Engineering area intelligent risk inspection method and system based on aerial survey of unmanned aerial vehicle

Abstract

The invention relates to the technical field of engineering area risk inspection, in particular to an intelligent engineering area risk inspection method and system based on unmanned aerial vehicle aerial survey, and discloses an unmanned aerial vehicle regular inspection module, a risk identification and analysis module and a risk early warning feedback module. According to the invention, the unmanned aerial vehicle aerial survey technology is combined with the hyperspectral imaging technology to carry out regular risk inspection on the target engineering area, and the damage position in the inspection process is further analyzed; according to the method, the damage position is deeply analyzed in combination with the basic risk, the risk time sequence and the risk relevance of the damage position and the risk variability under each load condition, the damage risk existing in the engineering area is accurately locked, whether the corresponding damage position needs to be repaired or not is efficiently judged, accurate risk control over the engineering area is guaranteed, and the engineering safety is improved. And it can be guaranteed that limited resources can be accurately input to the damaged position really needing to be repaired, and the resource utilization efficiency is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of engineering area risk inspection, and more specifically, to an intelligent risk inspection method and system for engineering areas based on unmanned aerial vehicle (UAV) aerial survey. Background Art

[0002] With the rapid advancement of modernization, various large-scale engineering facilities (such as high-rise buildings, bridges, water conservancy projects, industrial factories, etc.) have emerged continuously, and they play a crucial role in social and economic development. However, during the long-term use of these projects, they will inevitably be affected by various factors, resulting in varying degrees of damage to the structure, and thus various risk hazards are generated. For example, in terms of natural environmental factors, long-term exposure to wind, sun, temperature changes, rainfall erosion, and natural disasters such as earthquakes and floods will gradually deteriorate the material properties of the engineering structure, causing damage such as cracks, deformations, and rust; in terms of service loads, the distribution changes of personnel and equipment inside buildings, the frequent passage of vehicles on bridges and other dynamic load effects will also generate continuous stress impacts on the structure, accelerating the formation and development of damage.

[0003] To ensure the safety and reliability of projects, risk inspection of engineering areas has become an essential key link. In recent years, with the continuous development of technology, some advanced detection technologies have gradually been applied to the field of engineering risk inspection. For example, UAV aerial survey technology has begun to stand out in engineering inspection with its advantages of flexibility, high efficiency, and the ability to cover large areas. By carrying various shooting devices on UAVs, high-definition image data of engineering areas can be obtained, facilitating the inspection of the structure appearance and other situations. At the same time, hyperspectral imaging technology has also received increasing attention. It can detect deep changes inside engineering materials and some damages that are difficult to distinguish by the naked eye based on the principle that different substances have unique spectral reflection, absorption, and transmission characteristics, providing strong support for more accurate damage detection. However, simply applying these technologies for damage detection is only the first step in risk inspection. In actual engineering risk management, many challenges still remain. Currently, for the detected damage locations, there is often a lack of systematic and comprehensive analysis methods, making it difficult to deeply understand the essential risk characteristics of the damage. In most cases, only a superficial record of the damage is made, without fully considering the basic riskiness of the damage location (such as factors like the severity of the damage itself and whether it is in a key part of the structure), risk timeliness (the development and change trend of the damage over time), risk correlation (the relationship of mutual influence between different damage locations), and risk variability under various load conditions (the dynamic change of damage risk under different load conditions). This incomplete and in-depth analysis makes it lack a scientific basis when determining whether to repair the damage location and how to repair it, and it is prone to misjudgment. Either excessive repair is carried out on some damages that do not need to be repaired urgently, wasting a large amount of human, material, and financial resources; or some damages with potential high risks that urgently need to be repaired are ignored, resulting in the continuous accumulation of risks and ultimately possibly triggering serious engineering safety accidents, bringing huge losses to society.

[0004] In view of the defects and deficiencies existing in the above-mentioned existing engineering area risk inspection and management methods, there is an urgent need for an innovative method and system that can combine advanced detection technologies, conduct in-depth and comprehensive analysis of the damage locations, thereby accurately controlling the risks in the engineering area, reasonably allocating repair resources, and effectively reducing the probability of engineering safety accidents to meet the growing demand for engineering safety guarantee. This is exactly the technical problem and the original intention of research and development that this invention aims to solve. Summary of the Invention

[0005] Aiming at the deficiencies of the existing technology, the purpose of the present invention is to provide an intelligent risk inspection method and system for engineering areas based on UAV aerial survey.

