Communication iron tower intelligent inspection system and method based on artificial intelligence

Abstract

The invention discloses a communication iron tower intelligent inspection system and method based on artificial intelligence, and relates to the technical field of intelligent inspection, and the inspection system comprises an inspection event extraction module, an equipment distinguishing module, a key inspection event marking module, and an equipment response uniqueness judgment module. The routing inspection event extraction module is used for extracting routing inspection events recorded by routing inspection of the communication iron tower based on the artificial intelligence equipment; the equipment distinguishing module is used for carrying out response distinguishing under the influence of the environment state on the artificial intelligence equipment based on each type of distinguished target inspection events; the key inspection event marking module is used for marking the target inspection event responding to the abnormal characteristics as a key inspection event; and the equipment response uniqueness judgment module is used for retrieving key inspection data recorded by the security execution equipment in the key inspection event, determining a feature response range of the security execution equipment and judging the response uniqueness of the security execution equipment.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent inspection, and in particular to an intelligent inspection system and method for a communication tower based on artificial intelligence. Background Art

[0002] When the communication tower has environmental adaptability problems, such as extreme weather such as heavy snow and fog, the drone inspection may not be able to fly normally, and the image acquisition effect of the intelligent image acquisition device may be affected, resulting in an inability to clearly obtain the tower situation. In addition, in extremely cold or high temperature environments, the sensor performance may be unstable, resulting in data deviation or false alarms; In the existing technology, when intelligent inspections cannot be accurate due to the environment, the improvement often implemented is to replace more accurate image acquisition equipment or drone equipment, etc., but this also increases the cost of intelligent inspections to a certain extent, and the frequency of extreme environments is low, and the high-performance usage attributes of the equipment are less utilized; in most cases, basic equipment is sufficient for daily intelligent inspections, so it is very important to establish monitoring and inspection analysis under extreme weather conditions from a data level; so that on the basis of energy saving, it can ensure that daily inspection equipment can respond reasonably based on data when encountering extreme weather. Summary of the invention

[0003] The purpose of the present invention is to provide a communication tower intelligent inspection system and method based on artificial intelligence to solve the problems raised in the prior art.

[0004] To achieve the above object, the present invention provides the following technical solution: an intelligent inspection method for communication towers based on artificial intelligence, characterized in that the method comprises the following specific steps: Step S100: extracting inspection events recorded by the communication tower based on the artificial intelligence equipment for inspection, marking the inspection events in extreme environments as target inspection events, and distinguishing the target inspection events according to the extreme environment types, and distinguishing the responses of the artificial intelligence equipment under the influence of the environmental state based on each type of distinguished target inspection event; the distinguished categories include non-state execution equipment, semi-state execution equipment and safe execution equipment; Step S200: marking the target inspection event with the abnormal response feature as a key inspection event, retrieving the key inspection data recorded by the security execution device in the key inspection event, determining the characteristic response range of the security execution device, and judging the uniqueness of the response of the security execution device; Step S300: when the output safety execution device response is unique, the extreme environment recorded in the key inspection event is associated with the safety execution device and the corresponding characteristic response range to generate a first association structure signal; when the output safety execution device response is not unique, the association priority of other execution devices based on the safety execution device is analyzed; Step S400: analyzing an association feature model for an execution device at a first priority level in the association priority level, and generating a second association structure signal based on the association feature model; Step S500: In real-time response to an extreme environment, triggering the first associated structure signal or the second associated structure signal to perform intelligent inspection and monitoring of the communication tower.

[0005] Furthermore, distinguishing the responses of the artificial intelligence device under the influence of the environmental state includes the following specific steps: Step S110: The target inspection events recorded in each type of extreme environment are the same type of events, and the target inspection events of the same type are divided into a first set and a second set according to whether there are inspection abnormality records or no inspection abnormality records in the extreme environment; Step S120: Compare the artificial intelligence devices with target inspection event records in the first set and the second set in each type of extreme environment; extract the artificial intelligence devices with records in both the first set and the second set and whose output data error range is less than or equal to the threshold when no inspection anomaly is identified as safe execution devices; Step S130: Obtain all artificial intelligence devices constituting set A that record the history of inspection events, constitute set B for safety execution devices, constitute set C for artificial intelligence devices excluding set B recorded in the second set, and mark set D for non-state execution devices in corresponding type of extreme environment, where D=A-(B+C); and judge the artificial intelligence devices in set C, the judgment process is: calculate the error range of the output data of each artificial intelligence device in set C when the inspection anomaly is not identified, and mark the devices with an error range greater than the threshold as semi-state execution devices.

