A gas detection method and system for dangerous workplaces

Abstract

The present invention relates to the field of gas detection technology, and in particular to a gas detection method and system for hazardous workplaces. The method comprises: obtaining the gas concentration and position information of each detection point in real time; dividing the entire detection area into a plurality of grids, and using the kernel density analysis method to calculate the concentration density of each grid based on the gas concentration and position information of each detection point; constructing a gas concentration plane distribution map of the entire detection area based on the concentration density of all grids; and performing manual inspections on locations where the gas concentration exceeds the limit based on the gas concentration plane distribution map to ensure the reliability and accuracy of gas detection in a working environment where the gas diffusion is complex or uneven.

Description

Technical Field

[0001] The present invention relates to the technical field of gas detection, and in particular to a gas detection method and system for dangerous workplaces. Background Art

[0002] In recent years, with the advancement of industrialization and the annual growth of the processing scale of the chemical industry, many industrial activities involve the production, use, storage and transportation of toxic and hazardous chemicals, and these industrial activities are gradually developing towards scale and large-scale development, which increases the risk of leakage and emission of toxic and hazardous gases, causing significant harm to the environment and human health. Therefore, how to accurately detect toxic and hazardous gases and ensure the safety of detection personnel in dangerous working environments has become a current research hotspot.

[0003] To this end, Chinese patent 202211486924.8, a method for intelligent monitoring of toxic gases, detects the toxic gas concentration at each point in the monitoring area according to the gas detection sensors arranged at each point in the monitoring area and performs inverse distance weighted data fitting to obtain the toxic gas concentration distribution map at any position in the monitoring area, which makes up for the shortcomings of the gas detection sensors and can more intuitively obtain the concentration of harmful gases in the area; then, drone inspections are carried out according to the harmful gas distribution map, which can quickly locate abnormal points, which is safer and faster than manual inspections. Among them, the inverse distance weighted data fitting method used in the existing patent is an interpolation technology, which estimates the value at the fitting point by calculating the distance between each sampling point and the fitting point and assigning different weights to these distances. This method is very effective when processing continuous data. However, in the actual detection operation environment, the diffusion of toxic and harmful gases is affected by many factors, including temperature, humidity, wind speed and physical and chemical properties of the gas. These factors will cause the distribution of gas in space to be very complex and uneven, resulting in points at a longer distance having a greater adverse effect on the fitting point. The inverse distance weighted data fitting method assumes that there is a simple mathematical relationship between data points and cannot accurately capture the adverse effects of complex or uneven gas diffusion, which leads to a large difference between the fitting result and the actual gas distribution, making it difficult to ensure the reliability and accuracy of gas detection. At the same time, although drone inspections can provide efficient and low-cost solutions in some cases, drones may be difficult to operate in dangerous working environments, such as in small spaces or complex terrain structures. In addition, in the event of an emergency, drones rely on preset programs and sensor data and cannot make the best decisions in response to environmental changes in a timely manner, making it difficult to ensure a timely response to emergencies. Therefore, manual inspections are still indispensable, and the safety of inspectors is particularly important. Summary of the invention

[0004] In response to the above technical problems, the present invention proposes a gas detection method and system for hazardous workplaces, aiming to ensure the reliability and accuracy of gas detection in working environments with complex or uneven gas diffusion, and to dynamically monitor the safety of inspectors.

[0005] In a first aspect, the present application provides a method for detecting gas in a hazardous workplace, comprising the following steps:

[0006] Step 102, obtaining the gas concentration and location information of each detection point in real time;

[0007] Step 104, dividing the entire detection area into a plurality of grids, and using a kernel density analysis method to calculate the concentration density of each grid based on the gas concentration and position information of each detection point;

[0008] Step 106, constructing a gas concentration plane distribution map of the entire detection area based on the concentration density of all grids;

[0009] Step 108: Based on the gas concentration plane distribution map, manually inspect the locations where the gas concentration exceeds the limit.

[0010] In some embodiments, in step 104, based on the gas concentration and position information of each detection point, a kernel density analysis method is used to calculate the concentration density of each grid, including:

[0011] Get the center point position of each grid;

[0012] Based on the position of the center point of the grid and the position information of each detection point, the x-axis distance and y-axis distance between the center point of each grid and each detection point are calculated and recorded as the distance data of the grid;

[0013] Based on the distance data of each grid and the gas concentration of each detection point, the concentration density of each grid is calculated using a preset kernel function.

[0014] In some embodiments, based on the distance data of each grid and the gas concentration of each detection point, a preset kernel function is used to calculate the concentration density of each grid, including:

[0015] Use Silverman's rule of thumb to determine bandwidth;

[0016] Based on the determined bandwidth, the distance data of each grid and the gas concentration of each detection point, the concentration density of each grid is calculated using the Gaussian kernel function.

[0017] In some embodiments, based on the determined bandwidth, the distance data of each grid and the gas concentration of each detection point, a Gaussian kernel function is used to calculate the concentration density of each grid, including:

[0018] Based on the determined bandwidth and the distance data of each grid, the Gaussian kernel function is used to calculate the weight between the grid center point and each detection point;

[0019] The weight between the grid center point and each detection point is multiplied by the gas concentration at each detection point to obtain the weighted concentration between the grid center point and each detection point;

[0020] All weighted concentrations are averaged to obtain the concentration density of the grid.