[0006] To achieve the above purpose, the present invention provides the following technical solutions: An intelligent risk inspection system for engineering areas based on UAV aerial survey includes a UAV regular inspection module, a risk identification and analysis module, and a risk early warning and feedback module; The UAV regular inspection module determines the target engineering area, conducts regular inspections on the target engineering area based on the UAV aerial survey path, and generates a damage detection log of the actual damage locations; The risk identification and analysis module, when the UAV completes one inspection work, constructs a finite element analysis model of the engineering area for the target engineering area, obtains all the damage detection logs generated in this inspection, updates the finite element analysis model of the engineering area, and further obtains the diverse risk indices of each actual damage location; The risk warning feedback module determines the damage repair requirement index of the actual damage location based on the damage appearance risk index and the diversified risk index of the actual damage location, and determines whether to repair the corresponding actual damage location.

[0007] Furthermore, the damage detection log of the actual damage location is generated as follows: The target engineering area is detected in real time through hyperspectral imaging technology. When an actual damage location appears in the target engineering area, a damage detection log of the actual damage location is generated; The damage detection log includes the damage location coordinates, the damage location characteristics, and the damage appearance risk index.

[0008] Furthermore, the damage location characteristics of the damage detection log are obtained as follows: The hyperspectral image data of the actual damage location is obtained, and feature extraction is performed on the hyperspectral image data to obtain the damage location characteristics; The damage appearance risk index of the damage detection log is obtained as follows: Obtain a damage detection log of this inspection, obtain the damage location characteristics of the damage detection log, input the damage location characteristics into the damage basic risk model, the damage basic risk model outputs the damage basic risk index, obtain all the damage detection logs generated before this actual damage location, sort all the damage detection logs in chronological order of generation, calculate the difference between the damage appearance risk index of the adjacent and later damage detection logs after sorting and the damage appearance risk index to obtain the damage appearance risk change index, set the damage appearance risk change threshold index, increase the number of times of damage appearance risk aggravation by one, and obtain the damage appearance risk index based on the damage basic risk index and the number of times of damage appearance risk aggravation.

[0009] Furthermore, update the finite element analysis model of the engineering area according to all the damage detection logs: Input the damage location characteristics and the damage location coordinates of all the damage detection logs into the finite element analysis model of the engineering area, and then add the damage location characteristics to the corresponding damage location in the three-dimensional model in the finite element analysis model of the engineering area, thereby completing the update of the finite element analysis model of the engineering area.

[0010] Furthermore, the diversification risk index of the actual damage location is obtained as follows: Set various load conditions in the target engineering area, select an actual damage location, obtain the load damage risk index of this actual damage location under various load conditions, compare the load damage risk indices of this actual damage location under various load conditions pairwise, calculate the absolute difference between the two compared load damage risk indices to obtain the load damage risk difference index, calculate the sum and average of all load damage risk difference indices to obtain the average load damage risk difference index, calculate the sum and average of the load damage risk indices of this actual damage location under various load conditions to obtain the average load damage risk index, and obtain the diversification risk index of the actual damage location based on the average load damage risk difference index and the average load damage risk index.

[0011] Furthermore, the load damage risk index of this actual damage location under a certain type of load condition is obtained as follows: Select a certain type of load condition, set various load parameters under this type of load condition, and then obtain the deterioration value of the damage characteristics of this actual damage location under various load parameters. Set the deterioration threshold of the damage characteristics. When the deterioration value of the damage characteristics of the actual damage location is greater than the deterioration threshold of the damage characteristics, increase the number of abnormal deterioration times of the damage characteristics by one. Calculate the sum and average of all deterioration values of the damage characteristics to obtain the average deterioration value of the damage characteristics. Obtain the load damage risk index of this actual damage location under this type of load condition based on the number of abnormal deterioration times of the damage characteristics and the average deterioration value of the damage characteristics.

