Communication tower inspection method and device, storage medium and computer device

Abstract

The application discloses a communication tower inspection method and device, a storage medium and computer equipment, relates to the field of communication safety technology, and can increase the inspection range, improve the inspection effect and the inspection accuracy. The application comprises the following steps: acquiring gas concentration data collected by a variety of gas detection devices when an unmanned flight device flies around a target communication tower in real time, and acquiring meteorological data in real time; judging whether the search condition of a fire source is met based on the gas concentration data, if yes, taking the collection position of the gas concentration data as a starting point of fire source search, starting from the starting point of fire source search, gradually searching for the fire source position based on the meteorological data and the gas concentration data, and determining an initial fire source position based on the gradual search result; determining a plurality of collection points in a first preset area of the initial fire source position, acquiring a variety of gas concentration values of each collection point, and determining the fire source position in the first preset area based on the variety of gas concentration values. The application is suitable for the communication tower inspection scene.

Description

Technical Field

[0001] This invention relates to the field of communication security technology, and in particular to a method, apparatus, storage medium, and computer equipment for inspecting communication towers. Background Technology

[0002] Communication towers and their ancillary equipment (such as antennas, feeders, and power cabinets) are typically made of metal or plastic. High temperatures can cause structural deformation, equipment meltdown, and even tower collapse. Furthermore, the power supply to these towers relies on mains electricity or backup generators; a fire could damage transmission lines, transformers, or diesel storage tanks, leading to power outages. Therefore, to ensure the safety of communication towers, regular inspections are necessary.

[0003] Currently, cameras are typically installed on communication towers for inspection purposes, using the images captured by these cameras. However, the monitoring range of cameras installed on communication towers is relatively small, limiting the scope of inspections. Furthermore, the cameras can only collect image data; if the location of a fire is obstructed by trees or other objects, it is difficult to detect the fire using image data alone, resulting in poor inspection effectiveness and low accuracy. Summary of the Invention

[0004] This invention provides a method, apparatus, storage medium, and computer equipment for inspecting communication towers, which mainly increases the inspection range, improves the inspection effect, and enhances the inspection accuracy.

[0005] According to a first aspect of the present invention, a method for inspecting communication towers is provided for an unmanned aerial vehicle (UAV) device, the UAV device being equipped with various gas detection devices, including:

[0006] The system acquires gas concentration data collected by the various gas detection devices during the flight of the unmanned aerial vehicle around the target communication tower in real time, and also acquires meteorological data around the target communication tower in real time.

[0007] Based on the gas concentration data, it is determined whether the search conditions for the fire source location are met. If so, the location where the gas concentration data was collected is taken as the starting point for the fire source search. Based on the real-time meteorological data and the real-time gas concentration data, the fire source location is searched step by step from the starting point. The initial fire source location is determined based on the search results.

[0008] Multiple collection points are determined within a first preset area centered on the initial fire source location, and the concentration values ​​of various gases corresponding to each collection point are obtained. Based on the concentration values ​​of various gases, the fire source location is determined within the first preset area.

[0009] Optionally, the unmanned aerial vehicle is also equipped with an image acquisition device. After determining the location of the fire source within the first preset area based on the diverse gas concentration values, the method further includes:

[0010] The image acquisition device acquires multi-view images of the target communication tower, and based on the multi-view images, the target communication tower is reconstructed in three dimensions to obtain a three-dimensional image of the target communication tower. Based on the three-dimensional image, the tower type of the target communication tower is determined.

[0011] The fire source location is captured by the image acquisition device, and the fire hazard level at the fire source location is determined based on the tower type and the fire point image.

[0012] Optionally, the step of progressively searching for the fire source location based on real-time collected meteorological data and real-time collected gas concentration data, starting from the fire source search starting point, and determining the initial fire source location based on the progressive search results, includes:

[0013] Multiple initial flight positions are determined within a second preset area centered on the fire source search starting point. The unmanned aerial vehicle is controlled to fly to each initial flight position to collect location gas concentration data and location meteorological data. A virtual flying body is configured at each initial flight position. The location meteorological data includes wind direction data and wind speed data.

[0014] The dynamic flight step size of the virtual flying body at the corresponding initial flight position is determined based on the gas concentration data at the location, the gas concentration correction factor at the corresponding initial flight position is determined based on the wind speed data, and the flight direction correction factor of the virtual flying body at the corresponding initial flight position is determined based on the wind direction data.

[0015] Based on the gas concentration correction factor, the position gas concentration data of the corresponding initial flight position is corrected to obtain the corrected gas concentration data for each initial flight position. In each initial flight position, multiple initial flight positions with corrected gas concentration data greater than a preset concentration threshold are selected as candidate flight guidance positions. Based on the position coordinates and corrected gas concentration data of each candidate flight guidance position, the global flight guidance position is determined.

[0016] The guidance position change of the virtual flying body is determined based on the global flight guidance position and the initial flight position. The random position change of the virtual flying body is determined based on the dynamic flight step size. The guidance position change and the random position change are corrected based on the flight direction correction factor to obtain the corrected guidance position change and the corrected random position change.

[0017] Based on the corrected guide position change and the corrected random position change, each virtual flying body is controlled to perform the first round of fire source location search flight. The position of each virtual flying body after the first round of flight is taken as the new initial flight position. New location gas concentration data and new location meteorological data are determined at each new initial flight position. Based on the new location gas concentration data and new location meteorological data, new guide position change and new random position change are determined. Based on the new guide position change and new random position change, each virtual flying body is controlled to perform multiple rounds of fire source location iterative search flight. The final position reached by each virtual flying body after the last round of fire source location search flight that meets the iterative search flight conditions is determined. The initial fire source position is determined based on each final position.

[0018] Optionally, determining the location of the ignition source within the first preset area based on the diverse gas concentration values ​​includes:

[0019] Acquire meteorological data of the collection locations corresponding to the concentration values ​​of the various gases, draw the pollution plume of the various gases at each collection point based on the concentration values ​​of the various gases at each collection point, and determine the shape information of the pollution plume of the various gases. Based on the meteorological data of each collection point and the shape information of the pollution plume of the various gases, determine the first initial fire source area in the first preset area.

[0020] For each collection point, the ratio of each gas concentration value is determined, and the ratio is matched with the standard gas concentration ratio in the preset fire source fingerprint database. Based on the matching result, the second initial fire source area is determined in the first preset area.

[0021] A gas concentration distribution map is drawn based on the diverse gas concentration values ​​of each of the collection points, and a meteorological flow field map is drawn based on the meteorological data collected at each of the collection points. The gas concentration distribution map and the meteorological flow field map are then superimposed, and a third initial fire source area is determined within the first preset area based on the superposition result.

[0022] Based on the first initial fire source area, the second initial fire source area, and the third initial fire source area, a fire source area is determined within the first preset area, and the fire source location is determined within the fire source area.

[0023] Optionally, determining the location of the fire source in the fire source area includes:

[0024] In the fire source area, a hypothetical fire source location is determined, the actual meteorological data corresponding to the hypothetical fire source location is determined, and based on the actual meteorological data, a monitoring point is determined with the hypothetical fire source location as a reference, and the actual concentration data of various gases at the monitoring point is determined.

[0025] A Gaussian diffusion model is obtained, and the assumed fire source location is input into the Gaussian diffusion model to predict the concentration of various gases at the monitoring point, thereby obtaining the predicted concentration of various gases at the monitoring point.

[0026] Based on the difference between the actual and predicted gas concentrations, it is determined whether the hypothetical fire source location needs to be adjusted. If so, the hypothetical fire source location is iteratively adjusted until the difference between the actual and predicted gas concentrations at the last adjusted hypothetical fire source location is less than a preset difference threshold. The hypothetical fire source location after the last adjustment is then taken as the first fire source location.

[0027] In the fire source area, the sampling point corresponding to the maximum concentration of diverse gases is determined, and the historical meteorological data of the sampling point in the past preset time period is obtained. Based on the historical meteorological data, the particle backtracking direction of the virtual particles corresponding to the diverse gases is determined.

