A harmful gas monitoring instrument configuration optimization method and monitoring instrument

Abstract

The present invention discloses a configuration optimization method for harmful gas monitoring instruments and monitoring instruments, which relate to the field of data processing technology, including: determining a preset monitoring area, mining regional risk points and determining a front-end configuration network; determining a monitoring trigger period, executing multi-source monitoring based on the front-end configuration network, and determining the regional flow field by performing spatial field conversion on multi-source monitoring data; constructing a target prediction function, and supervising the training of a flow field decision model; for the regional flow field, combining the flow field decision model, performing data fitting and flow field decomposition, and determining risk distribution information by performing risk over-limit judgment and trend prediction decision; optimizing the front-end configuration network based on the risk distribution information, and determining a configuration optimization strategy; and performing configuration optimization management of the monitoring instrument based on the configuration optimization strategy. The present invention solves the technical problem that the prior art cannot provide real-time warning of leakage risks, and achieves the technical effect of accurate and efficient monitoring and early warning of harmful gases.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and in particular to a harmful gas monitoring instrument configuration optimization method and a monitoring instrument. Background Art

[0002] In modern industrial and environmental monitoring, real-time and accurate detection and early warning of harmful gas leaks are important links in ensuring personnel safety and environmental protection. Many industrial sites, such as chemical plants, oil refineries, and mines, are at potential risk of harmful gas leaks. Once these gases, such as hydrogen sulfide, ammonia, and volatile organic compounds, leak, they not only pose a serious threat to the health of people inside the factory, but also have a significant impact on the surrounding environment and the safety of residents. Therefore, it is particularly important to establish an efficient and accurate harmful gas monitoring system.

[0003] However, traditional monitoring systems have a slow response speed and limited data processing and analysis capabilities. Faced with complex environmental data, traditional systems are often unable to quickly and accurately capture changes in gas concentrations and provide real-time warnings of leakage risks. Summary of the invention

[0004] The present application provides a harmful gas monitoring instrument configuration optimization method and a monitoring instrument, which are used to solve the technical problem that the prior art cannot provide real-time warning of leakage risks.

[0005] In view of the above problems, the present application provides a harmful gas monitoring instrument configuration optimization method and a monitoring instrument.

[0006] In a first aspect of the present application, a method for optimizing the configuration of a harmful gas monitoring instrument is provided, the method comprising:

[0007] Determine the preset monitoring area, mine the regional risk points and determine the front-end configuration network, wherein the regional risk points include risk location-risk type; determine the monitoring trigger period based on the monitoring needs, perform multi-source monitoring based on the front-end configuration network, and determine the regional flow field by performing spatial field conversion on the multi-source monitoring data; obtain the target prediction function, supervise the training of the flow field decision model, and the flow field decision model includes a real-time decision layer and a prediction decision layer; for the regional flow field, in combination with the flow field decision model, perform data fitting and flow field decomposition, and determine the risk distribution information by performing risk over-limit judgment and trend prediction decision, wherein the risk distribution information includes feature multi-tuples and tracking windows, wherein the regional flow field is used as the decision scenario of the flow field decision model; based on the risk distribution information, optimize the front-end configuration network and determine the configuration optimization strategy, and the configuration optimization strategy includes distribution configuration dimension and monitoring trigger dimension; based on the configuration optimization strategy, perform configuration optimization management of monitoring instruments.

[0008] A second aspect of the present application provides a monitoring instrument, the monitoring instrument comprising:

[0009] A front-end monitoring device, wherein the front-end monitoring device comprises a camera device, a sensor device and a fan, wherein the front-end monitoring device is distributed to constitute the front-end configuration network, and the front-end configuration network is adjustable; a control cabinet, wherein the control cabinet is the edge control center of the monitoring instrument, and can execute automated decision control and human-computer interaction control; wherein the front-end monitoring device control is integrated in the control cabinet.

[0010] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0011] The present application determines the preset monitoring area, mines the regional risk points and determines the front-end configuration network, wherein the regional risk points include the risk location-risk type; determines the monitoring trigger cycle based on the monitoring demand, performs multi-source monitoring based on the front-end configuration network, and determines the regional flow field by converting the multi-source monitoring data into a spatial field; constructs a target prediction function, supervises the training of the flow field decision model, and the flow field decision model includes a real-time decision layer and a prediction decision layer; for the regional flow field, combined with the flow field decision model, performs data fitting and flow field decomposition, and determines the risk distribution information by making risk over-limit judgments and trend prediction decisions, and the risk distribution information includes feature multi-groups and tracking windows; based on the risk distribution information, optimizes the front-end configuration network, determines the configuration optimization strategy, and the configuration optimization strategy includes the distribution configuration dimension and the monitoring trigger dimension; based on the configuration optimization strategy, performs configuration optimization management of the monitoring instrument. The present invention solves the technical problem that the prior art cannot warn of leakage risks in real time, determines the risk distribution information through multi-source data fusion, fluid dynamics simulation and time series analysis, and optimizes the sensor layout and monitoring strategy through real-time monitoring and data-driven risk assessment, so as to achieve the technical effect of accurate and efficient monitoring and early warning of harmful gases. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0013] Figure 1 A schematic diagram of a method for optimizing the configuration of a harmful gas monitoring instrument provided in an embodiment of the present application;