[0012] Furthermore, the deterioration value of the damage characteristics of this actual damage location under a certain load parameter is obtained as follows: Select a certain load parameter, input this load parameter under the load condition into the updated finite element analysis model of the engineering area, then control the finite element analysis model of the engineering area to perform a simulation for a risk inspection cycle. After the simulation ends, obtain the simulated damage location characteristics of this actual damage location in the finite element analysis model of the engineering area. Match the damage location characteristics of this actual damage location with the simulated damage location characteristics to form a damage characteristic comparison group. Obtain the damage characteristic comparison model corresponding to this type of load condition. Input the damage characteristic comparison group into the damage characteristic comparison model, and the damage characteristic comparison model outputs the deterioration value of the damage characteristics.

[0013] Furthermore, determine whether to repair the corresponding actual damage location: Set the damage repair requirement threshold index. When the damage repair requirement index of the actual damage location is greater than the damage repair requirement threshold index, repair the corresponding actual damage location. When the damage repair requirement index of the actual damage location is less than or equal to the damage repair requirement threshold index, mark the corresponding actual damage location as an un-repaired damage location.

[0014] Further, an intelligent risk inspection method for an engineering area based on UAV aerial survey includes the following steps: Step 1: Determine the target engineering area, conduct regular inspections on the target engineering area based on the UAV aerial survey path, and generate a damage detection log of the actual damage location. Step 2: When the UAV completes one inspection work, update the finite element analysis model of the engineering area, and then obtain the diverse risk indexes of each actual damage location. Step 3: Based on the damage appearance risk index and the diverse risk index of the actual damage location, determine the damage repair requirement index of the actual damage location, and determine whether to repair the corresponding actual damage location.

[0015] Compared with the prior art, the present invention has the following beneficial effects: The system of the present invention sets up a UAV regular inspection module, a risk identification and analysis module, and a risk early warning feedback module. Through the UAV aerial survey technology combined with the hyperspectral imaging technology, regular risk inspections are carried out on the target engineering area, and the damage locations in the inspection process are further analyzed. Combining the basic risk, risk timeliness, risk relevance of the damage location and the risk variability under each load condition, in-depth analysis is carried out on the damage location, accurately locking the damage risks existing in the engineering area, and efficiently determining whether it is necessary to repair the corresponding damage location, ensuring accurate risk control of the engineering area. The method of the present invention can ensure that limited resources can be accurately invested in the damage locations that really need to be repaired, improve the resource utilization efficiency, and at the same time effectively reduce the probability of engineering safety accidents caused by sudden risks, ensuring the stable operation of the project. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 is the principle block diagram of the system of the present invention; Figure 2 is the flowchart for obtaining the deterioration value of the damage characteristics of the actual damage location under a load parameter; Figure 3 is the flowchart of the method of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0017] Example 1: Refer to Figures 1 to 2 , an intelligent risk inspection system for an engineering area based on UAV aerial survey includes a UAV regular inspection module, a risk identification and analysis module, and a risk early warning feedback module.

[0018] The UAV regular inspection module determines the target engineering area, and sets the cycle duration of the risk inspection cycle to be , regularly inspect the target engineering area based on the UAV aerial survey path. During the inspection process, the UAV conducts inspections along the UAV aerial survey path, and uses hyperspectral imaging technology to detect the target engineering area in real time. When the actual damage location in the target engineering area is detected (whether it is an un-repaired damage location or a newly emerged damage location, collectively referred to as the actual damage location), a damage detection log of this actual damage location is generated. The damage detection log includes the damage location coordinates, the damage location characteristics, and the damage appearance risk index.

[0019] The specific method for obtaining the damage location characteristics of the damage detection log is as follows: Obtain the hyperspectral image data of the actual damage location, perform feature extraction on the hyperspectral image data, and extract the damage location characteristics.

[0020] The specific method for obtaining the damage appearance risk index of the damage detection log is as follows: Obtain a damage detection log of this inspection, obtain the damage location characteristics of this damage detection log, input the damage location characteristics into the damage basic risk model, and the damage basic risk model outputs the damage basic risk index Felix. Obtain all the damage detection logs generated previously for this actual damage location (if this actual damage location is an un-repaired damage location, there are previous damage detection logs). Arrange all the damage detection logs in chronological order of generation. Calculate the difference between the damage appearance risk index of the adjacent and later damage detection logs after sorting and the damage appearance risk index to obtain the damage appearance risk change index. Set the damage appearance risk change threshold index (the damage appearance risk change threshold index is a preset index used to compare with the damage appearance risk change index), increase the damage appearance risk exacerbation count by one, and mark the damage appearance risk exacerbation count as , through calculate the damage appearance risk index of the damage detection log.