[0006] By distinguishing the three different artificial intelligence devices in each type of environment, we can clearly and explicitly make different levels of inspection considerations for the inspection equipment of communication towers in different environments during the intelligent inspection process, analyze the equipment data generated by them in a certain direction, improve the analysis efficiency, and serve as effective preparation for subsequent data association.

[0007] Further, step S200 includes the following: Response abnormality features refer to the features corresponding to the inspection abnormalities recognized by the artificial intelligence equipment in the target inspection event; Extract the key inspection data when the identification device for response abnormality feature is a safety execution device, and the key inspection data is the output data of the safety execution device; construct the characteristic response range L based on the minimum or maximum value of the key inspection data, L=[Pmin,Pmin+a] or L=[Pmax-b,Pmax], a and b are any values ​​greater than zero; Pmin is the minimum value of the output data, and Pmax is the maximum value of the output data; Traverse other execution devices to see if they respond to the same abnormal characteristics. When there is an execution device that responds to the same abnormal characteristics as the security execution device, output a signal that the security execution device's response is not unique; when there is no execution device that responds to the same abnormal characteristics as the security execution device, output a signal that the security execution device responds uniquely.

[0008] Furthermore, step S300 includes the following specific steps: The first associated structure signal refers to a structure signal generated by data association of the extreme environment type, the corresponding safety execution device and the characteristic response range; when the system monitors the corresponding extreme environment type, it will respond to the safety execution device and the characteristic response range in the corresponding first associated structure signal; Obtain the output data of the semi-state execution device with the same characteristics as the safety execution device in the same extreme environment type corresponding to different key inspection events, using the formula: V={(1 / n)*[(p max -p min )1+(p max -p min )2+...+(p max -p min ) n ]} / p0; Calculate the environmental impact volatility V of the semi-state execution device, where (p max -p min ) n It represents the difference between the maximum and minimum values ​​of the output data of the semi-state execution device in the nth key inspection event under the same extreme environment type, n represents the total number of key inspection events recorded under the same extreme environment type, and p0 represents the average value of the difference between the maximum and minimum values ​​of the output data of the semi-state execution device under all extreme environment types; Extract the interval response difference T of the semi-state execution device with the same characteristics as the safety execution device, T=|t1-t2|, t1 represents the response time of the safety execution device identifying the abnormal characteristics, t2 represents the response time of the semi-state execution device identifying the abnormal characteristics; use the formula: U=T / tmax to calculate the association strength value U of the semi-state execution device, tmax represents the maximum value of the interval response difference between the safety execution device and the semi-state execution device; A comprehensive score Q is obtained by comprehensively scoring each half-state execution device, Q=k1*V+k2*U+k3*(1 / Y), where k1 represents the reference coefficient corresponding to the environmental impact volatility, k2 represents the reference coefficient corresponding to the association strength value, Y represents the number of half-state execution devices responding to abnormal features and non-state execution devices responding to the same abnormal features; k3 represents the reference coefficient corresponding to the number of features; Each semi-state execution device in various extreme environments is sorted from small to large according to the comprehensive score value to generate the associated priority of the corresponding safe execution device.

[0009] Further, step S400 includes the following: Step S410: extract the output data recorded by the semi-state execution device corresponding to the first priority when responding to abnormal characteristics in the extreme environment of the type in which it is located, draw a data fluctuation curve diagram from the monitoring period before the extreme environment occurs to the period when the extreme environment occurs, and draw the output data of each key inspection event record corresponding to the first priority in a data fluctuation curve diagram, and generate a correlation feature model composed of multiple data fluctuation curves; the correlation feature model refers to the minimum surface composed of multiple data fluctuation curves; Step S420: The second associated structure signal refers to a structure signal generated by data association between the extreme environment type, the corresponding safety execution device, the semi-state execution device and the associated feature model.

