An unmanned aerial vehicle-based limited space toxic gas gradient detection system and method

Abstract

The application discloses a limited space toxic gas gradient detection system and method based on a UAV, and the system comprises the following steps: a multi-sensor integrated module is installed at the bottom of a UAV platform, and the module comprises a hydrogen sulfide electrochemical sensor, an oxygen optical sensor, a positioning module and a laser radar; a controller is connected with the UAV platform, the multi-sensor integrated module and a communication module; and a data processing terminal is connected with the controller through wireless communication. The method comprises the following steps: a two-dimensional vertical plane model of a vertical detection plane is generated based on LiDAR data, a plurality of detection points arranged in a matrix form are divided from the two-dimensional vertical plane model, spatial coordinate information, hydrogen sulfide gas concentration data and oxygen concentration data of each detection point are collected, and hydrogen sulfide and oxygen concentration data of all horizontal layers in the vertical direction are calculated based on the detection data by using a Kriging interpolation algorithm. The system and method can automatically realize high-precision gradient detection of toxic gases in the vertical direction in a limited space.

Description

Technical Field

[0001] This invention belongs to the field of intelligent detection technology of toxic gases, specifically a toxic gas gradient detection system and method based on unmanned aerial vehicles in a confined space. Background Technology

[0002] In industries such as chemical engineering and municipal engineering, confined space operations are frequent. These spaces are often poorly ventilated and have complex environments, making them prone to accumulating toxic and harmful gases, especially hydrogen sulfide. Due to its large mass, hydrogen sulfide tends to accumulate at the bottom of confined spaces, forming a concentration gradient from top to bottom. In recent years, hydrogen sulfide poisoning accidents have occurred frequently, mainly because workers only conduct single-point sampling at the entrance or shallow areas before entering confined spaces, ignoring the gradient changes in hydrogen sulfide concentration within the space. This results in an inability to accurately assess the risks of deep, high-concentration areas. Existing gas detection devices are mostly handheld single-point measuring devices, lacking the ability to continuously detect vertically within the space. Furthermore, manually entering high-risk areas for sampling is not only inefficient but also poses significant safety risks. In addition, the complex distribution of obstacles within confined spaces makes traditional detection path planning difficult to adapt to dynamic obstacle avoidance requirements, leading to incomplete data collection. Moreover, current technologies cannot construct gas concentration gradient distribution models or delineate safe work zones based on real-time data, severely restricting the safety assurance capabilities of confined space operations. In recent years, although some UAV-based gas detection devices have emerged, these devices suffer from the following shortcomings: low sensor integration, making it impossible to detect multiple gases simultaneously; fixed detection step size, lack of adaptive sampling mechanisms, and a tendency to miss detections in regions of abrupt changes in gas concentration gradients; and inability to combine spatial structure and gas diffusion models to generate high-precision gas concentration distribution maps. Therefore, there is an urgent need to develop a system and method capable of efficiently, safely, and accurately detecting the gradient distribution of toxic gases within a confined space. Summary of the Invention

[0003] To address the problems existing in the prior art, this invention provides a confined space toxic gas gradient detection system and method based on unmanned aerial vehicles (UAVs). This system is simple in structure, low in manufacturing cost, highly automated, and has a high safety factor. It can automatically perform high-precision gradient detection of toxic gases in the vertical direction within a confined space, eliminating the need for personnel to personally enter high-risk areas to perform detection. It solves the technical problem that traditional single-point detection cannot identify deep dangerous areas and significantly improves the efficiency and safety of toxic gas detection in confined spaces. The method is simple to implement and highly automated, enabling high-precision gradient detection of toxic gases in the vertical direction within a confined space, and obtaining hydrogen sulfide and oxygen concentration data for all horizontal layers in the vertical direction.

[0004] To achieve the above objectives, the present invention provides a confined space toxic gas gradient detection system based on an unmanned aerial vehicle (UAV), comprising an UAV platform, a multi-sensor integration module, a communication module, a controller, and a data processing terminal;

[0005] The multi-sensor integrated module is installed at the bottom of the UAV platform and includes a hydrogen sulfide electrochemical sensor, an oxygen optical sensor, a positioning module, and a lidar.

[0006] The communication module is installed on the drone platform;

[0007] The controller is installed on the UAV platform and is connected to the UAV platform, the multi-sensor integration module and the communication module respectively.

[0008] The data processing terminal is installed in the control center and is connected to the controller via wireless communication.

[0009] Furthermore, in order to facilitate the collection of spatial coordinate information at different detection points, the multi-sensor integrated module also includes a positioning module.

[0010] Furthermore, to ensure detection accuracy, the hydrogen sulfide electrochemical sensor has a measurement range of 0-100ppm, a measurement accuracy of 0.01ppm, and a response time of ≤5s; the lidar is a LiDAR.

[0011] As a preferred embodiment, the controller is a PLC controller.

[0012] As a preferred embodiment, the data processing terminal is an industrial computer.