[0021] The specific construction method of the damage basic risk model: Collect multiple damage location characteristics, construct a deep learning model, use the damage location characteristics as the training data of the deep learning model, assign a damage basic risk index to each training data, and the value range of the damage basic risk index is (10 - 20). The larger the damage basic risk index, the greater the risk contained in the damage location characteristics. Divide the training data into a training set, a validation set, and a test set, with a ratio of 60%:20%:20%. Train the training set, the validation set, and the test set. Finally, construct the damage basic risk model.

[0022] Risk Identification and Analysis Module: When the drone completes an inspection mission, construct a finite element analysis model of the target engineering area, obtain all damage detection logs generated during this inspection, update the finite element analysis model of the engineering area based on all damage detection logs, and then obtain the diversified risk indices of each actual damage location.

[0023] Construct a finite element analysis model of the target engineering area: Obtain the design drawings of the target engineering area, create a 3D model of the target engineering area in general finite element analysis software based on the design drawings of the target engineering area, gradually construct the geometric models of each part according to the dimensions and shapes of the drawings to ensure the accuracy and integrity of the model. According to the design requirements and material properties, set parameters such as the overall dimensions of the model, material properties (such as elastic modulus, Poisson's ratio, density, etc.), connection methods (such as welding, bolt connection, etc.), structural thickness, structural strength, material density, etc. Ensure that the parameter settings conform to the actual engineering situation. After completing the parameter settings, conduct a preliminary verification and analysis of the model. By comparing the simulation results with the theoretical calculation values or actual test data, check the accuracy and rationality of the model. If any deviations or irrationalities are found in the model, promptly correct and adjust the model. Finally, construct a finite element analysis model of the engineering area.

[0024] Update the finite element analysis model of the engineering area according to all damage detection logs: Input the damage location characteristics and damage location coordinates of all damage detection logs into the finite element analysis model of the engineering area, and then add the damage location characteristics at the corresponding damage locations in the 3D model within the finite element analysis model of the engineering area, thereby completing the update of the finite element analysis model of the engineering area.

[0025] The diversified risk index of the actual damage location is obtained as follows: Set various load conditions of the target engineering area (load conditions include but are not limited to the following types: live load, wind load, snow load, temperature load), select an actual damage location, obtain the load damage risk index of this actual damage location under various load conditions, compare the load damage risk indices of this actual damage location under various load conditions pairwise, calculate the absolute difference between the two compared load damage risk indices to obtain the load damage risk difference index, calculate the sum and average of all load damage risk difference indices to obtain the average load damage risk difference index Calculate the sum and average of the load damage risk indices of this actual damage location under various load conditions to obtain the average load damage risk index , through Calculate to obtain the diversified risk index of this actual damage location , where L3 is the third coefficient, L4 is the fourth coefficient, the value of L3 is 2.81, and the value of L4 is 1.53.

[0026] The load damage risk index of the actual damage position under a type of load condition is obtained as follows: select a type of load condition, set multiple load parameters under this type of load condition (each load parameter is different, to simulate the situation under different load parameters, take wind load as an example, set different wind speed values, such as simulating breeze conditions, set the wind speed to 2m / s; simulate common urban strong wind weather, set the wind speed to 10m / s; for extreme weather scenes such as strong typhoons, the wind speed can be set to 30m / s or even higher. Different wind speeds will cause the structure to bear different lateral forces, and the effects on the displacement and stress of the structure will also be very different. Set multiple different wind direction angles degrees, such as 0° (north wind direction), 45°, 90° (east wind direction), etc. Because for most asymmetric structures, the force distribution and response of the structure are different when the wind blows from different directions. By changing the wind direction parameter, the mechanical properties of the structure when it is subjected to wind in various directions can be fully analyzed), and then the damage characteristic deterioration value of the actual damage position under various load parameters is obtained, and the damage characteristic deterioration threshold is set (the damage characteristic deterioration threshold is a preset value used for comparison with the damage characteristic deterioration value). When the damage characteristic deterioration value of the actual damage position is greater than the damage characteristic deterioration threshold, the number of abnormal damage characteristic deteriorations is increased by one, and the number of abnormal damage characteristic deteriorations is marked as , calculate the sum of all damage characteristic deterioration values and get the average damage characteristic deterioration value ,pass Calculate the load damage risk index of the actual damage position under this type of load condition , where L1 is the first coefficient, L2 is the second coefficient, the value of L1 is 1.21, and the value of L2 is 1.09.