[0028] The virtual particle is simulated to backtrack in reverse time according to the particle backtracking direction. During the reverse time backtracking process, the particle backtracking trajectory is determined, and the intersection of the particle backtracking trajectory is taken as the location of the second fire source.

[0029] The fire source area is divided into a three-dimensional grid, and the grid-specific gas concentration data and meteorological data are acquired in real time. Based on the grid meteorological data, the transport direction, transport speed and diffusion coefficient of the various gases in the three-dimensional grid are determined.

[0030] Based on the transport direction, the transport speed, and the diffusion coefficient, a multi-gas transport simulation and a chemical transformation simulation are performed in the three-dimensional grid. Based on the transport simulation results and the chemical transformation simulation results, the spatiotemporal distribution field of the multi-gas is determined, and the location of the third fire source is determined based on the spatiotemporal distribution field of the multi-gas.

[0031] Based on the first fire source location, the second fire source location, and the third fire source location, the fire source location is determined in the fire source area.

[0032] Optionally, after determining the location of the ignition source within the first preset area based on the diverse gas concentration values, the method further includes:

[0033] Obtain the locations of potential fire sources around the target communication tower in the preset map, and obtain fire source monitoring data of potential fire sources at the locations of potential fire sources.

[0034] The fire source location is overlaid with the potential fire source locations in a preset map, and the fire source location is verified based on the overlay result and the fire source monitoring data.

[0035] Optionally, the method further includes:

[0036] The location information of the unmanned aerial vehicle is acquired in real time, and the location information and the gas concentration data acquired in real time are transmitted back in real time. Based on the real-time transmitted location information and gas concentration data, the raster map is colored to different degrees to obtain a raster map with location information and gas concentration data.

[0037] According to a second aspect of the present invention, a communication tower inspection device is provided, applied to an unmanned aerial vehicle (UAV), wherein the UAV is equipped with various gas detection devices, including:

[0038] The acquisition unit is used to acquire in real time the gas concentration data collected by the multi-gas detection device during the flight of the unmanned aerial vehicle around the target communication tower, and to acquire in real time the meteorological data around the target communication tower.

[0039] The search unit is used to determine whether the search conditions for the fire source location are met based on the gas concentration data. If so, the collection location of the gas concentration data is used as the starting point for the fire source search, and the fire source location is searched step by step from the starting point based on the real-time meteorological data and the real-time gas concentration data. The initial fire source location is determined based on the step-by-step search results.

[0040] The determining unit is used to determine multiple collection points within a first preset area centered on the initial fire source location, and to obtain the concentration values ​​of various gases corresponding to each collection point, and to determine the fire source location within the first preset area based on the concentration values ​​of various gases.

[0041] According to a third aspect of the present invention, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the above-described communication tower inspection method.

[0042] According to a fourth aspect of the present invention, a computer device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the above-described communication tower inspection method.

[0043] According to the present invention, a method, apparatus, storage medium, and computer equipment for inspecting communication towers are provided. Compared with the current method of inspecting communication towers by capturing images from cameras installed on the towers, the present invention acquires gas concentration data collected in real time by an unmanned aerial vehicle (UAV) flying around the target communication tower using a variety of gas detection devices, and also acquires meteorological data around the target communication tower in real time. Based on the gas concentration data, it determines whether the search conditions for a fire source location are met. If so, the location where the gas concentration data was collected is taken as the starting point for the fire source search. Based on the real-time meteorological data and the real-time gas concentration data, the search for the fire source location is performed step by step from the starting point. The initial fire source location is determined based on the step-by-step search results. Finally, multiple collection points are determined within a first preset area centered on the initial fire source location, and the corresponding variety of gas concentration values ​​are acquired for each collection point. Based on the variety of gas concentration values, the fire source location is determined within the first preset area. Therefore, inspecting communication towers by using unmanned aerial vehicles to fly around them can increase the inspection range; searching for fires using real-time gas concentration data can avoid visual obstruction limitations and can detect potential fire sources caused by gas leaks that are not easily detected in the early stages, thereby improving the inspection effect of communication towers; and determining the final fire source location by gradually searching and analyzing the initial fire source location can improve the accuracy of fire source location determination. Attached Figure Description

[0044] The accompanying drawings, which are included to provide a further understanding of the invention and form part of this application, illustrate exemplary embodiments of the invention and, together with their description, serve to explain the invention and do not constitute an undue limitation thereof. In the drawings:

[0045] Figure 1 A flowchart of a communication tower inspection method provided by an embodiment of the present invention is shown;

[0046] Figure 2 This invention provides a flowchart of another communication tower inspection method according to an embodiment of the invention.

[0047] Figure 3 This diagram illustrates the structure of a communication tower inspection device according to an embodiment of the present invention.

[0048] Figure 4 This invention provides a schematic diagram of the structure of another communication tower inspection device according to an embodiment of the invention.

[0049] Figure 5 A schematic diagram of the physical structure of a computer device provided in an embodiment of the present invention is shown. Detailed Implementation

[0050] The present invention will be described in detail below with reference to the accompanying drawings and embodiments. It should be noted that, unless otherwise specified, the embodiments and features described in the present application can be combined with each other.

[0051] Currently, the method of inspecting iron towers by using images captured by cameras fixed on the towers has limitations in terms of vision. In addition, the camera device can only collect image data. If the location of the fire is obstructed by trees or other objects, it is difficult to detect the fire through image data, resulting in poor inspection effectiveness and low inspection accuracy.

[0052] To address the aforementioned problems, this invention provides a method for inspecting communication towers, applied to unmanned aerial vehicles (UAVs). The UAVs are equipped with various gas detection devices, such as… Figure 1 As shown, the method includes:

[0053] 101. Real-time acquisition of gas concentration data collected by the unmanned aerial vehicle through various gas detection devices during its flight around the target communication tower, and real-time acquisition of meteorological data around the target communication tower.

[0054] Among them, unmanned aerial vehicles include various drones and other devices; the multi-gas detection equipment integrates olfactory sensors for various gases, including olfactory sensors for monitoring various gases such as PM2.5, PM10, temperature, humidity, carbon monoxide, carbon dioxide, methane, and hydrogen; meteorological data includes wind direction data, wind speed data, temperature data, humidity data, and air pressure data around the communication tower.

[0055] In this embodiment of the invention, the unmanned aerial vehicle (UAV) is equipped with various gas detection devices and image acquisition devices. The image acquisition devices are used to collect surrounding images during the UAV's flight to enable flight path planning and obstacle avoidance, thereby ensuring stable flight. While flying around the target communication tower, the UAV uses its onboard various gas detection devices to collect real-time data on the concentration of various gases and meteorological data. This data is then used to search for fires. This embodiment of the invention uses real-time monitored gas concentration and meteorological data for fire searching, avoiding visual obstruction limitations and enabling the monitoring of potential fire sources caused by gas leaks that are not easily detected in the early stages, thus improving the effectiveness of communication tower inspections.

[0056] 102. Based on the gas concentration data, determine whether the search conditions for the fire source location are met. If so, take the gas concentration data collection location as the starting point for the fire source search, and based on the real-time collected meteorological data and real-time collected gas concentration data, start the fire source search from the starting point to gradually search for the fire source location, and determine the initial fire source location based on the gradual search results.