[0014] Figure 2 A schematic diagram of a process flow for determining a regional flow field in a method for optimizing configuration of a harmful gas monitoring instrument provided in an embodiment of the present application. DETAILED DESCRIPTION

[0015] This application provides a method for optimizing the configuration of a harmful gas monitoring instrument and a monitoring instrument, aiming to solve the technical problem that the existing technology cannot provide real-time warning of leakage risks. Through multi-source data fusion, fluid dynamics simulation and time series analysis, the risk distribution information is determined, and through real-time monitoring and data-driven risk assessment, the sensor layout and monitoring strategy are optimized to achieve the technical effect of accurate and efficient monitoring and early warning of harmful gases.

[0016] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.

[0017] It should be noted that any variations of the terms "include" and "have" are intended to cover non-exclusive inclusions. For example, a process, method, system, product or server that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or modules that are not explicitly listed or inherent to these processes, methods, products or devices.

[0018] Embodiment 1, as Figure 1 As shown, the present application provides a method for optimizing the configuration of a harmful gas monitoring instrument, the method comprising:

[0019] Step S100: determine a preset monitoring area, mine regional risk points and determine a front-end configuration network, wherein the regional risk points include risk location-risk type.

[0020] In the embodiment of the present application, the geographic information of the preset monitoring area is first obtained from the GIS database or the internal system of the enterprise. These areas are pre-set according to the safety management requirements and regulatory standards of the enterprise, such as chemical plant areas, storage areas or industrial parks. GIS technology is used to draw a detailed regional map, clarify the boundaries and internal structure of the monitoring area, and determine the preset monitoring area. Next, by analyzing historical safety accident records and environmental monitoring data, potential risk points in the area are automatically identified. Historical data is extracted from the database, and the accident sites and environmental monitoring data that have occurred are statistically analyzed and analyzed to find high-risk areas. By analyzing these data, specific risk locations and risk types are identified. For example, chemical leaks, exhaust gas outlets, storage tank areas, and production equipment failure points, each risk point records its specific geographic coordinates and risk types in detail, such as chemical leaks, exhaust gas emissions, or high-temperature equipment, to determine the risk location-risk type.

[0021] After identifying the risk points, the front-end configuration network is designed according to the location and type of these risk points. First, 3D modeling and simulation technology is used to automatically plan the installation points of monitoring instruments according to the location and type of risk points to ensure that these points can reasonably cover all identified risk points. Then, according to different risk types, appropriate monitoring equipment is selected, such as gas sensors, cameras, and anemometers. Gas sensors are used to detect specific harmful gases, such as volatile organic compounds, hydrogen sulfide, etc.; cameras are used to monitor the on-site situation in real time, especially chemical storage areas and exhaust gas outlets; anemometers are used to detect airflow changes and assist in determining gas diffusion paths. To ensure effective data transmission, the communication network topology is designed, and the appropriate communication method is selected according to the distribution and regional characteristics of the monitoring equipment. For example, when using wireless sensor network technology, low-power wide area network technologies such as ZigBee, LoRa, or NB-IoT are used. Through network simulation tools such as NS-3 or OPNET, different communication topologies are simulated, data transmission paths are optimized, and the reliability and low latency of the communication network are ensured.

[0022] Next, the front-end monitoring network is installed and debugged to ensure the normal operation of the monitoring equipment. A detailed equipment installation plan is generated for reference by manual installers to ensure that the equipment is reasonably located and the monitoring coverage is comprehensive. Then the communication parameters of all monitoring devices, such as IP address, communication protocol and encryption method, are configured, and the devices are connected to the monitoring network to ensure that the data can be stably transmitted to the central monitoring system. Network analysis tools such as Wireshark or SolarWinds are used to detect and optimize network performance, and comprehensive testing and verification are carried out through simulation data and actual operation data to ensure the reliability and stability of the system.

[0023] Through the above process, the preset monitoring area is determined, the regional risk points are discovered, and the front-end configuration network is determined.

[0024] Step S200: determining a monitoring trigger period based on monitoring requirements, executing multi-source monitoring based on the front-end configuration network, and determining a regional flow field by performing spatial field conversion on multi-source monitoring data.