[0010] Further, step S500 includes the following: Extract the extreme environment type of real-time response, match the structural signal recorded under the corresponding type, and when it is the first associated structural signal, respond to the corresponding artificial intelligence device and characteristic response range, and issue an early warning when the characteristic response range is met; When it is a second correlation structure signal, determine whether the output data of the artificial intelligence device in the second correlation structure signal is drawn in the correlation feature model. If it is satisfied, make an early warning response to the corresponding abnormal feature, and continue monitoring if it is not satisfied.

[0011] An intelligent inspection system for communication towers based on artificial intelligence, the inspection system includes an inspection event extraction module, an equipment differentiation module, a key inspection event marking module, and an equipment response uniqueness judgment module; The inspection event extraction module is used to extract the inspection events recorded by the communication tower based on artificial intelligence equipment; The device differentiation module is used to differentiate the responses of the artificial intelligence devices under the influence of the environmental status based on each type of target inspection event after differentiation; The key inspection event marking module is used to mark the target inspection events that respond to abnormal characteristics as key inspection events; The device response uniqueness judgment module is used to retrieve the key inspection data recorded by the security execution device in the key inspection event, determine the characteristic response range of the security execution device, and judge the response uniqueness of the security execution device.

[0012] Furthermore, the inspection system also includes an associated structure signal output module, an associated priority analysis module and an intelligent inspection monitoring module; The associated structure signal output module is used to associate and lock the extreme environment recorded in the key inspection event with the safety execution device and the corresponding characteristic response range to generate a first associated structure signal when outputting a unique response of the safety execution device; analyze the associated characteristic model for the execution device at the first priority in the associated priority, and generate a second associated structure signal based on the associated characteristic model; The association priority analysis module is used to analyze the association priorities of other execution devices based on the security execution device when the output security execution device response is not unique; The intelligent inspection and monitoring module is used to trigger the first associated structure signal or the second associated structure signal to perform intelligent inspection and monitoring on the communication tower when responding to extreme environments in real time.

[0013] Compared with the prior art, the beneficial effects of the present invention are: starting from the historical data of intelligent inspection of communication tower applications, the present invention analyzes the data in extreme environments, effectively builds abnormal response feature monitoring in the data dimension, saves resource utilization of high-end intelligent devices, and on the basis of ensuring the rational use of daily equipment, it can also achieve reasonable data response when extreme weather occurs and artificial intelligence equipment cannot be effectively used and data fluctuates. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] Figure 1 The present invention is a structural schematic diagram of an artificial intelligence-based communication tower intelligent inspection method. DETAILED DESCRIPTION

[0015] Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in the field without making any creative work shall fall within the scope of protection of the present invention.

[0016] Example: Figure 1 As shown, the present invention provides a communication tower intelligent inspection system and method technical solution based on artificial intelligence, and a communication tower intelligent inspection method based on artificial intelligence, characterized in that the method comprises the following specific steps: Step S100: extract the inspection events recorded by the communication tower based on the artificial intelligence equipment, mark the inspection events in extreme environments as target inspection events, and distinguish the target inspection events according to the extreme environment types, and distinguish the responses of the artificial intelligence equipment under the influence of the environmental state based on each type of differentiated target inspection events; the differentiated categories include non-state execution equipment, semi-state execution equipment and safety execution equipment; extreme environments include heavy rain, heavy snow, typhoon, etc.; Step S200: marking the target inspection event with the abnormal response feature as a key inspection event, retrieving the key inspection data recorded by the security execution device in the key inspection event, determining the characteristic response range of the security execution device, and judging the uniqueness of the response of the security execution device; Step S300: when the output safety execution device response is unique, the extreme environment recorded in the key inspection event is associated with the safety execution device and the corresponding characteristic response range to generate a first association structure signal; when the output safety execution device response is not unique, the association priority of other execution devices based on the safety execution device is analyzed; Step S400: analyzing an association feature model for an execution device at a first priority level in the association priority level, and generating a second association structure signal based on the association feature model; Step S500: In real-time response to an extreme environment, triggering the first associated structure signal or the second associated structure signal to perform intelligent inspection and monitoring of the communication tower.

