Gas detection system and device

Abstract

The invention provides a gas detection system and device, and relates to the technical field of gas detection, and the system comprises a configuration module which is used for carrying out the type selection configuration of a monitoring sensor according to the monitored gas; the reference construction module is used for constructing reference grid density based on the minimum coverage area of the monitoring sensor; the monitoring node establishment module is used for performing region importance calculation on the to-be-monitored region and configuring monitoring nodes; the collaborative analysis module is used for acquiring and establishing a monitoring sensor array and establishing an associated collaborative coefficient; and the detection module is used for calling a decision fusion network to carry out abnormal decision making on the monitoring data based on the association cooperation coefficient, and generating a gas detection result. By means of the gas detection method and device, the technical problem that in the prior art, due to the fact that selection and configuration of the sensors are not adjusted according to actual requirements, the gas detection precision is low can be solved, the sensors are reasonably configured according to the gas type and the region importance, and the gas detection precision is improved.

Description

Technical Field

[0001] This application relates to the technical field of gas detection, and particularly to a gas detection system and device. Background Art

[0002] Gas detection technology is widely used in industrial production, environmental monitoring, public safety and other fields, aiming to monitor and detect harmful gases in the air in real time to ensure personnel safety, environmental health and the normal operation of equipment. However, in complex scenarios, the selection and configuration of gas detection devices may not be flexible enough, resulting in unsatisfactory detection effects. The sensitivities and selectivities of different sensors to specific gases vary significantly. Selecting an inappropriate sensor may lead to a decrease in the accuracy of monitoring results, especially in cases where the gas types are complex, the concentration changes are drastic, or the regional environmental conditions vary greatly. Many sensors only work at a fixed precision, and existing gas detection often does not flexibly select and configure sensors according to actual application requirements (such as the types of gases to be monitored, the characteristics of the monitoring area, etc.).

[0003] In summary, there is a technical problem in the prior art that the accuracy of gas detection is relatively low due to the lack of adjustment of sensor selection and configuration according to actual needs. Summary of the Invention

[0004] The purpose of this application is to provide a gas detection system and device to solve the technical problem in the prior art that the accuracy of gas detection is relatively low due to the lack of adjustment of sensor selection and configuration according to actual needs.

[0005] In view of the above problems, this application provides a gas detection system and device.

[0006] In a first aspect, this application provides a gas detection system, wherein the gas detection system includes: a configuration module for selecting and configuring a monitoring sensor according to the monitored gas after determining the monitored gas, and the monitoring sensor includes a first-precision sensor and a second-precision sensor; a reference construction module for obtaining the minimum coverage range of the monitoring sensor and constructing a reference grid density based on the minimum coverage range; a monitoring node establishment module for calculating the importance degree of the area to be monitored and configuring monitoring nodes according to the calculation result of the area importance degree and the reference grid density; a collaborative analysis module for obtaining the precision difference between the first-precision sensor and the second-precision sensor, distributing the monitoring sensors according to the precision difference and the node importance degree of the monitoring nodes, establishing a monitoring sensor array, and establishing an associated collaboration coefficient of the monitoring sensors; a detection module for reading the monitoring data of the monitoring sensor array, calling a decision fusion network to perform abnormal decision on the monitoring data based on the associated collaboration coefficient, and generating a gas detection result.

[0007] Optionally, a fusion recognition unit is configured to call the decision fusion network to perform fusion decision analysis: a: After synchronizing and aligning the monitoring data, perform spatial anomaly correlation analysis based on the association cooperation coefficient through the spatial correlation channel to establish a first recognition anomaly; b: Perform temporal concentration anomaly analysis on the same-position monitoring results of the monitoring data to establish a second recognition anomaly; c: Perform joint anomaly decision-making on the first recognition anomaly and the second recognition anomaly.

[0008] Optionally, a positioning unit is configured to establish an anomaly positioning point based on the joint anomaly decision; a clustering unit is configured to perform fine-tuning clustering of the monitoring sensors based on the anomaly positioning point to establish a fine-tuning clustering result; an adjustment unit is configured to fine-tune the positions of the monitoring sensors based on the fine-tuning clustering result and obtain an additional monitoring data set after adjustment; a compensation unit is configured to perform gas detection result compensation according to the additional monitoring data set to generate an updated gas detection result.

[0009] Optionally, an importance calculation unit is configured to calculate the regional importance through the following formula: ; where represents the regional importance, represents the position at time the importance index under, represents the number of high-risk devices per unit area centered on the position , represents the highest device density, represents the real-time leakage probability obtained through simulation, is the exponential influence coefficient of the device density, is the exponential influence coefficient, represents the air pressure gradient intensity, represents the time the wind speed under, represents the reference wind speed for normalizing the wind speed influence, represents the angle between the wind speed direction and the main diffusion direction of the region, is the cosine coupling coefficient of the wind direction - regional main axis angle.

[0010] Optionally, an importance segmentation unit is configured to determine an importance segmentation constraint based on the precision difference by using an importance constraint network; an importance screening unit is configured to perform the node importance screening based on the importance segmentation constraint to determine an initial distribution of the first precision sensors; a screening unit is configured to perform a distribution weakness analysis based on the initial distribution of the first precision sensors, and perform additional screening based on the distribution weakness analysis result and the node importance to establish an additional distribution of the first precision sensors; a sensor distribution unit is configured to complete the monitoring sensor distribution according to the additional distribution and the initial distribution.

[0011] Optionally, a gas release module is configured to move to a corresponding random position point to perform gas release after configuring a random position point, and generate a release timing record; a feedback module is configured to perform a deviation verification based on the gas detection result and the release timing record after receiving the gas detection result. The deviation verification includes a position deviation verification and a sensitivity deviation verification, establish an identification feedback according to the deviation verification result, and perform system optimization management according to the identification feedback.

[0012] Optionally, a collaborative calculation unit is configured to calculate an associated collaborative coefficient through a formula as follows: ; where represents the associated collaborative coefficient of the monitoring sensors and the monitoring sensors ; represents a precision difference factor, , is the percentage of measurement error of the monitoring sensor ; is the percentage of measurement error of the monitoring sensor ; represents the Euclidean distance between the monitoring sensors and the monitoring sensors ; and represent the node importance, is a spatial attenuation coefficient.

[0013] In a second aspect, the present application further provides a gas detection device, including: at least one processor; a memory communicatively connected to the at least one processor; wherein, the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to implement any one of the gas detection systems in the first aspect above.

[0014] Optionally, an early warning module, which is configured to, after receiving the gas detection result, match the early warning level according to the gas detection result, configure an early warning plan according to the early warning level matching result and the gas detection result, and give an early warning according to the early warning plan.

[0015] Optionally, a self-monitoring module, which is configured to, after reading the monitoring data, perform co-location monitoring deviation verification on the first-precision sensor and the second-precision sensor according to the monitoring data, generate a co-location deviation verification result, and report a monitoring anomaly according to the co-location deviation verification result.