[0021] In some embodiments, after step 106 and before step 108, the process further includes:

[0022] Generate a gas concentration time variation curve of each grid based on the concentration density of each grid within a preset time period;

[0023] Periodically correcting the gas concentration time variation curve;

[0024] Based on the corrected gas concentration time variation curve, identify the gas concentration increase trend of each grid;

[0025] Based on the increasing trend of gas concentration in each grid, an early warning is issued for the gas concentration in the detection area.

[0026] In some embodiments, after step 108, the following steps are further included:

[0027] Step 110, using a handheld inspection device to monitor the acceleration and angular velocity during the inspection process in real time;

[0028] Step 112: Based on the acceleration and angular velocity during the inspection process, a safety monitoring alarm is issued to the inspector.

[0029] In some embodiments, in step 112, based on the acceleration and angular velocity during the inspection process, a safety monitoring alarm is performed on the inspector, including:

[0030] Based on the angular velocity during the inspection process, determine the magnitude of the angular velocity change;

[0031] The change range of the angular velocity is compared with the first preset threshold. If the change range of the angular velocity is less than the first preset threshold, it is determined that the inspector is currently safe. If the change range of the angular velocity is greater than or equal to the first preset threshold, the following steps are performed:

[0032] Preset the window period and determine the magnitude of acceleration change within the window period;

[0033] The acceleration change amplitude is compared with the second preset threshold. If the acceleration change amplitude is less than the second preset threshold, it is judged that the inspector is currently safe. If the acceleration change amplitude is greater than or equal to the second preset threshold, it is judged that the inspector has fallen and a safety alarm for the inspector is issued.

[0034] In a second aspect, the present application provides a gas detection system for hazardous workplaces, including a gas detector host, a plurality of gas detector slaves, and a handheld patrol device, wherein the gas detector host is located outside a detection area, and each gas detector slave is located at a detection point within the detection area, and a communication network link is formed between the gas detector host, each gas detector slave, and the handheld patrol device.

[0035] The gas detector slave is used to obtain the gas concentration and location information of the corresponding detection point in real time;

[0036] The gas detector host is used to divide the entire detection area into several grids, calculate the concentration density of each grid based on the gas concentration and location information of each detection point using the kernel density analysis method, and construct a gas concentration plane distribution map of the entire detection area based on the concentration density of all grids;

[0037] The handheld inspection device is used to be carried by inspectors into the detection area to conduct manual inspections of locations where gas concentration exceeds the limit based on the gas concentration plane distribution map.

[0038] In some embodiments, the handheld patroller includes a communication unit and a posture sensing unit. Several gas detector slaves and a gas detector host are all communicatively connected to the communication unit of the handheld patroller, and the posture sensing unit is connected to the communication unit. The posture sensing unit is used to monitor the acceleration and angular velocity during the inspection process in real time, and based on the acceleration and angular velocity during the inspection process, perform safety monitoring and alarm on the inspector, and send the alarm information to the gas detector host through the communication unit.

[0039] In some embodiments, the gas detector slave includes an infrared detection unit and a sensor detection unit. The infrared detection unit is used to obtain gas imaging information of the corresponding detection point in real time to provide early warning of gas concentration, and the sensor detection unit is used to obtain gas concentration and position information of the corresponding detection point in real time.

[0040] The beneficial technical effects of the present invention include at least:

[0041] 1. A gas detection method and system for hazardous workplaces is adopted. The detection area is divided into several grids. Based on the distance between the center point of each grid and each detection point and the gas concentration of each detection point, the concentration density of each grid is calculated by the kernel density analysis method. That is, the weight of each grid is calculated by the kernel function, and then the gas concentration data is spatially interpolated. This can provide good smoothness and reduce the influence of noise and outliers, so as to more accurately simulate the distribution of gas in space, overcome the gas concentration density estimation error caused by complex or uneven gas distribution to the greatest extent, improve the accuracy of gas concentration density estimation, and combine with the subsequent construction of the gas concentration plane distribution map of the entire detection area to conduct comprehensive and real-time monitoring of the gas concentration within the hazardous operation range, so that rescue work has a basis to rely on;

[0042] 2. Determine the bandwidth through Silverman's rule of thumb. While balancing the smoothness and deviation of the estimate, calculate the optimal bandwidth based on the statistical characteristics of the data, and achieve a more objective and accurate selection of the bandwidth of the Gaussian kernel function. Then, use the Gaussian kernel function to calculate the weight of the grid, which can make the grid weight more evenly distributed in the entire detection area, thereby forming a smoother curve in the subsequent gas concentration plane distribution map of the entire detection area, making the gas concentration plane distribution map closer to the actual working environment. At the same time, this characteristic better captures the local information of the data point, especially the concentration changes near the detection point, and improves the accuracy of the subsequent gas concentration plane distribution of the entire detection area in the local area.