[0017] Differentiating the responses of artificial intelligence devices under the influence of environmental conditions includes the following specific steps: Step S110: The target inspection events recorded in each type of extreme environment are the same type of events, and the target inspection events of the same type are divided into a first set and a second set according to whether there are inspection abnormality records or no inspection abnormality records in the extreme environment; Step S120: Compare the artificial intelligence devices with target inspection event records in the first set and the second set under each type of extreme environment; extract the artificial intelligence devices with records in both the first set and the second set and whose output data error range is less than or equal to the threshold when no inspection anomaly is identified as safe execution devices; such as recording temperature sensor devices, the extreme environment is heavy rain, no inspection anomaly is identified in target inspection event 1, and inspection anomaly is identified in target inspection event 2 but not temperature data, and the output data of the temperature sensor device in target inspection event 1 is 10 degrees, and the output data of the temperature sensor device in target inspection event 2 is 8 degrees, and the threshold is set to plus or minus five degrees, then it means that the temperature sensor device is a safe execution device, because it is not affected by the environment and can still work normally; Step S130: Obtain all artificial intelligence devices constituting set A that record the history of inspection events, constitute set B for safety execution devices, constitute set C for artificial intelligence devices excluding set B recorded in the second set, and mark set D for non-state execution devices in corresponding type of extreme environment, where D=A-(B+C); and judge the artificial intelligence devices in set C, the judgment process is: calculate the error range of the output data of each artificial intelligence device in set C when the inspection anomaly is not identified, and mark the devices with an error range greater than the threshold as semi-state execution devices.

[0018] Semi-state execution devices essentially refer to devices that exist in extreme environments and may have errors caused by unstable device performance due to environmental influences.

[0019] By distinguishing the three different artificial intelligence devices in each type of environment, we can clearly and explicitly make different levels of inspection considerations for the inspection equipment of communication towers in different environments during the intelligent inspection process, analyze the equipment data generated by them in a certain direction, improve the analysis efficiency, and serve as effective preparation for subsequent data association.

[0020] Step S200 includes the following: The response abnormality feature refers to the feature corresponding to the inspection abnormality recognized by the artificial intelligence device in the target inspection event; for example, when the temperature sensor detects that the temperature of the tower cable is abnormal, the response abnormality feature is the tower cable; Extract the key inspection data when the identification device that responds to abnormal features is a safety execution device. The key inspection data is the output data of the safety execution device. The feature response range L is constructed based on the minimum or maximum value of the key inspection data. L=[Pmin, Pmin+a] or L=[Pmax-b, Pmax], where a and b are any values ​​greater than zero. Pmin is the minimum value of the output data, and Pmax is the maximum value of the output data. When the key inspection data response is that the larger the abnormal feature, the more obvious it is, the range [Pmin, Pmin+a] is constructed, otherwise it is [Pmax-b, Pmax]. Traverse other execution devices to see if they respond to the same abnormal characteristics. When there is an execution device that responds to the same abnormal characteristics as the security execution device, output a signal that the security execution device's response is not unique; when there is no execution device that responds to the same abnormal characteristics as the security execution device, output a signal that the security execution device responds uniquely.

[0021] Step S300 includes the following specific steps: The first associated structure signal refers to a structure signal generated by data association of the extreme environment type, the corresponding safety execution device and the characteristic response range; when the system monitors the corresponding extreme environment type, it will respond to the safety execution device and the characteristic response range in the corresponding first associated structure signal; Obtain the output data of the semi-state execution device with the same characteristics as the safety execution device in the same extreme environment type corresponding to different key inspection events, using the formula: V={(1 / n)*[(p max -p min )1+(p max -p min )2+...+(p max -p min ) n ]} / p0; Calculate the environmental impact volatility V of the semi-state execution device, where (p max -p min ) n It represents the difference between the maximum and minimum values ​​of the output data of the semi-state execution device in the nth key inspection event under the same extreme environment type, n represents the total number of key inspection events recorded under the same extreme environment type, and p0 represents the average value of the difference between the maximum and minimum values ​​of the output data of the semi-state execution device under all extreme environment types; Extract the interval response difference T of the semi-state execution device with the same characteristics as the safety execution device, T=|t1-t2|, t1 represents the response time of the safety execution device identifying the abnormal characteristics, t2 represents the response time of the semi-state execution device identifying the abnormal characteristics; use the formula: U=T / tmax to calculate the association strength value U of the semi-state execution device, tmax represents the maximum value of the interval response difference between the safety execution device and the semi-state execution device; A comprehensive score Q is obtained by comprehensively scoring each half-state execution device, Q=k1*V+k2*U+k3*(1 / Y), where k1 represents the reference coefficient corresponding to the environmental impact volatility, k2 represents the reference coefficient corresponding to the association strength value, Y represents the number of half-state execution devices responding to abnormal features and non-state execution devices responding to the same abnormal features; k3 represents the reference coefficient corresponding to the number of features; Each semi-state execution device in various extreme environments is sorted from small to large according to the comprehensive score value to generate the associated priority of the corresponding safe execution device.