[0016] One or more technical solutions provided in this application have at least the following technical effects or advantages: Through a configuration module, which is configured to, after determining the monitored gas, select and configure monitoring sensors according to the monitored gas, where the monitoring sensors include a first-precision sensor and a second-precision sensor; a reference construction module, which is configured to obtain the minimum coverage range of the monitoring sensors and construct a reference grid density based on the minimum coverage range; a monitoring node establishment module, which is configured to calculate the importance of the area to be monitored, and configure monitoring nodes according to the area importance calculation result and the reference grid density; a collaborative analysis module, which is configured to obtain the precision difference between the first-precision sensor and the second-precision sensor, and distribute the monitoring sensors according to the precision difference and the node importance of the monitoring nodes, establish a monitoring sensor array, and establish an associated collaboration coefficient of the monitoring sensors; a detection module, which is configured to read the monitoring data of the monitoring sensor array, call a decision fusion network to perform anomaly decision on the monitoring data based on the associated collaboration coefficient, and generate a gas detection result. That is to say, by dynamically selecting sensors with different precisions according to the type of monitored gas, constructing a reference grid density based on the minimum coverage range of the monitoring sensors, configuring a sensor array in combination with the importance of the area to be monitored, and establishing an associated collaboration coefficient to perform gas detection, such as detecting gas components such as volatile organic compounds through various organic matter measuring instruments, potential gas anomaly problems can be effectively identified, and the accuracy of gas detection is improved.

[0017] The above description is only an overview of the technical solutions of this application. In order to be able to understand the technical means of this application more clearly, it can be implemented according to the content of the specification. And in order to make the above and other purposes, features and advantages of this application more obvious and understandable, the following specifically gives the specific implementation manners of this application. It should be understood that the content described in this part is not intended to identify the key or important features of the embodiments of this application, nor is it used to limit the scope of this application. Other features of this application will become easily understood through the following description. Description of the Drawings

[0018] To more clearly illustrate the technical solutions in the present application or the prior art, the following will briefly introduce the accompanying drawings required for the description of the embodiments or the prior art. Obviously, the accompanying drawings in the following description are only exemplary. For those of ordinary skill in the art, without creative efforts, other accompanying drawings can also be obtained based on the provided drawings.

[0019] Figure 1 This is a schematic structural diagram of a gas detection system according to the present application.

[0020] Figure 2 This is a schematic structural diagram of a gas detection device according to the present application.

[0021] Explanation of the reference numerals: Configuration module 11, reference construction module 12, monitoring node establishment module 13, collaborative analysis module 14, detection module 15. Detailed implementation manners

[0022] By providing a gas detection system and device, the present application solves the technical problem in the prior art that due to the lack of adjustment of the selection and configuration of sensors according to actual needs, the accuracy of gas detection is relatively low. By dynamically selecting sensors with different precisions according to the monitored gas type, constructing a reference grid density according to the minimum coverage range of the monitored sensors, configuring a sensor array in combination with the importance of the area to be monitored, and establishing an associated collaboration coefficient for gas detection, such as detecting gas components such as volatile organic compounds through various organic matter measuring instruments, potential gas anomaly problems can be effectively identified, and the accuracy of gas detection is improved.

[0023] Next, the technical solutions in the present application will be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments of the present application. It should be understood that the present application is not limited by the exemplary embodiments described here. Based on the embodiments of the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present application. Additionally, it should be noted that for the sake of description, only the parts related to the present application are shown in the accompanying drawings rather than all of them.

[0024] Embodiment 1. Please refer to the attached Figure 1 , the present application provides a gas detection system, wherein, the gas detection system includes: Configuration module 11, used for selecting and configuring monitoring sensors according to the monitored gas after determining the monitored gas, and the monitoring sensors include first-precision sensors and second-precision sensors.

[0025] Specifically, first determine the monitored gases, such as carbon monoxide, sulfur dioxide, nitrogen oxides, etc. Different gases have different molecular structures and characteristics, so there are different requirements for their monitoring. For example, for flammable gases, sensors using the infrared absorption method can be selected because gas molecules have unique absorption characteristics in the infrared spectrum. Therefore, by measuring the absorption degree of the gas to infrared light, the gas concentration can be accurately determined; for harmful gases, electrochemical sensors can be selected because electrochemical sensors have high sensitivity at low concentrations. Then, according to the characteristics of the selected gas and the environment, consult relevant technical specifications and sensor performance data sheets to determine the appropriate sensor type. That is, input the gas parameters of the monitored gas to automatically match the combination of the first-precision sensor and the second-precision sensor. Usually, one sensor has a higher precision and the other has a lower precision. Combining high-precision sensors and lower-precision but more economical sensors in the same system to achieve complementary advantages.

[0026] Exemplarily, input the gas parameters of the monitored gas to automatically match the first precision (TDLAS laser sensor) and the second precision (MEMS sensor). The TDLAS laser sensor is based on laser absorption spectroscopy technology and can accurately measure the concentration of gases, especially suitable for high-precision detection of low-concentration gases. By adjusting the laser wavelength to match the absorption line of the target gas, the light intensity attenuation is measured to calculate the gas concentration. The MEMS sensor (microelectromechanical system sensor) is a common low-cost and lower-precision gas sensor, widely used in monitoring conventional gases and monitoring based on the change of metal oxide resistance. The high and low precisions of the two complement each other, covering the full range, and can verify false alarms with each other. TDLAS is used as the gold standard to ensure the credibility of key data; MEMS expands the monitoring dimension and improves the system robustness.

[0027] After determining the sensors, configure the appropriate number and type of the first-precision sensors and the second-precision sensors according to the size of the monitoring area and the monitoring requirements. By selecting and configuring appropriate sensors, it is ensured to efficiently and accurately monitor specific gases, which helps to improve the accuracy and reliability of the monitoring data.

[0028] The reference building module 12 is used to obtain the minimum coverage range of the monitoring sensors and build a reference grid density based on the minimum coverage range.

[0029] Specifically, after determining the monitoring sensors, according to the working characteristics and monitoring capabilities of each monitoring sensor, the minimum coverage range of the monitoring sensor is obtained, that is, the minimum area range that the sensor can effectively monitor, which is usually affected by factors such as the sensitivity of the sensor, working conditions, detection range, and environmental factors. For example, some gas sensors have high sensitivity within a certain concentration range, but as the monitoring range expands, their detection ability may gradually weaken. Therefore, the minimum coverage range refers to the minimum area or minimum concentration range that the sensor can accurately and effectively detect. Usually, by reading the technical specification of the sensor, understanding the detection range and minimum detection concentration of the sensor, the effective detection radius or distance of the sensor can be determined, which is usually the minimum area where the sensor can reliably detect the gas. At the same time, according to the characteristics of the detection environment, such as wind speed, temperature, humidity, obstacles, etc., these factors may all affect the coverage range of the sensor, and jointly determine the minimum coverage range of the sensor.

[0030] Based on the minimum sensor coverage radius, the reference grid density is set according to the size of the monitoring area. The grid density determines the deployment spacing of the sensors, that is, how many sensors are needed per unit area to ensure the coverage of the entire area. By calculating the minimum coverage range of the monitoring sensor, a reasonable grid density can be determined, so as to ensure that within the entire monitoring area, the sensors can effectively cover and provide accurate monitoring data. After determining the grid density, the monitoring area can be divided into multiple grid cells, and corresponding sensors are configured in each grid cell. When configuring the sensors, factors such as the change trend of gas concentration in the area and the importance of the monitoring target are appropriately considered. For example, the configuration of sensors is encrypted near the gas leakage source. A reference grid is constructed in the monitoring area according to the calculated grid density, and the layout of the grid can be square, hexagonal or other shapes, depending on the geometric characteristics of the monitoring area and the layout of the sensors. Sensors are arranged at the center or appropriate positions of each grid cell to ensure that the entire monitoring area is covered by the sensors without monitoring blind spots.

