environmental prediction results;

[0045] The adaptive optimization module evaluates the risk level of environmental changes in the future based on the risk prediction results;

[0046] The adaptive optimization module evaluates the development direction of environmental parameters based on trend analysis results;

[0047] The adaptive optimization module generates new monitoring strategies based on the evaluation results. The monitoring strategies include increasing the monitoring frequency, adjusting the monitoring area, and activating the early warning mechanism.

[0048] The adaptive optimization module uploads the generated new environmental monitoring strategy to the environmental monitoring host to update the monitoring plan;

[0049] The adaptive optimization module evaluates the execution effect of the new strategy based on the new environmental prediction results; if the execution effect does not meet the expected results, a new monitoring strategy is regenerated.

[0050] Optionally, the analysis module is further used to construct a situational awareness model; the situational awareness model is used to adjust the environmental data in the virtual environment model in combination with external factors;

[0051] The method for constructing the situational awareness model includes:

[0052] Acquire external factor data; the external factor data includes weather forecast, equipment maintenance records, and historical abnormal event records;

[0053] Performing correlation analysis on the external factor data and the environmental data to determine the correlation between the external factor and the environmental parameter and selecting the most relevant M external factor features;

[0054] Constructing an initial situational awareness model; the initial situational awareness model is trained based on historical environmental data and the most relevant M external factor features are integrated into the initial situational awareness model to obtain a situational awareness model.

[0055] Optionally, the context awareness model is used to adjust the environmental data in the virtual environment model in combination with external factors, including:

[0056] fusing the external factor data with the environmental data to form a comprehensive data set;

[0057] inputting the comprehensive data set into the situational awareness model;

[0058] The context awareness model adjusts the environmental data according to external factors to reflect the impact of external factors on environmental parameters;

[0059] The situational awareness model outputs adjusted environmental data for generating environmental prediction results;

[0060] The analysis module performs anomaly detection, risk prediction and trend analysis based on the adjusted environmental data to obtain environmental prediction results.

[0061] In the second aspect, the environmental monitoring method based on digital twin technology provided in the present application adopts the following technical solution.

[0062] An environmental monitoring method based on digital twin technology, based on the above environmental monitoring system, is characterized in that the method includes:

[0063] Acquire environmental data of the monitoring area through a sensor array; the sensor array includes: temperature and humidity sensors, oxygen concentration sensors, combustible gas concentration sensors, vibration sensors, noise sensors and smoke sensors;

[0064] Acquiring and processing environmental data sent by each sensor in the sensor array through an environmental monitoring host to obtain an environmental prediction result;

[0065] Display environmental prediction results, display virtual environment models and provide user interaction interfaces through back-end monitoring terminals;

[0066] Wherein, the environmental monitoring host is configured with a digital twin module, an analysis module and an adaptive optimization module;

[0067] The digital twin module is used to construct a virtual mapping of the actual physical environment based on the various environmental data acquired in real time to obtain a virtual environment model that is updated in real time; the analysis module is used to perform anomaly detection, risk prediction and trend analysis on the environmental data in the virtual environment model to obtain environmental prediction results; the adaptive optimization module adjusts the environmental monitoring strategy according to the environmental prediction results.

[0068] In a third aspect, the present application discloses an electronic device, comprising a memory and a processor, wherein the memory stores a computer program that is loaded by the processor and executes the above method.