[0013] In this invention, an unmanned aerial vehicle (UAV) platform is used as the carrier of a multi-sensor integrated module. Simultaneously, the actions of the UAV platform are controlled by a controller, allowing for flexible adjustment of the detection position. This facilitates the controlled measurement of toxic gas concentration data at any detection point within a two-dimensional vertical plane, significantly improving the continuous detection capability in the vertical direction of space. By simultaneously integrating a hydrogen sulfide electrochemical sensor, an oxygen optical sensor, a positioning module, and a lidar into the multi-sensor integrated module, the concentrations of hydrogen sulfide and oxygen at different detection points can be detected synchronously. This allows for comprehensive perception of hydrogen sulfide and oxygen concentration data at different detection points throughout the entire detection range. Furthermore, it facilitates the construction of hydrogen sulfide and oxygen concentrations at all horizontal levels in the vertical direction using the detection data from each detection point, greatly improving the safety assurance capability during operations in confined spaces. On the other hand, lidar can be used to perform laser scanning detection within a confined space. This facilitates the construction of a two-dimensional vertical surface model based on the laser scanning data. Subsequently, it allows for the generation of hydrogen sulfide and oxygen concentration distribution curves at different heights within the confined space by combining hydrogen sulfide and oxygen concentration data from all horizontal layers in the vertical direction. Simultaneously, it facilitates the subsequent construction of a two-dimensional gas concentration distribution heat map within the confined space, thus more effectively guiding subsequent detection actions for personnel entering the detection environment and ensuring their personal safety. By connecting the data processing terminal to the controller via wireless communication, it is easy to receive detection data sent by the controller and perform efficient and accurate analysis and processing of the detection data using the data processing terminal.

[0014] The system is simple in structure, low in manufacturing cost, highly automated, and has a high safety factor. It can automatically perform high-precision gradient detection of toxic gases in the vertical direction in confined spaces without requiring operators to personally enter high-risk areas to perform detection operations. It solves the technical problem that traditional single-point detection cannot identify deep dangerous areas, significantly improves the efficiency and safety of toxic gas detection in confined spaces, and provides reliable safety protection for operators.

[0015] This invention also provides a method for detecting toxic gas gradients in a confined space based on unmanned aerial vehicles (UAVs), employing a UAV-based system for detecting toxic gas gradients in a confined space, comprising the following steps:

[0016] Step 1: First, select a vertical detection surface in the limited space to be detected, and control the UAV platform to hover at the center line of the top of the vertical detection surface. Then, use the lidar to emit laser pulses into the limited space and measure their reflection time to generate LiDAR data. Finally, send the LiDAR data to the controller.

[0017] Step 2: The controller generates a two-dimensional vertical plane model of the vertical detection surface based on LiDAR data. Then, the two-dimensional vertical plane model is divided into multiple detection planes from top to bottom with equal vertical step distances. On each detection plane, multiple detection points are divided from left to right with equal horizontal step distances, thus forming a matrix arrangement of multiple detection points on the two-dimensional vertical plane model.

[0018] Step 3: Control the drone platform to traverse each detection plane sequentially from top to bottom. At the same time, traverse each detection point sequentially from left to right on each detection plane, and make the drone platform hover at each detection point for a set detection time.

[0019] During the detection process at each detection point, the spatial coordinate information of the current detection point is collected by the positioning module and sent to the controller. The hydrogen sulfide electrochemical sensor collects the hydrogen sulfide gas concentration signal of the current detection point in real time and sends it to the controller. The oxygen optical sensor collects the oxygen concentration signal of the current detection point in real time and sends it to the controller. The hydrogen sulfide gas concentration data of the current detection point is obtained based on the hydrogen sulfide gas concentration signal, and the oxygen concentration data of the current detection point is obtained based on the oxygen concentration signal. The data is then recorded and stored.

[0020] After completing the detection work on two detection planes, the hydrogen sulfide gas concentration data and oxygen concentration data in the two adjacent detection planes are compared. When the change rate of hydrogen sulfide gas concentration data is >3ppm / 0.5m or the difference in oxygen concentration gradient is >5%VOL / 0.5m, the remaining undetected area is divided a second time, and the vertical step distance and the horizontal step distance are reduced to half of the original. Then, several detection points II are formed in a matrix arrangement on the remaining undetected area, and the detection work on several detection points II continues in the same way.

[0021] During the movement of the UAV platform, the LiDAR is used to collect obstacle signals on the flight path and send them to the controller. The controller generates an obstacle avoidance path based on the dynamic path planning algorithm, combined with the obstacle signals and the position of the target detection point one. Then, the controller controls the UAV platform to fly to the target detection point one according to the obstacle avoidance path.

[0022] After completing the detection work on the entire vertical detection surface, the controller sends the spatial coordinate information of each detection point, the hydrogen sulfide gas concentration data of each detection point, and the oxygen concentration data of each detection point to the data processing terminal located in the control center through the communication module.