[0031] Exemplarily, according to the minimum sensor coverage radius r, the side length of the grid is set to 2r. The purpose is to ensure that there are no blind spots between the sensors in the grids, while maintaining a reasonable spatial distribution and avoiding the deployment of redundant sensors. The Thiessen polygon algorithm is used to divide the area. Within each Thiessen polygon, the sensor is responsible for monitoring the gas concentration in the area. In the Thiessen polygon grid, the sensors are distributed at the center positions of each grid according to the previously set grid density, ensuring that each sensor can cover its responsible area and the entire monitoring area can be effectively monitored. The Thiessen polygon grid is a graph that divides the area according to certain rules. Within the area of each polygon, the sensor closest to the boundary points of the polygon is the central sensor of the polygon. Simply put, each sensor creates a responsible area - that is, the area where the sensor can most effectively monitor, and this area is defined by the Thiessen polygon boundary.

[0032] By setting a reasonable grid density, it can be ensured that there are no blind spots in the monitoring area and the gas concentration in all areas can be effectively monitored. For example, the minimum effective monitoring radius of the sensor is 20 meters, and based on this, the side length of the grid is determined to be 40 meters, that is, the monitoring area of each sensor is a square grid with a side length of 40 meters. A 120 m × 120 m area will be divided into 9 square grids with a side length of 40 meters, and a sensor is placed at the center of each grid. Each sensor is responsible for monitoring the area closest to it, forming a Voronoi diagram (Thiessen polygon grid) to ensure that the monitoring area has no overlap and no blind spots.

[0033] The monitoring node establishment module 13 is used to calculate the importance degree of the area to be monitored and configure the monitoring nodes according to the calculation result of the area importance degree and the reference grid density.

[0034] Furthermore, the monitoring node establishment module 13 in the gas detection system is further used for: Calculating the area importance degree through the formula as follows: ; where represents the area importance degree, indicating the importance index of the position at time , represents the number of high-risk devices per unit area centered on the position , represents the highest device density, represents the real-time leakage probability obtained through simulation, is the exponential influence coefficient of the device density, is the exponential influence coefficient, represents the air pressure gradient intensity, represents time wind speed below characterizes the reference wind speed for normalizing the wind speed impact characterizes the angle between the wind speed direction and the main diffusion direction of the region is the cosine coupling coefficient of the wind direction - regional main axis angle

[0035] Specifically, in gas monitoring and risk assessment, the regional importance is used to quantify the monitoring priority of a specific region. Areas with high regional importance require more monitoring resources because there may be higher risks. The formula for calculating the regional importance is as follows: ; where characterizes the regional importance, indicating the location at time the importance index below characterizes the number of high - risk devices per unit area centered on the location is a factor measuring the density of hazard sources in the region and reflects the contribution of the devices to the gas leakage risk. characterizes the highest device density, which is a constant representing the upper limit of the device density for normalization to ensure fair comparison of device densities between different regions. The larger the ratio of, the denser the devices in the region, and thus the higher its importance. α is the exponential impact coefficient of the device density, used to adjust the contribution degree of the device density to the regional importance. Different application scenarios may require different values of α to assign appropriate weights to the device density according to the actual risk.

[0036] characterizes the real - time leakage probability obtained through (Computational Fluid Dynamics) simulation, reflecting the likelihood of gas leakage at the location at time t. CFD is a numerical simulation technique that can be used to predict the flow of fluids. In practical applications, the diffusion of gas leakage is complex and affected by environmental factors (such as air pressure, temperature, wind speed, etc.). CFD simulation can calculate the leakage probability of a certain region based on these factors. CFD simulation usually predicts the process of gas diffusion through numerical methods (such as the finite element method). , 1 means that if there is no leakage risk, the contribution of the region is 1, The larger, the higher the leakage risk in the region, thereby increasing the risk value of the region.

[0037] is the exponential impact coefficient, used to adjust the impact of the air pressure gradient on the regional importance. By adjusting the value, the sensitivity of the air pressure gradient to the regional importance can be controlled. For example, when = 1, the impact of the air pressure gradient is linearly related to it, while Greater than 1, the influence of the pressure gradient will be amplified. Characterize the pressure gradient intensity, i.e., the position The pressure gradient intensity at time t. The pressure gradient refers to the degree of pressure change in space and is usually an important factor in gas leakage and diffusion. The stronger the pressure gradient, the faster the gas diffusion rate may be and the wider the leakage range may be. The exponential form of the calculation makes the influence of the pressure gradient on the regional importance non-linear, which means that the greater the pressure gradient, the stronger the effect of enhancing the regional importance.

[0038] Characterize time The wind speed at time t. The wind speed is an important factor in gas diffusion. The greater the wind speed, the faster the gas diffusion rate. Characterize the reference wind speed, which is used to normalize the influence of the wind speed so that the influences under different wind speed conditions can be compared. Characterize the angle between the wind speed direction and the main diffusion direction of the region, which reflects the actual influence of the wind speed direction on gas diffusion. If the wind speed direction is consistent with the gas diffusion direction, the gas will be quickly diffused and the risk of the region increases; if the directions are opposite, the gas diffusion rate will slow down and the risk of the region is lower. Is the cosine coupling coefficient of the wind direction - regional main axis angle, which is used to adjust the influence between the wind direction and the main diffusion direction of the region. When the angle between the wind speed direction and the diffusion direction is small, the gas diffusion rate is fast and the risk of the region increases; when the angle is large, the diffusion rate slows down and the risk decreases.

[0039] Exemplarily, assume that gas detection is carried out in a factory area. The equipment density in this area is 10 high-risk devices per square meter, the maximum equipment density is 15 per square meter, the exponential influence coefficient α = 2, the exponential influence coefficient β = 1.5, and the cosine coupling coefficient γ of the wind direction - regional main axis angle is 0.9. The real-time leakage probability obtained through CFD simulation is 0.4, the pressure gradient intensity is 0.1 Pa / m, the current wind speed is 5 m / s, the reference wind speed is 10 m / s, and the angle between the wind speed and the diffusion direction is 30°. Substituting these values into the above formula, the calculated importance value of this region Is approximately equal to 1.01, indicating that under the given parameter conditions, the monitoring priority of this region is relatively high and it belongs to the region that requires more attention.

[0040] Calculate the regional importance of the area to be detected using a formula. Based on the regional importance and the reference grid density, determine how many monitoring nodes should be deployed in the area. The reference grid density determines the minimum number of sensors required in each grid to ensure area coverage. For example, in areas of high importance, more sensors will be deployed to ensure a higher monitoring density; in areas of low importance, the number of sensors may be reduced. If the number of high-risk devices in an area is dense, and the wind speed in this area is low while the pressure gradient is large, the importance value of this area will be relatively high, so more sensors need to be deployed and monitored. Assume that the regional importance of a certain area is higher than that of other areas, then more sensors will be added in this area, and monitoring nodes will be preferentially deployed in places with higher regional importance. By calculating the regional importance and configuring sensors based on the result, it is ensured that high-risk areas receive more monitoring resources, improving the monitoring efficiency and detection accuracy.

[0041] The collaborative analysis module 14 is used to obtain the accuracy difference between the first-precision sensor and the second-precision sensor, perform monitoring sensor distribution according to the accuracy difference and the node importance of the monitoring nodes, establish a monitoring sensor array, and establish the associated collaboration coefficient of the monitoring sensors.