[0043] 3. By combining the two real-time monitoring parameters of acceleration and angular velocity as the basis for dynamic safety monitoring of inspectors, the safety status of inspectors can be evaluated more comprehensively. Specifically, the change in angular velocity can reflect whether the inspector has suddenly lost balance, and the change in acceleration can be combined to further confirm whether the inspector has encountered safety risk events such as collision or fall. The probability of misjudgment can be reduced through the layered threshold judgment method, which improves the accuracy and reliability of the dynamic safety monitoring alarm of inspectors. Timely warnings can be issued when inspectors are in danger, which helps rescue personnel to formulate rescue plans in a timely manner and improve the safety factor of inspectors.

[0044] 4. By forming a communication networking link with the gas detector host, each gas detector slave and the handheld patrol device, "one point to multiple transmissions" can be achieved regardless of environmental conditions. That is, each sensor module can serve as a backup for other modules. When a module fails or the signal is attenuated, other modules can continue to transmit data to ensure data continuity and integrity, enhance data redundancy and reliability, or when environmental conditions change, the communication path can be adjusted through the communication networking link to bypass obstacles or areas with severe attenuation, maintain stable data transmission, and enhance the system's adaptability. At the same time, each device can view each other's detection data and alarm status, thereby improving the system's emergency monitoring coordination;

[0045] 5. By setting up both the infrared detection unit and the sensor detection unit in the gas detector slave, it is possible to detect trace gas leaks early. Even if the gas concentration is low, an early warning can be issued quickly, adding a layer of protection before the sensor detection unit detects the gas concentration, thereby providing more accurate toxic and harmful gas detection results. It is more reliable and timely than the traditional single sensor detection method, and has higher safety. Moreover, by adding an infrared detection unit in conjunction with the sensor detection unit, while detecting the gas leakage concentration, it is also possible to obtain the location information of the specific gas leakage point or leakage source, which is helpful for tracing the leaked gas.

[0046] Other features and advantages of the present invention will be disclosed in detail in the following specific embodiments and drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] The present invention will be further described below in conjunction with the accompanying drawings:

[0048] Figure 1 This is a flow chart of a method for detecting gas in a hazardous workplace according to an embodiment of the present invention.

[0049] Figure 2 The figure is a schematic diagram of the structure of a gas detection system for a hazardous workplace according to an embodiment of the present invention. DETAILED DESCRIPTION

[0050] The technical solutions of the embodiments of the present invention are explained and described below in conjunction with the drawings of the embodiments of the present invention, but the following embodiments are only preferred embodiments of the present invention, not all. Based on the embodiments in the implementation mode, other embodiments obtained by those skilled in the art without creative work are all within the protection scope of the present invention.

[0051] In the following description, terms such as "inside", "outside", "up", "down", "left", "right", etc. that indicate directions or positional relationships are only used to facilitate the description of the embodiments and simplify the description, and do not indicate or imply that the referred device or element must have a specific direction, be constructed and operate in a specific direction. Therefore, it should not be understood as a limitation of the present invention.

[0052] Please refer to the attached Figure 1 , Figure 1 A flow chart of a gas detection method for a hazardous workplace provided by an embodiment of the present specification is shown.

[0053] like Figure 1 As shown, the gas detection method for hazardous workplaces may at least include the following steps:

[0054] Step 102, obtaining the gas concentration and position information of each detection point in real time.

[0055] It is understandable that several detection points are preset in the detection area, and a fixed gas detector is installed at each detection point, including but not limited to infrared gas detectors and gas detection sensors, etc. The fixed gas detector can also be equipped with a remote monitoring system, GPS or Beidou positioning system, such as video monitoring, to integrate the video monitoring screen with the gas concentration detection data to understand the status of each detection point in the detection area in real time, and provide accurate location information for each gas detector through GPS or Beidou positioning system, so as to provide a data basis for the subsequent fusion of location information and gas concentration data to realize the association between spatial position and gas concentration.

[0056] Step 104 , the entire detection area is divided into a number of grids, and based on the gas concentration and position information of each detection point, the concentration density of each grid is calculated using a kernel density analysis method.

[0057] Furthermore, in this embodiment, in step 104, based on the gas concentration and position information of each detection point, the concentration density of each grid is calculated using a kernel density analysis method, including:

[0058] Step 202, obtaining the center point position of each grid.

[0059] Specifically, in this embodiment, the method for obtaining the center point position of each grid may be:

[0060] 1. Initialization: Determine the spatial range and resolution of the entire detection area. According to the resolution required for the entire detection area, determine the number and size of the divided grids. The higher the resolution and the smaller the grid, the higher the accuracy.

[0061] 2. Coordinate conversion: Ensure that all test points are in the same coordinate system. If the actual terrain conditions require, you can use geographic information system (GIS) tools or libraries to convert geographic coordinates into plane coordinates. Generate the location information of each test point and the coordinates of each grid point in the same coordinate system.

[0062] 3. For each mesh, the coordinates of its center point can be calculated by taking the average of the coordinates of its four vertices.

[0063] Step 204, based on the position of the center point of each grid and the position information of each detection point, the x-axis distance and the y-axis distance between the center point of each grid and each detection point are calculated and recorded as the distance data of the grid.

[0064] Step 206: Based on the distance data of each grid and the gas concentration of each detection point, a preset kernel function is used to calculate the concentration density of each grid.