[0057] In this embodiment of the invention, a fire detection concentration threshold is pre-set for each gas according to actual needs. After the unmanned aerial vehicle (UAV) collects multi-gas concentration data in real time, it compares this multi-gas concentration data with the corresponding fire detection concentration threshold. If the concentration of a certain gas or multiple gases is greater than the corresponding fire detection concentration threshold, it can be determined that a suspected fire exists at the current moment, i.e., the search conditions for the fire source location are met. If the search conditions for the fire source location are met at the current moment, the collection location where the gas concentration data exceeds the fire detection concentration threshold is determined as the fire source search starting point. For example, if an olfactory sensor collects a carbon dioxide concentration exceeding 1% at coordinates (x1, y1, z1), then this coordinate location is taken as the fire source search starting point. Then, the fire source search path of the UAV is planned based on the fire source search starting point. For example, based on real-time wind direction data, the possible direction of fire spread is determined. If the fire source will spread along the wind direction, the search path should be planned preferentially along the direction of wind extension. Simultaneously, the gas concentration distribution around the fire source search starting point is analyzed. The search direction is determined based on the gas concentration gradient, and the search path can be planned in the direction of increasing gas concentration. For example, if a gradual increase in gas concentration is detected to the east of the fire source search starting point, the search path should be extended further eastward. Furthermore, wind speed affects the spread speed and range of the fire source, while terrain can obstruct or guide gas flow. When planning the search path, wind speed and terrain factors can be comprehensively considered, and the search step size and direction can be adjusted appropriately. For example, in cases of high wind speed, the search step size can be increased appropriately; when encountering obstacles such as mountains, the search should be continued around the obstacles. Thus, the search path for the unmanned aerial vehicle can be planned using the above method. The unmanned aerial vehicle (UAV) is then controlled to search step-by-step along the search path. During this process, various gas detection devices onboard the UAV collect real-time gas concentration and meteorological data from the surrounding environment. The collected data is analyzed to determine if the current location is closer to a fire source. If a significant increase in gas concentration is detected in a certain direction, the search path is adjusted, and the search continues in that direction. If the gas concentration does not change significantly or decreases, the search continues along the original planned path. The UAV then continues flying along the adjusted search path, repeating the gas concentration detection and analysis process to gradually narrow the search area until the location with the highest gas concentration is found as the initial fire source location. This embodiment of the invention starts from a single point and gradually narrows the search area based on clues such as gas concentration and meteorological data. Compared to searching a large area, this method can more quickly focus on areas where a fire source may exist, reducing unnecessary search workload and improving overall search efficiency.

[0058] 103. Within a first preset area centered on the initial fire source location, determine multiple collection points and obtain the concentration values ​​of various gases corresponding to each collection point. Based on the concentration values ​​of various gases, determine the fire source location within the first preset area.

[0059] In this embodiment of the invention, the first preset area can be set according to the fire monitoring location, environmental information, etc. For example, a circular area with a radius of n meters centered on the initial fire source location can be used as the first preset area. Alternatively, based on wind direction, wind speed, and the fire spread speed, the first preset area can be set as a square area with a side length of 20-50 meters. When setting multiple collection points within the first preset area, the collection points can follow the principles of uniform distribution and density in key areas. Uniform distribution ensures comprehensive monitoring of the entire area, while density in key areas allows for more detailed collection of data from areas that may be closer to the fire source or where gas concentration changes significantly. The number of collection points can be determined based on the size and complexity of the first preset area. For example, a smaller first preset area can have fewer collection points, while a larger first preset area can have more. It should be noted that the above process for determining the first preset area is only illustrative and does not specifically limit the embodiments of the invention. Then, for each collection point, various gas concentration values ​​are collected, and a distribution map of various gas concentration values ​​and spatial location is constructed. Finally, the fire source location is determined by analyzing the distribution map of various gas concentrations. By analyzing diverse gas concentration distribution maps, the correlation between different gas concentrations and the potential relationship between gas concentration and ignition source location are explored. For example, the concentrations of carbon monoxide and carbon dioxide typically show a high correlation near the ignition source, gradually decreasing with increasing distance. The ignition source location can be determined through correlation analysis. This embodiment of the invention improves the accuracy of ignition source location determination by progressively searching to determine the initial ignition source location and then analyzing the initial location to determine the final location.

[0060] According to the present invention, a method for inspecting communication towers, compared with the current method of inspecting communication towers by capturing images from cameras installed on the towers, the present invention acquires gas concentration data collected in real time by an unmanned aerial vehicle (UAV) flying around the target communication tower using a variety of gas detection devices, and also acquires meteorological data around the target communication tower in real time. Then, based on the gas concentration data, it is determined whether the search conditions for a fire source location are met. If so, the location where the gas concentration data was collected is taken as the starting point for the fire source search. Based on the real-time meteorological data and the real-time gas concentration data, the fire source location is searched step by step from the starting point. The initial fire source location is determined based on the step-by-step search results. Finally, multiple collection points are determined within a first preset area centered on the initial fire source location, and the corresponding variety of gas concentration values ​​are acquired for each collection point. Based on the variety of gas concentration values, the fire source location is determined within the first preset area. Therefore, using unmanned aerial vehicles (UAVs) to inspect communication towers by flying around them can increase the inspection range; using real-time monitoring of gas concentration data for fire detection can avoid visual obstruction limitations and can detect potential fire sources caused by gas leaks that are not easily detected in the early stages, thereby improving the inspection effect of communication towers; by gradually searching to determine the initial fire source location and then analyzing the initial fire source location to determine the final fire source location, the accuracy of fire source location determination can be improved.

[0061] Furthermore, to better illustrate the above-described process of inspecting communication towers, and as a refinement and extension of the above embodiments, this invention provides another method for inspecting communication towers, applied to an unmanned aerial vehicle (UAV). The UAV is equipped with various gas detection devices, such as... Figure 2 As shown, the method includes:

[0062] 201. Real-time acquisition of gas concentration data collected by the unmanned aerial vehicle through various gas detection devices during its flight around the target communication tower, and real-time acquisition of meteorological data around the target communication tower.

[0063] Specifically, the unmanned aerial vehicle (UAV) is equipped with a six-way binocular visual obstacle avoidance system (front, rear, left, right, up, and down) and an infrared perception obstacle avoidance system; it also features a GPS positioning module; and it is equipped with various gas detection devices for collecting information on PM2.5, PM10, temperature, humidity, carbon monoxide, and carbon dioxide, using these devices to collect real-time data on the concentrations of these gases. Furthermore, the UAV is also equipped with a multi-functional image acquisition device that combines thermal infrared, laser rangefinder, zoom, and wide-angle sensors. The data collected by this image acquisition device can assist in the precise location of fire sources based on the diverse gas concentration data.

[0064] 202. Based on the gas concentration data, determine whether the search conditions for the fire source location are met. If so, take the gas concentration data collection location as the starting point for the fire source search, and based on the real-time collected meteorological data and real-time collected gas concentration data, start the fire source search from the starting point to gradually search for the fire source location, and determine the initial fire source location based on the gradual search results.

[0065] In this embodiment of the invention, if the gas concentration is greater than the gas concentration value under normal environmental conditions such as no fire, it can be determined that the search conditions for the fire source location are met. At this time, it is necessary to search for the fire source location step by step. Based on this, step 202 specifically includes: determining multiple initial flight positions in a second preset area centered on the fire source search starting point; controlling the unmanned aerial vehicle to fly to each initial flight position to collect the location gas concentration data and location meteorological data of each initial flight position; configuring a virtual flying body at each initial flight position, wherein the location meteorological data includes wind direction data and wind speed data; determining the dynamic flight step size of the virtual flying body at the corresponding initial flight position based on the location gas concentration data; determining the gas concentration correction factor of the corresponding initial flight position based on the wind speed data; determining the flight direction correction factor of the virtual flying body at the corresponding initial flight position based on the wind direction data; correcting the location gas concentration data of the corresponding initial flight position based on the gas concentration correction factor to obtain the corrected gas concentration data corresponding to each initial flight position; selecting multiple initial flight positions with corrected gas concentration data greater than a preset concentration threshold as candidate flight guidance positions at each initial flight position; and based on the location coordinates corresponding to each candidate flight guidance position and... The corrected gas concentration data is used to determine the global flight guidance position. Based on the global flight guidance position and the initial flight position, the guidance position change of the virtual flying body is determined. Based on the dynamic flight step size, the random position change of the virtual flying body is determined. The guidance position change and the random position change are corrected based on the flight direction correction factor, resulting in the corrected guidance position change and the corrected random position change. Based on the corrected guidance position change and the corrected random position change, each virtual flying body is controlled to perform the first round of fire source location search flight, and... The position of each virtual flying body after the first round of flight is used as the new initial flight position. New location gas concentration data and new location meteorological data are determined at each new initial flight position. Based on the new location gas concentration data and new location meteorological data, new guidance position change and new random position change are determined. Based on the new guidance position change and new random position change, each virtual flying body is controlled to perform multiple rounds of iterative search flight for fire source positions. The final position reached by each virtual flying body after the last round of fire source position search flight that meets the iterative search flight conditions is determined. The initial fire source position is determined based on each final position.