[0025] In an embodiment of the present application, a monitoring trigger period is first determined based on monitoring requirements. The monitoring trigger period refers to the time interval at which the monitoring system starts data collection. High-risk time periods and frequencies are identified by statistical analysis of historical monitoring data and accident records. For example, a chemical plant may have a higher risk of leakage during certain specific time periods, such as peak production periods, so the monitoring frequency is increased during these time periods. Statistical models, such as the Poisson process or the Markov chain, are used to predict the time when possible abnormal events may occur, thereby optimizing the monitoring trigger period. After determining the monitoring trigger period, multi-source monitoring based on the front-end configuration network is performed, and data is collected through a variety of sensors deployed in the monitoring area, such as gas sensors, temperature sensors, and humidity sensors, to obtain multi-source monitoring data.

[0026] Next, the multi-source monitoring data is converted into a spatial field to determine the regional flow field. First, a spatial interpolation algorithm, such as Kriging interpolation or inverse distance weighted interpolation (IDW), is used to convert discrete data into continuous data. Then, the geographic coordinates of the sensor data are standardized using geographic information system technology to ensure the spatial consistency of the data. After the spatial processing is completed, flow field analysis and modeling are performed, and fluid dynamics simulation software, such as CFD, is used to simulate the gas flow in the region to generate a regional flow field. The regional flow field shows the concentration gradient and flow direction of the gas in the region.

[0027] Step S300: Obtain a target prediction function and supervise the training of a flow field decision model, wherein the flow field decision model includes a real-time decision layer and a prediction decision layer.

[0028] In the embodiment of the present application, the target prediction function is first obtained. Next, when the flow field decision model is supervised and trained, the historical monitoring data, including environmental data such as gas concentration, temperature, humidity, wind speed, etc., is first obtained through the historical database, and the recurrent neural network is trained using the prepared historical monitoring data to obtain the flow field decision model.

[0029] The flow field decision model includes a real-time decision layer and a prediction decision layer. The real-time decision layer is mainly responsible for real-time analysis and processing of monitoring data and immediate response. For example, when the gas sensor detects an abnormal increase in the concentration of harmful gases, the real-time decision layer immediately triggers an alarm and takes emergency measures. This layer usually uses fast response algorithms, such as real-time data processing through fast Fourier transform. The prediction decision layer predicts the gas concentration and flow conditions in the future based on current and historical monitoring data and combined with the target prediction function. The prediction decision layer uses time series analysis, such as the ARIMA model and dynamic simulation technology, such as CFD simulation, to estimate future environmental changes and provide early warning information.

[0030] Step S400: For the regional flow field, in combination with the flow field decision model, data fitting and flow field decomposition are performed, and risk distribution information is determined by performing risk over-limit judgment and trend prediction decisions. The risk distribution information includes a feature multi-tuple and a tracking window, wherein the regional flow field is used as a decision scenario for the flow field decision model.

[0031] In an embodiment of the present application, for the regional flow field, in combination with the flow field decision model, when performing data fitting, the model parameters are optimized by minimizing the error between the data points and the fitting curve through least squares fitting, so that the model can more accurately reflect the actual environment. After the data fitting is completed, flow field decomposition is performed. Flow field decomposition is to decompose complex flow field data into several simple patterns so that it is easier to understand and process. Flow field decomposition methods include principal component analysis and singular value decomposition. These methods reduce the data dimension and simplify the data processing process by extracting the main features of the flow field. For example, PCA projects high-dimensional data into a low-dimensional space by finding the principal components of the data, thereby extracting the main flow patterns.

[0032] After data fitting and flow field decomposition are completed, risk over-limit judgment and trend prediction decisions are made. When making risk over-limit judgments, the actual monitoring data is compared with the preset safety threshold to determine whether there is an over-limit situation. Specifically, a rule-based system or a threshold alarm system implements over-limit judgment. The specific steps include setting a safety threshold. When the real-time monitoring data exceeds the preset threshold, an alarm will be triggered. For example, the NumPy library in Python is used to compare the monitoring data with the threshold in real time, and an alarm will be triggered immediately when an over-limit situation is found. Subsequently, a time series analysis method is used to make trend prediction decisions. Using tools such as the ARIMA model or exponential smoothing method, a prediction model is trained based on historical monitoring data to predict the concentration of harmful gases and the trend of flow field changes in the future.

[0033] After completing the risk over-limit determination and trend prediction decision, determine the risk distribution information. Risk distribution information includes feature multi-groups and tracking windows. Feature multi-groups are a set of variable combinations that can describe risk characteristics, such as gas concentration, temperature, humidity, wind speed, etc. These variables can fully reflect the current environmental conditions and risk levels. The specific steps include using the Pandas library to organize and combine various monitoring data to form feature multi-groups. The tracking window refers to the time period used to monitor and analyze risk changes. Set a suitable time window to promptly detect and respond to risk changes. For example, set a one-hour tracking window and use real-time data for continuous monitoring and analysis.