[0023] Step 4: The data processing terminal summarizes the spatial coordinate information of each detection point, the hydrogen sulfide gas concentration data and oxygen concentration data of each detection point, and obtains the hydrogen sulfide gas concentration data and oxygen concentration data of different detection points on the two-dimensional vertical plane model. At the same time, based on the hydrogen sulfide gas concentration data and oxygen concentration data of different detection points, the Kriging interpolation algorithm is used to calculate the hydrogen sulfide gas concentration data and oxygen concentration data of all horizontal layers in the vertical direction of the limited space according to formula (1), and generates the hydrogen sulfide and oxygen concentration distribution curves in different height ranges within the limited space. At the same time, a gas concentration distribution heat map of the two-dimensional surface in the limited space is constructed. Then, combined with the hydrogen sulfide and oxygen concentration distribution curves and gas concentration distribution heat map in different height ranges, the red forbidden zone and green safe zone in the vertical direction of the limited space are delineated on the two-dimensional vertical plane model.

[0024]

[0025] In the formula, C(Z0) pre Let d(Z0) be the predicted concentration value at any position Z0 in the vertical direction. i Z0) is a known point Z i The perpendicular distance between the point and the point to be interpolated Z0, where Z is the perpendicular distance between the point and the point to be interpolated. i This refers to the detection point above the location of Z0; Given point Z i The absolute value of the concentration gradient at a given point; α is the gradient sensitivity coefficient, which is dynamically adjusted based on the hydrogen sulfide / oxygen concentration threshold.

[0026] Furthermore, in order to effectively delineate high-concentration hydrogen sulfide and low-oxygen restricted areas and safe working areas, the method for delineating the red restricted area and green safe area in the vertical direction of the confined space in step four is as follows:

[0027] Using 19.5% oxygen concentration and 7 ppm hydrogen sulfide concentration as critical values, red and green zones are marked vertically. When the oxygen concentration in any zone is below 19.5% or the hydrogen sulfide concentration is above 7 ppm, it is designated a red restricted area and displayed as a red zone on a 2D vertical model, with its coordinates marked. When the oxygen concentration in any zone is 19.5% or higher or the hydrogen sulfide concentration is 7 ppm or lower, it is designated a green safe zone and displayed as a green zone on the 2D vertical model. Using 19.5% oxygen concentration and 7 ppm hydrogen sulfide concentration as critical values ​​for red restricted areas and green safe zones maximizes the safety of workers and significantly reduces the risk of poisoning accidents.

[0028] This invention overcomes the shortcomings of existing technologies and provides a method for detecting toxic gas gradients in confined spaces based on unmanned aerial vehicles (UAVs). This method utilizes UAVs as carriers for multi-sensor integrated modules to perform detection operations at different detection points, significantly improving the continuous detection capability in both vertical and horizontal directions. Compared with traditional single-point detection, it can comprehensively capture spatial distribution changes in gas concentration, avoiding the risk of missed detections and significantly improving detection efficiency and safety. Simultaneously, it eliminates the need for personnel to personally enter high-risk areas for detection operations, resulting in a high safety factor. By integrating a positioning module, a hydrogen sulfide electrochemical sensor, an oxygen optical sensor, and a lidar into the multi-sensor integrated module, it possesses multiple detection functions. It can conveniently collect the spatial coordinates, hydrogen sulfide gas concentration, and oxygen gas concentration at different detection points. Furthermore, it can obtain LiDAR data within a confined space, and then generate a two-dimensional vertical surface model of the vertical detection surface based on the LiDAR data. Further, the vertical detection surface can be divided into a grid pattern based on the two-dimensional vertical surface model, resulting in several detection points arranged in a matrix. Each detection point is probed individually to obtain the spatial coordinates, hydrogen sulfide gas concentration, and oxygen gas concentration of several detection points. Simultaneously, when the rate of change of hydrogen sulfide gas concentration data in adjacent detection planes exceeds 3 ppm / 0.5m or the difference in oxygen concentration gradient exceeds 5% VOL / 0.5m, the subsequent vertical and horizontal step distances are automatically adjusted to half of the previous values. This dynamic adjustment of the step distance significantly improves the detection density in abnormal areas, accurately capturing the accumulation of hydrogen sulfide at the bottom and accurately assessing the risk of deep high-concentration areas, thus solving the problem of missed detection in deep areas with traditional single-point detection. Based on this, the hydrogen sulfide and oxygen gas concentrations at several detection points at different depths can be calculated using interpolation algorithms to obtain hydrogen sulfide and oxygen concentration data for all horizontal layers in the vertical direction. This generates hydrogen sulfide and oxygen concentration distribution curves within different height ranges in a confined space. Simultaneously, a two-dimensional gas concentration distribution heatmap is constructed within the confined space. Combining the hydrogen sulfide and oxygen concentration distribution curves and the gas concentration distribution heatmap within different height ranges, red restricted areas and green safe areas are delineated in the vertical direction of the confined space on the two-dimensional vertical surface model. Furthermore, to achieve accurate detection of the vertical gradient of toxic gas concentrations and precise delineation of hazardous areas, this invention innovates the interpolation calculation process. Traditional Kriging methods rely on semi-variograms to calculate weights; this invention innovatively adjusts the gradient-sensitive weights through attenuation terms. This invention directly incorporates the influence of concentration gradients, simplifying the calculation process and reducing computational load, thus improving computational efficiency. In high gradient regions (such as near toxic gas leak points), the gradient term automatically reduces the weight of distant points, prioritizing neighboring data, meeting the requirements of adaptive sampling and facilitating more accurate calculation results. Furthermore, based on the need for continuous detection in the vertical direction, this invention innovatively focuses on the vertical direction, requiring only the calculation of the vertical distance d(Z). i The method ignores horizontal differences, adapts to vertical layer interpolation scenarios, and further reduces the calculation process and workload, thus improving computational efficiency. Furthermore, this invention innovatively employs dynamic coefficient adaptation: the coefficient α can be adjusted according to the gradient threshold of hydrogen sulfide (3ppm / 0.5m) or oxygen (5% VOL / 0.5m), achieving parameterized adaptation and facilitating the construction of a more accurate complete gas concentration gradient distribution model. By delineating red restricted areas and green safe areas in the finite vertical direction on the two-dimensional vertical plane model, it is possible to label areas of abrupt concentration changes, effectively overcoming the limitations of traditional modeling. This provides real-time visual decision support for operators, effectively guiding their actions during subsequent detection operations, ensuring operator safety, and significantly reducing the risk of accidents in subsequent operations. During the movement of the UAV platform, obstacle signals on the flight path are collected using lidar. The controller generates obstacle avoidance paths based on a dynamic path planning algorithm, enabling the UAV to have real-time obstacle avoidance capabilities, ensuring a continuous and safe detection process, and effectively coping with various complex environments.