[0065] Among them, the preset kernel functions in this embodiment include but are not limited to Gaussian kernel function, quartic kernel function, triangular kernel function, etc.

[0066] For example, taking the triangular kernel function as the preset kernel function, the implementation method of calculating the concentration density of each grid is:

[0067] (1) First, in kernel density analysis, bandwidth is a key parameter that controls the width of the kernel function and thus affects the smoothness of the density estimate. For triangular kernel functions, the choice of bandwidth may be more dependent on the specific application scenario and the data itself. Therefore, the optimal bandwidth can be determined by cross-validation or by observing how the weight changes with bandwidth.

[0068] (2) Secondly, based on the determined bandwidth, the distance data of each grid and the gas concentration of each detection point, the concentration density of each grid is calculated using a triangular kernel function. Specifically, the concentration density calculation method of one grid is:

[0069] Based on the determined bandwidth and the distance data of each grid, the triangular kernel function is used to calculate the weight between the grid center point and each detection point, where the triangular kernel function can be expressed as:

[0070]

[0071] Among them, u represents the distance data of the grid, Represents the parameters of the kernel function, that is, the determined bandwidth;

[0072] The weight between the grid center point and each detection point is multiplied by the gas concentration at each detection point to obtain the weighted concentration between the grid center point and each detection point;

[0073] All weighted concentrations are averaged to obtain the concentration density of the grid.

[0074] This embodiment divides the entire detection area into several grids, and uses the kernel density analysis method to calculate the concentration density of each grid based on the distance between the center point of each grid and each detection point and the gas concentration of each detection point. That is, the weight of each grid is calculated by the kernel function, and then the gas concentration data is spatially interpolated. This can provide good smoothness and reduce the influence of noise and outliers, thereby more accurately simulating the distribution of gas in space, overcoming the gas concentration density estimation error caused by complex or uneven gas distribution to the greatest extent, improving the accuracy of gas concentration density estimation, and providing a reliable and accurate data basis for the subsequent construction of a gas concentration plane distribution map of the entire detection area.

[0075] Preferably, in this embodiment, based on the distance data of each grid and the gas concentration of each detection point, a preset kernel function is used to calculate the concentration density of each grid, including:

[0076] Use Silverman's rule of thumb to determine bandwidth;

[0077] Based on the determined bandwidth, the distance data of each grid and the gas concentration of each detection point, the concentration density of each grid is calculated using the Gaussian kernel function.

[0078] Specifically, the bandwidth can be determined using Silverman's rule of thumb as:

[0079]

[0080] in, It represents the standard deviation of the gas concentration data, which is used to describe the degree of deviation between each concentration value in the gas concentration detection data set and the average concentration value of the data set. m represents the total number of grid center points.

[0081] It is understandable that the goal of Silverman's rule of thumb is to find a suitable bandwidth so that the kernel density estimate is neither too smooth (resulting in loss of details) nor too rough (resulting in too much noise in the estimate), that is, to balance the smoothness and bias of the estimate. Therefore, using Silverman's rule of thumb in the kernel density estimation of univariate data sets such as gas concentration can achieve better monitoring effects.

[0082] Specifically, based on the determined bandwidth, the distance data of each grid and the gas concentration of each detection point, the Gaussian kernel function is used to calculate the concentration density of each grid, including:

[0083] Step 302, based on the determined bandwidth and the distance data of each grid, a Gaussian kernel function is used to calculate the weight between the grid center point and each detection point;

[0084] Step 304, multiplying the weight between the grid center point and each detection point by the gas concentration of each detection point to obtain the weighted concentration between the grid center point and each detection point;

[0085] Step 306, calculate the average of all weighted concentrations to obtain the concentration density of the grid. The expression of the concentration density of the grid is:

[0086]

[0087] Among them, (x, y) represents the center point position of the grid, f(x, y) represents the concentration density of the grid, h represents the preset bandwidth, i represents the serial number of the detection point, n represents the total number of detection points, (xi, yi) represents the position of each detection point, ci represents the gas concentration value obtained at each detection point, and K() represents the Gaussian kernel function.

[0088] Among them, the Gaussian kernel function can be expressed as:

[0089]

[0090] It can be understood that steps 302 to 306 of this embodiment only provide an example of using a Gaussian kernel function to calculate the concentration of a grid based on a determined bandwidth, distance data of each grid, and gas concentration at each detection point. The method of using a Gaussian kernel function to calculate the concentration density of other grids based on a determined bandwidth, distance data of each grid, and gas concentration at each detection point can refer to this embodiment.

[0091] Compared with other kernel functions (such as linear kernel function, quartic kernel function, triangular kernel function, etc.), this implementation first determines the bandwidth through Silverman's empirical rule, and calculates the optimal bandwidth based on the statistical characteristics of the data while balancing the smoothness and deviation of the estimate, so as to achieve a more objective and accurate selection of the bandwidth of the Gaussian kernel function; then, the Gaussian kernel function is used to calculate the weight of the grid. The grid weight calculated by the Gaussian kernel function can gradually decrease with the increase of distance, but the rate of decrease is slower. This characteristic can make the grid weight more evenly distributed in the entire detection area, thereby forming a smoother curve in the subsequently constructed gas concentration plane distribution map of the entire detection area, making the gas concentration plane distribution map closer to the actual working environment. At the same time, this characteristic better captures the local information of the data points, especially the concentration changes near the detection points, and improves the accuracy of the subsequently constructed gas concentration plane distribution of the entire detection area in the local area.

