A Multi-Gas Sensing Fusion System and Method Based on Artificial Intelligence

Abstract

The present invention discloses a multi-gas perception fusion system and method based on artificial intelligence, which belongs to the field of artificial intelligence. The present invention comprises the following steps: S1, establishing a map model; S2, analyzing the predicted obstacles on the path when maintenance personnel go to the maintenance point, and marking the obstacle turning point group; S3, monitoring dangerous gases, and marking the target maintenance point; S4, collecting data, establishing a Gaussian model, and marking dangerous turning points; S5, screening safe path points and unobstructed paths, calculating the length of each unobstructed path, and selecting the best maintenance path corresponding to each entrance; the present invention realizes the monitoring of dangerous gases in the monitored place and predicts the spread of dangerous gases through the comprehensive application of multiple sensors, and according to the prediction results, combined with the analysis of obstacles in the monitored place, avoids dangerous areas and obstacle areas, and plans the best maintenance path, which not only increases the safety of personnel but also improves the maintenance efficiency.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence, and in particular to an artificial intelligence-based multi-gas sensing fusion system and method. Background Art

[0002] With the development of society, the use of electricity is indispensable to our lives, and cable tunnels are closely related to electricity. Cable tunnels are corridors or tunnel-type structures that accommodate a large number of cables laid on cable racks. In addition to allowing the tunnel to better protect the cables, cable tunnels also facilitate people's inspection and maintenance of cables. However, in cable tunnels, some toxic and harmful gases to the human body will be produced. If the human body is in this environment for a long time, it may even be life-threatening. Therefore, it is very important to monitor the gas in the cable tunnel;

[0003] The current gas monitoring system in cable tunnels can detect and warn when dangerous gases appear, but it cannot predict the spread of dangerous gases. Maintenance personnel rush to the maintenance site through the path of dangerous gases, which increases the contact time. At the same time, the system cannot provide maintenance personnel with convenient on-site information, which prolongs the maintenance time and increases the possibility of casualties.

[0004] Therefore, people urgently need an artificial intelligence-based multi-gas sensing fusion system to solve the above problems. Summary of the invention

[0005] The purpose of the present invention is to provide a multi-gas sensing fusion system based on artificial intelligence to solve the problems raised in the above background technology.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0007] A multi-gas sensing fusion method based on artificial intelligence, the method comprising the following steps:

[0008] S1. Collecting map data of the place to be monitored, establishing a map model, and collecting the coordinate points of each entrance of the place to be monitored in the map model and the coordinate points of each inflection point in the map model;

[0009] S2. Collecting object data and building data of the monitored site, wherein the object data includes location data, placement data and size data of each object, and the building data includes the path width of the building, analyzing the predicted obstacles on the path when the maintenance personnel go to the maintenance point, and marking the front and rear turning points of the path where the predicted obstacles are located as an obstacle turning point group;

[0010] S3. When the sensor detects that the concentration of any dangerous gas exceeds the safety standard, the system will alarm and mark the coordinate point of this sensor as the target maintenance point in the map model;

[0011] S4, collecting the concentration of the gas at the time of the alarm and the concentration of the gas after a unit time, calculating the release rate of the dangerous gas, collecting the average time for the maintenance personnel to reach the target maintenance point after receiving the alarm, and the airflow data of the target maintenance point, the airflow data including the airflow direction and airflow speed, establishing a Gaussian model, calculating all safety critical points, and marking the inflection point on the path from the target maintenance point to the safety critical point as a dangerous inflection point;

[0012] S5. According to the dangerous turning points, the safe path points are screened out, and the paths from each entrance to the target maintenance point are planned. According to the obstacle turning point group, the smooth paths are screened out, the length of each smooth path is calculated, and the best maintenance path corresponding to each entrance is selected.

