Intelligent remote monitoring harmful gas leakage early warning system

Abstract

The invention relates to the technical field of gas monitoring, and discloses an intelligent remote monitoring harmful gas leakage early warning system which is composed of a sensing layer, an edge computing layer and a cloud computing layer, and harmful gas concentration data are collected in real time and preliminarily processed through a sensor module and a data collection module. The edge calculation layer is responsible for further data processing, early warning judgment and low-power-consumption wireless communication, and real-time analysis and early warning triggering of data are achieved. And the cloud computing layer is responsible for data storage, intelligent analysis and user interface display, identifying potential leakage sources and high-risk areas through a clustering algorithm, and providing an intuitive monitoring interface. According to the distributed technology architecture, the overall performance and reliability of the system are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of gas monitoring, and in particular to an intelligent remote monitoring and early warning system for harmful gas leakage. Background Art

[0002] In many fields such as industrial production, chemical storage, and environmental monitoring, the leakage of harmful gases may cause serious safety accidents and environmental pollution.

[0003] At present, the hazardous gas leak warning system mainly relies on the traditional centralized data processing mode, that is, after collecting field data through sensors, it is directly transmitted to a remote data center for processing and analysis. Since the data needs to be transmitted remotely to the data center, factors such as network bandwidth, transmission distance, and server processing capacity will lead to a decrease in the real-time nature of data processing and analysis, which in turn affects the accuracy and timeliness of the warning. In emergency situations such as hazardous gas leaks, this delay may lead to irreversible consequences. The centralized data processing model is difficult to cope with the concurrent data transmission of large-scale sensors, and is prone to data congestion and processing bottlenecks, affecting the stability and reliability of the overall system.

[0004] Moreover, most existing systems can only provide basic data monitoring and alarm functions, lack in-depth data analysis and mining capabilities, and cannot effectively identify potential leakage sources and high-risk areas. Therefore, we propose an intelligent remote monitoring hazardous gas leakage early warning system. Summary of the invention

[0005] The purpose of the present invention is to provide an intelligent remote monitoring and early warning system for harmful gas leakage to solve the problems raised in the above background technology.

[0006] To achieve the above-mentioned purpose, the present invention provides the following technical solutions: an intelligent remote monitoring and early warning system for harmful gas leakage, comprising a perception layer, an edge computing layer and a cloud computing layer;

[0007] The perception layer includes a sensor module and a data acquisition module;

[0008] The sensor module is used to collect the concentration of harmful gases in a specific area in real time; the data acquisition module is connected to the gas sensor and is used to convert the analog signal collected by the sensor into a digital signal and perform preliminary data filtering processing;

[0009] The edge computing layer includes a data processing module, an early warning judgment module and a communication module;

[0010] The data processing module receives the digital signal from the data acquisition module and performs further data processing, including data fusion and compression;

[0011] The warning judgment module performs real-time analysis on the fused data based on preset thresholds and algorithms to determine whether the concentration of harmful gases exceeds the standard, and triggers a warning if it does;

[0012] The communication module is used to send the compressed data to the cloud computing layer through low-power wireless communication technology;

[0013] The cloud computing layer includes a data storage module, an intelligent analysis module and a user interface module;

[0014] The data storage module receives data from the edge computing layer and performs persistent storage;

[0015] The intelligent analysis module uses the K-means clustering algorithm to perform cluster analysis on the harmful gas concentration data collected by real-time monitoring to identify potential leakage sources and high-risk areas;

[0016] The user interface module provides an intuitive monitoring interface, displaying real-time data of harmful gas concentration, historical trend charts and warning information;

[0017] Using a distributed technology architecture, the data processing module, early warning judgment module and communication module are deployed on the edge computing node near the perception layer, and are connected to the sensor module and data acquisition module of the perception layer wirelessly.

[0018] Preferably, the sensor module comprises a high voltage gas sensor array, and the high voltage gas sensor array is composed of a plurality of high voltage gas sensors for simultaneously monitoring a plurality of gas components and their concentrations.

[0019] Preferably, the data acquisition module includes a signal conditioning circuit and an analog-to-digital converter; the signal conditioning circuit is used to amplify, filter and calibrate the analog electrical signal transmitted by the high-voltage gas sensor to eliminate interference; the analog-to-digital converter converts the analog electrical signal processed by the signal conditioning circuit into a digital signal.