[0066] The second preset area is set according to actual needs, such as a circular area with a radius of m meters centered on the fire source search starting point; the preset concentration threshold is set according to actual needs.

[0067] Specifically, based on actual needs, multiple initial flight positions are determined within the second preset area, and a virtual flight body is configured for each initial flight position. First, the virtual flight bodies conduct a first round of search flight. At the start of the first round of search flight, an initial flight step size is randomly set for each virtual flight body. And based on the initial flight step size Each virtual flying body is determined according to the following formula. Dynamic flight step length :

[0068]

[0069] in, For virtual flight bodies The location-based concentration variation gradient. Based on the wind speed at the initial flight location. The gas concentration correction factor during the first round of flight is determined according to the following formula. :

[0070]

[0071] in, This is the gas correction factor. The flight direction correction factor for the first round of flight is then determined using the following formula. :

[0072]

[0073] in, Initial flight position The wind direction at the location is determined. Then, the gas concentration data for the initial flight position corresponding to each virtual flying object is multiplied by the corresponding gas concentration correction factor to obtain the corrected gas concentration data. During the first round of search, the corrected gas concentration data for each virtual flying object is sorted in descending order, and the initial flight positions corresponding to the top n concentration data are determined as candidate flight guidance positions. The global flight guidance position is then determined according to the following formula:

[0074]

[0075] in, For global flight guidance position coordinates, k This serves as the initial flight position identifier. For the first k Corrected gas concentration data corresponding to each initial flight position For the first k The position coordinates of the initial flight position nThe total number of initial flight positions. The change in the guiding position of each virtual flight object is determined using the following formula. :

[0076]

[0077]

[0078]

[0079] in, For the first The coordinates of the initial flight position, For the first The change in gas concentration gradient at each initial flight position. Simultaneously, the change in gas concentration gradient for each virtual flight body is determined according to the following formula. random position change :

[0080]

[0081]

[0082]

[0083] in, For virtual flight bodies Dynamic flight stride, The angle is randomly determined. Then, the flight direction correction factor is multiplied by the guide position change and the random position change, respectively, to obtain the corrected guide position change and the corrected random position change. Next, the weighting coefficients corresponding to the guide position change and the random position change are determined. Based on these weighting coefficients, the corrected guide position change and the corrected random position change are weighted and summed to obtain the search position change for each virtual flying object. Then, each virtual flying object is controlled according to this search position change to perform the first round of fire source location search flight. Once a certain position is reached, it is used as the new initial flight position for the second round of search flight. At the beginning of the second round of search flight, gas concentration data and meteorological data at the new initial flight position are collected according to the same process as the first round. Based on the gas concentration data, the dynamic flight step size of the virtual flying object at the corresponding new initial flight position is determined. The gas concentration correction factor for the corresponding new initial flight position is determined based on wind speed data, and the flight direction correction factor for the virtual flying object at the corresponding new initial flight position is determined based on wind direction data. The position gas concentration data at the corresponding initial flight position is corrected based on the gas concentration correction factor to obtain the corrected guide position change and the corrected random position change for each virtual flying object. For each new initial flight position, the corrected gas concentration data is used. Multiple new initial flight positions with corrected gas concentration data greater than a preset concentration threshold (the preset concentration threshold varies for each round) are selected as new candidate flight guidance positions. Based on the position coordinates and corrected gas concentration data of each new candidate flight guidance position, a new global flight guidance position is determined. Based on the new global flight guidance position and the new initial flight position, the new guidance position change of the virtual flight body is determined. Based on the dynamic flight step size, the new random position change of the virtual flight body is determined. Based on the flight direction correction factor, the new guidance position change and the new random position change are corrected, resulting in the corrected new guidance position change and the corrected new random position change. Based on the new corrected guidance position change and the new corrected random position change, each virtual flight body is controlled to perform a second round of fire source location search flight. This process is repeated to control each virtual flight body to perform multiple rounds of iterative fire source location search flight until the iteration stop condition is met, such as the required number of iterations or a high global gas concentration data in consecutive iterations. Finally, the ultimate position reached by each virtual flying object after the last round of fire source location search flight is determined, and the initial fire source position is determined based on each ultimate position, such as using the centroid of each ultimate position as the initial fire source position. This embodiment of the invention improves the accuracy of fire source location determination by iteratively updating the group position, balancing global exploration and local development, and avoiding the pitfalls of getting trapped in local optima.

[0084] 203. Within a first preset area centered on the initial fire source location, determine multiple collection points and obtain the concentration values ​​of various gases corresponding to each collection point. Based on the concentration values ​​of various gases, determine the fire source location within the first preset area.

[0085] In this embodiment of the invention, after determining the initial ignition source location, it is also necessary to determine the final ignition source location based on the initial ignition source location. Therefore, step 203 specifically includes: acquiring meteorological data of the collection locations corresponding to the diverse gas concentration values; drawing a diverse gas concentration pollution plume for each collection point based on the diverse gas concentration values ​​of each collection point, and determining the shape information of the diverse gas concentration pollution plume; determining a first initial ignition source area within the first preset area based on the meteorological data of each collection point and the shape information of the diverse gas concentration pollution plume; and for each collection point, determining the ratio of each gas concentration value, and comparing the ratio with a preset value. The standard gas concentration ratios in the fire source fingerprint database are matched, and based on the matching results, a second initial fire source region is determined within the first preset region. A gas concentration distribution map is drawn based on the diverse gas concentration values ​​of each collection point, and a meteorological flow field map is drawn based on the collected meteorological data of each collection point. The gas concentration distribution map and the meteorological flow field map are superimposed, and a third initial fire source region is determined within the first preset region based on the superposition result. Based on the first initial fire source region, the second initial fire source region, and the third initial fire source region, a fire source region is determined within the first preset region, and the fire source location is determined within the fire source region.

[0086] Specifically, the multi-gas concentration values ​​and meteorological data collected at each sampling point are preprocessed, including interpolation and spatiotemporal alignment. For each sampling point, a dynamic concentration curve is plotted with the multi-gas concentration values ​​as the ordinate and the time series as the abscissa. Combined with spatial coordinates, a three-dimensional concentration distribution heatmap is generated, thus forming a multi-gas concentration pollution plume. When the shape information of the multi-gas concentration pollution plume is a thin plume, the extension direction of the multi-gas concentration pollution plume is determined based on the collected meteorological data, and the direction of the ignition source is determined based on the extension direction. Based on the direction of the ignition source, a first initial ignition source area is determined within a first preset area. When the shape information is a diffuse cloud, the overlapping area of ​​the multi-gas diffuse cloud is determined, and the first initial ignition source area is determined within a first preset area based on the overlapping area. Specifically, in the case of a thin plume, the following formula is used to calculate the concentration of each sampling point. Wind direction consistency index :

[0087]

[0088] in, For collection points Various gas concentration values, For collection points Various gas concentration values, The total number of collection points. For collection points Wind direction data, for Average wind direction data from each collection point. Select the wind direction consistency index. The largest and first few sampling points are designated as target sampling points, and the vectors and directions of these target sampling points are used as the extension directions of the fine plume. The intersection of the extension direction with the boundary of the first preset area is used as a reference point, and the first initial ignition source area is delineated based on this reference point. When the shape information of the diverse gas concentration pollution plume is that it is a diffuse cloud, the overlapping area of ​​each gas diffuse cloud is determined, and this overlapping area is used as the first initial ignition source area.

[0089] Furthermore, the concentration ratios of various gases at different sampling points (e.g., SO2 / NO) were calculated. X (VOCs / CO). Compare these ratios with the standard gas concentration ratios in a pre-set fire source fingerprint database (such as SO2 / NO from coal-fired power plants). X The ratio is usually much higher than that of vehicle exhaust. The preset ignition source fingerprint database stores the concentration ratios of various gases in multiple areas (factories, enterprises). If the ratio of a certain area highly matches the gas concentration ratio of the corresponding gas at the sampling point, then such a source is likely to exist in that area, and this area is then designated as the second initial ignition source area.