[0013] According to the above technical solution, in step S2, the predicted obstacles on the path of the maintenance personnel going to the maintenance point are analyzed, and the two turning points before and after the path where the predicted obstacles are located are marked as an obstacle turning point group. The specific process is as follows:

[0014] S2-1, collecting the location data, placement data and size data of each object in the monitored place, and obtaining the path width of the building corresponding to the location of each object according to the location data of each object, and forming a numerical set of path widths recorded as set A;

[0015] S2-2, according to the size data of each item and the placement data of each item, obtain the size occupied by the item on the path, and form a numerical set of the occupied sizes, which is recorded as set B;

[0016] S2-3. Each element in set A and set B satisfies Then mark this item as a predicted obstacle; if it does not meet The item is not marked, where A a represents the width of the ath path, represents the occupied size of the bth item in the ath path, α is a fixed parameter preset by the system, a∈[1, a'], b∈[1, a b '], a' represents the total number of paths, a b ' represents the total number of items in the ath path;

[0017] According to the predicted obstacle, the two turning points before and after the corresponding path are marked as an obstacle turning point group, and the set of coordinate points forming the obstacle turning point group is recorded as set C.

[0018] According to the above technical solution, in step S4, the concentration of the gas at the alarm time and the concentration of the gas after a unit time are collected, the release rate of the dangerous gas is calculated, and the average time for the historical maintenance personnel to arrive at the target maintenance point after receiving the alarm and the airflow data of the target maintenance point are collected. The specific process is as follows:

[0019] S3-1. The alarm time is recorded as t0, the concentration of the gas at the time of t0 is recorded as c0, the unit time is set to Δt, Δt is a fixed parameter preset by the system, the concentration of the gas at the time of t0+Δt is recorded as c1, and the release rate of the dangerous gas is calculated according to the following formula:

[0020]

[0021] Wherein, Q represents the release rate of the hazardous gas;

[0022] S3-2. Collecting history: After receiving the alarm, the average time for the maintenance personnel to reach the target maintenance point is recorded as t, and the airflow direction and airflow speed of the target maintenance point are collected, and the obtained airflow speed is recorded as u;

[0023] When the system is initialized, the maintenance time is long because there is no path planning for historical maintenance. After using it for a period of time, the maintenance time will be reduced and the system screening will be more accurate.

[0024] According to the above technical solution, in step S4, a Gaussian model is established to calculate all safety critical points, and the turning points on the path from the target maintenance point to the safety critical point are marked as dangerous turning points. The specific process is as follows:

[0025] S4-1. Establish a Gaussian model, as follows:

[0026]

[0027] Where C(x, y) represents the gas concentration at the coordinate point (x, y), σ x and σ y is the standard deviation of the Gaussian distribution obtained by the Gaussian model;

[0028] In this model, the hazardous gas propagates in the direction of the airflow velocity u over time t. Therefore, in the x direction, the gas moves in the direction of the airflow over time, which is represented by x-ut, and in the y direction, the Gaussian distribution of the natural diffusion of the gas is adopted.

[0029] S4-2. According to the safety standard, the C(x, y) value is set to β, where β is a fixed parameter preset by the system. β is used as input, and all coordinate points are calculated through the Gaussian model, and the obtained coordinate points are marked as safety critical points;

[0030] S4-3. Record the turning points on the path from the target maintenance point to each safety critical point as dangerous turning points, and record the set of coordinate points forming the dangerous turning points as set D.

[0031] According to the above technical solution, in step S5, according to the dangerous turning points, safe path points are screened out, and paths from each entrance to the target maintenance point are planned. The specific process is as follows:

[0032] S5-1. The coordinate points of each turning point in the collected map model are formed into a coordinate point set and recorded as set E. The coordinate points shared by set E and set D are removed from set E. After the removal, all elements in set E are safe path points.

[0033] S5-2, taking the coordinate points of each entrance as the path initial point, taking the target maintenance point as the path end point, planning the path, sequentially passing through the path initial point, the coordinate points required by the path included in set E, and the path end point, and obtaining N path point sets.