[0020] Preferably, the data processing module uses a weighted average method to fuse and calculate the data collected by multiple sensors to obtain accurate gas concentration data. The data of each high-voltage gas sensor is assigned different weights according to accuracy and reliability, and all weighted data are added together to obtain the final fusion result; the calculation method is as follows: Among them, agg_result is the final fusion result, n is the number of high voltage gas sensors, is the weight of the i-th high voltage gas sensor, is the measurement value of the i-th high voltage gas sensor.

[0023] Preferably, the data processing module uses a discrete cosine transform algorithm DCT to compress the data.

[0024] Preferably, the method in which the intelligent analysis module performs cluster analysis using a K-means clustering algorithm is:

[0025] Receive harmful gas concentration data from the edge computing layer, including timestamp, gas type and concentration value;

[0026] The steps to apply the K-means clustering algorithm include:

[0027] Step 1: Initialization step: randomly select K data points as the initial cluster centers, where K is the preset number of clusters;

[0028] Step 2: Assignment step: For each data point, calculate its Euclidean distance to each cluster center and assign it to the nearest cluster center. The calculation formula of the Euclidean distance is: Among them, x and y are two data points respectively, and n is the dimension of the data point;

[0029] Step 3, Update step: For each cluster, calculate the mean of all its points and use the mean as the new cluster center. The calculation formula of the mean is: Where N is the number of data points in the cluster, x i is the data point in the cluster;

[0030] Step 4, iterative step: repeat the assignment step and the update step until the cluster center no longer changes or the preset number of iterations is reached;

[0031] Finally, based on the clustering results, potential leakage sources and high-risk areas are identified and output.

[0032] Preferably, the intelligent analysis module analyzes each cluster, specifically: for each cluster, calculating the average and standard deviation of the harmful gas concentration of its internal data points; according to a preset concentration threshold, identifying the cluster whose average concentration exceeds the threshold as a potential leakage source; for the cluster identified as the leakage source, further analyzing the data points it contains to determine the specific location of the leakage source and the high-risk area.

[0033] Preferably, ZooKeeper is used as a distributed coordination service tool to achieve coordination and service registration and discovery between edge computing nodes.

[0034] Preferably, Kafka is used as a message queue tool to implement asynchronous message transmission between edge computing nodes and the cloud computing layer.

[0035] Preferably, Kubernetes is used as a containerized deployment tool to achieve lightweight deployment and rapid expansion of edge computing nodes.

[0036] Compared with the prior art, the present invention has the following beneficial effects:

[0037] The present invention realizes local processing and analysis of data by deploying data processing modules and early warning judgment modules at the edge computing layer, significantly reduces the delay of remote data transmission, and improves the real-time performance of data processing and early warning. This enables the system to respond quickly and issue early warnings in emergency situations such as harmful gas leaks, effectively avoiding or reducing safety accidents and environmental pollution.

[0038] The application of distributed technology architecture enables the system to better cope with the concurrent data transmission of large-scale sensors, effectively avoiding data congestion and processing bottlenecks. The deployment of edge computing nodes reduces the burden on the cloud computing layer, improves the stability and reliability of the overall system, and ensures the continuous availability of early warning services.

[0039] The intelligent analysis module uses the K-means clustering algorithm to conduct in-depth analysis and mining of harmful gas concentration data collected through real-time monitoring, and can accurately identify potential leakage sources and high-risk areas. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] Figure 1 It is the overall structure diagram of the present invention;

[0041] Figure 2 This is a diagram of the internal workflow of the data acquisition module;

[0042] Figure 3 This is a working step diagram of the intelligent analysis module. DETAILED DESCRIPTION

[0043] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0044] See also Figure 1-3 ,The present invention provides a technical solution: an intelligent remote monitoring and early warning system for harmful gas leakage, including a perception layer, an edge computing layer and a cloud computing layer.

[0045] The perception layer includes sensor modules and data acquisition modules.

[0046] The sensor module is used to collect the concentration of harmful gases in a specific area in real time. The sensor has high sensitivity and high precision and can accurately measure the concentration of harmful gases in the environment.

[0047] The data acquisition module is connected to the gas sensor and is responsible for converting the analog signal collected by the sensor into a digital signal and performing preliminary data filtering to improve the accuracy and reliability of the data.

[0048] The edge computing layer includes a data processing module and an early warning judgment module.

[0049] The data processing module receives the digital signal from the data acquisition module and performs further data processing, including data fusion and compression. Data fusion technology is used to integrate data from multiple sensors to improve data integrity and accuracy; data compression technology is used to reduce the amount of data transmission and reduce communication costs.