[0029] This method is simple to implement and highly automated. It innovatively achieves high-precision gradient detection of toxic gases in the vertical direction within a confined space. It can obtain hydrogen sulfide and oxygen concentration data for all horizontal layers in the vertical direction, solving the problems of missed gradient detection risk and low efficiency of manual sampling in traditional single-point detection. It realizes three-dimensional dynamic assessment of toxic gases in confined spaces and accurate definition of safe areas. It is characterized by high efficiency, safety, and strong engineering applicability, and can provide reliable safety protection for operators. Attached Figure Description

[0030] Figure 1 This is a schematic diagram illustrating the present invention's method of detecting toxic gases in a confined space using a drone;

[0031] Figure 2 This is a schematic diagram showing the distribution of various detection points in the vertical direction within a limited space in this invention;

[0032] Figure 3 This is a schematic diagram of the vertical two-dimensional surface model range constructed based on LiDAR data according to the present invention;

[0033] Figure 4 This is a schematic diagram of the assembly of the multi-sensor integration module and the UAV platform in this invention;

[0034] Figure 5 for Figure 4 Top view;

[0035] Figure 6 This is a schematic diagram of the multi-sensor integrated module in this invention;

[0036] Figure 7 This is a schematic diagram of the system control section in this invention;

[0037] Figure 8 This is a schematic diagram of the gas concentration distribution curve (height-concentration comparison curve) in this invention.

[0038] In the diagram: 1. Unmanned aerial vehicle platform; 2. Multi-sensor integrated module, 2-1. Hydrogen sulfide electrochemical sensor, 2-2. Oxygen optical sensor, 2-3. Positioning module, 2-4. LiDAR, 2-5. Positioning module; 3. Limited space. Detailed Implementation

[0039] The invention will now be further described with reference to the accompanying drawings.

[0040] like Figures 1 to 7 As shown, the present invention provides a toxic gas gradient detection system in a confined space based on an unmanned aerial vehicle (UAV), comprising an UAV platform 1, a multi-sensor integrated module 2, a communication module, a controller, and a data processing terminal;

[0041] The multi-sensor integrated module 2 is installed at the bottom of the UAV platform 1, and includes a hydrogen sulfide electrochemical sensor 2-1, an oxygen optical sensor 2-2, a positioning module 2-3, and a lidar 2-4.

[0042] The communication module is installed on the UAV platform 1;

[0043] The controller is installed on the UAV platform 1 and is connected to the UAV platform 1, the multi-sensor integration module 2 and the communication module respectively.

[0044] The data processing terminal is installed in the control center and is connected to the controller via wireless communication.

[0045] As a preferred option, a power supply module is also included, which is used to supply power.

[0046] To facilitate the collection of spatial coordinate information at different detection points, the multi-sensor integrated module 2 also includes a positioning module 2-3.