[0034] According to the above technical solution, in step S5, according to the obstacle turning point group, the clear path is screened out, the length of each clear path is calculated, and the best maintenance path corresponding to each entrance is selected. The specific process is as follows:

[0035] S6-1. The N path point sets obtained are combined with the set C for screening. If there are two coordinate points in the path point set that correspond to any obstacle turning point group in the set C, then this path point set is removed; if there are no two coordinate points in the path point set that correspond to any obstacle turning point group in the set C, then this path point set is retained;

[0036] After the screening is completed, a set of path points of n clear paths is obtained;

[0037] S6-2. Calculate the length of each clear path according to the following formula:

[0038]

[0039] Among them, S d represents the length of the dth clear path, f d represents the total number of coordinate points in the path point set of the d-th clear path, represents the horizontal coordinate value of the e-th coordinate point in the path point set of the d-th clear path, represents the ordinate value of the e-th coordinate point in the path point set of the d-th clear path, represents the horizontal coordinate value of the e+1th coordinate point in the path point set of the dth clear path, Represents the ordinate value of the e+1th coordinate point in the path point set of the dth clear path, d∈[1,n], e d ∈[0,f d-1], n represents the total number of clear paths; in particular, when e d =0, x0 represents the abscissa value of the entrance corresponding to the clear path, and y0 represents the ordinate value of the entrance corresponding to the clear path;

[0040] According to the above formula, the calculated S d , according to the entrances corresponding to the clear paths, the shortest clear path corresponding to each entrance is selected as the optimal maintenance path corresponding to the entrance.

[0041] An artificial intelligence-based multi-gas sensing fusion system, the system comprising an information acquisition module, an information processing module and an information display module;

[0042] The information acquisition module is used to acquire various information required by the system; the information processing module is used to store, analyze and transmit various information of each module; and the information display module is used to display various information of the information processing module.

[0043] According to the above technical solution, the information collection module includes a field monitoring unit and a location monitoring unit;

[0044] The on-site monitoring unit is used to collect information about various objects in the place to be monitored, and the location monitoring unit is used to collect location information about various objects in the place to be monitored.

[0045] According to the above technical solution, the information processing module includes an information storage unit, an information analysis unit and an information transmission unit;

[0046] The information storage unit is used to store various information acquired by the information acquisition module; the information analysis unit is used to analyze various information acquired by the information acquisition module; and the information transmission unit is used for information transmission among various modules in the system.

[0047] According to the above technical solution, the information display module includes an alarm unit and a line display unit;

[0048] The alarm unit is used to display information about dangerous gases in the place to be monitored, and the line display unit is used to display information about the best maintenance path.

[0049] Compared with the prior art, the beneficial effects achieved by the present invention are:

[0050] The present invention realizes the monitoring of hazardous gases in the monitored places and predicts the spread of hazardous gases through the comprehensive application of multiple sensors. Based on the prediction results and combined with the analysis of obstacles in the monitored places, the present invention avoids hazardous areas and obstacle areas and plans the best maintenance path, which not only increases the safety of personnel but also improves the maintenance efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0052] Figure 1 It is a flow chart of a multi-gas sensing fusion method based on artificial intelligence of the present invention;

[0053] Figure 2 It is a structural schematic diagram of an artificial intelligence-based multi-gas sensing fusion system of the present invention. DETAILED DESCRIPTION

[0054] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0055] See also Figure 1 , the present invention provides a technical solution:

[0056] A multi-gas sensing fusion method based on artificial intelligence, the method comprising the following steps:

[0057] S1. Collecting map data of the place to be monitored, establishing a map model, and collecting the coordinate points of each entrance of the place to be monitored in the map model and the coordinate points of each inflection point in the map model;

[0058] S2. Collecting object data and building data of the monitored site, wherein the object data includes location data, placement data and size data of each object, and the building data includes the path width of the building, analyzing the predicted obstacles on the path when the maintenance personnel go to the maintenance point, and marking the front and rear turning points of the path where the predicted obstacles are located as an obstacle turning point group;

[0059] In step S2, the predicted obstacles on the path of the maintenance personnel going to the maintenance point are analyzed, and the two inflection points before and after the path where the predicted obstacles are located are marked as an obstacle inflection point group. The specific process is as follows:

[0060] S2-1, collecting the location data, placement data and size data of each object in the monitored place, and obtaining the path width of the building corresponding to the location of each object according to the location data of each object, and forming a numerical set of path widths recorded as set A;