[0022] Step S400 includes the following: Step S410: extract the output data recorded by the semi-state execution device corresponding to the first priority when responding to abnormal characteristics in the extreme environment of the type in which it is located, draw a data fluctuation curve diagram from the monitoring period before the extreme environment occurs to the period when the extreme environment occurs, and draw the output data of each key inspection event record corresponding to the first priority on a data fluctuation curve diagram, and generate a correlation feature model composed of multiple data fluctuation curves; the horizontal axis of the data fluctuation curve diagram is in time units, and the vertical axis is in units of the corresponding dimension of the output data; the correlation feature model refers to the minimum surface composed of multiple data fluctuation curves; Step S420: The second associated structure signal refers to a structure signal generated by data association between the extreme environment type, the corresponding safety execution device, the semi-state execution device and the associated feature model.

[0023] Step S500 includes the following: Extract the extreme environment type of real-time response, match the structural signal recorded under the corresponding type, and when it is the first associated structural signal, respond to the corresponding artificial intelligence device and characteristic response range, and issue an early warning when the characteristic response range is met; When it is a second correlation structure signal, determine whether the output data of the artificial intelligence device in the second correlation structure signal is drawn in the correlation feature model. If it is satisfied, make an early warning response to the corresponding abnormal feature, and continue monitoring if it is not satisfied.

[0024] When analyzing the second correlation structure signal, all different abnormal response characteristics recorded in the same extreme environment will correspond to a correlation feature model. Therefore, when analyzing the extreme environment type of the second correlation structure signal, the data fluctuation curve constructed by the first priority analyzed from different abnormal response characteristics can be analyzed in multiple dimensions at the same time, thereby more comprehensively and quickly analyzing and warning the data when artificial equipment cannot effectively monitor in extreme environments.

[0025] An intelligent inspection system for communication towers based on artificial intelligence, the inspection system includes an inspection event extraction module, an equipment differentiation module, a key inspection event marking module, and an equipment response uniqueness judgment module; The inspection event extraction module is used to extract the inspection events recorded by the communication tower based on artificial intelligence equipment; The device differentiation module is used to differentiate the responses of the artificial intelligence devices under the influence of the environmental status based on each type of target inspection event after differentiation; The key inspection event marking module is used to mark the target inspection events that respond to abnormal characteristics as key inspection events; The device response uniqueness judgment module is used to retrieve the key inspection data recorded by the security execution device in the key inspection event, determine the characteristic response range of the security execution device, and judge the response uniqueness of the security execution device.

[0026] The inspection system also includes a correlation structure signal output module, a correlation priority analysis module and an intelligent inspection monitoring module; The associated structure signal output module is used to associate and lock the extreme environment recorded in the key inspection event with the safety execution device and the corresponding characteristic response range to generate a first associated structure signal when outputting a unique response of the safety execution device; analyze the associated characteristic model for the execution device at the first priority in the associated priority, and generate a second associated structure signal based on the associated characteristic model; The association priority analysis module is used to analyze the association priorities of other execution devices based on the security execution device when the output security execution device response is not unique; The intelligent inspection and monitoring module is used to trigger the first associated structure signal or the second associated structure signal to perform intelligent inspection and monitoring on the communication tower when responding to extreme environments in real time.