[0092] Step 106, constructing a gas concentration plane distribution map of the entire detection area based on the concentration density of all grids.

[0093] For example, the concentration density of all grids can be visualized as a heat map, and the visualized heat map can be used as a gas concentration plane distribution map. Therefore, the gas concentration distribution in the overall dangerous working environment can be analyzed based on the gas concentration plane distribution map, providing a basis for formulating emergency rescue measures.

[0094] It is understandable that the gas concentration at each detection point is obtained in real time, so the constructed gas concentration plane distribution map is also updated in real time. Further, in this embodiment, after step 106 and before step 108, the following steps may also be included:

[0095] Based on the concentration density of each grid within a preset time period, a gas concentration time variation curve of each grid is generated. Specifically, the concentration density data of each grid within the preset time period are arranged in chronological order; a time series graph is drawn using data points, with the horizontal axis being time and the vertical axis being the concentration density data of the grid, and gas concentration time variation curves of different grids are drawn respectively; the gas concentration is smoothed or predicted by a suitable smoothing curve algorithm or machine learning method to generate a gas concentration time variation curve that can reflect the gas concentration variation trend of the actual detection process of toxic and harmful gases at the grid location.

[0096] The gas concentration time variation curve is corrected periodically. Specifically, a periodic window (for example, every day) can be selected, and the concentration density value at the corresponding time point of each new period is inserted into the original curve as a correction value, replacing the concentration density value at the corresponding time point.

[0097] Based on the corrected gas concentration time variation curve, the gas concentration increase trend of each grid is identified. Specifically, the corrected gas concentration time variation curve is subjected to trend analysis, such as using linear regression, moving average and other methods; a rising trend threshold is preset, and when the slope of the corrected curve exceeds the threshold, it is identified that the gas concentration increase trend of the corresponding grid is too fast; when the slope of the corrected curve does not exceed the threshold, it is identified that the gas concentration increase trend of the corresponding grid is within a controllable range.

[0098] Based on the gas concentration increase trend of each grid, an early warning is issued for the gas concentration in the detection area. That is, when the identification result shows that the gas concentration of a grid increases too fast, an alarm is issued to warn the gas concentration in the detection area.

[0099] Step 108: Based on the gas concentration plane distribution map, manually inspect the locations where the gas concentration exceeds the limit.

[0100] It can be understood that in this embodiment, an upper limit threshold of gas concentration is set, and based on the gas concentration values ​​in the gas concentration plane distribution map, the locations where the gas concentration exceeds the upper limit threshold are identified, and based on the locations where the gas concentration exceeds the limit, inspectors are further arranged to conduct manual inspections of these areas to confirm whether there are actual safety risks.

[0101] Furthermore, in order to ensure the safety of the inspectors, in this embodiment, after step 108, the following steps are also included:

[0102] Step 110, using a handheld inspection device to monitor the acceleration and angular velocity during the inspection process in real time;

[0103] Step 112: Based on the acceleration and angular velocity during the inspection process, a safety monitoring alarm is issued to the inspector.

[0104] Specifically, in this embodiment, in step 112, based on the acceleration and angular velocity during the inspection process, a safety monitoring alarm is performed on the inspector, including:

[0105] Based on the angular velocity during the inspection process, determine the magnitude of the angular velocity change;

[0106] The change range of the angular velocity is compared with the first preset threshold. If the change range of the angular velocity is less than the first preset threshold, it is determined that the inspector is currently safe. If the change range of the angular velocity is greater than or equal to the first preset threshold, the following steps are performed:

[0107] Preset the window period and determine the magnitude of acceleration change within the window period;

[0108] The acceleration change amplitude is compared with the second preset threshold. If the acceleration change amplitude is less than the second preset threshold, it is judged that the inspector is currently safe. If the acceleration change amplitude is greater than or equal to the second preset threshold, it is judged that the inspector has fallen and a safety alarm for the inspector is issued.

[0109] For example, in this embodiment, the implementation method of performing safety monitoring and alarming on inspectors based on acceleration and angular velocity during the inspection process is:

[0110] First, the attitude sensor installed in the handheld patrol device collects the acceleration and angular velocity values ​​of the accelerometer and gyroscope in real time, which is used as the basis for monitoring the patrolman's motion status. Assume that the acquisition frequency of the attitude sensor is 100Hz, the cutoff frequency of the low-pass filter is set to 5Hz, and the collected signal is filtered and corrected, using Kalman filtering, where the dynamic noise covariance matrix Q value is 1e-5. When the angular velocity change is collected, determine the change amplitude of the angular velocity;

[0111] Secondly, the change of angular velocity is compared with the first preset threshold (taking 50° / s as an example). If the change of angular velocity is less than 50° / s, it is judged that no fall or other conditions have occurred, the inspector is currently safe, and the handheld inspector returns to the gas detection data collection process; if the change of angular velocity is greater than or equal to 50° / s, the window period is set to 2s, and the change of acceleration within 2s is detected;

[0112] Next, the acceleration change amplitude is compared with the second preset threshold (taking 0.8G as an example). If the acceleration change amplitude is greater than or equal to 0.8G, it is judged that the inspector has a safety risk event such as falling, and the handheld inspector will immediately send a safety alarm of the inspector to the host; if the acceleration change amplitude is less than 0.8G, it is judged as a normal activity event of the inspector.