[0061] S2-2, according to the size data of each item and the placement data of each item, obtain the size occupied by the item on the path, and form a numerical set of the occupied sizes, which is recorded as set B;

[0062] S2-3. Each element in set A and set B satisfies Then mark this item as a predicted obstacle; if it does not meet The item is not marked, where A a represents the width of the ath path, represents the occupied size of the bth item in the ath path, α is a fixed parameter preset by the system, a∈[1, a'], b∈[1, a b '], a' represents the total number of paths, a b ' represents the total number of items in the ath path;

[0063] The predicted obstacles are screened out based on the spatial proportion of the objects in the path, and the subsequent paths are chosen to avoid the predicted obstacles.

[0064] According to the predicted obstacle, the two turning points before and after the corresponding path are marked as an obstacle turning point group, and the set of coordinate points forming the obstacle turning point group is recorded as set C.

[0065] S3. When the sensor detects that the concentration of any dangerous gas exceeds the safety standard, the system will alarm and mark the coordinate point of this sensor as the target maintenance point in the map model;

[0066] Establish a map model and mark the acquired information in the map model to facilitate the planning and screening of subsequent paths;

[0067] S4, collecting the concentration of the gas at the time of the alarm and the concentration of the gas after a unit time, calculating the release rate of the dangerous gas, collecting the average time for the maintenance personnel to reach the target maintenance point after receiving the alarm, and the airflow data of the target maintenance point, the airflow data including the airflow direction and airflow speed, establishing a Gaussian model, calculating all safety critical points, and marking the inflection point on the path from the target maintenance point to the safety critical point as a dangerous inflection point;

[0068] In step S4, the concentration of the gas at the alarm time and the concentration of the gas after a unit time are collected, the release rate of the dangerous gas is calculated, and the average time for the maintenance personnel to arrive at the target maintenance point after receiving the alarm and the airflow data of the target maintenance point are collected. The specific process is as follows:

[0069] S3-1. The alarm time is recorded as t0, the concentration of the gas at the time of t0 is recorded as c0, the unit time is set to Δt, Δt is a fixed parameter preset by the system, the concentration of the gas at the time of t0+Δt is recorded as c1, and the release rate of the dangerous gas is calculated according to the following formula:

[0070]

[0071] Wherein, Q represents the release rate of the hazardous gas;

[0072] S3-2. Collecting history: After receiving the alarm, the average time for the maintenance personnel to reach the target maintenance point is recorded as t, and the airflow direction and airflow speed of the target maintenance point are collected, and the obtained airflow speed is recorded as u;

[0073] When the system is initially used, the maintenance time is long because there is no path planning for historical maintenance. After using it for a period of time, the maintenance time will be reduced and the system screening will be more accurate.

[0074] In step S4, a Gaussian model is established to calculate all safety critical points, and the turning points on the path from the target maintenance point to the safety critical point are marked as dangerous turning points. The specific process is as follows:

[0075] S4-1. Establish a Gaussian model, as follows:

[0076]

[0077] Where C(x, y) represents the gas concentration at the coordinate point (x, y), σ x and σ y is the standard deviation of the Gaussian distribution obtained by the Gaussian model;

[0078] In this model, the hazardous gas propagates in the direction of the airflow velocity u over time t. Therefore, in the x direction, the gas moves in the direction of the airflow over time, which is represented by x-ut, and in the y direction, the Gaussian distribution of the natural diffusion of the gas is adopted.

[0079] S4-2. According to the safety standard, the C(x, y) value is set to β, where β is a fixed parameter preset by the system. β is used as input, and all coordinate points are calculated through the Gaussian model, and the obtained coordinate points are marked as safety critical points;

[0080] S4-3, recording the turning points on the path from the target maintenance point to each safety critical point as dangerous turning points, and recording the set of coordinate points forming the dangerous turning points as set D;

[0081] The safety critical point is obtained through the Gaussian model, the dangerous turning points are screened out, and the subsequent path selection avoids the dangerous turning points.