[0090] Furthermore, the gas concentration distribution map based on diverse gas concentration values ​​and the meteorological flow field map based on collected meteorological data are overlaid to obtain an overlay map. In the overlay map, regions with gas concentrations greater than a set value are selected as candidate regions. For each candidate region, its centroid coordinates are calculated, and the streamline closest to the centroid in the meteorological flow field map is retrieved. If the angle between the direction of the streamline and the vector direction from the centroid to the boundary of the candidate region is greater than a preset angle threshold (wherein the preset angle threshold is set according to actual needs), then the candidate region is selected as the third initial ignition source region.

[0091] Furthermore, the fire source region can be determined based on at least one of the first, second, and third initial fire source regions. For example, if the overlapping area of ​​the first, second, and third initial fire source regions is taken as the fire source region, then the location of the fire source needs to be determined within this fire source region. Based on this, the method includes: determining a hypothetical fire source location in the fire source area; determining the actual meteorological data corresponding to the hypothetical fire source location; and, based on the actual meteorological data, determining a monitoring point with the hypothetical fire source location as a reference, and determining the actual multi-gas concentration data of the monitoring point; acquiring a Gaussian diffusion model, inputting the hypothetical fire source location into the Gaussian diffusion model to predict the multi-gas concentration of the monitoring point, and obtaining the predicted multi-gas concentration of the monitoring point; based on the difference between the actual multi-gas concentration data and the predicted multi-gas concentration, determining whether the hypothetical fire source location needs to be adjusted; if so, iteratively adjusting the hypothetical fire source location until the difference between the actual multi-gas concentration data and the predicted multi-gas concentration at the last adjusted hypothetical fire source location is less than a preset difference threshold, and taking the last adjusted hypothetical fire source location as the first fire source location; determining the sampling point corresponding to the maximum multi-gas concentration value in the fire source area, and acquiring the sampling point's... Historical meteorological data within a preset time period is used to determine the particle backtracking direction of virtual particles corresponding to various gases. The virtual particles are simulated to backtrack in reverse time according to the particle backtracking direction, and their backtracking trajectories are determined during this process. The intersection of these trajectories is used as the location of the second fire source. The fire source area is divided into a three-dimensional grid, and the grid-specific gas concentration data and meteorological data are acquired in real time. Based on the meteorological data, the transport direction, transport speed, and diffusion coefficient of the various gases in the three-dimensional grid are determined. Based on the transport direction, transport speed, and diffusion coefficient, the transport and chemical transformation simulations of the various gases are performed in the three-dimensional grid. Based on the transport simulation results and chemical transformation simulation results, the spatiotemporal distribution field of the various gases is determined, and the location of the third fire source is determined based on the spatiotemporal distribution field. Based on the locations of the first, second, and third fire sources, the location of the fire source is determined within the fire source area.

[0092] Specifically, firstly, a hypothetical fire source location is randomly determined in the fire source area, and the predicted concentration of various gases at the hypothetical fire source location is predicted using a Gaussian diffusion model. If the difference between the predicted concentration of various gases at the hypothetical fire source location and the actual concentration of various gases is large, the hypothetical fire source location is continuously adjusted until the difference between the predicted concentration of various gases at the adjusted hypothetical fire source location and the actual concentration of various gases is less than a preset difference threshold (the preset difference threshold is set according to actual needs). The hypothetical fire source location after the last adjustment is then taken as the first fire source location.

[0093] Meanwhile, based on the historical meteorological data of the sampling point with the maximum diversity gas concentration value within a preset time period (the preset time period is set according to actual needs), the particle backtracking direction of the virtual particles is determined. Then, starting from the sampling point corresponding to the maximum diversity gas concentration value, the simulated particles backtrack in reverse time according to the particle backtracking direction. The area where all backtracking trajectories converge is the location of the second fire source.

[0094] Meanwhile, the fire source area is divided into a three-dimensional grid based on an appropriate grid size. Based on real-time meteorological data and considering factors such as topography and building structure in the fire source area, the gas transport direction within the three-dimensional grid is analyzed. In simple terrain and open areas, the gas transport direction is mainly influenced by wind direction; however, in complex terrain (such as mountains, cities, canyons, etc.) or where buildings obstruct the flow, the gas transport direction may change. The gas transport velocity within the three-dimensional grid is determined based on wind speed from meteorological data. The diffusion coefficient is appropriately set based on gas stability and wind speed. Subsequently, an Eulerian method can be used to establish a gas transport model, treating the gas as a continuous medium and solving it on a fixed three-dimensional grid to simulate gas transport. Based on the types of gases involved in the fire and possible chemical reactions, a suitable chemical reaction mechanism is determined, and chemical transformation simulations of the gas are performed based on this mechanism. The results from gas transport simulations and chemical transformation simulations are integrated to analyze the concentration changes of different gases at different time points within each grid. Based on these concentration changes, the spatiotemporal distribution characteristics of the gases within the ignition source region are determined, i.e., the spatiotemporal distribution field. The concentration gradient changes of characteristic gases in the spatiotemporal distribution field are analyzed. For example, near the ignition source location, the concentration gradient of characteristic gases is usually large, indicating drastic concentration changes. By identifying areas with large concentration gradients, possible ignition source locations can be preliminarily determined, and these locations are designated as secondary ignition source locations.

[0095] Ultimately, any one of the first, second, and third fire source positions can be used as the final fire source position, or the final fire source position can be determined based on at least two of the first, second, and third fire source positions, such as using the centroid position of at least two of the first, second, and third fire source positions as the final fire source position.

[0096] Furthermore, after determining the location of the fire source, in order to improve the accuracy of the determination, it is also necessary to verify the location of the fire source. Based on this, the method includes: obtaining the potential fire source locations around the target communication tower in a preset map, and obtaining fire source monitoring data of the potential fire sources at the potential fire source locations; overlaying the fire source locations with the potential fire source locations in the preset map, and verifying the fire source locations based on the overlay result and the fire source monitoring data.

[0097] Potential ignition sources refer to areas prone to fire, such as power plants, chemical plants, and oil refineries. Ignition source monitoring data includes the concentration of pollutants around potential ignition sources and temperature changes on key components such as equipment surfaces, pipes, and cables at potential ignition sources.

[0098] Specifically, the location of the fire source is overlaid with potential fire source locations on a preset map. If the fire source location coincides with a power plant on the preset map, and the power plant's fire source monitoring data indicates that a fire is highly likely to have occurred, then the fire source location is determined to be accurate. If the power plant's fire source monitoring data indicates that a fire is unlikely to have occurred, then the fire source location is determined to be incorrect, and the fire source location needs to be repositioned. This embodiment of the invention, by verifying the fire source location, can further improve the accuracy of fire source location positioning.

[0099] 204. Acquire multi-view images of the target communication tower using image acquisition equipment, and reconstruct the target communication tower in three dimensions based on the multi-view images to obtain a three-dimensional image of the target communication tower. Determine the tower type based on the three-dimensional image.

[0100] Specifically, the target communication tower is photographed from various angles using image acquisition equipment, such as conducting a 720-degree three-dimensional surround inspection of the tower structure, photographing each key part, and generating multi-angle, multi-view ultra-high-definition images through background processing. A panoramic link is then created, and a three-dimensional model of the tower, i.e., a three-dimensional image, is constructed based on these multi-view images. Tower type recognition is then performed based on this three-dimensional image. In another embodiment of the invention, a preset tower type recognition model can be pre-constructed (using a sample dataset composed of multi-view images of sample towers with various tower type labels), and the multi-view images are input into the preset tower type recognition model for tower type recognition.

[0101] 205. Acquire fire point images of the fire source location using image acquisition equipment, and determine the fire hazard level at the fire source location based on the tower type and fire point images.