[0113] Furthermore, the data of the change amplitude of the angular velocity and the change amplitude of the acceleration can be added to the training samples of the support vector machine and marked as normal activities (where the training coefficient C=1, gamma=0.1) to assist in the dynamic safety monitoring of the inspectors. By using the previous data of the change amplitude of the acceleration and the change amplitude of the acceleration to construct a support vector machine, the reliability of the handheld patrol device's dynamic safety monitoring alarm for the inspectors can be improved.

[0114] It can be understood that only an example of safety monitoring and alarming of inspectors based on acceleration and angular velocity during the inspection process is given here. The specific first preset threshold, second preset threshold and window period can be adjusted according to actual needs to adapt to different toxic and harmful gas monitoring requirements and dangerous working environment conditions. This embodiment does not limit this.

[0115] This embodiment combines the two real-time monitoring parameters of acceleration and angular velocity as the basis for dynamic safety monitoring of inspectors, and can more comprehensively evaluate the safety status of inspectors. Specifically, the amplitude of change in angular velocity reflects whether the inspector suddenly loses balance, and the amplitude of change in acceleration is used to further confirm whether the inspector has encountered safety risk events such as collision or fall. The layered threshold judgment method is used to reduce the probability of misjudgment, improve the accuracy and reliability of dynamic safety monitoring alarms for inspectors, and make timely warnings when inspectors are in danger, which helps rescue personnel to formulate rescue plans in a timely manner and improve the safety factor of inspectors.

[0116] See also Figure 2 , Figure 2 A schematic diagram of the structure of a gas detection system for a hazardous workplace is provided as an embodiment of this specification.

[0117] like Figure 2As shown, the gas detection system of the hazardous workplace may at least include a gas detector host, several gas detector slaves and a handheld patrol device. The gas detector host is located outside the detection area, and each gas detector slave is located at a detection point within the detection area. A communication network link is formed between the gas detector host, each gas detector slave and the handheld patrol device, wherein:

[0118] The gas detector slave is used to obtain the gas concentration and location information of the corresponding detection point in real time;

[0119] The gas detector host is used to divide the entire detection area into several grids, calculate the concentration density of each grid based on the gas concentration and location information of each detection point using the kernel density analysis method, and construct a gas concentration plane distribution map of the entire detection area based on the concentration density of all grids;

[0120] The handheld inspection device is used to be carried by inspectors into the detection area to conduct manual inspections of locations where gas concentration exceeds the limit based on the gas concentration plane distribution map.

[0121] It is understood that the gas detector host, each gas detector slave and the handheld patrol device form a communication network link. Figure 2 It means that each gas detector slave is connected to each other in communication, each gas detector slave is connected to the gas detector host in communication, each gas detector slave is connected to the handheld patrol device in communication, the gas detector host is connected to the handheld patrol device in communication, and the gas detector host can also be connected to the external network in communication.

[0122] Specifically, within the communication networking link, data is exchanged between multiple slaves and alarm information is synchronized. When an alarm occurs in a slave, other slaves, the host, and the handheld patrollers will all receive the alarm information. The host collects the gas concentration information and location information monitored by the slaves to construct a plane distribution map of the overall gas concentration in the detection area. It can comprehensively and real-time control the changing state of gas concentration within the dangerous operation range, and send the plane distribution map of the overall gas concentration in the detection area to the handheld patroller. In conjunction with the display screen of the handheld patroller, an internal detection network that can work collaboratively and is not affected by the detection environment is formed.

[0123] Common multi-parameter toxic and harmful gas detection systems are usually equipped with a small number of sensor modules to form an array, detect a small number of gas factors, and transmit data within a certain distance through wired and wifi / 4G methods. Multiple sensor modules are uniformly controlled and issued by uploading cloud platforms, which makes it difficult to achieve collaborative operation between multiple sensor modules. In addition, wireless signals will be attenuated due to obstacles (such as buildings, metal structures, etc.) during transmission. Especially in long-distance transmission, the signal attenuation is more serious, affecting the accuracy and integrity of the data. To this end, this embodiment forms a communication networking link to achieve "one point multiple transmission" regardless of environmental conditions, that is, each sensor module can be used as a backup for other modules. When a module fails or the signal is attenuated, other modules can continue to transmit data to ensure the continuity and integrity of the data, enhance data redundancy and reliability, or when environmental conditions change, the communication path can be adjusted through the communication networking link to bypass obstacles or areas with severe attenuation, maintain stable data transmission, and enhance the adaptability of the system. At the same time, each device can view the detection data and alarm status of each other, thereby improving the emergency monitoring coordination of the system.