[0042] Furthermore, the collaborative analysis module 14 in the gas detection system is also used for: The importance segmentation unit is used to determine the importance segmentation constraint based on the accuracy difference using the importance constraint network; the importance screening unit is used to screen the node importance based on the importance segmentation constraint to determine the initial distribution of the first-precision sensors; the screening unit is used to perform distribution weakness analysis according to the initial distribution of the first-precision sensors, and perform additional screening based on the distribution weakness analysis result and the node importance to establish the additional distribution of the first-precision sensors; the sensor distribution unit is used to complete the monitoring sensor distribution according to the additional distribution and the initial distribution.

[0043] Specifically, calculate the accuracy difference between the first-precision sensor and the second-precision sensor according to their accuracies, which is obtained by comparing the test results of the two sensors under the same environmental conditions. For example, at the same gas concentration, the error of the first-precision sensor may be ±0.5%, while the error of the second-precision sensor is ±5%, then the accuracy difference is 5% - 0.5% = 4.5%. The accuracy difference refers to the accuracy difference in the output results of the first-precision sensor and the second-precision sensor in actual detection.

[0044] According to the specific requirements and objectives of the monitoring area, an importance constraint network is established. In the importance constraint network, different monitoring areas will be divided into multiple importance levels according to their importance (such as equipment density, leakage risk, wind speed, air pressure difference, etc.). The importance constraint network is a mathematical model used to analyze and set the relationship between regional importance and sensor distribution. In this network, the monitoring requirements of the area will affect the sensor configuration, including the distribution of first-precision sensors and second-precision sensors. The importance segmentation constraint is used to limit the allocation of sensor types and quantities according to different importance values of the area. High-importance areas will be constrained to deploy more high-precision sensors, while low-importance areas may only deploy low-precision sensors.

[0045] Each monitoring node (i.e., the location where the sensor is located) has a certain node importance, usually calculated through an importance formula, including the number of high-risk devices, gas leakage probability, air pressure, wind speed, etc. Each monitoring node (i.e., each grid or location within the monitoring area) is screened according to the importance value of the node. After screening by node importance, it is determined whether to deploy first-precision sensors according to the importance of each node, so as to determine the initial distribution of first-precision sensors. That is, in the preliminary configuration process, according to the screening results of node importance, high-precision sensors (such as TDLAS laser sensors) are deployed into the preliminary layout of the monitoring area. For example, through precision difference analysis and node importance screening, it is determined to deploy 25 first-precision sensors in high-risk areas, 10 first-precision sensors in medium-risk areas, and 1 first-precision sensor in low-risk areas.

[0046] Based on the initial distribution of first-precision sensors, a distribution weakness analysis is carried out on the sensor layout to identify the deficiencies in the sensor distribution, that is, those areas with less coverage or insufficient precision. Through a comprehensive analysis of the sensor distribution within the monitoring area, it is determined which areas have insufficient monitoring precision or insufficient monitoring coverage, and further optimization and supplementation are carried out. Data analysis methods, such as spatial analysis, nearest neighbor analysis, etc., are used to evaluate the coverage of sensors in each area. Compare the response precision of sensors in different areas, and combine the regional importance to identify high-risk areas and areas with lower monitoring precision. For example, assume that in a monitoring area with an area of 1000 meters × 1000 meters, 70 first-precision sensors are initially deployed. Area A (a high-risk area for gas leakage) has only 2 sensors and they are far apart, unable to effectively monitor the gas concentration in this area; in Area B (an area with concentrated equipment), due to large wind speed changes, there are only 3 sensors and the monitoring effect is poor.

[0047] Based on the results of the vulnerability analysis and combined with the node importance, the areas where sensors need to be added are screened. When arranging sensors in vulnerable areas, not only the insufficient coverage is considered, but also the number and accuracy of the added sensors are determined according to the importance of the nodes. According to the importance of the area, it is decided whether more sensors need to be arranged; according to the accuracy requirements of the vulnerable area, it is decided whether to arrange first-precision sensors; for areas with higher risks, first-precision sensors are preferentially selected for addition. For example, Area A is a high-risk area with a very high node importance, so more first-precision sensors need to be added in this area. On the other hand, Area B may be in a medium-risk area with a lower node importance, so only some second-precision sensors may need to be added.

[0048] After screening the areas where sensors need to be added, the next step is to perform additional distribution, that is, in vulnerable areas or places with insufficient accuracy, more first-precision sensors are arranged. After additional distribution, the final sensor layout will include the initial layout and the additional distribution, forming an efficient and accurate monitoring array. Additional distribution refers to the new sensor layout based on the results of the distribution vulnerability analysis on the basis of the original sensor layout. Usually, it is to supplement the vulnerable areas under the existing sensor configuration to ensure the comprehensiveness and accuracy of the monitoring system. By allocating sensors with different accuracies to different areas, resources are maximally utilized, the overall efficiency of the monitoring system is improved, unnecessary investment in high-precision sensors is reduced, and at the same time, accurate monitoring of the monitored area is ensured.

[0049] Through the initial distribution and the additional distribution, the arrangement of the monitoring sensors for the entire monitored area is completed. The monitoring sensor array refers to a sensor network formed by reasonably distributing all sensors within the monitored area, with a wide coverage area and efficient monitoring capabilities. According to conditions such as node importance and area risk, first-precision sensors are initially arranged to form the initial distribution. The placement location of each sensor depends on the risk level and monitoring accuracy requirements of the selected node. Through vulnerability analysis and node importance screening, the monitoring areas that need to be further strengthened are found, and more sensors are added in these areas, especially first-precision sensors, second-precision sensors, or the arrangement of sensors is adjusted, so that the deficiencies in the original layout are supplemented to form the additional distribution.

[0050] After completing the preliminary sensor distribution, it is also necessary to consider whether the position of each sensor can maximize its monitoring effectiveness. During this process, the position of the sensor can be finely adjusted. Each sensor can be finely adjusted at the center of the monitoring node where it is located, moving a certain distance to optimize the coverage or improve the monitoring accuracy. The direction and amplitude of the fine adjustment can be adjusted based on factors such as the specific requirements of the area, the mutual interference between sensors, and the air flow direction. Specifically, by analyzing the real-time monitoring data of the sensors, identify those sensors with low monitoring accuracy or unstable data, which are usually located in areas with large interference or are greatly affected by environmental conditions (such as wind speed, air flow direction, etc.). On the premise of maintaining the original monitoring range, slightly adjust the positions of these sensors, usually moving a certain degree around the center of the node where they are located. For example, the sensor can be finely adjusted along the direction of the air flow or the blank area of the coverage area to ensure more comprehensive and accurate coverage.

[0051] After the initial distribution, additional distribution, and fine adjustment, a complete monitoring sensor array is finally formed. At this time, each node in the array is equipped with a suitable sensor and is reasonably configured according to the requirements of the monitoring area. The monitoring sensor array not only includes high-precision first-precision sensors but also second-precision sensors to ensure that the needs of different risk areas are met. The layout of the array is continuously optimized through continuous monitoring and adaptively adjusted as the environment changes or the requirements change to ensure that each high-risk area is adequately monitored with high precision, while the monitoring of other low-risk areas is also fully guaranteed. By deploying the sensor array and continuously optimizing the arrangement of the sensors, ensure that the monitoring accuracy and coverage rate reach the best state. The sensor array will be arranged according to the regional importance and risk level to ensure that high-risk areas are covered by more sensors.