[0034] Step S500: Based on the risk distribution information, the front-end configuration network is optimized and a configuration optimization strategy is determined. The configuration optimization strategy includes a distribution configuration dimension and a monitoring trigger dimension.

[0035] In an embodiment of the present application, when the front-end configuration network is optimized based on risk distribution information, the purpose is to ensure that the monitoring equipment can cover high-risk areas more effectively and respond to potential risk changes in a timely manner.

[0036] Specifically, first use the risk distribution information and data analysis methods to determine high-risk areas and key monitoring points. Use heat maps or risk maps to display risk distribution. These charts can help identify areas that need to be monitored. Then use optimization algorithms to determine the best sensor layout scheme. Optimization algorithms include genetic algorithms and particle swarm optimization algorithms. These algorithms iteratively search the optimization space to find the sensor configuration that can maximize coverage of high-risk areas. After determining the optimized sensor configuration scheme, simulation tests are carried out. By using fluid dynamics simulation software such as ANSYSFluent, the optimized sensor configuration is simulated and tested to verify its effectiveness and coverage. Through simulation tests, it is ensured that the optimized configuration scheme can effectively monitor the changes in gas concentration in high-risk areas in actual applications.

[0037] Next, determine the monitoring trigger dimension. The monitoring trigger dimension refers to the time interval and conditions for the sensor to start data collection. In order to dynamically adjust the monitoring trigger period, the monitoring frequency is dynamically adjusted according to the risk distribution information. Increase the monitoring frequency during high-risk periods, such as peak production periods, and reduce the monitoring frequency during low-risk periods. Statistical models such as Poisson processes or Markov chains can be used to predict risk changes and optimize monitoring trigger strategies. In addition, set up an event-driven monitoring mechanism. Event-driven monitoring means that when monitoring data is abnormal, such as a sharp increase in gas concentration, additional monitoring activities are immediately triggered. For example, using a threshold alarm system, high-frequency monitoring is initiated when real-time monitoring data exceeds a preset threshold. The specific implementation can be achieved by using the NumPy library in Python to compare monitoring data with thresholds in real time. When data exceeds the limit, an alarm and additional monitoring are immediately triggered.

[0038] The configuration optimization strategy is implemented by optimizing the distribution configuration dimension and the monitoring trigger dimension.

[0039] Step S600: Based on the configuration optimization strategy, perform configuration optimization management of monitoring instruments.

[0040] In the embodiment of the present application, the configuration optimization strategy includes the distribution configuration dimension and the monitoring trigger dimension. In the distribution configuration dimension, according to the optimization results, the sensors are actually installed to ensure that they accurately cover the high-risk areas. In the monitoring trigger dimension, the monitoring frequency is dynamically adjusted according to the risk distribution information. Through this process, the configuration optimization management of the monitoring instrument is completed.

[0041] Furthermore, the method provided in the application embodiment also includes:

[0042] The communication network topology is determined by communicating with the monitoring instrument, wherein the communication network access is the connection between the internal network and the external network of the monitoring instrument; based on the communication network topology, data communication decisions are made and communication strategies are determined, wherein the communication strategies include communication paths and synchronization constraints; wherein the data communication decisions include: based on the network time protocol, multi-target synchronization constraints are performed by setting synchronization timestamps; and communication path decisions are made based on channel status and channel type.

[0043] In an embodiment of the present application, the communication network topology needs to be determined first for the monitoring instrument to communicate with the network, which involves connecting the monitoring instrument to the intranet and the extranet to achieve the interconnection of the intranet and the extranet. Communication network access refers to connecting the monitoring device to the existing network infrastructure so that the device can send and receive data. Specifically, network topology discovery tools such as SNMP and LLDP are used to identify and record the locations of all monitoring instruments and the network nodes to which they are connected, thereby determining the communication network topology. Next, the monitoring instrument is connected to the intranet and the extranet respectively to achieve the interconnection of the two. This step is achieved by configuring network switches and routers to ensure that data can be smoothly transmitted between the intranet and the extranet. During the configuration process, switch configuration tools (such as Cisco's IOS) and routing protocols (such as OSPF or BGP) are used to set the routing and switching rules of the device.