[0124] Furthermore, in this embodiment, the handheld patroller includes a communication unit and a posture sensing unit, and several gas detector slaves and a gas detector host are all communicatively connected to the communication unit of the handheld patroller, and the posture sensing unit is connected to the communication unit, wherein the posture sensing unit is used to monitor the acceleration and angular velocity during the inspection process in real time, and based on the acceleration and angular velocity during the inspection process, perform safety monitoring and alarm on the inspectors, and send the alarm information to the gas detector host through the communication unit.

[0125] Optionally, the handheld patrol device can also be equipped with a call unit and a satellite positioning unit. The satellite positioning unit is used to obtain the patrolman's positioning information and provide a location basis for rescue personnel to formulate rescue plans. During the patrol inspection, if the posture sensor unit detects that the patrolman is in danger, the handheld patrol device will immediately issue an alarm, and the alarm information will be sent to the gas detector host through the communication unit in a timely manner. After receiving the alarm information, the host can inquire about the patrolman's situation through the call module.

[0126] It can be understood that other technical concepts of a gas detection system for a hazardous workplace provided in this embodiment are similar to the technical concepts of the aforementioned gas detection method for a hazardous workplace, and this embodiment will not be repeated here.

[0127] In another embodiment of the present specification, the gas detector slave may include at least an infrared detection unit and a sensor detection unit. The infrared detection unit is used to obtain gas imaging information of the corresponding detection point in real time to provide early warning of the gas concentration, and the sensor detection unit is used to obtain the gas concentration and position information of the corresponding detection point in real time.

[0128] It is understandable that during the real-time detection process, the infrared detection unit and the sensor detection unit perform monitoring simultaneously.

[0129] Furthermore, the gas detector slave can also be equipped with a video unit and a satellite positioning unit, which can provide real-time visual monitoring of the accident scene when an early warning occurs.

[0130] Specifically, the gas detector slave can be used to perform the following operations:

[0131] If the infrared detection unit gives priority to issuing an early warning, taking the infrared thermal imaging camera as an infrared detection unit as an example, when the infrared thermal imaging camera detects a gas leak in a limited space based on the real-time gas imaging information of the corresponding detection point, the slave where the infrared thermal imaging camera is located issues an early warning and sends it to the host and the handheld inspection device synchronously, and the slave automatically connects to the video unit of the host, so that the host can grasp the on-site situation through the video unit as soon as it receives the early warning;

[0132] If the subsequent sensor detection unit does not issue an alarm, and the infrared detection unit does not detect a continuous warning, the warning state is automatically eliminated;

[0133] If the subsequent sensor detection unit does not sound an alarm, but the infrared detection unit continues to sound an alarm, the warning status of the slave device needs to be manually eliminated based on the judgment of the on-site inspector;

[0134] If the subsequent sensor detection unit issues an alarm for over-limit concentration, the alarm information will be synchronized between each slave, and all slave alarm information will be summarized on the host and handheld inspection device. The alarm information requires the inspector to manually eliminate the alarm status of the slave according to the on-site situation. After an over-limit alarm occurs, the host will provide a reliable and accurate reference for rescue and evacuation work based on the real-time updated gas concentration plane distribution map of the detection area.

[0135] It is understandable that for some gases, the corresponding electrochemical sensors may have technical problems of slow response, or because the detection area is large, resulting in a long gas diffusion time, the electrochemical sensor may actually leak in a short period of time, but the sensor has no corresponding alarm. To this end, this embodiment sets an infrared detection unit and a sensor detection unit in the gas detector slave at the same time. Since the infrared imaging response is faster than the electrochemical sensor concentration response speed, some trace gas leaks can be discovered early, and even if the gas concentration is low, an early warning can be quickly issued, and an additional layer of protection is added before the sensor detection unit detects the gas concentration, thereby providing more accurate toxic and harmful gas detection results, which is more reliable and timely than the traditional single sensor detection method, and has higher safety. Moreover, by adding an infrared detection unit to cooperate with the sensor detection unit, while detecting the gas leakage concentration, it is also possible to obtain the location information of the specific gas leakage point or leakage source, which is very helpful for tracing the leaked gas.

[0136] The above description is only the preferred embodiment disclosed in this application and the description of the technical principle used. Those skilled in the art should understand that the scope of protection involved in this disclosure is not limited to the technical solution formed by a specific combination of the above technical features, but also should cover other technical solutions formed by any combination of the above technical features or their equivalent features without departing from the above disclosed concept. For example, the above features are replaced with the technical features with similar functions disclosed in this disclosure (but not limited to) to form a technical solution.

[0137] In addition, although each operation is described in a specific order, this should not be understood as requiring these operations to be performed in the specific order shown or in a sequential order. Under certain circumstances, multitasking and parallel processing may be advantageous. Similarly, although some specific implementation details are included in the above discussion, these should not be interpreted as limiting the scope of the present disclosure. Some features described in the context of a separate embodiment can also be implemented in a single embodiment in combination. On the contrary, the various features described in the context of a single embodiment can also be implemented in multiple embodiments individually or in any suitable sub-combination mode.