[0052] Furthermore, the collaborative analysis module 14 in the gas detection system is also used for: Calculating the correlation and collaboration coefficient through the formula as follows: ; where represents the correlation and collaboration coefficient of the monitoring sensors and the monitoring sensor , represents the precision difference factor, , is the percentage of the measurement error of the monitoring sensor , is the percentage of the measurement error of the monitoring sensor , represents the Euclidean distance between the monitoring sensors and the monitoring sensor , and Characterize the node importance is the spatial attenuation coefficient

[0053] Specifically, the correlation and collaboration coefficient is used to measure the degree of collaboration between two monitoring sensors, reflecting the collaboration relationship between the monitoring sensor and the monitoring sensor taking into account factors such as their spatial distance, accuracy difference, and node importance. A high correlation and collaboration coefficient means that the two sensors can collaborate highly when monitoring the same target, and a lower coefficient indicates a poor collaboration effect between them. By calculating the accuracy difference factor between the monitoring sensor and the monitoring sensor that is, by calculating the percentage of measurement error for each sensor and then calculating the difference in the percentage of measurement error between the two sensors

[0054] Calculating the Euclidean distance between the monitoring sensor and the monitoring sensor to quantify the relative position of the two sensors in space. For example, assuming the coordinates of the monitoring sensor are (100, 200) and the coordinates of the monitoring sensor are (200, 300), then (100 - 200)² + (200 - 300)² = 20000, and taking the square root of 20000 gives the Euclidean distance of 141.42 meters. The node importance and represent the importance of the nodes where the monitoring sensor and the monitoring sensor are located, reflecting the monitoring priority of the nodes where the sensors are located. Usually, through the analysis of the monitoring area, it is determined that certain areas have higher requirements for safety or monitoring, so the node importance in this area is also higher. The higher the importance of the node, the more urgent and important the monitoring task of the sensor

[0055] is the spatial attenuation coefficient, which determines the attenuation rate. As the distance between the two sensors increases, the collaboration coefficient will decrease. Therefore, the spatial attenuation coefficient controls this attenuation rate indicates that as the accuracy difference factor increases, the collaboration coefficient will decrease, meaning that the greater the accuracy difference between the two sensors, the worse their collaboration effect indicates that as the distance between the two sensors increases, the collaboration coefficient will decrease, and the spatial attenuation coefficient λ determines the attenuation rate It represents the average value of node importance and the overall importance of the node where the sensor is located. By calculating and adjusting the correlation synergy coefficient of the monitoring sensor, the cooperation ability between the two sensors can be effectively evaluated and optimized accordingly.

[0056] The detection module 15 is used to read the monitoring data of the monitoring sensor array, call the decision fusion network to make an abnormal decision on the monitoring data based on the correlation synergy coefficient, and generate a gas detection result.

[0057] Specifically, the monitoring sensor array refers to a network array composed of multiple sensors arranged in the monitoring area. Each sensor is responsible for collecting gas data in a specific area. These sensors work together to achieve real-time monitoring of a large area. The decision fusion network is called to perform fusion decision analysis. The decision fusion network is a system that integrates multiple data sources and multiple monitoring point information and performs comprehensive analysis. It generates the final decision output through multi-dimensional analysis of monitoring data. The data collected by different sensors are integrated to identify abnormal situations and provide more accurate gas detection results.

[0058] Read the detection data of the monitoring sensor array and synchronize the monitoring data. Since each monitoring sensor may have time delay or different sampling intervals, it is necessary to align the monitoring data of all sensors to ensure that their timestamps are consistent. After completing data synchronization, enter the spatial correlation channel analysis, and by calculating the correlation synergy coefficient between sensors, identify which areas have abnormal data relationships between monitoring sensors. Perform time series concentration anomaly analysis on the monitoring results at the same location to identify gas concentration fluctuations at different time points at the same location (or monitoring node). Under normal circumstances, gas concentration should have a certain regular change. If the concentration fluctuation at a certain time point is abnormally drastic, or the concentration changes at several consecutive time points do not meet expectations, there may be gas leakage or other abnormal conditions. After spatial anomaly correlation analysis and time series concentration anomaly analysis, the first identification anomaly and the second identification anomaly are generated respectively. At this time, the joint anomaly decision method is used to merge the anomaly information of the two, make the final anomaly judgment, and generate gas detection results. By integrating spatial anomaly and time series concentration anomaly analysis, gas leakage or other abnormal conditions can be comprehensively and accurately identified, thereby improving the reliability of detection.

[0059] Furthermore, the detection module 15 in the gas detection system further includes: The fusion recognition unit is used to call the decision fusion network to perform fusion decision analysis: a: After synchronizing and aligning the monitoring data, perform spatial anomaly correlation analysis based on the correlation coefficient through the spatial correlation channel to establish the first recognition anomaly; b: Perform temporal concentration anomaly analysis on the same-location monitoring results of the monitoring data to establish the second recognition anomaly; c: Perform joint anomaly decision on the first recognition anomaly and the second recognition anomaly.

[0060] Specifically, call the decision fusion network to perform fusion decision analysis on the monitoring data, identify the anomalies of the monitored gas, determine whether there is an actual gas leakage or abnormal situation, and generate gas detection results. Different sensors may record data at different timestamps, so data synchronization and alignment are required to adjust the monitoring data of all sensors to the same time basis so that they can be compared and analyzed at the same time point. Perform data synchronization and alignment on the detection data to unify the data to the same time scale. For missing data, linear interpolation or spline interpolation can be used for data synchronization.

[0061] Spline interpolation is an interpolation method used to generate a smooth curve or function through known data points, especially suitable for cases where the differences between data points are large or the distribution of data points is uneven. The common spline interpolation method is cubic spline interpolation, which approximates the changes between data points through multiple cubic polynomial interpolation functions and ensures the continuity of the interpolation curve and the continuity of the first and second derivatives at the nodes, effectively eliminating the abrupt changes between data.

[0062] Exemplarily, assume that the gas concentration data collected by two sensors are not at the same time point. In order to compare and analyze this data, it is necessary to align their timestamps so that the data of all sensors have corresponding values at the same moment. The data of sensor i are time: [0s, 5s, 10s, 15s, 20s], concentration: [5ppm, 8ppm, 12ppm, 18ppm, 20ppm]; the data of sensor j are time: [2s, 6s, 8s, 14s, 19s], concentration: [4ppm, 9ppm, 10ppm, 16ppm, 22ppm]. For each pair of adjacent time points, find a cubic polynomial that can connect the two data points, and use the concentration values of each data point and their timestamps to solve the coefficients of these polynomials. Repeat this process continuously until the interpolation curve is constructed for all data points. Cubic polynomials are common knowledge and will not be elaborated here. Through spline interpolation, the concentration data of sensor j can be interpolated to the timestamps aligned with the data of sensor i to generate new data for comparison. The data of sensor i are time: [0s, 5s, 10s, 15s, 20s], and after spline interpolation, the time of sensor j is the same as that of sensor i, concentration: [5.5ppm, 8.3ppm, 12.2ppm, 17.8ppm, 19.6ppm]. After spline interpolation, the time series of sensor i and sensor j can now be compared on a unified time axis. Through spline interpolation, the difference in data acquisition time can be eliminated, ensuring that the data of multiple sensors can be effectively compared at the same time point.