[0044] Next, data communication decisions are made based on the determined communication network topology, including the determination of communication paths and synchronization constraints. The communication strategy determines the transmission path and time of data in the network to ensure the reliability and timeliness of data transmission. Specifically, the network time protocol is used to synchronize the time of all monitoring instruments. NTP is a protocol for synchronizing computer system time. It implements multi-target synchronization constraints by synchronizing timestamp settings. The system configures an NTP server so that all monitoring instruments regularly synchronize time with the NTP server, thereby ensuring that the time of all devices is consistent and eliminating the data inconsistency problem caused by time differences. Next, the best communication path is selected based on channel status and channel type. Channel status refers to the real-time status of each communication link in the network, including indicators such as bandwidth, delay, and packet loss rate. Channel type refers to different network types, such as wired networks and wireless networks. Use dynamic routing protocols such as OSPF or BGP to monitor network conditions in real time, dynamically adjust communication paths according to channel status and channel type, and ensure efficient and stable data transmission. When making communication decisions, consider multi-target synchronization constraints, including data transmission delay, bandwidth utilization, and data integrity. Through multi-objective optimization algorithms, such as genetic algorithm and multi-objective particle swarm optimization algorithm, the optimal balance point is found among multiple objectives to ensure the best performance of data transmission while meeting the synchronization constraints.

[0045] In summary, by connecting monitoring instruments to the communication network and determining the communication network topology, combined with the network time protocol and dynamic routing protocol, multi-objective synchronization constraints and communication path decisions are achieved.

[0046] Furthermore, the method provided in the application embodiment also includes:

[0047] Based on data sensitivity and time limit priority, data decision allocation is performed on the edge and the cloud to determine the first monitoring data and the second monitoring data; based on the first monitoring data, an edge decision is performed to determine the first decision data; based on the second monitoring data, a cloud decision is performed to determine the second decision data; in combination with the network time protocol, the first decision data and the second decision data are coordinated to perform configuration optimization management of the front-end configuration device through timestamp constraints.

[0048] In an embodiment of the present application, the monitoring data is first classified. Data sensitivity refers to the importance of the data and the need for privacy protection, while time limit priority refers to the urgency and real-time requirements of data processing. Using a rule-based classification algorithm, the monitoring data is divided into two categories: first monitoring data and second monitoring data according to preset sensitivity and time limit priority rules. Data with high sensitivity and high time limit priority are assigned to edge processing as first monitoring data; data with low sensitivity and low time limit priority are assigned to cloud processing as second monitoring data.

[0049] Next, based on the first monitoring data, edge decision making is performed to determine the first decision data. Edge decision making refers to local data processing and analysis to quickly respond to real-time events. Specifically, the edge device uses IoT sensor technology to collect the first monitoring data in real time, uses stream processing technology or complex event processing system to quickly analyze the data, and generates the first decision data based on the analysis results.

[0050] Based on the second monitoring data, a decision is made in the cloud to determine the second decision data. Cloud decision-making is suitable for non-real-time processing that requires a large amount of computing resources and storage space. Specifically, the second monitoring data is uploaded to the cloud server through the data transmission service of a cloud service provider (such as AWS, Azure, Google Cloud), and the data is batch processed and analyzed using a big data processing platform (such as Apache Hadoop, Apache Spark) in the cloud, and finally the second decision data is generated based on the analysis results.

[0051] In order to ensure the consistency and collaboration of the edge decision data and the cloud decision data, the first decision data and the second decision data are coordinated in combination with the network time protocol, and the configuration optimization management of the front-end configuration device is performed through timestamp constraints. First, use NTP to synchronize the time of all edge devices and cloud servers to ensure the time consistency of all devices. During the data collection and processing process, use time synchronization services such as Chrony or NTP clients to add timestamps to each data point and decision result. Then, data collaboration is performed based on the timestamp, and the first decision data and the second decision data are compared and integrated by using database management systems and data processing tools to ensure the consistency and synchronization of the two. Based on the coordinated decision data, the front-end configuration device is optimized and managed.

[0052] Further, such as Figure 2 As shown, in the method provided in the embodiment of the application, the determining of the regional flow field further includes:

[0053] Based on the front-end configuration network, multiple field spaces are determined, wherein the multiple field spaces correspond one-to-one to the front-end configuration devices, with the distribution position as the field center; a global space is determined, and the global space covers the preset monitoring area; based on the global space, the multiple field spaces are converted into spatial coordinate systems to determine the regional flow field constructed by multi-source monitoring data.

[0054] In an embodiment of the present application, firstly, multiple field spaces are determined based on the front-end configuration network. The front-end configuration network is composed of a sensor network deployed in the monitoring area, and these sensors collect environmental data in real time. Multiple field spaces refer to different monitoring areas covered by sensors, and each field space corresponds to a front-end configuration device one by one. Specifically, the monitoring area of ​​each sensor is determined by the geographical location and coverage of the sensor, and these monitoring areas are defined as field spaces, with the distribution location of the sensor as the field center.