[0027] The damage characteristic deterioration value of the actual damage position under a load parameter is specifically obtained as follows: select a load parameter, input the load parameter under the load condition into the updated engineering area finite element analysis model, then control the engineering area finite element analysis model to simulate a risk inspection cycle, and after the simulation, obtain the simulated damage position characteristics of the actual damage position in the engineering area finite element analysis model, match the (actual) damage position characteristics of the actual damage position with the simulated damage position characteristics into a damage characteristic comparison group, obtain the damage characteristic comparison model corresponding to this type of load condition, input the damage characteristic comparison group into the damage characteristic comparison model, and the damage characteristic comparison model outputs the damage characteristic deterioration value.

[0028] Different damage feature comparison models correspond to various load conditions, and each damage feature comparison model is constructed based on a deep learning model. In this embodiment, taking the live load as an example, the specific construction method of the damage feature comparison model for the live load is disclosed: collect multiple damage feature comparison groups regarding the live load, each damage feature comparison group includes an (actual) damage location feature under the live load and a simulated damage location feature, construct a deep learning model, use the damage feature comparison groups of the live load as the training data of the deep learning model, assign a damage feature deterioration value to each training data, and the value range of the damage feature deterioration value is (0.1 - 9.9). The larger the damage feature deterioration value, the more serious the damage of the simulated damage location feature compared to the (actual) damage location feature in the damage feature comparison group. Divide the training data into a training set, a validation set, and a test set with a ratio of 70%:15%:15%, and train the training set, the validation set, and the test set. Finally, construct the damage feature comparison model for the live load.

[0029] The risk early warning feedback module determines the damage repair requirement index of the actual damage location based on the damage appearance risk index and the diversified risk index of the actual damage location, sets a damage repair requirement threshold index (the damage repair requirement threshold index is a preset index used to compare with the damage repair requirement index). When the damage repair requirement index of the actual damage location is greater than the damage repair requirement threshold index, repair the corresponding actual damage location. When the damage repair requirement index of the actual damage location is less than or equal to the damage repair requirement threshold index, mark the corresponding actual damage location as an un-repaired damage location (that is, no repair is performed this time).

[0030] The damage repair requirement index of the actual damage location is determined specifically as follows: select an actual damage location, and obtain the damage appearance risk index of this actual damage location and the diversified risk index , and through calculate to obtain the damage repair requirement index of this actual damage location , where L5 is the fifth coefficient, L6 is the sixth coefficient, the value of L5 is 0.24, and the value of L6 is 0.46.

[0031] Set up the UAV regular inspection module, the risk identification and analysis module, and the risk early warning feedback module. Through the UAV aerial survey technology combined with the hyperspectral imaging technology, conduct regular risk inspections on the target engineering area, and further analyze the damage locations during the inspection process. Deeply analyze the damage locations in combination with the basic riskiness, risk timeliness, risk relevance of the damage locations, and the risk variability under various load conditions, accurately lock the damage risks existing in the engineering area, and efficiently determine whether it is necessary to repair the corresponding damage locations, ensuring precise risk control of the engineering area.

[0032] Example 2: Refer to Figure 3 , an intelligent risk inspection method for engineering areas based on UAV aerial survey, includes the following steps: Step 1: Determine the target engineering area, conduct regular inspections on the target engineering area based on the UAV aerial survey path, and generate a damage detection log of the actual damage location.

[0033] Step 2: When the UAV completes an inspection, update the finite element analysis model of the engineering area, and then obtain the diverse risk indices of each actual damage location.

[0034] Step 3: Based on the damage appearance risk index and the diverse risk index of the actual damage location, determine the damage repair requirement index of the actual damage location, and determine whether to repair the corresponding actual damage location.