[0069] In a fourth aspect, the present application discloses a computer-readable storage medium storing a computer program that can be loaded by a processor and execute the above method. BRIEF DESCRIPTION OF THE DRAWINGS

[0070] Figure 1 It is a system block diagram of an environmental monitoring system based on digital twin technology in an embodiment of the present application;

[0071] Figure 2 It is a flow chart of an environmental monitoring method based on digital twin technology in an embodiment of the present application;

[0072] In the figure, 101 is a sensor array; 102 is an environmental monitoring host; 103 is a back-end monitoring terminal. DETAILED DESCRIPTION

[0073] The following is combined with Figure 1-2 The present application is further described with specific embodiments:

[0074] First of all, it should be noted that in the description of this application, if the terms "center", "upper", "lower", "left", "right", "vertical", "horizontal", "inside", "outside" and other directional words appear, the orientation or position relationship indicated is based on the orientation or position relationship shown in the drawings, which is only for the convenience of description, and does not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation of this application; in addition, if the terms "first", "second", "third" and other numerical quantifiers appear, they are only used for descriptive purposes and cannot be understood as indicating or implying relative importance. In addition, in this application, unless otherwise clearly specified and limited, if the terms "installed", "connected", and "connected" appear, they should be understood in a broad sense, for example, it can be a fixed connection, or a detachable connection, a limited connection such as an interference fit, a transition fit, or an integral connection; it can be directly connected or indirectly connected through an intermediate medium; therefore, for ordinary technicians in this field, the specific meanings of the above terms in this application can be understood according to the specific circumstances.

[0075] The present application embodiment discloses an environmental monitoring system based on digital twin technology. Figure 1 As an implementation method of an environmental monitoring system based on digital twin technology, an environmental monitoring system based on digital twin technology includes:

[0076] The sensor array 101 is used to obtain environmental data of the monitoring area; the sensor array includes: temperature and humidity sensors, oxygen concentration sensors, combustible gas concentration sensors, vibration sensors, noise sensors and smoke sensors. Specifically, various sensors are arranged in the area to be monitored. For example, the temperature and humidity sensors can be used to monitor the temperature and humidity of the air; the oxygen concentration sensor can be used to monitor the oxygen content in the air; the combustible gas concentration sensor is used to detect whether there is a combustible gas leak; the vibration sensor is used to monitor whether there is abnormal vibration; the noise sensor is used to monitor the noise level; the smoke sensor is used to detect whether there is smoke concentration. Each sensor transmits the collected data to the environmental monitoring host wirelessly or manually.

[0077] The environment monitoring host 102 is used to obtain and process the environment data sent by each sensor in the sensor array to obtain an environment prediction result.

[0078] The back-end monitoring terminal 103 is used to display the environmental prediction results, show the virtual environment model and provide a user interaction interface. Specifically, the back-end monitoring terminal has a user interaction interface, which is convenient for users to view the environmental prediction results and can intuitively display the virtual environment model constructed by the digital twin module.

[0079] Users can interact with the system through this user interface, for example, setting alarm thresholds, querying historical data, and adjusting monitoring parameters. The back-end monitoring terminal also supports remote access, so that managers can monitor the environment status through the network from anywhere.

[0080] Wherein, the environmental monitoring host is configured with a digital twin module, an analysis module and an adaptive optimization module;

[0081] The digital twin module is used to construct a virtual mapping of the actual physical environment based on various environmental data acquired in real time to obtain a virtual environment model that is updated in real time;

[0082] The analysis module is used to perform anomaly detection, risk prediction and trend analysis on the environmental data in the virtual environment model to obtain an environmental prediction result;

[0083] The adaptive optimization module adjusts the environmental monitoring strategy according to the environmental prediction result.

[0084] Specifically, the digital twin module acquires data from the sensor array in real time and updates the virtual environment model in real time: the analysis module can identify abnormal patterns and predict risk trends by analyzing the data in the virtual environment model, thereby providing early warning. The adaptive optimization module adjusts the monitoring strategy based on the environmental prediction results, and the system can dynamically adjust its working mode according to the actual situation. The user interaction interface provided by the back-end monitoring terminal enables managers to understand the environmental conditions more intuitively and respond quickly. By displaying the prediction results through the user interaction interface, managers can more clearly understand the trends and potential risks of environmental changes.