[0102] Specifically, after determining the location of the fire source, the image acquisition equipment on the unmanned aerial vehicle (UAV) is used to collect images of the fire at that location. These images are used to determine the fire's coverage area, and tower type identification is used to determine the impact of the fire around the tower on the surrounding environment. Finally, the fire hazard level is determined based on the fire's coverage area and its impact on the surrounding environment. A larger fire coverage area and a greater impact on the surrounding environment result in a higher fire hazard level, requiring appropriate measures to be implemented for firefighting.

[0103] In another embodiment of the present invention, during the safety inspection of the tower, the flight trajectory of the unmanned aerial vehicle and the monitored various gas concentration data can be transmitted back to the ground for display. Based on this, the method includes: acquiring the location information of the unmanned aerial vehicle in real time, transmitting the location information and the acquired gas concentration data back in real time, and performing different degrees of coloring processing on the raster map based on the transmitted location information and gas concentration data to obtain a raster map with location information and gas concentration data.

[0104] Specifically, this invention has developed a data cloud platform for unmanned aerial vehicles (UAVs), enabling real-time transmission and display of unmanned detection data. The platform uses a raster map as its base map, which has both satellite and vector map modes. By reading the UAV's onboard GPS information and coloring the raster map, and simultaneously reading diverse gas concentration data collected by the UAV and coloring the raster map in another style, the colored map clearly distinguishes the UAV's flight trajectory and gas concentration changes.

[0105] According to another communication tower inspection method provided by the present invention, compared with the current method of inspecting communication towers by taking images captured by cameras installed on the towers, the present invention acquires gas concentration data collected in real time by an unmanned aerial vehicle (UAV) flying around the target communication tower using a variety of gas detection devices, and acquires meteorological data around the target communication tower in real time; then, based on the gas concentration data, it is determined whether the search conditions for the fire source location are met. If so, the location where the gas concentration data was collected is taken as the starting point for the fire source search, and the fire source location is searched step by step from the starting point based on the real-time meteorological data and the real-time gas concentration data. The initial fire source location is determined based on the step-by-step search results; finally, multiple collection points are determined within a first preset area centered on the initial fire source location, and the corresponding variety of gas concentration values ​​are acquired for each collection point. Based on the variety of gas concentration values, the fire source location is determined within the first preset area. Therefore, using unmanned aerial vehicles (UAVs) to inspect communication towers by flying around them can increase the inspection range; using real-time monitoring of gas concentration data for fire detection can avoid visual obstruction limitations and can detect potential fire sources caused by gas leaks that are not easily detected in the early stages, thereby improving the inspection effect of communication towers; by gradually searching to determine the initial fire source location and then analyzing the initial fire source location to determine the final fire source location, the accuracy of fire source location determination can be improved.

[0106] Furthermore, as Figure 1 In a specific implementation, this invention provides a communication tower inspection device applied to an unmanned aerial vehicle (UAV). The UAV is equipped with various gas detection devices, such as... Figure 3 As shown, the device includes: an acquisition unit 31, a search unit 32, and a determination unit 33.

[0107] The acquisition unit 31 can be used to acquire gas concentration data collected by the multi-gas detection device during the flight of the unmanned aerial vehicle around the target communication tower in real time, and to acquire meteorological data around the target communication tower in real time.

[0108] The search unit 32 can be used to determine whether the search conditions for the fire source location are met based on the gas concentration data. If so, the collection location of the gas concentration data is taken as the starting point for the fire source search, and the fire source location is searched step by step from the starting point based on the real-time collected meteorological data and the real-time collected gas concentration data. The initial fire source location is determined based on the step-by-step search results.

[0109] The determining unit 33 can be used to determine multiple collection points within a first preset area centered on the initial fire source location, and obtain the diverse gas concentration values ​​corresponding to each collection point, and determine the fire source location within the first preset area based on the diverse gas concentration values.

[0110] In specific application scenarios, the unmanned aerial vehicle is also equipped with image acquisition equipment to determine the fire hazard level, such as... Figure 4 As shown, the device also includes a level division unit 34.

[0111] The classification unit 34 can be used to acquire multi-view images of the target communication tower through the image acquisition device, and to perform three-dimensional reconstruction of the target communication tower based on the multi-view images to obtain a three-dimensional image of the target communication tower. Based on the three-dimensional image, the tower type of the target communication tower can be determined. The fire point image of the fire source location can be acquired through the image acquisition device, and the fire hazard level at the fire source location can be determined based on the tower type and the fire point image.

[0112] In specific application scenarios, in order to gradually search for the location of the fire source, the search unit 32 includes a data acquisition module 321, a first determination module 322, and a correction module 323.

[0113] The acquisition module 321 can be used to determine multiple initial flight positions within a second preset area centered on the fire source search starting point, control the unmanned aerial vehicle to fly to each initial flight position to collect location gas concentration data and location meteorological data for each initial flight position, and configure a virtual flying body at each initial flight position, wherein the location meteorological data includes wind direction data and wind speed data.

[0114] The first determining module 322 can be used to determine the dynamic flight step size of the virtual flying body at the corresponding initial flight position based on the position gas concentration data, determine the gas concentration correction factor at the corresponding initial flight position based on the wind speed data, and determine the flight direction correction factor of the virtual flying body at the corresponding initial flight position based on the wind direction data.

[0115] The first determining module 322 can be specifically used to correct the position gas concentration data of the corresponding initial flight position based on the gas concentration correction factor, to obtain the corrected gas concentration data corresponding to each initial flight position, to select multiple initial flight positions whose corrected gas concentration data is greater than a preset concentration threshold as candidate flight guidance positions, and to determine the global flight guidance position based on the position coordinates corresponding to each candidate flight guidance position and the corrected gas concentration data.

[0116] The correction module 323 can be used to determine the change in the guidance position of the virtual flying body based on the global flight guidance position and the initial flight position, determine the change in the random position of the virtual flying body based on the dynamic flight step size, and correct the change in the guidance position and the change in the random position based on the flight direction correction factor, so as to obtain the corrected change in the guidance position and the corrected change in the random position.

[0117] The first determining module 322 can be specifically used to control each virtual flying body to perform a first round of fire source location search flight based on the corrected guide position change and the corrected random position change, and to take the position of each virtual flying body after the first round of flight as a new initial flight position, determine the new position gas concentration data and new position meteorological data at each new initial flight position, and determine the new guide position change and new random position change based on the new position gas concentration data and new position meteorological data, and control each virtual flying body to perform multiple rounds of fire source location iterative search flight based on the new guide position change and new random position change, determine the final position reached by each virtual flying body after the last round of fire source location search flight that meets the iterative search flight conditions, and determine the initial fire source position based on each final position.

[0118] In a specific application scenario, in order to determine the location of the fire source within a first preset area, the determining unit 33 includes a second determining module 331, a matching module 332, and a superimposing module 333.

[0119] The second determining module 331 can be used to acquire meteorological data of the collection location corresponding to the multi-gas concentration value, draw the multi-gas concentration pollution plume of the corresponding collection point based on the multi-gas concentration value of each collection point, and determine the shape information of the multi-gas concentration pollution plume. Based on the meteorological data of each collection point and the shape information of the multi-gas concentration pollution plume, the first initial fire source area is determined in the first preset area.

[0120] The matching module 332 can be used to determine the ratio of each gas concentration value for each collection point, and match the ratio with the standard gas concentration ratio in the preset fire source fingerprint database. Based on the matching result, a second initial fire source area is determined in the first preset area.

[0121] The overlay module 333 can be used to draw a gas concentration distribution map based on the diverse gas concentration values ​​of each of the collection points, draw a meteorological flow field map based on the collected meteorological data of each of the collection points, and overlay the gas concentration distribution map and the meteorological flow field map, and determine the third initial fire source area in the first preset area based on the overlay result.

[0122] The second determining module 331 can be specifically used to determine a fire source area within the first preset area based on the first initial fire source area, the second initial fire source area, and the third initial fire source area, and to determine the fire source location within the fire source area.