[0063] After completing the data synchronization and alignment, enter the spatial correlation channel analysis. According to the spatial positions of the sensors, calculate the correlation and cooperation coefficient to determine which sensors have spatial anomalies. Calculate the correlation and cooperation coefficient between the sensors, analyze the spatial correlation between the sensors, and identify the abnormal correlations between the sensors. If the correlation and cooperation coefficient is lower than a certain threshold and the data shows abnormal fluctuations, it indicates that the monitoring results between these two sensors are inconsistent, and there may be regional anomalies or faults, which are recorded as the first identified anomalies.

[0064] Compare all the calculated correlation and cooperation coefficients. If the correlation and cooperation coefficient between a pair of sensors is lower than a certain threshold and their concentration data fluctuates greatly, it is considered that gas leakage or other abnormalities may have occurred in this area. Based on the spatial correlation analysis, mark the areas with a high probability of abnormality. The sensors in these areas show significant correlation differences, which may be caused by gas leakage, equipment failure, etc. According to the results of the spatial correlation analysis, establish the first identified abnormality, indicating that this area requires further investigation and monitoring. Spatial anomaly correlation analysis uses the spatial relationship between sensors (such as position, cooperation coefficient, etc.) to judge whether there is an abnormality in a certain area, and can identify potential gas leakage points, pollution sources, etc. based on the similarity or difference of sensor data.

[0065] Select the monitoring data at the same location in the monitoring data and analyze the time series of the concentration. That is, collect the concentration data of multiple sensors at the same location at the same time point. These sensors monitor the same gas, so there should be a certain correlation in their concentration data, and there is a high consistency in the abnormal changes at the same location. Time-series concentration anomaly analysis is to analyze the monitoring data of multiple sensors at the same location in the time dimension to identify abnormal fluctuations in gas concentration during a certain period. Usually, the gas concentration will have a certain fluctuation range under normal circumstances, but if the concentration fluctuation exceeds the expected range, it may indicate some abnormal situation (such as leakage, equipment failure, etc.). For each time point, the concentration data of these sensors should be similar, assuming that their concentration data fluctuates within a certain range. If the concentration data of a certain sensor suddenly deviates from the data of other sensors, it may indicate that an abnormality has occurred in this area at this moment.

[0066] Analyze the concentration data at each time point and calculate the volatility between them. By calculating the concentration average value, standard deviation, and comparing with a preset normal concentration fluctuation range, judge whether there is an abnormality here. The normal concentration fluctuation range is set according to historical experience, etc. When the standard deviation or concentration change exceeds the preset threshold, it can be considered that a concentration anomaly has occurred at this time point, and it is recorded as the second identified abnormality.

[0067] Make a comprehensive judgment by combining the first recognition anomaly (based on spatial correlation analysis) and the second recognition anomaly (based on temporal concentration anomaly analysis). If the spatial anomaly and the temporal anomaly results are consistent, that is, both anomaly analyses point to the same region or time period, it is judged as a reliable anomaly event. Determine the final decision plan by combining the severity of each anomaly (such as anomaly amplitude, duration, number of sensors involved, etc.). If the anomaly results of the two are inconsistent, determine the final result through weighting, priority ranking or other decision rules. For example, assume that at a certain monitoring node, the temporal analysis shows that the concentration of this node suddenly increases, while the spatial analysis shows that the coordination coefficient among multiple sensors in this area is low and their data fluctuates greatly. Since both point to the same anomaly location, the gas detection result in this area is an anomaly such as gas leakage.

[0068] By fusing spatial correlation analysis and temporal concentration analysis, comprehensively analyze the anomalies in gas monitoring data, process and make decisions on the data from each sensor in real time, and ensure that in the event of gas leakage or other safety hazards, a quick response can be made to avoid the occurrence of safety accidents.

[0069] Furthermore, the detection module 15 in the gas detection system further includes: A positioning unit for establishing an anomaly positioning point according to the joint anomaly decision; a clustering unit for performing fine-tuning clustering of monitoring sensors based on the anomaly positioning point to establish a fine-tuning clustering result; an adjustment unit for fine-tuning the positions of the monitoring sensors based on the fine-tuning clustering result and obtaining an additional monitored data set after adjustment; a compensation unit for compensating the gas detection result according to the additional monitored data set to generate an updated gas detection result.

[0070] Specifically, by combining multiple anomaly detection methods (such as spatial anomaly correlation and temporal concentration anomaly), a joint anomaly decision is obtained to accurately identify gas leakage or other dangerous situations, and reduce false alarms and missed alarms of single anomaly recognition. According to the joint anomaly decision result, determine the anomaly positioning point, that is, the area pointed to by the joint anomaly decision result. The anomaly positioning point refers to determining the specific location or range of a certain anomaly event in space or time, which helps to clarify the specific area where the anomaly occurs.

[0071] According to the abnormal positioning points, fine-tuning clustering is performed on the sensors in this area, that is, the positions of the sensors are optimized and adjusted according to the data volatility of each sensor, the distance from other sensors, and the distribution of the abnormal positioning points, and the sensors are moved or redeployed to optimize the monitoring coverage. For example, assuming a leak occurs in a certain area (such as near a storage tank), the fine-tuning clustering module will concentrate the surrounding sensors towards this area, thereby ensuring that the sensors can monitor the leak point more densely. The fine-tuning clustering result is the clustering information after adjusting the distribution of the sensors based on the correlation between the data of the sensors and the abnormal recognition result, reflecting the optimization of the sensor positions and configurations, ensuring that there are more sensors or more precise arrangements in key areas.

[0072] According to the fine-tuning clustering result, the positions of the monitoring sensors are fine-tuned to make them more reasonably distributed within the monitoring area. The fine-tuning of the sensor positions can be automatically executed based on an algorithm or completed through manual control of the device for correction. For example, after fine-tuning, new sensors may be arranged in areas with higher leakage risks, thereby optimizing the layout of the monitoring network. The gas in this area is monitored again through the adjusted and added sensor positions, and data is collected again to generate a new additional monitoring dataset, reflecting the gas concentration changes in the optimized monitoring area. For example, if the original sensor configuration was relatively sparse, resulting in insufficient monitoring of the gas concentration in some areas, after fine-tuning, the new sensor positions may cover more high-risk areas, thereby providing more accurate monitoring data.

[0073] The gas detection results are compensated and corrected through the additional monitoring dataset. The purpose of compensation is to ensure that the final gas detection results are more accurate. Especially in the event of anomalies such as leaks, the compensated results can provide more precise position and concentration information, thereby generating updated gas detection results. For example, assuming in a certain leak area, the original gas detection results show a relatively low concentration in this area, but due to the sparse arrangement of sensors, the center point of the leak was not fully detected. After fine-tuning, the new sensor dataset reflects a significant increase in the gas concentration in this area, and the compensated gas detection results will show a higher concentration, thereby giving an early warning of the leak event.

[0074] By fine-tuning the distribution of the sensors, high-risk areas can be covered more densely, improving the accuracy of gas detection, especially the monitoring ability in abnormal areas. Through fine-tuning clustering and position adjustment, a dynamic and flexible sensor layout can be achieved, enabling the monitoring system to automatically optimize according to the actual situation, making the gas detection results more accurate, thereby reducing the possibility of missed reports and false alarms.