[0085] For example:

[0086] Suppose such an environmental monitoring system is deployed in a factory. When a large amount of unknown gas suddenly leaks one day, the digital twin module captures this change based on the environmental data sent by the sensor array and reflects this emergency in the virtual environment. By comparing historical data, the analysis module finds that the sharp increase in gas concentration may mean that equipment failure or an accident is about to occur. The system then issues an early warning to the factory's safety managers. The adaptive optimization module automatically adjusts the monitoring strategy, increases the sampling frequency of sensors in the relevant area, and starts a higher-level monitoring program. The safety manager directly sees this change through the user interface on the back-end monitoring terminal, and can take quick action.

[0087] As a specific implementation of an environmental monitoring system based on digital twin technology, the digital twin module is used to construct a virtual mapping of the actual physical environment based on various environmental data acquired in real time to obtain a real-time updated virtual environment model, including:

[0088] Obtain environmental data from each sensor in the sensor array. Specifically, each sensor should have a unique identifier to facilitate data management. The frequency of data collection needs to be set according to monitoring needs to ensure that changes can be reflected in a timely manner without generating excessive data burden.

[0089] The environmental data sent by each sensor is fused to improve the consistency of the data. Specifically, the data collected by the sensor may be inconsistent, so data fusion processing is required to improve the consistency and reliability of the data. Data fusion algorithms, such as Kalman filtering, particle filtering or other suitable methods, can be used to process the data differences between different sensors. The fused data is closer to the actual environmental state and reduces the impact of single sensor errors.

[0090] Generate a virtual representation of the corresponding actual physical environment based on the fused environmental data; the virtual representation is a digital representation that reflects the distribution of various environmental parameters of the actual physical environment. Specifically, use 3D modeling software or platforms, such as Unity, UnrealEngine, etc., to create a 3D space model and map the sensor data to the 3D space model. Different color codes or symbols are used according to different types of environmental data, so that users can understand the environmental status at a glance.

[0091] A real-time virtual environment model including various environmental parameters in the monitoring area is constructed based on the virtual representation; the virtual environment model is used to display the spatial distribution of environmental parameters and their changing trends over time. Specifically, the virtual environment model displays the changing trends of environmental parameters over time. Through the virtual representation, the user can intuitively see the spatial distribution of environmental parameters, which helps to quickly locate the problem area.

[0092] The virtual environment model is updated in real time to reflect changes in environmental parameters in the actual physical environment.

[0093] As a specific implementation of an environmental monitoring system based on digital twin technology, the analysis module performs anomaly detection on environmental data in the virtual environment model, including:

[0094] The analysis module calculates the importance score of each environmental data feature; the importance score is calculated based on mutual information. Specifically, mutual information is a method to measure the degree of mutual dependence between two random variables, which is used to measure the strength of the relationship between a feature and other features. The larger the mutual information value, the stronger the dependence between the two features. For each environmental data feature, the mutual information value between it and other features is calculated.

[0095] The analysis module selects the top N features with the highest importance scores as the feature set for anomaly detection according to the importance scores. Specifically, an importance score is assigned to each feature according to the mutual information value. The higher the score, the more critical the feature is for describing the environmental state. N is a predefined parameter, and the optimal number of features can be selected based on experience or through cross-validation, and the top N features are selected according to the importance score sorting to form a feature set.

[0096] The analysis module obtains a baseline model based on the environmental data during historical normal operation, compares the real-time environmental data with the baseline module, and calculates the distance between the real-time environmental data point and the normal state environmental data point through the Euclidean distance calculation method to obtain the difference between the two. Specifically, Euclidean distance calculation: for each real-time data point, calculate the Euclidean distance between it and the data point representing the normal state in the baseline model, and evaluate the difference between the real-time data point and the normal state data point based on the value obtained by the Euclidean distance.

[0097] The analysis module calculates the anomaly score for each environmental data point based on the difference and in combination with the feature importance score in the feature set. Specifically, the anomaly score calculation: the difference value is combined with the feature importance score to form a comprehensive score. The difference can be multiplied by the feature importance score, or a more complex weighted average formula can be used. A preset value is set as the threshold for the anomaly score. When the calculated anomaly score exceeds this threshold, the corresponding environmental data point is considered to be abnormal.