[0050] The early warning module analyzes the fused data in real time based on the preset threshold and algorithm to determine whether the concentration of harmful gases exceeds the standard. If it does, an early warning signal will be triggered immediately so that timely countermeasures can be taken.

[0051] The communication module is responsible for sending the compressed data to the cloud computing layer through low-power wireless communication technology.

[0052] The cloud computing layer includes data storage module, intelligent analysis module and user interface module.

[0053] The data storage module receives data from the edge computing layer and performs persistent storage. This module uses distributed storage technology to ensure data security and reliability.

[0054] The intelligent analysis module uses the K-means clustering algorithm to perform cluster analysis on the harmful gas concentration data collected through real-time monitoring, identify potential leakage sources and high-risk areas, and discover hidden patterns and associations in the data through data mining functions.

[0055] The user interface module provides an intuitive monitoring interface that displays real-time data of harmful gas concentrations, historical trend charts, and warning information. Users can easily view and analyze data and receive warning notifications through this interface.

[0056] System architecture and deployment:

[0057] Using a distributed technology architecture, the data processing module, early warning judgment module and communication module are deployed on edge computing nodes near the perception layer, thereby reducing data transmission delays and improving the real-time performance of data processing and early warning.

[0058] The edge computing nodes are wirelessly connected to the sensor modules and data acquisition modules of the perception layer to form a complete distributed monitoring network. The network structure is highly scalable and flexible and can adapt to monitoring needs of different scales and complexities.

[0059] The present invention will be further described below in conjunction with Examples 1 to 3:

[0060] Embodiment 1:

[0061] The sensor module includes a high-voltage gas sensor array, which is composed of multiple independent high-voltage gas sensors, each of which has high sensitivity and selectivity for specific harmful gases. The sensors are connected by circuits to ensure that they can work simultaneously and transmit the collected data to the data acquisition module. Suppose that three harmful gases, carbon dioxide, methane and hydrogen sulfide, need to be monitored simultaneously. Select high-voltage gas sensors that are highly sensitive and selective to these three gases, respectively, and assemble them into a sensor array. The sensor array is deployed in the monitoring area and connected to the data acquisition module through circuits. When the concentration of harmful gases in the monitoring area changes, the sensor array will collect these data in real time and transmit them to the data acquisition module for further processing.

[0062] The data acquisition module includes a signal conditioning circuit and an analog-to-digital converter.

[0063] The main function of the signal conditioning circuit is to pre-process the analog electrical signal transmitted by the high-voltage gas sensor, including amplification, filtering and calibration, to eliminate interference and improve the accuracy and stability of the signal. Since the signal output by the sensor is usually weak, it needs to be amplified by an amplifier for subsequent processing. The amplifier uses an operational amplifier with low noise, high precision and stability.

[0064] In order to eliminate the noise and interference in the signal, a filter is needed for filtering. The filter includes low-pass filter, high-pass filter and band-pass filter. Due to the non-ideal characteristics of the sensor and circuit, the output signal may have deviations. Therefore, calibration is required to eliminate the deviation and make the output signal accurately proportional to the actual concentration of harmful gases.

[0065] Assume that the output signal range of the high-voltage gas sensor used is 0-5V, corresponding to the harmful gas concentration of 0-100ppm. The amplifier in the signal conditioning circuit uses an operational amplifier with a gain of 10 to amplify the sensor output signal to the range of 0-50V. Then, a low-pass filter is used to filter out high-frequency noise. Finally, the output signal is adjusted to a digital signal that is accurately proportional to the actual harmful gas concentration through a calibration circuit.

[0066] The main function of an analog-to-digital converter (ADC) is to convert analog electrical signals processed by signal conditioning circuits into digital signals for subsequent digital processing and analysis. The ADC samples analog signals at a certain sampling rate, converts continuously changing analog signals into discrete analog signals, and further converts the sampled discrete analog signals into a finite number of discrete values, i.e., quantization. The quantization process is to divide the amplitude range of the analog signal into a finite number of intervals and map the analog signal in each interval to a fixed digital value. The quantized discrete values ​​are converted into binary codes for processing and storage by digital systems.

[0067] Assume that the analog signal range after the signal conditioning circuit is 0-50V, and the corresponding digital signal range is 0-1023. The ADC uses a 10-bit resolution quantization method to divide the analog signal into 1024 intervals. When the analog signal is 25V, the corresponding digital signal after quantization is 512. Then, this digital signal is converted into binary code for storage and processing.