[0047] To ensure detection accuracy, the hydrogen sulfide electrochemical sensor 2-1 has a range of 0-100ppm, a measurement accuracy of 0.01ppm, and a response time of ≤5s; the lidar 2-4 is a LiDAR.

[0048] As a preferred embodiment, the controller is a PLC controller.

[0049] As a preferred embodiment, the data processing terminal is an industrial computer.

[0050] In this invention, an unmanned aerial vehicle (UAV) platform is used as the carrier of a multi-sensor integrated module. Simultaneously, the actions of the UAV platform are controlled by a controller, allowing for flexible adjustment of the detection position. This facilitates the controlled measurement of toxic gas concentration data at any detection point within a two-dimensional vertical plane, significantly improving the continuous detection capability in the vertical direction of space. By simultaneously integrating a hydrogen sulfide electrochemical sensor, an oxygen optical sensor, a positioning module, and a lidar into the multi-sensor integrated module, the concentrations of hydrogen sulfide and oxygen at different detection points can be detected synchronously. This allows for comprehensive perception of hydrogen sulfide and oxygen concentration data at different detection points throughout the entire detection range. Furthermore, it facilitates the construction of hydrogen sulfide and oxygen concentrations at all horizontal levels in the vertical direction using the detection data from each detection point, greatly improving the safety assurance capability during operations in confined spaces. On the other hand, lidar can be used to perform laser scanning detection within a confined space. This facilitates the construction of a two-dimensional vertical surface model based on the laser scanning data. Subsequently, it allows for the generation of hydrogen sulfide and oxygen concentration distribution curves at different heights within the confined space by combining hydrogen sulfide and oxygen concentration data from all horizontal layers in the vertical direction. Simultaneously, it facilitates the subsequent construction of a two-dimensional gas concentration distribution heat map within the confined space, thus more effectively guiding subsequent detection actions for personnel entering the detection environment and ensuring their personal safety. By connecting the data processing terminal to the controller via wireless communication, it is easy to receive detection data sent by the controller and perform efficient and accurate analysis and processing of the detection data using the data processing terminal.

[0051] The system is simple in structure, low in manufacturing cost, highly automated, and has a high safety factor. It can automatically perform high-precision gradient detection of toxic gases in the vertical direction in confined spaces without requiring operators to personally enter high-risk areas to perform detection operations. It solves the technical problem that traditional single-point detection cannot identify deep dangerous areas, significantly improves the efficiency and safety of toxic gas detection in confined spaces, and provides reliable safety protection for operators.

[0052] This invention also provides a method for detecting toxic gas gradients in a confined space based on unmanned aerial vehicles (UAVs), employing a UAV-based system for detecting toxic gas gradients in a confined space, comprising the following steps:

[0053] Step 1: First, select the vertical detection surface in the limited space 3 to be detected, and control the UAV platform 1 to hover at the center line of the top of the vertical detection surface. Then, use the lidar 2-4 to emit laser pulses into the limited space 3 and measure their reflection time to generate LiDAR data. Then, send the LiDAR data to the controller.

[0054] Step 2: The controller generates a two-dimensional vertical plane model of the vertical detection surface based on LiDAR data. Then, the two-dimensional vertical plane model is divided into multiple detection planes (horizontal detection plane A, horizontal detection plane B, horizontal detection plane C, horizontal detection plane D, horizontal detection plane E, horizontal detection plane F, and horizontal detection plane G) from top to bottom with equal vertical step distances. Then, multiple detection points are divided into multiple detection points 1 from left to right with equal horizontal step distances on each detection plane. This forms a matrix arrangement of multiple detection points 1 on the two-dimensional vertical plane model.

[0055] Step 3: Control the UAV platform 1 to traverse each detection plane sequentially from top to bottom. At the same time, on each detection plane, traverse each detection point sequentially from left to right, and make the UAV platform 1 hover at each detection point for a set detection time.

[0056] During the detection process at each detection point, the positioning module 2-3 collects the spatial coordinate information of the current detection point and sends it to the controller. The hydrogen sulfide electrochemical sensor 2-1 collects the hydrogen sulfide gas concentration signal of the current detection point in real time and sends it to the controller. The oxygen optical sensor 2-2 collects the oxygen concentration signal of the current detection point in real time and sends it to the controller. The hydrogen sulfide gas concentration data of the current detection point is obtained based on the hydrogen sulfide gas concentration signal, and the oxygen concentration data of the current detection point is obtained based on the oxygen concentration signal. The data is then recorded and stored.

[0057] After completing the detection work on two detection planes, the hydrogen sulfide gas concentration data and oxygen concentration data in the two adjacent detection planes are compared. When the change rate of hydrogen sulfide gas concentration data is >3ppm / 0.5m or the difference in oxygen concentration gradient is >5%VOL / 0.5m, the remaining undetected area is divided a second time, and the vertical step distance and the horizontal step distance are reduced to half of the original. Then, several detection points II are formed in a matrix arrangement on the remaining undetected area, and the detection work on several detection points II continues in the same way.