[0098] When the abnormality score is greater than a preset value, it is determined that the environmental data point is abnormal.

[0099] Specifically, in the environmental monitoring system based on digital twin technology, the analysis module calculates the importance score of each environmental data feature, selects the top N most important features to form a feature set, and uses the data during historical normal operation to establish a baseline model. The real-time environmental data is compared with the baseline model, and the Euclidean distance is used to calculate the difference between the real-time data points and the normal state data points. The anomaly score of each data point is calculated based on the importance and difference of the features. When the anomaly score exceeds the preset threshold, it is judged as an anomaly, thereby realizing anomaly detection of environmental data.

[0100] As a specific implementation of an environmental monitoring system based on digital twin technology, the analysis module performs risk prediction and trend analysis on environmental data in the virtual environment model, including:

[0101] The analysis module is based on the historical environmental data, and analyzes the change trend of the historical environmental data to obtain the analysis results of the historical environmental data;

[0102] The analysis module builds a risk prediction model based on the analysis results of historical environmental data;

[0103] The analysis module inputs the real-time environmental data into the risk prediction model to obtain the prediction results of the change trend of the environmental parameters in the future period of time output by the risk prediction model;

[0104] The analysis module evaluates possible future risks based on the forecast results and classifies the risks into different levels;

[0105] The analysis module generates a risk prediction report including prediction results and risk levels;

[0106] The analysis results of the historical environmental data obtained by analyzing the changing trend of the historical environmental data include:

[0107] Use time series analysis methods to capture the changing patterns of environmental parameters over time;

[0108] Analyze the conditions and patterns of historical abnormal events and identify the combination of environmental parameters that may lead to abnormalities;

[0109] The analysis module builds a risk prediction model based on the analysis results of historical environmental data, including:

[0110] A generator network is constructed, wherein the generator network is used to predict future prediction results based on historical data; the generator network includes: multiple convolutional layers for extracting features from historical data, an upsampling layer for increasing the time step to predict future time points, a residual block for alleviating gradient disappearance and enhancing feature learning capabilities, and a jump connection module for maintaining information transfer of low-level features during upsampling; wherein the loss function of the generator network is: ;in, A function that measures the difference between the future environment parameters predicted by the generator network and the actual environment parameters; ; is the predicted value of the generator network for the i-th time point, is the actual value of i time points, and n is the total number of time points; Used to characterize the environment data generated by the generator network After passing through the discriminator network D, the probability that the discriminator network marks it as real data; ; When D(G(Y)) is close to 1, that is, when the discriminator believes that the generated data G(Y) is real data, the loss tends to zero; The degree of match between the generated data G(Y) and the feature importance score W; ; represents the importance score of the i-th feature; Used to measure and ensure the similarity between the interdependencies between features in the generated data G(Y) and the real data; , and is the weight coefficient;

[0111] Constructing a discriminator network, the discriminator network is used to evaluate whether the prediction result of the generator network is consistent with the actual risk status; the discriminator network includes: a plurality of convolutional layers for extracting features from the prediction result and a fully connected layer for evaluating whether the risk prediction is consistent with the actual risk status;

[0112] Constructing a loss function module, the loss function module is used to quantify the difference between the generated risk prediction and the actual risk status; the loss function module includes: a content loss module for measuring the difference between the generated risk prediction and the actual risk status and an adversarial loss module for encouraging the generator to generate predictions that are closer to the actual risk status;

[0113] The initial model is obtained based on the constructed generator network, discriminator network and loss function module;

[0114] Use historical data to train the initial model and use a validation set to evaluate the prediction performance of the initial model and update the generator network and the discriminator network during the validation process;

[0115] When the confidence of the initial model is greater than the preset value, the risk prediction model is obtained.