Claims

1. A method for detecting gas in a hazardous workplace, characterized in that: The following steps are involved: Step 102, obtaining the gas concentration and location information of each detection point in real time; Step 104, dividing the entire detection area into a plurality of grids, and using a kernel density analysis method to calculate the concentration density of each grid based on the gas concentration and position information of each detection point; Step 106, constructing a gas concentration plane distribution map of the entire detection area based on the concentration density of all grids; Step 108, based on the gas concentration plane distribution map, manually inspect the location where the gas concentration exceeds the limit; In step 104, based on the gas concentration and position information of each detection point, the concentration density of each grid is calculated using a kernel density analysis method, including: Get the center point position of each grid; Based on the position of the center point of the grid and the position information of each detection point, the x-axis distance and y-axis distance between the center point of each grid and each detection point are calculated and recorded as the distance data of the grid; Use Silverman's rule of thumb to determine bandwidth; Based on the determined bandwidth and the distance data of each grid, the Gaussian kernel function is used to calculate the weight between the grid center point and each detection point; The weight between the grid center point and each detection point is multiplied by the gas concentration at each detection point to obtain the weighted concentration between the grid center point and each detection point; All weighted concentrations are averaged to obtain the concentration density of the grid.

2. A method for detecting gas in a hazardous workplace as claimed in claim 1, characterized in that: After step 106 and before step 108, the method further includes: Generate a gas concentration time variation curve of each grid based on the concentration density of each grid within a preset time period; Periodically correcting the gas concentration time variation curve; Based on the corrected gas concentration time variation curve, identify the gas concentration increase trend of each grid; Based on the increasing trend of gas concentration in each grid, an early warning is issued for the gas concentration in the detection area.

3. A method for detecting gas in a hazardous workplace as claimed in claim 1, characterized in that: After step 108, the following steps are also included: Step 110, using a handheld inspection device to monitor the acceleration and angular velocity during the inspection process in real time; Step 112: Based on the acceleration and angular velocity during the inspection process, a safety monitoring alarm is issued to the inspector.

4. A method for detecting gas in a hazardous workplace as claimed in claim 3, characterized in that: In step 112, based on the acceleration and angular velocity during the inspection process, a safety monitoring alarm is performed on the inspector, including: Based on the angular velocity during the inspection process, determine the magnitude of the angular velocity change; The change range of the angular velocity is compared with the first preset threshold. If the change range of the angular velocity is less than the first preset threshold, it is determined that the inspector is currently safe. If the change range of the angular velocity is greater than or equal to the first preset threshold, the following steps are performed: Preset the window period and determine the magnitude of acceleration change within the window period; The acceleration change amplitude is compared with the second preset threshold. If the acceleration change amplitude is less than the second preset threshold, it is judged that the inspector is currently safe. If the acceleration change amplitude is greater than or equal to the second preset threshold, it is judged that the inspector has fallen and a safety alarm for the inspector is issued.

5. A gas detection system for hazardous workplaces, characterized in that: It includes a gas detector host, several gas detector slaves and a handheld patrol device. The gas detector host is located outside the detection area, and each gas detector slave is located at a detection point in the detection area. A communication network link is formed between the gas detector host, each gas detector slave and the handheld patrol device. The gas detector slave is used to obtain the gas concentration and location information of the corresponding detection point in real time; The gas detector host is used to divide the entire detection area into several grids, calculate the concentration density of each grid based on the gas concentration and location information of each detection point using the kernel density analysis method, and construct a gas concentration plane distribution map of the entire detection area based on the concentration density of all grids; The handheld inspection device is used to be carried by inspectors into the detection area to conduct manual inspections of locations where gas concentration exceeds the limit based on the gas concentration plane distribution map; Among them, the gas detector host is used to perform the following steps: Get the center point position of each grid; Based on the position of the center point of the grid and the position information of each detection point, the x-axis distance and y-axis distance between the center point of each grid and each detection point are calculated and recorded as the distance data of the grid; Use Silverman's rule of thumb to determine bandwidth; Based on the determined bandwidth and the distance data of each grid, the Gaussian kernel function is used to calculate the weight between the grid center point and each detection point; The weight between the grid center point and each detection point is multiplied by the gas concentration at each detection point to obtain the weighted concentration between the grid center point and each detection point; All weighted concentrations are averaged to obtain the concentration density of the grid.

6. A gas detection system for hazardous workplaces as claimed in claim 5, characterized in that: The handheld patroller includes a communication unit and a posture sensing unit. Several gas detector slaves and a gas detector host are all communicatively connected to the communication unit of the handheld patroller. The posture sensing unit is connected to the communication unit. The posture sensing unit is used to monitor the acceleration and angular velocity during the inspection process in real time. Based on the acceleration and angular velocity during the inspection process, safety monitoring and alarms are performed on the inspectors, and the alarm information is sent to the gas detector host through the communication unit.

7. A gas detection system for hazardous workplaces as claimed in claim 5, characterized in that: The gas detector slave includes an infrared detection unit and a sensor detection unit. The infrared detection unit is used to obtain gas imaging information of the corresponding detection point in real time to provide early warning of gas concentration. The sensor detection unit is used to obtain gas concentration and position information of the corresponding detection point in real time.

Images