[0075] Furthermore, the gas detection system further includes a gas release module and a feedback module. The gas release module is used to configure random position points and then move to the corresponding random position points to perform gas release, generating a release timing record. The feedback module is used to receive the gas detection result and perform deviation verification based on the gas detection result and the release timing record. The deviation verification includes position deviation verification and sensitivity deviation verification. An identification feedback is established according to the deviation verification result, and system optimization management is performed according to the identification feedback.

[0076] Specifically, the random position points selected within the monitoring area can be places where gas leakage may occur in a simulated environment (such as inside a factory, a storage area, etc.). By moving to the corresponding random position points, gas release operations are performed, and at the same time, information such as the timing, concentration, and type of the release is recorded, generating a release timing record. The release timing record of the gas is a record of the specific time and location of the gas release, reflecting the dynamic process of gas leakage. For example, the release timing record may include: at time T1, 0.03 ppm of gas was released at location A.

[0077] After the gas is released, the change in the gas concentration in the monitoring area is detected, and the gas detection result is generated through the above-mentioned complete steps. The gas detection result is compared with the release timing record to perform deviation verification. Deviation verification is to check whether there is a deviation by comparing the gas detection result with the release timing record. The deviation verification includes position deviation verification and sensitivity deviation verification. The position deviation verification is used to check the deviation between the gas detection result and the actual gas release position; the sensitivity deviation verification is used to evaluate the sensitivity of the monitoring sensor to see whether it can correctly reflect the change in gas concentration.

[0078] By comparing the positions of the actual release point and the sensor detection point, it is judged whether there is a deviation. For example, if the actual leakage point A is 100 meters away from the sensor B, the position deviation is recorded. In addition to the deviation in position, it is also compared whether the detected gas concentration matches the actual concentration. For example, if the gas release timing record shows that 10 cubic meters of gas has leaked, but the sensor B only detects a concentration of 5 ppm, it may be because the sensor sensitivity is insufficient, resulting in inaccurate detection results, and sensitivity deviation verification is required.

[0079] After completing the deviation verification, recognition feedback is generated to provide detailed feedback on the current gas detection capabilities, pointing out potential problems in the system, such as positioning errors and insufficient sensitivity. Based on the recognition feedback, system optimization management is carried out to adjust the position of the monitoring sensors or calibrate the sensitivity of the sensors to improve the accuracy of the system. The operation process of the gas release module is optimized to ensure the reliability and repeatability of the test. Through position deviation verification and sensitivity deviation verification, potential problems in the monitoring system are discovered, and the detection accuracy of gas leakage is improved through optimization and adjustment. The data provided by the recognition feedback can help optimize the position and layout of the sensors to ensure sufficient monitoring coverage in high-risk areas and improve the overall monitoring capabilities.

[0080] Furthermore, the gas detection system further includes an early warning module, which is configured to, after receiving the gas detection result, match the early warning level according to the gas detection result, configure an early warning plan according to the early warning level matching result and the gas detection result, and issue an early warning according to the early warning plan.

[0081] Specifically, after receiving the gas detection result, including information such as gas type, concentration value, location, and time, the real-time state of gas leakage in the target area is understood. According to the concentration data and leakage risk in the gas detection result, the level of the early warning is determined. For example, if the detected gas concentration is higher than a certain set threshold, different levels of early warnings will be triggered (such as low-level, medium-level, and high-level early warnings). For example, it is set that gas leakage with a concentration below 10 ppm triggers a low-level early warning; gas leakage with a concentration between 10 ppm and 50 ppm triggers a medium-level early warning; and gas leakage with a concentration exceeding 50 ppm triggers a high-level early warning.

[0082] Based on the detected gas concentration, the corresponding early warning level can be determined. According to the matched early warning level, a suitable early warning plan is automatically configured. Each early warning level corresponds to different emergency treatment plans. For example, a low-level early warning requires the monitoring personnel to increase close monitoring; a medium-level early warning requires starting equipment inspections and increasing the vigilance of personnel in the area; a high-level early warning requires more stringent emergency response measures such as starting emergency shutdown, evacuating personnel, and activating the fire protection system.

[0083] According to the configured early warning plan, an early warning is issued, and an early warning is sent to relevant personnel through audible and visual alarms (sending audible and visual signals through alarm devices) and digital notifications (transmitting information through methods such as text messages, emails, or system notifications). By real-time matching the gas detection result and the early warning level, the most appropriate emergency response plan is automatically generated to ensure that all levels of personnel can receive information in the shortest possible time and take corresponding measures in a timely manner, while reducing unnecessary early warnings and avoiding resource waste caused by false alarms or missed alarms.

[0084] Further, the gas detection system further includes a self-monitoring module. After reading the monitoring data, the self-detection module is configured to perform co-location monitoring deviation verification on the first-precision sensor and the second-precision sensor according to the monitoring data, generate a co-location deviation verification result, and report a monitoring anomaly according to the co-location deviation verification result.

[0085] Specifically, real-time monitoring data in the sensor array is read, including gas concentration, gas type, sensor location, time, etc. According to the monitoring data, the monitoring data of the first-precision sensor and the second-precision sensor at the same location is extracted, and the results of the two are compared. If the difference between the detection results of the two is large, a deviation verification will be generated, indicating that there may be a problem.

[0086] Verification is performed based on the deviation between the data of the two sensors to generate a co-location deviation verification result, including deviation value, occurrence time, location information, etc. If the deviation exceeds the preset threshold, it means that the results of the two sensors differ too much, indicating that there may be a problem with the sensors. Once the deviation verification result exceeds the threshold, a monitoring anomaly is reported, indicating equipment failure, environmental factor interference, or sensor accuracy problems, etc. And according to the monitoring anomaly, an anomaly report is generated, including sensor number, anomaly type (such as excessive deviation, sensor failure, etc.), anomaly level (such as warning, severe warning, etc.). According to the anomaly report, the operator or the system automatically executes corresponding emergency measures, such as sensor calibration, system inspection, or alarm notification.

[0087] Exemplarily, the data of two sensors at the same location and the same time are read. The first-precision sensor shows a concentration of 10 ppm, and the second-precision sensor shows a concentration of 75 ppm. The deviation between the two readings is calculated and found to be 65 ppm, which seriously exceeds the preset deviation threshold (such as ±5 ppm), generating a co-location deviation verification result and reporting a monitoring anomaly. After receiving the anomaly report, the operator checks and calibrates the sensor to ensure the accuracy and reliability of the monitoring system.

[0088] Through co-location deviation verification, when the sensor fails or is interfered by environmental factors, early warning and measures are taken, such as recalibrating the sensor or replacing the faulty equipment, to ensure the accuracy of the monitoring data and the stable operation of the system.