[0027] 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 communication tower intelligent inspection method based on artificial intelligence, characterized in that: The method comprises the following specific steps: Step S100: extracting inspection events recorded by the artificial intelligence equipment during the inspection of the communication tower, marking the inspection events in extreme environments as target inspection events, and distinguishing the target inspection events according to the extreme environment types, and distinguishing the responses of the artificial intelligence equipment under the influence of the environmental state based on each type of distinguished target inspection events; the distinguished categories include non-state execution equipment, semi-state execution equipment and safe execution equipment; Step S200: marking the target inspection event with the abnormal response feature as a key inspection event, retrieving the key inspection data recorded by the security execution device in the key inspection event, determining the characteristic response range of the security execution device, and judging the uniqueness of the response of the security execution device; Step S300: when the output safety execution device response is unique, the extreme environment recorded in the key inspection event is associated with the safety execution device and the corresponding characteristic response range to generate a first associated structure signal; When the output safety execution device response is not unique, analyzing the association priority of other execution devices based on the safety execution device; Step S400: analyzing an association feature model for an execution device at a first priority level in the association priority level, and generating a second association structure signal based on the association feature model; Step S500: In real-time response to an extreme environment, triggering the first associated structure signal or the second associated structure signal to perform intelligent inspection and monitoring of the communication tower.

2. The method for intelligent inspection of communication towers based on artificial intelligence according to claim 1, characterized in that: The step of distinguishing the responses of the artificial intelligence device under the influence of the environmental state includes the following specific steps: Step S110: The target inspection events recorded in each type of extreme environment are the same type of events, and the target inspection events of the same type are divided into a first set and a second set according to whether there are inspection abnormality records or no inspection abnormality records in the extreme environment; Step S120: Compare the artificial intelligence devices with target inspection event records in the first set and the second set in each type of extreme environment; extract the artificial intelligence devices with records in both the first set and the second set and whose output data error range is less than or equal to the threshold when no inspection anomaly is identified as safe execution devices; Step S130: Obtain all artificial intelligence devices constituting set A that record the history of inspection events, constitute set B for safety execution devices, constitute set C for artificial intelligence devices excluding set B recorded in the second set, and mark set D for non-state execution devices in corresponding type of extreme environment, where D=A-(B+C); and judge the artificial intelligence devices in set C, the judgment process is: calculate the error range of the output data of each artificial intelligence device in set C when the inspection anomaly is not identified, and mark the devices with an error range greater than the threshold as semi-state execution devices.

3. The method for intelligent inspection of communication towers based on artificial intelligence according to claim 1, characterized in that: The step S200 includes the following: The response abnormality feature refers to the feature corresponding to the inspection abnormality recognized by the artificial intelligence device in the target inspection event; Extracting key inspection data when the identification device for responding to abnormal characteristics is a safety execution device, wherein the key inspection data is the output data of the safety execution device; constructing a characteristic response range L based on the minimum or maximum value of the key inspection data, where L=[Pmin, Pmin+a] or L=[Pmax-b, Pmax], where a and b are any values ​​greater than zero; Pmin is the minimum value of the output data, and Pmax is the maximum value of the output data; Traverse other execution devices to see if they respond to the same abnormal characteristics. When there is an execution device that responds to the same abnormal characteristics as the security execution device, output a signal that the security execution device's response is not unique; when there is no execution device that responds to the same abnormal characteristics as the security execution device, output a signal that the security execution device responds uniquely.