[0123] In a specific application scenario, to determine the location of a fire source in a fire source area, the second determining module 331 can be specifically used to determine a hypothetical fire source location in the fire source area, determine the actual meteorological data corresponding to the hypothetical fire source location, and based on the actual meteorological data, determine a monitoring point with the hypothetical fire source location as a benchmark, and determine the actual multi-gas concentration data of the monitoring point; acquire a Gaussian diffusion model, input the hypothetical fire source location into the Gaussian diffusion model to predict the multi-gas concentration of the monitoring point, and obtain the predicted multi-gas concentration of the monitoring point; based on the difference between the actual multi-gas concentration data and the predicted multi-gas concentration, determine whether the hypothetical fire source location needs to be adjusted; if so, iteratively adjust the hypothetical fire source location until the difference between the actual multi-gas concentration data and the predicted multi-gas concentration at the last adjusted hypothetical fire source location is less than a preset difference threshold, and take the hypothetical fire source location after the last adjustment as the first fire source location; determine the maximum multi-gas concentration value corresponding to the fire source area. The sampling points are selected, and historical meteorological data of the sampling points within a preset time period are obtained. Based on the historical meteorological data, the particle backtracking direction of the virtual particles corresponding to the various gases is determined. The virtual particles are simulated to backtrack in reverse time according to the particle backtracking direction. During the reverse time backtracking process, the particle backtracking trajectory is determined, and the intersection of the particle backtracking trajectory is taken as the second fire source location. The fire source area is divided into a three-dimensional grid, and the grid-specific gas concentration data and grid meteorological data corresponding to the three-dimensional grid are obtained in real time. Based on the grid meteorological data, the transport direction, transport speed, and diffusion coefficient of the various gases in the three-dimensional grid are determined. Based on the transport direction, the transport speed, and the diffusion coefficient, the transport and chemical transformation simulation of the various gases are performed in the three-dimensional grid. Based on the transport simulation results and the chemical transformation simulation results, the spatiotemporal distribution field of the various gases is determined, and the third fire source location is determined based on the spatiotemporal distribution field of the various gases. Based on the first fire source location, the second fire source location, and the third fire source location, the fire source location is determined in the fire source area.

[0124] In specific application scenarios, in order to verify the location of the fire source, the device also includes a verification unit 35.

[0125] The verification unit 35 can be used to obtain the location of potential fire sources around the target communication tower in a preset map, and obtain the fire source monitoring data of the potential fire sources at the potential fire source locations; overlay the fire source locations with the potential fire source locations in the preset map, and verify the fire source locations based on the overlay result and the fire source monitoring data.

[0126] In specific application scenarios, in order to display the trajectory of the unmanned aerial vehicle and gas concentration information in the grid map, the device also includes a coloring unit 36.

[0127] The coloring unit 36 ​​can be used to acquire the location information of the unmanned aerial vehicle in real time, transmit the location information and the gas concentration data acquired in real time, and perform different degrees of coloring processing on the raster map based on the real-time transmitted location information and gas concentration data to obtain a raster map with location information and gas concentration data.

[0128] It should be noted that other corresponding descriptions of the functional modules involved in the communication tower inspection device provided in this embodiment of the invention can be found in [reference needed]. Figure 1 The corresponding description of the method shown will not be repeated here.

[0129] Based on the above, Figure 1 Accordingly, this embodiment of the invention also provides a computer-readable storage medium storing a computer program that, when executed by a processor, performs the following steps: real-time acquisition of gas concentration data collected by an unmanned aerial vehicle (UAV) through a multi-gas detection device during its flight around a target communication tower, and real-time acquisition of meteorological data around the target communication tower; based on the gas concentration data, determining whether the search conditions for a fire source location are met; if so, taking the location where the gas concentration data was collected as the starting point for the fire source search, and based on the real-time acquired meteorological data and the real-time acquired gas concentration data, progressively searching for the fire source location from the starting point, and determining the initial fire source location based on the progressive search results; determining multiple collection points within a first preset area centered on the initial fire source location, and acquiring the multi-gas concentration values ​​corresponding to each collection point; and determining the fire source location within the first preset area based on the multi-gas concentration values.

[0130] Based on the above, Figure 1 The method shown and as Figure 3 The embodiment of the device shown in the invention also provides a physical structure diagram of a computer device, such as... Figure 5As shown, the computer device includes: a processor 41, a memory 42, and a computer program stored in the memory 42 and executable on the processor. Both the memory 42 and the processor 41 are mounted on a bus 43. When the processor 41 executes the program, it performs the following steps: real-time acquisition of gas concentration data collected by the unmanned aerial vehicle (UAV) through a multi-gas detection device during its flight around the target communication tower, and real-time acquisition of meteorological data around the target communication tower; based on the gas concentration data, determining whether the search conditions for a fire source location are met; if so, using the gas concentration data acquisition location as the starting point for the fire source search, and based on the real-time acquired meteorological data and gas concentration data, progressively searching for the fire source location from the starting point, and determining the initial fire source location based on the progressive search results; determining multiple acquisition points within a first preset area centered on the initial fire source location, and acquiring the multi-gas concentration value corresponding to each acquisition point; and determining the fire source location within the first preset area based on the multi-gas concentration values.

[0131] The present invention acquires gas concentration data collected in real time by an unmanned aerial vehicle (UAV) during its flight around a target communication tower using various gas detection devices, and also acquires meteorological data around the target communication tower in real time. Based on the gas concentration data, it determines whether the search conditions for a fire source location are met. If so, the location where the gas concentration data was collected is taken as the starting point for the fire source search. Based on the real-time meteorological and gas concentration data, the fire source location is searched step-by-step from the starting point. The initial fire source location is determined based on the search results. Finally, multiple collection points are determined within a first preset area centered on the initial fire source location, and the concentration values ​​of various gases corresponding to each collection point are acquired. Based on these concentration values, the fire source location is determined within the first preset area. Therefore, using unmanned aerial vehicles (UAVs) to inspect communication towers by flying around them can increase the inspection range; using real-time monitoring of gas concentration data for fire detection can avoid visual obstruction limitations and can detect potential fire sources caused by gas leaks that are not easily detected in the early stages, thereby improving the inspection effect of communication towers; by gradually searching to determine the initial fire source location and then analyzing the initial fire source location to determine the final fire source location, the accuracy of fire source location determination can be improved.

[0132] It is obvious to those skilled in the art that the modules or steps of the present invention described above can be implemented using general-purpose computing devices. They can be centralized on a single computing device or distributed across a network of multiple computing devices. Optionally, they can be implemented using computer-executable program code, thereby storing them in a storage device for execution by a computing device. In some cases, the steps shown or described can be performed in a different order than those presented herein, or they can be fabricated as separate integrated circuit modules, or multiple modules or steps can be fabricated as a single integrated circuit module. Thus, the present invention is not limited to any particular combination of hardware and software.

[0133] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for inspecting communication towers, characterized in that, This technology is applied to unmanned aerial vehicles (UAVs) equipped with various gas detection devices, including: The system acquires gas concentration data collected by the various gas detection devices during the flight of the unmanned aerial vehicle around the target communication tower in real time, and also acquires meteorological data around the target communication tower in real time. Based on the gas concentration data, it is determined whether the search conditions for the fire source location are met. If so, the location where the gas concentration data was collected is taken as the starting point for the fire source search. Based on the real-time meteorological data and the real-time gas concentration data, the fire source location is searched step by step from the starting point. The initial fire source location is determined based on the search results. Multiple collection points are determined within a first preset area centered on the initial fire source location, and the concentration values ​​of various gases corresponding to each collection point are obtained. Based on the concentration values ​​of various gases, the fire source location is determined within the first preset area. The process of determining whether the search criteria for the location of a fire source are met based on the gas concentration data includes: Pre-set the fire detection concentration threshold for each gas, and compare the gas concentration data with the corresponding fire detection concentration threshold in real time. When the concentration of a certain gas or multiple gases is greater than the corresponding fire detection concentration threshold, it is determined that the search conditions for the fire source location are met. The unmanned aerial vehicle is also equipped with an image acquisition device. After determining the location of the fire source within the first preset area based on the diverse gas concentration values, the method further includes: The image acquisition device acquires multi-view images of the target communication tower, and based on the multi-view images, performs three-dimensional reconstruction of the target communication tower to obtain a three-dimensional image of the target communication tower. Based on the three-dimensional image, the tower type of the target communication tower is determined. The image acquisition device also acquires fire point images of the fire source location, and based on the tower type and the fire point images, the fire hazard level at the fire source location is determined.