[0055] Next, determine the global space. The global space refers to a unified monitoring space covering the entire preset monitoring area. According to the geographic boundaries of the preset monitoring area, all field spaces are integrated to form a complete global space. Then, based on the global space, multiple field spaces are converted into spatial coordinate systems. Spatial coordinate system conversion refers to converting the data of each field space into a unified coordinate system for overall analysis and processing. Specifically, using geographic information system technology, the monitoring data of each field space is converted according to a unified coordinate system. By processing and converting geographic spatial data through GIS technology, data from different sources can be aligned and integrated in the same coordinate system.

[0056] Finally, determine the regional flow field constructed by multi-source monitoring data. The regional flow field refers to the gas flow conditions covering the entire monitoring area, which is constructed through multi-source monitoring data. Multi-source monitoring data includes data such as gas concentration, temperature, humidity, wind speed, etc. collected by various sensors. Using fluid mechanics simulation software, based on the converted spatial coordinate system, the multi-source monitoring data is simulated for flow field to generate a regional flow field model.

[0057] Furthermore, in the method provided in the embodiment of the application, the constructing of the target prediction function further includes:

[0058] Get the target prediction function:

[0059] ;

[0060] in, is the fitting function, is the monitoring value of the first gas data at the first position, is the risk level of the first gas, is the total influence coefficient of the flow field trend-related factors of the first gas; For the The weight value of the factors related to the flow field trend change, For the Monitoring values ​​of flow field trend-related factors, For the The coefficient of the trend of the flow field trend related factors, is the total amount of factors related to flow field trend variation; Is a constant.

[0061] In the embodiment of the present application, the monitoring value of the first gas data at the first location is obtained in real time through the sensor network deployed in the monitoring area. These data include environmental parameters such as gas concentration, temperature, humidity, wind speed, etc. The risk level is determined by analyzing historical monitoring data and risk assessment models. The total impact coefficient refers to the combined impact of all flow field trend-related factors on gas flow and concentration. This coefficient is calculated through fluid mechanics simulation. The weight values ​​of the factors related to the flow field trend are determined by multiple regression analysis to determine the optimal weight value. By performing multiple regression analysis on historical monitoring data, the contribution of each factor to the change in gas concentration is found, and the weight value is determined based on the contribution. The monitoring value of the flow field trend-related factors refers to the actual measurement data of each flow field trend-related factor, which is also collected in real time by the sensor network. The trend coefficient represents the effect strength of each flow field trend-related factor on the gas flow change. This coefficient is determined by fitting and optimizing experimental data or historical data using nonlinear regression or other fitting algorithms.

[0062] After all parameters are determined, all monitoring data and calculated parameters are integrated together to form a complete data set, and the target prediction function is constructed using the above formula.

[0063] Furthermore, in the method provided in the embodiment of the application, the risk distribution information includes a tracking window, and further includes:

[0064] Based on the flow field prediction trend, the time tracking feature is determined with the monitoring trigger time as the goal; based on the flow field prediction trend, the space tracking feature is determined with the monitoring flow field position as the goal; the time tracking feature and the space tracking feature are fitted to determine the tracking window.

[0065] In an embodiment of the present application, the flow field prediction trend is first obtained through fluid dynamics simulation and historical monitoring data analysis. Based on the flow field prediction trend, the time tracking characteristics are determined with the monitoring trigger time as the goal. The monitoring trigger time refers to when the system starts data collection. To determine these key time points, time series analysis methods such as the ARIMA model or exponential smoothing method are first used to analyze the historical monitoring data to identify the patterns and mutation points of gas concentration changes. Through these analyses, it is determined which time points in the future are key moments for monitoring. These moments are often the time points when the gas concentration changes significantly and need to be focused on and monitored.

[0066] Next, the spatial tracking features are determined with the monitoring flow field location as the target. The monitoring flow field location refers to the area that needs to be monitored. In order to identify these areas, the geographic information and sensor locations within the monitoring area are analyzed using geographic information system technology. The geospatial data is determined by GIS software to help the system identify high-risk areas and key monitoring points. By analyzing the flow field prediction results, areas with significant changes in gas concentration are identified, which are the main paths and locations for gas leakage or diffusion.

[0067] After determining the time tracking features and the space tracking features, the two are fitted to determine the tracking window. The tracking window refers to the time period and spatial region used to monitor and analyze gas changes. To this end, the time tracking features and the space tracking features are first integrated into a comprehensive data set, and the data are sorted and merged using data processing tools. Then, multiple regression analysis or other fitting algorithms, such as the least squares method, are used to construct a fitting model to fit the time tracking features and the space tracking features. Through multiple regression analysis, the weight of the influence of each feature on the tracking window is determined, so as to accurately determine the optimal tracking window. The tracking window includes a specific time period and spatial region.