[0082] S5. Filter out safe path points according to dangerous turning points, plan the paths from each entrance to the target maintenance point, and filter out smooth paths according to the obstacle turning point group, calculate the length of each smooth path, and select the best maintenance path corresponding to each entrance;

[0083] In step S5, according to the dangerous turning points, safe path points are screened out, and paths from each entrance to the target maintenance point are planned. The specific process is as follows:

[0084] S5-1. The coordinate points of each turning point in the collected map model are formed into a coordinate point set and recorded as set E. The coordinate points shared by set E and set D are removed from set E. After the removal, all elements in set E are safe path points.

[0085] S5-2, taking the coordinate points of each entrance as the path initial point, taking the target maintenance point as the path end point, planning the path, sequentially passing through the path initial point, the coordinate points required by the path included in the set E, and the path end point, and obtaining N path point sets;

[0086] The N path point sets obtained at this time only avoid dangerous turning points, and subsequent avoidance analysis of predicted obstacles will be performed.

[0087] In step S5, according to the obstacle inflection point group, the clear paths are screened out, the lengths of the clear paths are calculated, and the best maintenance paths corresponding to the entrances are selected. The specific process is as follows:

[0088] S6-1. The N path point sets obtained are combined with the set C for screening. If there are two coordinate points in the path point set that correspond to any obstacle turning point group in the set C, then this path point set is removed; if there are no two coordinate points in the path point set that correspond to any obstacle turning point group in the set C, then this path point set is retained;

[0089] After the screening is completed, a set of path points of n clear paths is obtained;

[0090] The n path point sets obtained at this time not only avoid dangerous turning points, but also avoid predicted obstacles;

[0091] S6-2. Calculate the length of each clear path according to the following formula:

[0092]

[0093] Among them, S d represents the length of the dth clear path, f d represents the total number of coordinate points in the path point set of the d-th clear path, represents the horizontal coordinate value of the e-th coordinate point in the path point set of the d-th clear path, represents the ordinate value of the e-th coordinate point in the path point set of the d-th clear path, represents the horizontal coordinate value of the e+1th coordinate point in the path point set of the dth clear path, Represents the ordinate value of the e+1th coordinate point in the path point set of the dth clear path, d∈[1,n], e d ∈[0,f d -1], n represents the total number of clear paths; in particular, when ed =0, x0 represents the abscissa value of the entrance corresponding to the clear path, and y0 represents the ordinate value of the entrance corresponding to the clear path;

[0094] According to the above formula, the calculated S d , according to the entrances corresponding to the clear paths, the shortest clear path corresponding to each entrance is selected as the optimal maintenance path corresponding to the entrance.

[0095] See also Figure 2 , a multi-gas sensing fusion system based on artificial intelligence, the system includes an information acquisition module, an information processing module and an information display module;

[0096] The information collection module is used to collect various information required by the system;

[0097] The information processing module includes an information storage unit, an information analysis unit and an information transmission unit;

[0098] The information storage unit is used to store various information acquired by the information acquisition module; the information analysis unit is used to analyze various information acquired by the information acquisition module; and the information transmission unit is used for information transmission among various modules in the system.

[0099] The information processing module is used to store, analyze and transmit various information of each module;

[0100] The information collection module includes a field monitoring unit and a location monitoring unit;

[0101] The on-site monitoring unit is used to collect information about various objects in the place to be monitored, and the location monitoring unit is used to collect location information about various objects in the place to be monitored.

[0102] The information display module is used to display various information of the information processing module;

[0103] The information display module includes an alarm unit and a line display unit;

[0104] The alarm unit is used to display information about dangerous gases in the place to be monitored, and the line display unit is used to display information about the best maintenance path.

[0105] Example:

[0106] There is a production site to be monitored, and dangerous gas appears. After system analysis, the site has only one entrance, and the entrance coordinates are (x0=0, y0=0). The coordinates of the target maintenance point are There are n=2 path point sets of the clear path from the entrance to the target maintenance point. The total number of coordinate points in the path point set of the first clear path is f1=4. The coordinates contained in the path point set are (x0=0, y0=0), and The total number of coordinate points in the path point set of the second clear path is f2=5, and the coordinates contained in the path point set are (x0=0, y0=0), and Calculate the length of the maintenance line according to the following formula:

[0107]

[0108] According to the above formula, the entrances corresponding to the clear paths are compared, and the shortest clear path corresponding to each entrance, that is, the clear path corresponding to S1, is selected as the optimal maintenance path corresponding to the entrance.