2. The method according to claim 1, characterized in that, The method involves progressively searching for the location of a fire source, starting from the initial search point, based on real-time meteorological data and real-time gas concentration data. The initial fire source location is determined based on the progressively collected search results, including: Multiple initial flight positions are determined within a second preset area centered on the fire source search starting point. The unmanned aerial vehicle is controlled to fly to each initial flight position to collect location gas concentration data and location meteorological data. A virtual flying body is configured at each initial flight position. The location meteorological data includes wind direction data and wind speed data. The dynamic flight step size of the virtual flying body at the corresponding initial flight position is determined based on the gas concentration data at the location, the gas concentration correction factor at the corresponding initial flight position is determined based on the wind speed data, and the flight direction correction factor of the virtual flying body at the corresponding initial flight position is determined based on the wind direction data. Based on the gas concentration correction factor, the position gas concentration data of the corresponding initial flight position is corrected to obtain the corrected gas concentration data for each initial flight position. In each initial flight position, multiple initial flight positions with corrected gas concentration data greater than a preset concentration threshold are selected as candidate flight guidance positions. Based on the position coordinates and corrected gas concentration data of each candidate flight guidance position, the global flight guidance position is determined. The guidance position change of the virtual flying body is determined based on the global flight guidance position and the initial flight position. The random position change of the virtual flying body is determined based on the dynamic flight step size. The guidance position change and the random position change are corrected based on the flight direction correction factor to obtain the corrected guidance position change and the corrected random position change. Based on the corrected guide position change and the corrected random position change, each virtual flying body is controlled to perform the first round of fire source location search flight. The position of each virtual flying body after the first round of flight is taken as the new initial flight position. New location gas concentration data and new location meteorological data are determined at each new initial flight position. Based on the new location gas concentration data and new location meteorological data, new guide position change and new random position change are determined. Based on the new guide position change and new random position change, each virtual flying body is controlled to perform multiple rounds of fire source location iterative search flight. The final position reached by each virtual flying body after the last round of fire source location search flight that meets the iterative search flight conditions is determined. The initial fire source position is determined based on each final position.

3. The method according to claim 1, characterized in that, Determining the location of the ignition source within the first preset area based on the diverse gas concentration values ​​includes: Acquire meteorological data of the collection locations corresponding to the concentration values ​​of the various gases, draw the pollution plume of the various gases at each collection point based on the concentration values ​​of the various gases at each collection point, and determine the shape information of the pollution plume of the various gases. Based on the meteorological data of each collection point and the shape information of the pollution plume of the various gases, determine the first initial fire source area in the first preset area. For each collection point, the ratio of each gas concentration value is determined, and the ratio is matched with the standard gas concentration ratio in the preset fire source fingerprint database. Based on the matching result, the second initial fire source area is determined in the first preset area. A gas concentration distribution map is drawn based on the diverse gas concentration values ​​of each of the collection points, and a meteorological flow field map is drawn based on the meteorological data collected at each of the collection points. The gas concentration distribution map and the meteorological flow field map are then superimposed, and a third initial fire source area is determined within the first preset area based on the superposition result. Based on the first initial fire source area, the second initial fire source area, and the third initial fire source area, a fire source area is determined within the first preset area, and the fire source location is determined within the fire source area.

4. The method according to claim 3, characterized in that, Determining the location of the fire source within the fire source area includes: In the fire source area, a hypothetical fire source location is determined, the actual meteorological data corresponding to the hypothetical fire source location is determined, and based on the actual meteorological data, a monitoring point is determined with the hypothetical fire source location as a reference, and the actual concentration data of various gases at the monitoring point is determined. A Gaussian diffusion model is obtained, and the assumed fire source location is input into the Gaussian diffusion model to predict the concentration of various gases at the monitoring point, thereby obtaining the predicted concentration of various gases at the monitoring point. Based on the difference between the actual and predicted gas concentrations, it is determined whether the hypothetical fire source location needs to be adjusted. If so, the hypothetical fire source location is iteratively adjusted until the difference between the actual and predicted gas concentrations at the last adjusted hypothetical fire source location is less than a preset difference threshold. The hypothetical fire source location after the last adjustment is then taken as the first fire source location. In the fire source area, the sampling point corresponding to the maximum concentration of diverse gases is determined, and the historical meteorological data of the sampling point in the past preset time period is obtained. Based on the historical meteorological data, the particle backtracking direction of the virtual particles corresponding to the diverse gases is determined. The virtual particle is simulated to backtrack in reverse time according to the particle backtracking direction. During the reverse time backtracking process, the particle backtracking trajectory is determined, and the intersection of the particle backtracking trajectory is taken as the location of the second fire source. The fire source area is divided into a three-dimensional grid, and the grid-specific gas concentration data and meteorological data are acquired in real time. Based on the grid meteorological data, the transport direction, transport speed and diffusion coefficient of the various gases in the three-dimensional grid are determined. Based on the transport direction, the transport speed, and the diffusion coefficient, a multi-gas transport simulation and a chemical transformation simulation are performed in the three-dimensional grid. Based on the transport simulation results and the chemical transformation simulation results, the spatiotemporal distribution field of the multi-gas is determined, and the location of the third fire source is determined based on the spatiotemporal distribution field of the multi-gas. Based on the first fire source location, the second fire source location, and the third fire source location, the fire source location is determined in the fire source area.

5. The method according to claim 1, characterized in that, After determining the location of the ignition source within the first preset area based on the diverse gas concentration values, the method further includes: Obtain the locations of potential fire sources around the target communication tower in the preset map, and obtain fire source monitoring data of potential fire sources at the locations of potential fire sources. The fire source location is overlaid with the potential fire source locations in a preset map, and the fire source location is verified based on the overlay result and the fire source monitoring data.

6. The method according to claim 1, characterized in that, The method further includes: The location information of the unmanned aerial vehicle is acquired in real time, and the location information and the gas concentration data acquired in real time are transmitted back in real time. Based on the real-time transmitted location information and gas concentration data, the raster map is colored to different degrees to obtain a raster map with location information and gas concentration data.

7. A communication tower inspection device, characterized in that, This technology is applied to unmanned aerial vehicles (UAVs) equipped with various gas detection devices, including: The acquisition unit is used to acquire in real time the gas concentration data collected by the multi-gas detection device during the flight of the unmanned aerial vehicle around the target communication tower, and to acquire in real time the meteorological data around the target communication tower. The search unit is used to determine whether the search conditions for the fire source location are met based on the gas concentration data. If so, the location where the gas concentration data was collected is taken as the starting point for the fire source search, and the fire source location is searched step by step from the starting point based on real-time meteorological data and real-time gas concentration data. The initial fire source location is determined based on the step-by-step search results. The determination of whether the search conditions for the fire source location are met based on the gas concentration data includes: pre-setting a fire detection concentration threshold for each gas, comparing the gas concentration data with the corresponding fire detection concentration threshold in real time, and determining that the search conditions for the fire source location are met when the concentration of a certain gas or multiple gases is greater than the corresponding fire detection concentration threshold. The determining unit is used to determine multiple collection points within a first preset area centered on the initial fire source location, and to obtain the concentration values ​​of various gases corresponding to each collection point, and to determine the fire source location within the first preset area based on the concentration values ​​of various gases. The classification unit is used to acquire multi-view images of the target communication tower through an image acquisition device, and to perform three-dimensional reconstruction of the target communication tower based on the multi-view images to obtain a three-dimensional image of the target communication tower. Based on the three-dimensional image, the tower type of the target communication tower is determined. The unit also acquires fire point images of the fire source location through the image acquisition device, and determines the fire hazard level at the fire source location based on the tower type and the fire point images.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 6.

9. A computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 6.

Images