[0035] The above method can ensure that limited resources can be accurately invested in the damage locations that really need to be repaired, improve the resource utilization efficiency, and at the same time effectively reduce the probability of engineering safety accidents caused by sudden risks, ensuring the stable operation of the project.

[0036] The above formulas are all dimensionless and take their numerical values for calculation. The preset parameters in the formulas are set by those skilled in the art according to the actual situation.

[0037] The above embodiments can be implemented in whole or in part by software, hardware, firmware, or any other combination. When implemented using software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on the computer, the processes or functions described in the embodiments of the present application are generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center in a wired or wireless (such as infrared, wireless, microwave, etc.) manner. The computer-readable storage medium can be any available medium that the computer can access, or a data storage device such as a server or data center that contains one or more collections of available media. The available media can be magnetic media (such as floppy disks, hard disks, magnetic tapes), optical media (such as DVDs), or semiconductor media. The semiconductor media can be a solid-state drive.

[0038] It should be understood that in various embodiments of the present application, the sequence numbers of the above processes do not imply the order of execution, and the order of execution of each process should be determined by its function and internal logic, and should not constitute any limitation to the implementation process of the embodiments of the present application.

[0039] Those of ordinary skill in the art can realize that the units and algorithm steps of each example described in combination with the embodiments disclosed herein can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are executed in a hardware or software manner depends on the specific application and design constraints of the technical solution. Professional technicians can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of the present application.

[0040] Those skilled in the art can clearly understand that for the convenience and conciseness of description, the specific working processes of the systems, devices, and units described above can refer to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0041] In several embodiments provided by the present application, it should be understood that the disclosed systems, devices, and methods can be implemented in other ways. For example, the device embodiments described above are only illustrative. For example, the division of the units is only a logical function division, and there can be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed couplings, direct couplings, or communication connections to each other can be through some interfaces, and the indirect couplings or communication connections of the devices or units can be in electrical, mechanical, or other forms.

[0042] If the function is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on such an understanding, the technical solution of the present application, in essence, or the part that contributes to the prior art, or a part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in various embodiments of the present application. The foregoing storage medium includes: USB flash drives, mobile hard disks, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks, and other various media that can store program codes.

[0043] As described above, it is only the specific implementation manner of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present application can easily think of changes or substitutions, which should all be covered within the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims described above.

Claims

1. An intelligent risk inspection system for engineering areas based on UAV aerial survey, characterized in that, It includes a regular inspection module for drones, a risk identification and analysis module, and a risk warning and feedback module; The regular inspection module for drones determines the target engineering area, conducts regular inspections on the target engineering area based on the drone aerial survey path, and generates a damage detection log of the actual damage location; When the drone completes one inspection work, the risk identification and analysis module constructs a finite element analysis model of the engineering area for the target engineering area, obtains all the damage detection logs generated by this inspection, and updates the finite element analysis model of the engineering area, thereby obtaining the diverse risk indices of each actual damage location; Based on the damage appearance risk index and the diverse risk index of the actual damage location, the risk warning and feedback module determines the damage repair requirement index of the actual damage location and determines whether to repair the corresponding actual damage location.

2. The intelligent risk inspection system for engineering areas based on UAV aerial survey according to claim 1, characterized in that, The process of generating the damage detection log of the actual damage location is as follows: The target engineering area is detected in real time through hyperspectral imaging technology. When an actual damage location appears in the target engineering area, a damage detection log of this actual damage location is generated; The damage detection log includes the damage location coordinates, the damage location characteristics, and the damage appearance risk index.

3. The intelligent risk inspection system for engineering areas based on UAV aerial survey according to claim 2, wherein The method for specifically obtaining the damage location characteristics of the damage detection log is as follows: Obtain the hyperspectral image data of the actual damage location, perform feature extraction on the hyperspectral image data, and extract the damage location characteristics; The method for specifically obtaining the damage appearance risk index of the damage detection log is as follows: Obtain a damage detection log of this inspection, obtain the damage location characteristics of this damage detection log, input the damage location characteristics into the damage basic risk model, the damage basic risk model outputs the damage basic risk index, obtain all the damage detection logs generated before this actual damage location, sort all the damage detection logs in the order of generation time, calculate the difference between the damage appearance risk indices of the adjacent and later damage detection logs after sorting, calculate the damage appearance risk change index, set the damage appearance risk change threshold index, increase the number of times of damage appearance risk aggravation by one, and obtain the damage appearance risk index based on the damage basic risk index and the number of times of damage appearance risk aggravation.