[0068] Furthermore, in the method provided in the embodiment of the application, after determining the risk distribution information, the method further includes:

[0069] Identify the risk distribution information, perform risk localization based on the feature multi-group, and determine the first warning information; based on the tracking window, perform risk-based time localization and space localization to determine the second warning information, wherein the localization characteristics include spatiotemporal risk nodes and trend change rates, and issue a hazardous gas risk warning based on the first warning information and the second warning information.

[0070] In an embodiment of the present application, risk distribution information is first identified, including spatiotemporal distribution data of variables such as gas concentration, temperature, humidity, and wind speed. These variables are combined to form a feature multivariate group. Next, risk localization is performed based on the feature multivariate group. Risk localization refers to determining high-risk areas and time periods. Specifically, through multivariate regression analysis or statistical methods, the over-limit conditions of key variables in the feature multivariate group are identified, such as the spatial expansion rate of gas concentration and the growth rate in the time dimension. When a variable or multiple variables exceed a preset threshold, the point is identified as a risk point. By integrating the data of these risk points, the first warning information is determined.

[0071] Next, time localization and space localization based on risk are performed based on the tracking window. Time localization refers to monitoring risk changes within a specific time period, while space localization refers to monitoring risk diffusion within a specific spatial range. Specifically, the risk point data within the tracking window is analyzed, and the spatial expansion rate of gas concentration changes and the growth rate in the time dimension are evaluated through fluid mechanics simulation and time series analysis. If these change rates exceed the preset thresholds, these points are marked as spatiotemporal risk nodes. Based on the above analysis, the second warning information is determined. The second warning information includes spatiotemporal risk nodes and trend change rates. The trend change rate refers to the rate of change of gas concentration in a specific space and time dimension.

[0072] Finally, the first warning information and the second warning information are combined to issue a hazardous gas risk warning. The hazardous gas risk warning means that the system sends an alarm to the monitoring personnel, indicating the existence of potential high-risk areas and time periods. Specifically, the risk distribution map is displayed in real time through data visualization tools, and warning information is sent to relevant personnel through the alarm system, such as SMS, email, and push notifications. The warning information includes detailed data such as the specific location, time, gas concentration, expansion rate, and growth rate of the risk point.

[0073] In the embodiments of the present application, in summary, the embodiments of the present application have at least the following technical effects:

[0074] The present application determines the preset monitoring area, mines the regional risk points and determines the front-end configuration network, wherein the regional risk points include the risk location-risk type; determines the monitoring trigger cycle based on the monitoring demand, performs multi-source monitoring based on the front-end configuration network, and determines the regional flow field by converting the multi-source monitoring data into a spatial field; constructs a target prediction function, supervises the training of the flow field decision model, and the flow field decision model includes a real-time decision layer and a prediction decision layer; for the regional flow field, combined with the flow field decision model, performs data fitting and flow field decomposition, and determines the risk distribution information by making risk over-limit judgments and trend prediction decisions, and the risk distribution information includes feature multi-groups and tracking windows; based on the risk distribution information, optimizes the front-end configuration network, determines the configuration optimization strategy, and the configuration optimization strategy includes the distribution configuration dimension and the monitoring trigger dimension; based on the configuration optimization strategy, performs configuration optimization management of the monitoring instrument. The present invention solves the technical problem that the prior art cannot warn of leakage risks in real time, determines the risk distribution information through multi-source data fusion, fluid dynamics simulation and time series analysis, and optimizes the sensor layout and monitoring strategy through real-time monitoring and data-driven risk assessment, so as to achieve the technical effect of accurate and efficient monitoring and early warning of harmful gases.

[0075] Embodiment 2, based on the same inventive concept as a water-photovoltaic storage smart energy management method in the aforementioned embodiment, the present application provides a monitoring instrument, the monitoring instrument implements a monitoring configuration of a harmful gas monitoring instrument configuration optimization method, the monitoring instrument establishes a communication connection with the cloud, and the monitoring instrument includes:

[0076] A front-end monitoring device, wherein the front-end monitoring device comprises a camera device, a sensor device and a fan, wherein the front-end monitoring device is distributed to constitute the front-end configuration network, and the front-end configuration network is adjustable; a control cabinet, wherein the control cabinet is the edge control center of the monitoring instrument, and can execute automated decision control and human-computer interaction control; wherein the front-end monitoring device control is integrated in the control cabinet.

[0077] It should be noted that the above-mentioned sequence of the embodiments of the present application is only for description and does not represent the advantages and disadvantages of the embodiments. And the above-mentioned specific embodiments of this specification are described. The processes depicted in the accompanying drawings do not necessarily require the specific order and continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0078] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application shall be included in the protection scope of the present application.

[0079] This specification and drawings are merely exemplary illustrations of the present application and are deemed to cover any and all modifications, variations, combinations or equivalents within the scope of the present application. Obviously, a person skilled in the art may make various modifications and variations to the present application without departing from the scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalents, the present application intends to include these modifications and variations.