[0116] Specifically, the analysis module uses time series analysis methods to capture the changing patterns of environmental parameters in historical environmental data over time, and analyzes the conditions and patterns of historical abnormal events, so as to identify the combination of environmental parameters that may cause anomalies, and then selects and trains the time series prediction model based on the characteristics of the feature set and historical data. After the prediction performance of the model reaches the preset confidence standard, the final risk prediction model is obtained, and real-time environmental data is input into the model to predict the changing trend of environmental parameters in the future period, and based on this, the possible risk level in the future is evaluated, and finally a risk prediction report containing the prediction results and risk level is generated.

[0117] As one implementation of an environmental monitoring system based on digital twin technology, the adaptive optimization module adjusts the environmental monitoring strategy according to the environmental prediction result, including:

[0118] The adaptive optimization module evaluates the safety level of the current monitored environment state based on the received environmental prediction results;

[0119] The adaptive optimization module evaluates the risk level of environmental changes in the future based on the risk prediction results;

[0120] The adaptive optimization module evaluates the development direction of environmental parameters based on trend analysis results;

[0121] The adaptive optimization module generates new monitoring strategies based on the evaluation results. The monitoring strategies include increasing the monitoring frequency, adjusting the monitoring area, and activating the early warning mechanism.

[0122] The adaptive optimization module uploads the generated new environmental monitoring strategy to the environmental monitoring host to update the monitoring plan;

[0123] The adaptive optimization module evaluates the execution effect of the new strategy based on the new environmental prediction results; if the execution effect does not meet the expected results, a new monitoring strategy is regenerated.

[0124] Specifically, the adaptive optimization module determines the safety level of the current environmental status, the risk level of future environmental changes, and the development direction of environmental parameters by evaluating the environmental prediction results, and generates new monitoring strategies based on this, such as increasing the monitoring frequency, adjusting the monitoring area, or activating the early warning mechanism. The new strategy is then uploaded to the environmental monitoring host to update the monitoring plan, and the execution effect of the new strategy is continuously evaluated. If the execution effect does not meet expectations, a new monitoring strategy is regenerated, thereby realizing dynamic adjustment and optimization of the environmental monitoring strategy and ensuring the real-time and effectiveness of the monitoring system.

[0125] As one implementation of an environmental monitoring system based on digital twin technology, the analysis module is also used to construct a situational awareness model; the situational awareness model is used to adjust the environmental data in the virtual environment model in combination with external factors;

[0126] The method for constructing the situational awareness model includes:

[0127] Acquire external factor data; the external factor data includes weather forecast, equipment maintenance records, and historical abnormal event records;

[0128] Performing correlation analysis on the external factor data and the environmental data to determine the correlation between the external factor and the environmental parameter and selecting the most relevant M external factor features;

[0129] Constructing an initial situational awareness model; the initial situational awareness model is trained based on historical environmental data and the most relevant M external factor features are integrated into the initial situational awareness model to obtain a situational awareness model.

[0130] Specifically, the analysis module obtains external factor data such as weather forecasts, equipment maintenance records, and historical abnormal event records, and performs correlation analysis with environmental data to determine the most relevant M external factor features, and then constructs an initial situational awareness model. The model is trained using historical environmental data and the selected external factor features are integrated into it to obtain the final situational awareness model, thereby achieving dynamic adjustment of environmental data in the virtual environment model in combination with external factors, thereby enhancing the prediction accuracy and situational adaptability of the environmental monitoring system.

[0131] As one implementation of an environmental monitoring system based on digital twin technology, the context awareness model is used to adjust environmental data in the virtual environment model in combination with external factors, including:

[0132] fusing the external factor data with the environmental data to form a comprehensive data set;

[0133] inputting the comprehensive data set into the situational awareness model;

[0134] The context awareness model adjusts the environmental data according to external factors to reflect the impact of external factors on environmental parameters;

[0135] The situational awareness model outputs adjusted environmental data for generating environmental prediction results;

[0136] The analysis module performs anomaly detection, risk prediction and trend analysis based on the adjusted environmental data to obtain environmental prediction results.