[0068] Embodiment 2: The data processing module uses the weighted average method to fuse and calculate the data collected by multiple sensors to obtain accurate gas concentration data. The data of each high-voltage gas sensor is assigned different weights according to accuracy and reliability. All weighted data are added together to obtain the final fusion result. The calculation method is as follows: Among them, agg_result is the final fusion result, n is the number of high voltage gas sensors, is the weight of the i-th high voltage gas sensor, is the measured value of the i-th high voltage gas sensor. Assume that there are three high voltage gas sensors used to monitor carbon monoxide concentration, and their measured values ​​are 400ppm, 410ppm and 405ppm respectively. According to the accuracy and reliability of the sensors, the weights assigned to them are 0.5, 0.3 and 0.2 respectively.

[0072] Calculated using the weighted average method:

[0073] agg_result=(0.5×400ppm)+(0.3×410ppm)+(0.2×405ppm)=404ppm

[0074] The fused CO2 concentration is therefore 404 ppm. This result is more accurate and reliable than any single sensor measurement because it eliminates the errors and uncertainties of the individual sensors.

[0075] The data processing module uses discrete cosine transform (DCT) algorithm to compress data. DCT is a transformation method related to Fourier transform, which is used to convert signals or images from the spatial domain to the frequency domain. In the frequency domain, the energy of the signal or image is usually concentrated in the low-frequency part, and data compression is achieved by discarding the high-frequency part. The specific steps are:

[0076] The gas concentration data collected by a group of sensors are preprocessed to obtain an 8x8 data matrix, and the data matrix is ​​transformed using the DCT algorithm to obtain an 8x8 frequency domain matrix.

[0077] The elements in the frequency domain matrix are quantized and each element is rounded to the nearest integer. Then, the quantized matrix is ​​Huffman encoded.

[0078] Store the encoded data to hard disk or transfer it to a remote server.

[0079] When necessary, decompress the compressed data. Decode the encoded data to obtain the quantized frequency domain matrix. Perform the inverse DCT to restore the data to the original spatial domain matrix.

[0080] The method for the intelligent analysis module to perform cluster analysis using the K-means clustering algorithm is as follows:

[0081] Receive harmful gas concentration data from the edge computing layer, including timestamp, gas type and concentration value;

[0082] The steps to apply the K-means clustering algorithm include:

[0083] Step 1: Initialization step: randomly select K data points as the initial cluster centers, where K is the preset number of clusters;

[0084] Step 2: Assignment step: For each data point, calculate its Euclidean distance to each cluster center and assign it to the nearest cluster center. The calculation formula of the Euclidean distance is: Among them, x and y are two data points respectively, and n is the dimension of the data point;

[0085] Step 3, Update step: For each cluster, calculate the mean of all its points and use the mean as the new cluster center. The calculation formula of the mean is: Where N is the number of data points in the cluster, x i is the data point in the cluster;

[0086] Step 4, iterative step: repeat the assignment step and the update step until the cluster center no longer changes or the preset number of iterations is reached;

[0087] Finally, based on the clustering results, potential leakage sources and high-risk areas are identified and output.

[0088] The intelligent analysis module conducts in-depth analysis on each cluster. The specific steps are as follows:

[0089] Calculate statistical indicators: For each cluster, calculate the average harmful gas concentration of all data points within it. This step aims to obtain the central trend of the gas concentration in each cluster. At the same time, calculate the standard deviation of the harmful gas concentration of the data points within each cluster. The standard deviation is used to measure the degree of dispersion of the data points, that is, the fluctuation of the concentration value.

[0090] Identify potential leak sources: Based on the preset concentration threshold, determine whether the average concentration of each cluster exceeds the threshold, which is used to distinguish normal concentration from potential dangerous concentration. If the average concentration of a cluster exceeds the preset threshold, the cluster is identified as a potential leak source. This means that the area represented by the cluster may have a hazardous gas leak.

[0091] Identify leak source locations and high-risk areas: For clusters identified as leak sources, further analysis of all the data points they contain aims to determine the specific location of the leak source through detailed analysis. By analyzing the spatiotemporal distribution characteristics of data points, such as concentration values, timestamps, and geographic location information, the intelligent analysis module is able to locate the approximate location of the leak source. Based on the concentration values ​​and distribution of data points, high-risk areas are identified. High-risk areas are those areas where the concentration of hazardous gases is high and there may be continuous leakage or accumulation.

[0092] Output results: The intelligent analysis module integrates the location of potential leak sources, gas types, concentration averages, standard deviations, and information on high-risk areas.