Claims

1. A method for optimizing the configuration of harmful gas monitoring instruments, characterized in that: The method comprises: Determine the preset monitoring area, dig out the regional risk points and determine the front-end configuration network, wherein the regional risk points include risk location-risk type; Determine a monitoring trigger period based on monitoring requirements, perform multi-source monitoring based on the front-end configuration network, and determine a regional flow field by performing spatial field conversion on multi-source monitoring data; Obtaining a target prediction function and supervising the training of a flow field decision model, wherein the flow field decision model includes a real-time decision layer and a prediction decision layer; For the regional flow field, in combination with the flow field decision model, data fitting and flow field decomposition are performed, and risk distribution information is determined by making risk over-limit judgments and trend prediction decisions, wherein the risk distribution information includes a feature multi-tuple and a tracking window, wherein the regional flow field is used as a decision scenario of the flow field decision model; Based on the risk distribution information, the front-end configuration network is optimized to determine a configuration optimization strategy, wherein the configuration optimization strategy includes a distribution configuration dimension and a monitoring trigger dimension; Based on the configuration optimization strategy, perform configuration optimization management of monitoring instruments; The determining of the regional flow field comprises: Based on the front-end configuration network, a plurality of field spaces are determined, wherein the plurality of field spaces correspond to the front-end configuration devices one by one, and the distribution position is the field center; Determine a global space, where the global space covers the preset monitoring area; Based on the global space, the multiple field spaces are converted into spatial coordinate systems to determine the regional flow field constructed by multi-source monitoring data; Construct the target prediction function, including: Get the target prediction function: ； in, is the fitting function, is the monitoring value of the first gas data at the first position, is the risk level of the first gas, is the total influence coefficient of the flow field trend-related factors of the first gas; For the The weight value of the factors related to the flow field trend change, is the monitoring value of the factors related to the trend change of the flow field, For the The coefficient of the trend of the flow field trend related factors, is the total amount of factors related to flow field trend variation; Is a constant.

2. A method for optimizing the configuration of harmful gas monitoring instruments as claimed in claim 1, characterized in that: The method further comprises: The monitoring instrument is connected to the communication network to determine the communication network topology, wherein the communication network connection is the connection between the internal network and the external network of the monitoring instrument; In combination with the communication network topology, a data communication decision is made to determine a communication strategy, wherein the communication strategy includes a communication path and synchronization constraints; Among them, making data communication decisions includes: Based on the network time protocol, multi-objective synchronization constraints are performed by setting synchronization timestamps; The communication path decision is made based on the channel status and channel type.

3. A method for optimizing the configuration of harmful gas monitoring instruments as claimed in claim 2, characterized in that: The method further comprises: Based on data sensitivity and time priority, data decision allocation is performed on the edge and cloud to determine the first monitoring data and the second monitoring data; Based on the first monitoring data, performing edge decision-making to determine first decision data; Based on the second monitoring data, performing cloud-based decision-making to determine second decision data; In combination with the network time protocol, the first decision data and the second decision data are coordinated to perform configuration optimization management of the front-end configuration device through timestamp constraints.

4. A method for optimizing the configuration of harmful gas monitoring instruments according to claim 1, characterized in that: The risk distribution information includes a tracking window, including: Based on the flow field prediction trend, the time tracking characteristics are determined with the monitoring trigger time as the goal; Based on the flow field prediction trend, the spatial tracking characteristics are determined with the goal of monitoring the flow field position; The time tracking feature and the space tracking feature are fitted to determine the tracking window.

5. A method for optimizing the configuration of harmful gas monitoring instruments as claimed in claim 1, characterized in that: After the risk distribution information is determined, the following steps are included: Identify the risk distribution information, perform risk localization based on the characteristic multi-tuple, and determine first warning information; Based on the tracking window, time localization and space localization based on risk are performed to determine the second warning information, wherein the localization features include time-space risk nodes and trend change rates, Based on the first warning information and the second warning information, a harmful gas risk warning is issued.

6. A monitoring instrument, characterized in that: The monitoring instrument implements the monitoring configuration of a harmful gas monitoring instrument configuration optimization method according to any one of claims 1 to 5, the monitoring instrument establishes a communication connection with the cloud, and the monitoring instrument includes: A front-end monitoring device, wherein the front-end monitoring device comprises a camera device, a sensor device and a fan, and the front-end monitoring device is distributed to form the front-end configuration network, and the front-end configuration network is adjustable; A control cabinet, which is the edge control center of the monitoring instrument and can perform automated decision control and human-computer interaction control; Wherein, the front-end monitoring device control is integrated into the control cabinet.

Images