[0137] Specifically, the situational awareness model forms a comprehensive data set by fusing external factor data with environmental data, and inputs the comprehensive data set into the situational awareness model. The model adjusts the environmental data according to the external factors to reflect the impact of external factors on environmental parameters, and outputs the adjusted environmental data for generating environmental prediction results. The analysis module performs anomaly detection, risk prediction and trend analysis based on the adjusted environmental data, thereby obtaining more accurate environmental prediction results. Overall, it achieves accurate perception and prediction of the environmental status under the influence of external factors, and improves the reliability and prediction accuracy of the environmental monitoring system.

[0138] This application also provides an environmental monitoring method based on digital twin technology, including:

[0139] S201, obtaining environmental data of the monitoring area through a sensor array; the sensor array includes: a temperature and humidity sensor, an oxygen concentration sensor, a combustible gas concentration sensor, a vibration sensor, a noise sensor and a smoke sensor;

[0140] S202, obtaining and processing environmental data sent by each sensor in the sensor array through an environmental monitoring host to obtain an environmental prediction result;

[0141] S203, displaying the environment prediction results, displaying the virtual environment model and providing a user interaction interface through the back-end monitoring terminal;

[0142] Wherein, the environmental monitoring host is configured with a digital twin module, an analysis module and an adaptive optimization module;

[0143] The digital twin module is used to construct a virtual mapping of the actual physical environment based on the various environmental data acquired in real time to obtain a virtual environment model that is updated in real time; the analysis module is used to perform anomaly detection, risk prediction and trend analysis on the environmental data in the virtual environment model to obtain environmental prediction results; the adaptive optimization module adjusts the environmental monitoring strategy according to the environmental prediction results.

[0144] The embodiment of the present application also discloses an electronic device.

[0145] Specifically, the device includes a memory and a processor, and the memory stores a computer program that can be loaded by the processor and execute the above-mentioned environmental monitoring method based on digital twin technology.

[0146] The embodiment of the present application also discloses a computer-readable storage medium. Specifically, the computer-readable storage medium stores a computer program that can be loaded by a processor and executes the above-mentioned environmental monitoring method based on digital twin technology, and the computer-readable storage medium includes, for example: a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, and other media that can store program codes.

[0147] It should be noted that the above embodiments are only used to illustrate the present application and are not intended to limit the technical solutions described in the present application. Although the present application has been described in detail in this specification with reference to the above embodiments, a person of ordinary skill in the art should understand that a person of ordinary skill in the art can still modify or make equivalent substitutions to the present application, and all technical solutions and improvements thereof that do not depart from the spirit and scope of the present application should be included in the scope of the claims of the present application.