[0058] During the movement of the UAV platform 1, the LiDAR 2-4 is used to collect obstacle signals on the flight path and send them to the controller. The controller generates an obstacle avoidance path based on the dynamic path planning algorithm, combined with the obstacle signals and the position of the target detection point 1, and then controls the UAV platform 1 to fly to the target detection point 1 according to the obstacle avoidance path.

[0059] After completing the detection work on the entire vertical detection surface, the controller sends the spatial coordinate information of each detection point, the hydrogen sulfide gas concentration data of each detection point, and the oxygen concentration data of each detection point to the data processing terminal located in the control center through the communication module.

[0060] Step 4: The data processing terminal summarizes the spatial coordinate information of each detection point, the hydrogen sulfide gas concentration data and oxygen concentration data of each detection point, and obtains the hydrogen sulfide gas concentration data and oxygen concentration data of different detection points on the two-dimensional vertical plane model. At the same time, based on the hydrogen sulfide gas concentration data and oxygen concentration data of different detection points, the Kriging interpolation algorithm is used to calculate the hydrogen sulfide gas concentration data and oxygen concentration data of all horizontal layers in the vertical direction of the confined space 3 according to formula (1), and generates the hydrogen sulfide and oxygen concentration distribution curves in different height ranges within the confined space 3. At the same time, a gas concentration distribution heat map of the two-dimensional surface in the confined space 3 is constructed. Then, combined with the hydrogen sulfide and oxygen concentration distribution curves and the gas concentration distribution heat map in different height ranges, the red forbidden zone and green safe zone in the vertical direction of the confined space 3 are delineated on the two-dimensional vertical plane model.

[0061]

[0062] In the formula, C(Z0) pre Let d(Z0) be the predicted concentration value at any position Z0 in the vertical direction. i Z0) is a known point Z i The perpendicular distance between the point and the point to be interpolated Z0, where Z is the perpendicular distance between the point and the point to be interpolated. i This refers to the detection point above the location of Z0; Given point Z i The absolute value of the concentration gradient at a given point; α is the gradient sensitivity coefficient, dynamically adjusted according to the hydrogen sulfide / oxygen concentration thresholds. Preferably, the gradient sensitivity coefficient for hydrogen sulfide is 0.5, and the gradient sensitivity coefficient for oxygen is 0.3.

[0063] In order to effectively delineate restricted and safe working areas for high-concentration hydrogen sulfide and low-oxygen zones, the method for delineating the red restricted area and green safe area in the vertical direction of the confined space (3) in step four is as follows:

[0064] like Figure 8As shown, using 19.5% oxygen concentration and 7 ppm hydrogen sulfide concentration as critical values, red and green zones are marked vertically. When the oxygen concentration in any zone is below 19.5% or the hydrogen sulfide concentration is above 7 ppm, it is designated as a red restricted area and displayed as such on the 2D vertical model, with its coordinates also marked. When the oxygen concentration in any zone is equal to or higher than 19.5% or the hydrogen sulfide concentration is equal to or lower than 7 ppm, it is designated as a green safe zone and displayed as such on the 2D vertical model. Using 19.5% oxygen concentration and 7 ppm hydrogen sulfide concentration as critical values ​​for red restricted areas and green safe zones maximizes the safety of workers and significantly reduces the risk of poisoning accidents.