[0089] In summary, the gas detection system provided by the present application has the following technical effects: A configuration module is used to select and configure a monitoring sensor according to the monitored gas after determining the monitored gas. The monitoring sensor includes a first-precision sensor and a second-precision sensor. A reference construction module is used to obtain the minimum coverage range of the monitoring sensor and construct a reference grid density based on the minimum coverage range. A monitoring node establishment module is used to calculate the regional importance of the area to be monitored and configure monitoring nodes according to the calculation result of the regional importance and the reference grid density. A collaborative analysis module is used to obtain the precision difference between the first-precision sensor and the second-precision sensor, distribute the monitoring sensors according to the precision difference and the node importance of the monitoring nodes, establish a monitoring sensor array, and establish an associated collaborative coefficient for the monitoring sensors. A detection module is used to read the monitoring data of the monitoring sensor array, call a decision fusion network to perform an anomaly decision on the monitoring data based on the associated collaborative coefficient, and generate a gas detection result. That is to say, by dynamically selecting sensors with different precisions according to the type of monitored gas, constructing a reference grid density based on the minimum coverage range of the monitoring sensors, configuring a sensor array in combination with the importance of the area to be monitored, and establishing an associated collaborative coefficient, gas detection is performed. For example, gas components such as volatile organic compounds are detected by various organic matter measuring instruments, effectively identifying potential gas anomaly problems and improving the accuracy of gas detection.

[0090] Embodiment 2. Based on the same inventive concept as a gas detection system in the foregoing Embodiment 1, the present application further provides a gas detection device, including: at least one processor; a memory communicatively connected to the at least one processor; wherein, the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to implement any one of the foregoing gas detection systems in Embodiment 1.

[0091] Appendix Figure 2 This is a schematic structural diagram of an exemplary gas detection device of the present application. In Figure 2Among them, the bus architecture is represented by bus 300. Bus 300 may include any number of interconnected buses and bridges. Bus 300 connects various circuits of one or more processors represented by processor 302 and memory represented by memory 304 together. Bus 300 may also connect various other circuits such as peripheral devices, voltage regulators, and power management circuits, etc. These are well known in the art, so they will not be further described herein. Bus interface 305 provides an interface between bus 300 and receiver 301 and transmitter 303. Receiver 301 and transmitter 303 may be the same element, i.e., a transceiver, providing a unit for communicating with various other devices over a transmission medium. Processor 302 is responsible for managing bus 300 and general processing, while memory 304 may be used to store data used by processor 302 when performing operations.

[0092] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be apparent to those skilled in the art. The general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to these embodiments shown herein, but rather will be accorded the widest scope consistent with the principles and novel features disclosed herein.

[0093] Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the spirit and 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 equivalent technologies, the present application also intends to include these changes and modifications.

Claims

1. A gas detection system, characterized in that, The described gas detection system includes: A configuration module, which is used to select and configure a monitoring sensor according to the monitored gas after determining the monitored gas. The monitoring sensor includes a first-precision sensor and a second-precision sensor; A reference construction module, which is used to obtain the minimum coverage range of the monitoring sensor and construct a reference grid density based on the minimum coverage range; A monitoring node establishment module, which is used to calculate the importance degree of the area to be monitored and configure monitoring nodes according to the calculation result of the area importance degree and the reference grid density; A collaborative analysis module, which is used to obtain the precision difference between the first-precision sensor and the second-precision sensor, distribute the monitoring sensors according to the precision difference and the node importance degree of the monitoring node, establish a monitoring sensor array, and establish an associated collaboration coefficient of the monitoring sensor; A detection module, which is used to read the monitoring data of the monitoring sensor array, call a decision fusion network to perform an anomaly decision on the monitoring data based on the associated collaboration coefficient, and generate a gas detection result.

2. The gas detection system according to claim 1, characterized in that, The detection module includes: A fusion recognition unit, which is used to call the decision fusion network and perform fusion decision analysis: a: After synchronizing and aligning the monitoring data, perform spatial anomaly correlation analysis based on the associated collaboration coefficient through a spatial correlation channel to establish a first recognition anomaly; b: Perform temporal concentration anomaly analysis on the monitoring results at the same location of the monitoring data to establish a second recognition anomaly; c: Perform a joint anomaly decision on the first recognition anomaly and the second recognition anomaly.

3. A gas detection system according to claim 2, wherein The detection module further includes: A positioning unit, which is used to establish an anomaly positioning point according to the joint anomaly decision; A clustering unit, which is used to perform fine-tuning clustering of the monitoring sensor based on the anomaly positioning point to establish a fine-tuning clustering result; An adjustment unit, which is used to fine-tune the position of the monitoring sensor based on the fine-tuning clustering result and obtain an additional monitoring data set after adjustment; A compensation unit, which is used to compensate the gas detection result according to the additional monitoring data set to generate an updated gas detection result.

4. A gas detection system according to claim 1, characterized in that, The monitoring node establishment module further includes: An importance degree calculation unit, which is used to calculate the area importance degree through a formula as follows: ； Among them, Characterize the regional importance degree, indicating the position at a certain time The importance index, Characterize the number of high-risk devices per unit area centered on the position Characterize the highest device density, Characterize the Real-time leakage probability obtained through simulation, Is the exponential influence coefficient of device density, Is the exponential influence coefficient, Characterize the air pressure gradient intensity, Characterize the time The wind speed at Characterize the reference wind speed, used to normalize the wind speed influence, Characterize the angle between the wind speed direction and the main diffusion direction of the region, Is the cosine coupling coefficient of the wind direction - regional main axis angle.​ 5. A gas detection system according to claim 1, characterized in that, The collaborative analysis module further includes: An importance degree segmentation unit, which is used to determine an importance degree segmentation constraint based on the precision difference by using an importance degree constraint network; An importance degree screening unit, which is used to screen the node importance degree based on the importance degree segmentation constraint to determine the initial distribution of the first-precision sensor; A screening unit, which is used to perform distribution weakness analysis according to the initial distribution of the first-precision sensor, perform additional screening based on the distribution weakness analysis result and the node importance degree, and establish an additional distribution of the first-precision sensor; A sensor distribution unit, which is used to complete the distribution of the monitoring sensor according to the additional distribution and the initial distribution.

6. The gas detection system according to claim 1, wherein The described gas detection system further includes: A gas release module, which is used to configure a random position point and then move to the corresponding random position point to perform gas release and generate a release time sequence record; A feedback module, which is configured to, after receiving the gas detection result, perform deviation verification according to the gas detection result and the release timing record. The deviation verification includes position deviation verification and sensitivity deviation verification. An identification feedback is established according to the deviation verification result, and system optimization management is performed according to the identification feedback.

7. The gas detection system according to claim 1, characterized in that, The collaborative analysis module further includes: A collaborative calculation unit, which is configured to calculate the associated collaborative coefficient through a formula as follows: ； Among them, characterizes the correlation and cooperation coefficient of the monitoring sensor and the monitoring sensor ; characterizes the accuracy difference factor, , is the percentage of measurement error of the monitoring sensor ; monitor , characterizes the Euclidean distance between the monitoring sensor and the monitoring sensor ; and characterize the node importance, is the spatial attenuation coefficient.

8. The gas detection system according to claim 1, wherein, The gas detection system further includes: An early warning module, which is configured to, after receiving the gas detection result, perform early warning level matching according to the gas detection result, configure an early warning plan according to the early warning level matching result and the gas detection result, and issue an early warning according to the early warning plan.

9. The gas detection system according to claim 1, wherein, The gas detection system further includes: A self-monitoring module, which is configured to, after reading the monitoring data, perform co-location monitoring deviation verification of the first precision sensor and the second precision sensor according to the monitoring data, generate a co-location deviation verification result, and report a monitoring anomaly according to the co-location deviation verification result.

10. A gas detection device, characterized in that, It includes: At least one processor; A memory communicatively connected to the at least one processor; Wherein, the memory stores instructions executable by the at least one processor, and when the at least one processor executes the instructions, the gas detection system according to any one of claims 1 to 9 is implemented.

Images