Claims

1. An environmental monitoring system based on digital twin technology, characterized in that: include: A sensor array is used to obtain environmental data of the monitoring area; the sensor array includes a temperature and humidity sensor, an oxygen concentration sensor, a combustible gas concentration sensor, a vibration sensor, a noise sensor and a smoke sensor; The environment monitoring host is used to: acquire and process the environment data sent by each sensor in the sensor array to obtain an environment prediction result; The back-end monitoring terminal is used to display the environmental prediction results, present the virtual environment model and provide a user interaction interface; Wherein, the environmental monitoring host is configured with a digital twin module, an analysis module and an adaptive optimization module; The digital twin module is used to construct a virtual mapping of the actual physical environment based on various environmental data acquired in real time to obtain a virtual environment model that is updated in real time; The analysis module is used to perform anomaly detection, risk prediction and trend analysis on the environmental data in the virtual environment model to obtain an environmental prediction result; The adaptive optimization module adjusts the environmental monitoring strategy according to the environmental prediction result; The analysis module performs risk prediction and trend analysis on the environmental data in the virtual environment model, including: The analysis module is based on the historical environmental data, and analyzes the change trend of the historical environmental data to obtain the analysis results of the historical environmental data; The analysis module builds a risk prediction model based on the analysis results of historical environmental data; The analysis module inputs the real-time environmental data into the risk prediction model to obtain the prediction results of the change trend of the environmental parameters in the future period of time output by the risk prediction model; The analysis module evaluates possible future risks based on the forecast results and classifies the risks into different levels; The analysis module generates a risk prediction report including prediction results and risk levels; The analysis results of the historical environmental data obtained by analyzing the changing trend of the historical environmental data include: Use time series analysis methods to capture the changing patterns of environmental parameters over time; Analyze the conditions and patterns of historical abnormal events and identify the combination of environmental parameters that may lead to abnormalities; The analysis module builds a risk prediction model based on the analysis results of historical environmental data, including: A generator network is constructed, wherein the generator network is used to predict future prediction results based on historical data; the generator network includes: multiple convolutional layers for extracting features from historical data, an upsampling layer for increasing the time step to predict future time points, a residual block for alleviating gradient disappearance and enhancing feature learning capabilities, and a jump connection module for maintaining information transfer of low-level features during upsampling; wherein the loss function of the generator network is: ;in, A function that measures the difference between the future environment parameters predicted by the generator network and the actual environment parameters; ; is the predicted value of the generator network for the i-th time point, is the actual value of i time points, and n is the total number of time points; Used to characterize the environment data generated by the generator network After passing through the discriminator network D, the probability that the discriminator network marks it as real data; ; When D(G(Y)) is close to 1, that is, when the discriminator believes that the generated data G(Y) is real data, the loss tends to zero; The degree of match between the generated data G(Y) and the feature importance score W; ; represents the importance score of the i-th feature; Used to measure and ensure the interdependence between features in the generated data G(Y) and the similarity between the real data; , and is the weight coefficient; Constructing a discriminator network, the discriminator network is used to evaluate whether the prediction result of the generator network is consistent with the actual risk status; the discriminator network includes: a plurality of convolutional layers for extracting features from the prediction result and a fully connected layer for evaluating whether the risk prediction is consistent with the actual risk status; Constructing a loss function module, the loss function module is used to quantify the difference between the generated risk prediction and the actual risk status; the loss function module includes: a content loss module for measuring the difference between the generated risk prediction and the actual risk status and an adversarial loss module for encouraging the generator to generate predictions that are closer to the actual risk status; The initial model is obtained based on the constructed generator network, discriminator network and loss function module; Use historical data to train the initial model and use a validation set to evaluate the prediction performance of the initial model and update the generator network and the discriminator network during the validation process; When the confidence of the initial model is greater than the preset value, the risk prediction model is obtained.

2. According to claim 1, an environmental monitoring system based on digital twin technology is characterized in that: The digital twin module is used to construct a virtual mapping of the actual physical environment based on the real-time acquired environmental data to obtain a real-time updated virtual environment model, including: Acquire environmental data sent from each sensor of the sensor array; Fuse the environmental data sent by each sensor to improve data consistency; Generate a virtual representation of the corresponding actual physical environment based on the fused environmental data; the virtual representation is a digital representation that reflects the distribution of various environmental parameters of the actual physical environment; Constructing a real-time virtual environment model including various environmental parameters in the monitoring area according to the virtual representation; the virtual environment model is used to display the spatial distribution of environmental parameters and their changing trends over time; The virtual environment model is updated in real time to reflect changes in environmental parameters in the actual physical environment.

3. The environmental monitoring system based on digital twin technology according to claim 2 is characterized in that: The analysis module performs anomaly detection on the environmental data in the virtual environment model, including: The analysis module calculates the importance score of each environmental data feature; the importance score is calculated based on the mutual information; The analysis module selects the top N features with the highest importance scores as a feature set for anomaly detection; The analysis module obtains a baseline model based on the environmental data during historical normal operation, compares the real-time environmental data with the baseline module, and calculates the distance between the real-time environmental data point and the normal state environmental data point using the Euclidean distance calculation method to obtain the difference between the two; The analysis module calculates an anomaly score for each environmental data point based on the difference and in combination with the feature importance scores in the feature set; When the abnormality score is greater than a preset value, it is determined that the environmental data point is abnormal.