[0093] Here is a specific example of the calculation:

[0094] Receive harmful gas concentration data from the edge computing layer, where the data includes timestamp, gas type, concentration value and other information, for example: “2023-04-01 12:00:00, methane, 50ppm”.

[0095] Initialization step: preset the number of clusters K, for example K=3.

[0096] Randomly select K data points from the received data as the initial clustering centers, for example, select "2023-04-01 12:00:00, methane, 50ppm", "2023-04-01 12:15:00, carbon dioxide, 80ppm", and "2023-04-01 12:30:00, methane, 70ppm" as the initial clustering centers.

[0097] Assignment step: For each data point, calculate its Euclidean distance from each cluster center. For example, calculate the distance between the data point "2023-04-01 12:45:00, methane, 60ppm" and the three cluster centers.

[0098] Assign the data point to the nearest cluster center. Assuming that "2023-04-01 12:45:00,methane,60ppm" is closest to "2023-04-01 12:00:00,methane,50ppm", it will be assigned to that cluster center.

[0099] Update step: For each cluster, calculate the mean of all its points and use the mean as the new cluster center. For example, for the first cluster, calculate the mean concentration of all data points belonging to the cluster.

[0100] Iteration step: Repeat the assignment step and the update step until the cluster center no longer changes or the preset number of iterations is reached. For example, the cluster center is stable after 10 iterations.

[0101] Based on the clustering results, potential leakage sources and high-risk areas are identified. For example, if the concentration of data points near the first cluster center is generally high, the area is judged to be a high-risk area and a possible leakage source. The output includes information such as the location of the leakage source, gas type, concentration range, etc. For example: "High-risk area: Warehouse A, main gas: methane, concentration range: 50ppm-70ppm".

[0102] Embodiment 3:

[0103] The implementation steps of using ZooKeeper as a distributed coordination service tool are as follows:

[0104] Install the ZooKeeper service on the edge computing node server and ensure that all edge computing nodes can access the ZooKeeper service. Configure ZooKeeper service parameters, including cluster settings, data storage path, client connection port, etc.

[0105] When each edge computing node is started, it will register its service information with ZooKeeper, including node ID, IP address, port number, and service type provided. When other edge computing nodes or cloud computing layers need to call a service, they can query the service registration information through ZooKeeper to obtain the address and port of the edge computing node that provides the service.

[0106] The distributed lock and notification mechanism of ZooKeeper is used to achieve coordination between edge computing nodes. For example, when multiple nodes need to access shared resources at the same time, they can obtain distributed locks through ZooKeeper to avoid resource conflicts.

[0107] When the status of an edge computing node changes, such as crashing or restarting, ZooKeeper can promptly notify other nodes so that they can handle the situation accordingly.

[0108] The implementation steps of using Kafka as a message queue tool are as follows:

[0109] Install the Kafka service on the cloud computing layer or edge computing nodes, and configure Kafka's cluster parameters, topic settings, producer and consumer configurations, etc. Ensure that the edge computing nodes can send messages to Kafka as producers, and the cloud computing layer can receive messages from Kafka as consumers.

[0110] The edge computing node encapsulates the collected data or processing results into messages and sends them to the specified topic of Kafka. The cloud computing layer subscribes to the corresponding topic in Kafka and receives messages from the edge computing node from the topic.

[0111] The cloud computing layer can further process, analyze or store the received messages.

[0112] The implementation steps of using Kubernetes as a containerized deployment tool are:

[0113] Install the Kubernetes cluster on the edge computing node server and configure the cluster's network, storage, security and other parameters. Prepare the application code for the edge computing node and package it into a container image.

[0114] Use Kubernetes's Deployment or StatefulSet controllers to deploy containerized applications on edge computing nodes. Use Kubernetes's Autoscaling feature to automatically adjust the number of application replicas based on the load of edge computing nodes to achieve lightweight deployment and rapid expansion. Monitor resource usage of edge computing nodes, such as CPU, memory, and storage, to schedule and optimize resources when needed.