[0065] This invention overcomes the shortcomings of existing technologies and provides a method for detecting toxic gas gradients in confined spaces based on unmanned aerial vehicles (UAVs). This method utilizes UAVs as carriers for multi-sensor integrated modules to perform detection operations at different detection points, significantly improving the continuous detection capability in both vertical and horizontal directions. Compared with traditional single-point detection, it can comprehensively capture spatial distribution changes in gas concentration, avoiding the risk of missed detections and significantly improving detection efficiency and safety. Simultaneously, it eliminates the need for personnel to personally enter high-risk areas for detection operations, resulting in a high safety factor. By integrating a positioning module, a hydrogen sulfide electrochemical sensor, an oxygen optical sensor, and a lidar into the multi-sensor integrated module, it possesses multiple detection functions. It can conveniently collect the spatial coordinates, hydrogen sulfide gas concentration, and oxygen gas concentration at different detection points. Furthermore, it can obtain LiDAR data within a confined space, and then generate a two-dimensional vertical surface model of the vertical detection surface based on the LiDAR data. Further, the vertical detection surface can be divided into a grid pattern based on the two-dimensional vertical surface model, resulting in several detection points arranged in a matrix. Each detection point is probed individually to obtain the spatial coordinates, hydrogen sulfide gas concentration, and oxygen gas concentration of several detection points. Simultaneously, when the rate of change of hydrogen sulfide gas concentration data in adjacent detection planes exceeds 3 ppm / 0.5m or the difference in oxygen concentration gradient exceeds 5% VOL / 0.5m, the subsequent vertical and horizontal step distances are automatically adjusted to half of the previous values. This dynamic adjustment of the step distance significantly improves the detection density in abnormal areas, accurately capturing the accumulation of hydrogen sulfide at the bottom and accurately assessing the risk of deep high-concentration areas, thus solving the problem of missed detection in deep areas with traditional single-point detection. Based on this, the hydrogen sulfide and oxygen gas concentrations at several detection points at different depths can be calculated using interpolation algorithms to obtain hydrogen sulfide and oxygen concentration data for all horizontal layers in the vertical direction. This generates hydrogen sulfide and oxygen concentration distribution curves within different height ranges in a confined space. Simultaneously, a two-dimensional gas concentration distribution heatmap is constructed within the confined space. Combining the hydrogen sulfide and oxygen concentration distribution curves and the gas concentration distribution heatmap within different height ranges, red restricted areas and green safe areas are delineated in the vertical direction of the confined space on the two-dimensional vertical surface model. Furthermore, to achieve accurate detection of the vertical gradient of toxic gas concentrations and precise delineation of hazardous areas, this invention innovates the interpolation calculation process. Traditional Kriging methods rely on semi-variograms to calculate weights; this invention innovatively adjusts the gradient-sensitive weights through attenuation terms. This invention directly incorporates the influence of concentration gradients, simplifying the calculation process and reducing computational load, thus improving computational efficiency. In high gradient regions (such as near toxic gas leak points), the gradient term automatically reduces the weight of distant points, prioritizing neighboring data, meeting the requirements of adaptive sampling and facilitating more accurate calculation results. Furthermore, based on the need for continuous detection in the vertical direction, this invention innovatively focuses on the vertical direction, requiring only the calculation of the vertical distance d(Z). i The method ignores horizontal differences, adapts to vertical layer interpolation scenarios, and further reduces the calculation process and workload, thus improving computational efficiency. Furthermore, this invention innovatively employs dynamic coefficient adaptation: the coefficient α can be adjusted according to the gradient threshold of hydrogen sulfide (3ppm / 0.5m) or oxygen (5% VOL / 0.5m), achieving parameterized adaptation and facilitating the construction of a more accurate complete gas concentration gradient distribution model. By delineating red restricted areas and green safe areas in the finite vertical direction on the two-dimensional vertical plane model, it is possible to label areas of abrupt concentration changes, effectively overcoming the limitations of traditional modeling. This provides real-time visual decision support for operators, effectively guiding their actions during subsequent detection operations, ensuring operator safety, and significantly reducing the risk of accidents in subsequent operations. During the movement of the UAV platform, obstacle signals on the flight path are collected using lidar. The controller generates obstacle avoidance paths based on a dynamic path planning algorithm, enabling the UAV to have real-time obstacle avoidance capabilities, ensuring a continuous and safe detection process, and effectively coping with various complex environments.

[0066] This method is simple to implement and highly automated. It innovatively achieves high-precision gradient detection of toxic gases in the vertical direction within a confined space. It can obtain hydrogen sulfide and oxygen concentration data for all horizontal layers in the vertical direction, solving the problems of missed gradient detection risk and low efficiency of manual sampling in traditional single-point detection. It realizes three-dimensional dynamic assessment of toxic gases in confined spaces and accurate definition of safe areas. It is characterized by high efficiency, safety, and strong engineering applicability, and can provide reliable safety protection for operators.