[0109] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0110] Finally, it should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art can still modify the technical solutions described in the aforementioned embodiments or replace some of the technical features therein by equivalents. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A multi-gas sensing fusion method based on artificial intelligence, characterized by: The method comprises the following steps: S1. Collecting map data of the place to be monitored, establishing a map model, and collecting the coordinate points of each entrance of the place to be monitored in the map model and the coordinate points of each inflection point in the map model; S2. Collect the data of items and buildings at the place to be monitored, analyze the predicted obstacles on the path when the maintenance personnel go to the maintenance point, and mark the two turning points before and after the path where the predicted obstacles are located as an obstacle turning point group; S3. When the sensor detects that the concentration of any dangerous gas exceeds the safety standard, the system will alarm and mark the coordinate point of this sensor as the target maintenance point in the map model; S4. Collect the concentration of the gas at the time of the alarm and the concentration of the gas after a unit of time, calculate the release rate of the dangerous gas, collect the average time for the historical maintenance personnel to reach the target maintenance point after receiving the alarm, and the airflow data of the target maintenance point, establish a Gaussian model, calculate all safety critical points, and mark the turning point on the path from the target maintenance point to the safety critical point as a dangerous turning point; S5. According to the dangerous turning points, the safe path points are screened out, and the paths from each entrance to the target maintenance point are planned. According to the obstacle turning point group, the smooth paths are screened out, the length of each smooth path is calculated, and the best maintenance path corresponding to each entrance is selected.

2. The multi-gas sensing fusion method based on artificial intelligence according to claim 1 is characterized by: In step S2, the predicted obstacles on the path of the maintenance personnel going to the maintenance point are analyzed, and the two inflection points before and after the path where the predicted obstacles are located are marked as an obstacle inflection point group. The specific process is as follows: S2-1, collecting the location data, placement data and size data of each object in the monitored place, and obtaining the path width of the building corresponding to the location of each object according to the location data of each object, and forming a numerical set of path widths recorded as set A; S2-2, according to the size data of each item and the placement data of each item, obtain the size occupied by the item on the path, and form a numerical set of the occupied sizes, which is recorded as set B; S2-3. Each element in set A and set B satisfies Then mark this item as a predicted obstacle; if it does not meet The item is not marked, where A a represents the width of the ath path, represents the occupied size of the bth item in the ath path, α is a fixed parameter preset by the system, a∈[1, a'], b∈[1, a b '], a' represents the total number of paths, a b ' represents the total number of items in the ath path; According to the predicted obstacle, the two turning points before and after the corresponding path are marked as an obstacle turning point group, and the set of coordinate points forming the obstacle turning point group is recorded as set C.

3. The multi-gas sensing fusion method based on artificial intelligence according to claim 2 is characterized by: In step S4, the concentration of the gas at the alarm time and the concentration of the gas after a unit time are collected, the release rate of the dangerous gas is calculated, and the average time for the maintenance personnel to arrive at the target maintenance point after receiving the alarm and the airflow data of the target maintenance point are collected. The specific process is as follows: S3-1. The alarm time is recorded as t0, the concentration of the gas at the time of t0 is recorded as c0, the unit time is set to Δt, Δt is a fixed parameter preset by the system, the concentration of the gas at the time of t0+Δt is recorded as c1, and the release rate of the dangerous gas is calculated according to the following formula: Wherein, Q represents the release rate of the hazardous gas; S3-2. Collecting History After receiving the alarm, the average time for the maintenance personnel to reach the target maintenance point is recorded as t, and the airflow direction and airflow speed of the target maintenance point are collected, and the obtained airflow speed is recorded as u.