[0115] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0116] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. Intelligent remote monitoring and early warning system for harmful gas leakage, characterized by: Includes perception layer, edge computing layer and cloud computing layer; The perception layer includes a sensor module and a data acquisition module; The sensor module is used to collect the concentration of harmful gases in a specific area in real time; the data acquisition module is connected to the gas sensor and is used to convert the analog signal collected by the sensor into a digital signal and perform preliminary data filtering processing; The edge computing layer includes a data processing module, an early warning judgment module and a communication module; The data processing module receives the digital signal from the data acquisition module and performs further data processing, including data fusion and compression; The warning judgment module performs real-time analysis on the fused data based on preset thresholds and algorithms to determine whether the concentration of harmful gases exceeds the standard, and triggers a warning if it does; The communication module is used to send the compressed data to the cloud computing layer through low-power wireless communication technology; The cloud computing layer includes a data storage module, an intelligent analysis module and a user interface module; The data storage module receives data from the edge computing layer and performs persistent storage; The intelligent analysis module uses the K-means clustering algorithm to perform cluster analysis on the harmful gas concentration data collected by real-time monitoring to identify potential leakage sources and high-risk areas; The user interface module provides an intuitive monitoring interface, displaying real-time data of harmful gas concentration, historical trend charts and warning information; Using a distributed technology architecture, the data processing module, early warning judgment module and communication module are deployed on the edge computing node near the perception layer, and are connected to the sensor module and data acquisition module of the perception layer wirelessly.

2. The intelligent remote monitoring hazardous gas leakage early warning system according to claim 1 is characterized by: The sensor module comprises a high voltage gas sensor array, which is composed of a plurality of high voltage gas sensors and is used for simultaneously monitoring a plurality of gas components and their concentrations.

3. The intelligent remote monitoring and early warning system for harmful gas leakage according to claim 2 is characterized in that: The data acquisition module includes a signal conditioning circuit and an analog-to-digital converter; the signal conditioning circuit is used to amplify, filter and calibrate the analog electrical signal transmitted by the high-voltage gas sensor to eliminate interference; the analog-to-digital converter converts the analog electrical signal processed by the signal conditioning circuit into a digital signal.

4. The intelligent remote monitoring and early warning system for harmful gas leakage according to claim 3 is characterized in that: The data processing module uses the weighted average method to fuse and calculate the data collected by multiple sensors to obtain accurate gas concentration data. The data of each high-voltage gas sensor is assigned different weights according to accuracy and reliability. All weighted data are added together to obtain the final fusion result. The calculation method is as follows: Among them, agg_result is the final fusion result, n is the number of high voltage gas sensors, is the weight of the i-th high voltage gas sensor, is the measurement value of the i-th high voltage gas sensor.

5. The intelligent remote monitoring and early warning system for harmful gas leakage according to claim 4 is characterized in that: The data processing module uses discrete cosine transform algorithm DCT to compress data.

6. The intelligent remote monitoring and early warning system for harmful gas leakage according to claim 1 is characterized in that: The method for the intelligent analysis module to perform cluster analysis using the K-means clustering algorithm is as follows: Receive harmful gas concentration data from the edge computing layer, including timestamp, gas type and concentration value; The steps to apply the K-means clustering algorithm include: Step 1: Initialization step: randomly select K data points as the initial cluster centers, where K is the preset number of clusters; Step 2: Assignment step: For each data point, calculate its Euclidean distance to each cluster center and assign it to the nearest cluster center. The calculation formula of the Euclidean distance is: Among them, x and y are two data points respectively, and n is the dimension of the data point; Step 3, Update step: For each cluster, calculate the mean of all its points and use the mean as the new cluster center. The calculation formula of the mean is: Where N is the number of data points in the cluster, x i is the data point in the cluster; Step 4, iterative step: repeat the assignment step and the update step until the cluster center no longer changes or the preset number of iterations is reached; Finally, based on the clustering results, potential leakage sources and high-risk areas are identified and output.

7. The intelligent remote monitoring and early warning system for harmful gas leakage according to claim 1 is characterized in that: The intelligent analysis module analyzes each cluster, specifically: for each cluster, the average and standard deviation of the harmful gas concentration of its internal data points are calculated; based on a preset concentration threshold, the cluster whose average concentration exceeds the threshold is identified as a potential leakage source; for the cluster identified as a leakage source, the data points it contains are further analyzed to determine the specific location of the leakage source and the high-risk area.

8. The intelligent remote monitoring hazardous gas leakage early warning system according to claim 1 is characterized by: ZooKeeper is used as a distributed coordination service tool to achieve coordination, service registration and discovery between edge computing nodes.

9. The intelligent remote monitoring and early warning system for harmful gas leakage according to claim 8 is characterized in that: Kafka is used as a message queue tool to realize asynchronous message transmission between edge computing nodes and the cloud computing layer.

10. The intelligent remote monitoring hazardous gas leakage early warning system according to claim 8 is characterized by: Kubernetes is used as a containerized deployment tool to achieve lightweight deployment and rapid expansion of edge computing nodes.

Images