4. The environmental monitoring system based on digital twin technology according to claim 3 is characterized in that: The adaptive optimization module adjusts the environmental monitoring strategy according to the environmental prediction result, including: The adaptive optimization module evaluates the safety level of the current monitored environment state based on the received environmental prediction results; The adaptive optimization module evaluates the risk level of environmental changes in the future based on the risk prediction results; The adaptive optimization module evaluates the development direction of environmental parameters based on trend analysis results; The adaptive optimization module generates new monitoring strategies based on the evaluation results. The monitoring strategies include increasing the monitoring frequency, adjusting the monitoring area, and activating the early warning mechanism. The adaptive optimization module uploads the generated new environmental monitoring strategy to the environmental monitoring host to update the monitoring plan; The adaptive optimization module evaluates the execution effect of the new strategy based on the new environmental prediction results; if the execution effect does not meet the expected results, a new monitoring strategy is regenerated.

5. The environmental monitoring system based on digital twin technology according to claim 4 is characterized in that: The analysis module is also used to construct a situational awareness model; the situational awareness model is used to adjust the environmental data in the virtual environment model in combination with external factors; The method for constructing the situational awareness model includes: Acquire external factor data; the external factor data includes weather forecast, equipment maintenance records, and historical abnormal event records; Performing correlation analysis on the external factor data and the environmental data to determine the correlation between the external factor and the environmental parameter and selecting the most relevant M external factor features; Constructing an initial situational awareness model; the initial situational awareness model is trained based on historical environmental data and the most relevant M external factor features are integrated into the initial situational awareness model to obtain a situational awareness model.

6. The environmental monitoring system based on digital twin technology according to claim 5 is characterized in that: The situational awareness model is used to adjust the environmental data in the virtual environment model in combination with external factors, including: fusing the external factor data with the environmental data to form a comprehensive data set; inputting the comprehensive data set into the situational awareness model; The context awareness model adjusts the environmental data according to external factors to reflect the impact of external factors on environmental parameters; The situational awareness model outputs adjusted environmental data for generating environmental prediction results; The analysis module performs anomaly detection, risk prediction and trend analysis based on the adjusted environmental data to obtain environmental prediction results.

7. An environmental monitoring method based on digital twin technology, based on the environmental monitoring system according to any one of claims 1 to 6, characterized in that: The method comprises: Acquire environmental data of the monitoring area through a sensor array; the sensor array includes: temperature and humidity sensors, oxygen concentration sensors, combustible gas concentration sensors, vibration sensors, noise sensors and smoke sensors; Acquiring and processing environmental data sent by each sensor in the sensor array through an environmental monitoring host to obtain an environmental prediction result; Display environmental prediction results, display virtual environment models and provide user interaction interfaces through back-end monitoring terminals; Wherein, the environmental monitoring host is configured with a digital twin module, an analysis module and an adaptive optimization module; The digital twin module is used to construct a virtual mapping of the actual physical environment based on the various environmental data acquired in real time to obtain a virtual environment model that is updated in real time; the analysis module is used to perform anomaly detection, risk prediction and trend analysis on the environmental data in the virtual environment model to obtain environmental prediction results; the adaptive optimization module adjusts the environmental monitoring strategy according to the environmental prediction results.

8. An electronic device, characterized in that: The method comprises a memory and a processor, wherein the memory stores a computer program of the method according to claim 7 which is loaded and executed by the processor.

9. A computer-readable storage medium, characterized in that: A computer program is stored which can be loaded by a processor and execute the method as claimed in claim 7.

Images