4. The multi-gas sensing fusion method based on artificial intelligence according to claim 3 is characterized by: In step S4, a Gaussian model is established to calculate all safety critical points, and the turning points on the path from the target maintenance point to the safety critical point are marked as dangerous turning points. The specific process is as follows: S4-1. Establish a Gaussian model, as follows: Where C(x, y) represents the gas concentration at the coordinate point (x, y), σ x and σ y is the standard deviation of the Gaussian distribution obtained by the Gaussian model; S4-2. According to the safety standard, the C(x, y) value is set to β, where β is a fixed parameter preset by the system. β is used as input, and all coordinate points are calculated through the Gaussian model, and the obtained coordinate points are marked as safety critical points; S4-3. Record the turning points on the path from the target maintenance point to each safety critical point as dangerous turning points, and record the set of coordinate points forming the dangerous turning points as set D.

5. The multi-gas sensing fusion method based on artificial intelligence according to claim 4 is characterized by: In step S5, according to the dangerous turning points, safe path points are screened out, and paths from each entrance to the target maintenance point are planned. The specific process is as follows: S5-1. The coordinate points of each turning point in the collected map model are formed into a coordinate point set and recorded as set E. The coordinate points shared by set E and set D are removed from set E. After the removal, all elements in set E are safe path points. S5-2, taking the coordinate points of each entrance as the path initial point, taking the target maintenance point as the path end point, planning the path, sequentially passing through the path initial point, the coordinate points required by the path included in set E, and the path end point, and obtaining N path point sets.

6. The multi-gas sensing fusion method based on artificial intelligence according to claim 5 is characterized by: In step S5, according to the obstacle inflection point group, the clear paths are screened out, the lengths of the clear paths are calculated, and the best maintenance paths corresponding to the entrances are selected. The specific process is as follows: S6-1, the obtained N path point sets are combined with the set C for screening, if there are two coordinate points in the path point set that correspond to any obstacle turning point group in the set C, then this path point set is removed; If there are no two coordinate points in the path point set that correspond to any obstacle turning point group in set C, then this path point set is retained; After the screening is completed, a set of path points of n clear paths is obtained; S6-2. Calculate the length of each clear path according to the following formula: Among them, S d represents the length of the dth clear path, f d represents the total number of coordinate points in the path point set of the d-th clear path, represents the horizontal coordinate value of the e-th coordinate point in the path point set of the d-th clear path, represents the ordinate value of the e-th coordinate point in the path point set of the d-th clear path, represents the horizontal coordinate value of the e+1th coordinate point in the path point set of the dth clear path, Represents the ordinate value of the e+1th coordinate point in the path point set of the dth clear path, d∈[1,n], e d ∈[0,f d -1], n represents the total number of clear paths; in particular, when e d =0, x0 represents the abscissa value of the entrance corresponding to the clear path, and y0 represents the ordinate value of the entrance corresponding to the clear path; According to the above formula, the calculated S d , according to the entrances corresponding to the clear paths, the shortest clear path corresponding to each entrance is selected as the optimal maintenance path corresponding to the entrance.

7. An artificial intelligence-based multi-gas sensing fusion system, such as a multi-gas sensing fusion method based on artificial intelligence according to any one of claims 1 to 6, characterized in that: The system includes an information acquisition module, an information processing module and an information display module; The information acquisition module is used to acquire various information required by the system; the information processing module is used to store, analyze and transmit various information of each module; and the information display module is used to display various information of the information processing module.

8. The multi-gas sensing fusion system based on artificial intelligence according to claim 7 is characterized by: The information collection module includes a field monitoring unit and a location monitoring unit; The on-site monitoring unit is used to collect information about various objects in the place to be monitored, and the location monitoring unit is used to collect location information about various objects in the place to be monitored.

9. The multi-gas sensing fusion system based on artificial intelligence according to claim 7 is characterized by: The information processing module includes an information storage unit, an information analysis unit and an information transmission unit; The information storage unit is used to store various information acquired by the information acquisition module; the information analysis unit is used to analyze various information acquired by the information acquisition module; and the information transmission unit is used for information transmission among various modules in the system.

10. The multi-gas sensing fusion system based on artificial intelligence according to claim 7 is characterized by: The information display module includes an alarm unit and a line display unit; The alarm unit is used to display information about dangerous gases in the place to be monitored, and the line display unit is used to display information about the best maintenance path.

Images