Claims

1. A method for detecting toxic gas gradients in a confined space based on unmanned aerial vehicles (UAVs), comprising a UAV-based system for detecting toxic gas gradients in a confined space, characterized in that, Includes the following steps: Step 1: First, select the vertical detection surface in the limited space (3) to be detected, and control the UAV platform (1) to hover at the center line of the top of the vertical detection surface. Then, use the lidar (2-4) to emit laser pulses into the limited space (3) and measure its reflection time, and generate LiDAR data. Then, send the LiDAR data to the controller. Step 2: The controller generates a two-dimensional vertical plane model of the vertical detection surface based on LiDAR data. Then, the two-dimensional vertical plane model is divided into multiple detection planes from top to bottom with equal vertical step distances. On each detection plane, multiple detection points are divided from left to right with equal horizontal step distances, thus forming a matrix arrangement of multiple detection points on the two-dimensional vertical plane model. Step 3: Control the UAV platform (1) to traverse each detection plane from top to bottom, and at the same time, traverse each detection point from left to right on each detection plane, and make the UAV platform (1) hover at each detection point to set the detection time; During the detection process at each detection point, the positioning module (2-3) collects the spatial coordinate information of the current detection point and sends it to the controller. The hydrogen sulfide electrochemical sensor (2-1) collects the hydrogen sulfide gas concentration signal of the current detection point in real time and sends it to the controller. The oxygen optical sensor (2-2) collects the oxygen concentration signal of the current detection point in real time and sends it to the controller. The hydrogen sulfide gas concentration data of the current detection point is obtained based on the hydrogen sulfide gas concentration signal of the current detection point, and the oxygen concentration data of the current detection point is obtained based on the oxygen concentration signal of the current detection point. The data is recorded and stored. After completing the detection work on two detection planes, the hydrogen sulfide gas concentration data and oxygen concentration data in the two adjacent detection planes are compared. When the change rate of hydrogen sulfide gas concentration data is >3ppm / 0.5m or the oxygen concentration gradient difference is >5%VOL / 0.5m, the remaining undetected area is divided a second time, and the vertical step distance and the horizontal step distance are reduced to half of the original. Then, several detection points II are formed in a matrix arrangement on the remaining undetected area, and the detection work on several detection points II continues in the same way. During the movement of the UAV platform (1), the LiDAR (2-4) is used to collect obstacle signals on the flight path and send them to the controller. The controller generates an obstacle avoidance path based on the dynamic path planning algorithm, combining the obstacle signals and the position of the target detection point one, and then controls the UAV platform (1) to fly to the target detection point one according to the obstacle avoidance path. After completing the detection work on the entire vertical detection surface, the controller sends the spatial coordinate information of each detection point, the hydrogen sulfide gas concentration data of each detection point, and the oxygen concentration data of each detection point to the data processing terminal located in the control center through the communication module. Step 4: The data processing terminal summarizes the spatial coordinate information of each detection point, the hydrogen sulfide gas concentration data and oxygen concentration data of each detection point, and obtains the hydrogen sulfide gas concentration data and oxygen concentration data at different detection points on the two-dimensional vertical plane model. At the same time, based on the hydrogen sulfide gas concentration data and oxygen concentration data at different detection points, the Kriging interpolation algorithm is used to calculate the hydrogen sulfide gas concentration data and oxygen concentration data of all horizontal layers in the vertical direction of the finite space (3) according to formula (1), and generates the hydrogen sulfide and oxygen concentration distribution curves in different height ranges within the finite space (3). At the same time, a gas concentration distribution heat map of the two-dimensional surface in the finite space (3) is constructed. Then, combined with the hydrogen sulfide and oxygen concentration distribution curves and gas concentration distribution heat map in different height ranges, the red forbidden zone and green safe zone in the vertical direction of the finite space (3) are delineated on the two-dimensional vertical plane model. （1）； In the formula, This represents the predicted concentration value at any position Z0 in the vertical direction. For known points Location and interpolation point The vertical distance at point, where, for The detection point above the current location; Given point Z i The absolute value of the concentration gradient at that location; This is a gradient sensitivity coefficient, dynamically adjusted based on hydrogen sulfide / oxygen concentration thresholds.

2. The method for detecting toxic gas gradients in a confined space based on an unmanned aerial vehicle (UAV) according to claim 1, characterized in that, In step four, the method for delineating the red restricted area and green safe area in the vertical direction of the confined space (3) is as follows: Using 19.5% oxygen concentration and 7ppm hydrogen sulfide concentration as critical values, red and green areas are marked in the vertical direction. When the oxygen concentration in any area is lower than 19.5% or the hydrogen sulfide concentration is greater than 7ppm, it is designated as a red restricted area and displayed as a red area on the two-dimensional vertical plane model. At the same time, the coordinates of the area are marked in the red area. When the oxygen concentration in any area is higher than or equal to 19.5% or the hydrogen sulfide concentration is lower than or equal to 7ppm, it is designated as a green safe area and displayed as a green area on the two-dimensional vertical plane model.

3. The method for detecting toxic gas gradients in a confined space based on an unmanned aerial vehicle (UAV) according to claim 1, wherein the UAV-based confined space toxic gas gradient detection system comprises an UAV platform (1), characterized in that, It also includes a multi-sensor integration module (2), a communication module, a controller, and a data processing terminal; The multi-sensor integrated module (2) is installed at the bottom of the UAV platform (1), and includes a hydrogen sulfide electrochemical sensor (2-1), an oxygen optical sensor (2-2), a positioning module (2-3), and a lidar (2-4). The communication module is installed on the unmanned aerial vehicle platform (1); The controller is installed on the UAV platform (1) and is connected to the UAV platform (1), the multi-sensor integration module (2) and the communication module respectively; The data processing terminal is installed in the control center and is connected to the controller via wireless communication.

4. The method for detecting toxic gas gradients in a confined space based on an unmanned aerial vehicle (UAV) according to claim 3, characterized in that, The hydrogen sulfide electrochemical sensor (2-1) has a measurement range of 0-100ppm, a measurement accuracy of 0.01ppm, and a response time of ≤5s; the lidar (2-4) is a LiDAR.

5. The method for detecting toxic gas gradients in a confined space based on an unmanned aerial vehicle (UAV) according to claim 3, characterized in that, The controller is a PLC controller.

6. The method for detecting toxic gas gradients in a confined space based on an unmanned aerial vehicle (UAV) according to claim 3, characterized in that, The data processing terminal is an industrial computer.

Images