4. The method for intelligent inspection of communication towers based on artificial intelligence according to claim 1, characterized in that: The step S300 includes the following specific steps: The first associated structure signal refers to a structure signal generated by data association of the extreme environment type, the corresponding safety execution device and the characteristic response range; when the system monitors the corresponding extreme environment type, it will respond to the safety execution device and the characteristic response range in the corresponding first associated structure signal; Obtain the output data of the semi-state execution device with the same characteristics as the safety execution device in the same extreme environment type corresponding to different key inspection events, using the formula: V={(1 / n)*[(p max -p min )1+(p max -p min )2+...+(p max -p min ) n ]} / p0; Calculate the environmental impact volatility V of the semi-state execution device, where (p max -p min ) n It represents the difference between the maximum and minimum values ​​of the output data of the semi-state execution device in the nth key inspection event under the same extreme environment type, n represents the total number of key inspection events recorded under the same extreme environment type, and p0 represents the average value of the difference between the maximum and minimum values ​​of the output data of the semi-state execution device under all extreme environment types; Extract the interval response difference T of the semi-state execution device with the same characteristics as the safety execution device, T=|t1-t2|, t1 represents the response time of the safety execution device identifying the abnormal characteristics, t2 represents the response time of the semi-state execution device identifying the abnormal characteristics; use the formula: U=T / tmax to calculate the association strength value U of the semi-state execution device, tmax represents the maximum value of the interval response difference between the safety execution device and the semi-state execution device; A comprehensive score Q is obtained by comprehensively scoring each half-state execution device, Q=k1*V+k2*U+k3*(1 / Y), where k1 represents the reference coefficient corresponding to the environmental impact volatility, k2 represents the reference coefficient corresponding to the association strength value, Y represents the number of half-state execution devices responding to abnormal features and non-state execution devices responding to the same abnormal features; k3 represents the reference coefficient corresponding to the number of features; Each semi-state execution device in various extreme environments is sorted from small to large according to the comprehensive score value to generate the associated priority of the corresponding safe execution device.

5. The method for intelligent inspection of communication towers based on artificial intelligence according to claim 4, characterized in that: The step S400 includes the following: Step S410: extract the output data recorded by the semi-state execution device corresponding to the first priority when responding to abnormal characteristics in the extreme environment of the type in which it is located, draw a data fluctuation curve diagram from the monitoring period before the extreme environment occurs to the period when the extreme environment occurs, and draw the output data of each key inspection event record corresponding to the first priority in a data fluctuation curve diagram, and generate a correlation feature model composed of multiple data fluctuation curves; the correlation feature model refers to the minimum surface composed of multiple data fluctuation curves; Step S420: The second associated structure signal refers to a structure signal generated by data association between the extreme environment type, the corresponding safety execution device, the semi-state execution device and the associated feature model.

6. The method for intelligent inspection of communication towers based on artificial intelligence according to claim 1, characterized in that: The step S500 includes the following: Extract the extreme environment type of real-time response, match the structural signal recorded under the corresponding type, and when it is the first associated structural signal, respond to the corresponding artificial intelligence device and characteristic response range, and issue an early warning when the characteristic response range is met; When it is a second correlation structure signal, determine whether the output data of the artificial intelligence device in the second correlation structure signal is drawn in the correlation feature model. If it is satisfied, make an early warning response to the corresponding abnormal feature, and continue monitoring if it is not satisfied.

7. An artificial intelligence-based communication tower intelligent inspection system, such as using an artificial intelligence-based communication tower intelligent inspection method according to any one of claims 1 to 6, characterized in that: The inspection system includes an inspection event extraction module, an equipment differentiation module, a key inspection event marking module, and an equipment response uniqueness judgment module; The inspection event extraction module is used to extract the inspection events recorded by the communication tower based on the artificial intelligence equipment; The device differentiation module is used to differentiate the responses of the artificial intelligence device under the influence of the environmental state based on each type of target inspection event after differentiation; The key inspection event marking module is used to mark the target inspection event that responds to abnormal characteristics as a key inspection event; The device response uniqueness judgment module is used to retrieve the key inspection data recorded by the security execution device in the key inspection event, determine the characteristic response range of the security execution device, and judge the response uniqueness of the security execution device.

8. The communication tower intelligent inspection system based on artificial intelligence according to claim 7 is characterized in that: The inspection system also includes an associated structure signal output module, an associated priority analysis module and an intelligent inspection monitoring module; The associated structure signal output module is used to associate and lock the extreme environment recorded in the key inspection event with the safety execution device and the corresponding characteristic response range to generate a first associated structure signal when outputting a unique response of the safety execution device; analyze the associated characteristic model for the execution device at the first priority in the associated priority, and generate a second associated structure signal based on the associated characteristic model; The association priority analysis module is used to analyze the association priorities of other execution devices based on the security execution device when the output security execution device response is not unique; The intelligent patrol and monitoring module is used to trigger the first associated structure signal or the second associated structure signal to perform intelligent patrol and monitoring on the communication tower when responding to extreme environments in real time.

Images