4. The intelligent risk inspection system for engineering areas based on UAV aerial survey according to claim 1, wherein, Update the finite element analysis model of the engineering area according to all the damage detection logs: Input the damage location characteristics and the damage location coordinates of all the damage detection logs into the finite element analysis model of the engineering area, and then add the damage location characteristics to the corresponding damage locations in the three-dimensional model in the finite element analysis model of the engineering area, thereby completing the update of the finite element analysis model of the engineering area.

5. The intelligent risk inspection system for engineering areas based on UAV aerial survey according to claim 1, characterized in that The diversification risk index of the actual damage location is obtained as follows: Set various load conditions in the target engineering area, select an actual damage location, obtain the load damage risk index of this actual damage location under various load conditions, compare the load damage risk indices of this actual damage location under various load conditions pairwise, calculate the absolute difference between the two compared load damage risk indices to obtain the load damage risk difference index, calculate the sum and average of all load damage risk difference indices to obtain the average load damage risk difference index, calculate the sum and average of the load damage risk indices of this actual damage location under various load conditions to obtain the average load damage risk index, and obtain the diversification risk index of the actual damage location based on the average load damage risk difference index and the average load damage risk index.

6. The intelligent risk inspection system for engineering areas based on UAV aerial survey according to claim 5, characterized in that, The load damage risk index of the actual damage location under a certain type of load condition is obtained as follows: Select a certain type of load condition, set various load parameters under this type of load condition, and then obtain the deterioration value of the damage characteristics of this actual damage location under various load parameters. Set the deterioration threshold of the damage characteristics. When the deterioration value of the damage characteristics of the actual damage location is greater than the deterioration threshold of the damage characteristics, increase the number of abnormal deterioration times of the damage characteristics by one. Calculate the sum and average of all deterioration values of the damage characteristics to obtain the average deterioration value of the damage characteristics. Obtain the load damage risk index of this actual damage location under this type of load condition based on the number of abnormal deterioration times of the damage characteristics and the average deterioration value of the damage characteristics.

7. The intelligent risk inspection system for engineering areas based on UAV aerial survey according to claim 6, characterized in that, The deterioration value of the damage characteristics of the actual damage location under a certain load parameter is obtained as follows: Select a certain load parameter, input this load parameter under the load condition into the updated finite element analysis model of the engineering area, then control the finite element analysis model of the engineering area to perform a risk inspection cycle simulation. After the simulation is completed, obtain the simulated damage location characteristics of this actual damage location in the finite element analysis model of the engineering area. Match the damage location characteristics of this actual damage location with the simulated damage location characteristics to form a damage characteristic comparison group. Obtain the damage characteristic comparison model corresponding to this type of load condition. Input the damage characteristic comparison group into the damage characteristic comparison model, and the damage characteristic comparison model outputs the deterioration value of the damage characteristics.

8. The intelligent risk inspection system for engineering areas based on UAV aerial survey according to claim 1, characterized in that, Determine whether to repair the corresponding actual damage location: Set the damage repair requirement threshold index. When the damage repair requirement index of the actual damage location is greater than the damage repair requirement threshold index, repair the corresponding actual damage location. When the damage repair requirement index of the actual damage location is less than or equal to the damage repair requirement threshold index, mark the corresponding actual damage location as an un-repaired damage location.

9. The intelligent risk inspection method for an engineering area based on UAV aerial survey is applied to the intelligent risk inspection system for an engineering area based on UAV aerial survey according to any one of claims 1-8, and is characterized in that, It includes the following steps: Step 1: Determine the target engineering area, regularly inspect the target engineering area based on the UAV aerial survey path, and generate a damage detection log of the actual damage location. Step 2: When the UAV completes an inspection work, update the finite element analysis model of the engineering area, and then obtain the diversification risk index of each actual damage location. Step 3: Based on the damage appearance risk index and the diversification risk index of the actual damage location, determine the damage repair requirement index of the actual damage location, and determine whether to repair the corresponding actual damage location.

Images