Building use process maintenance management system and method based on artificial intelligence

Abstract

The invention relates to the technical field of building maintenance management, and discloses a building use process maintenance management system and method based on artificial intelligence, and the system comprises a building maintenance control module, a building maintenance problem recognition processing module, and a building maintenance problem output module. An unmanned aerial vehicle carries a cloud lens to accurately collect an outer wall state image of a real-time inspection position of a building outer wall, and intelligent identification of a damage problem type of the building outer wall is carried out by combining an intelligent search algorithm and scientifically preset image information of different building outer wall damage types; based on an artificial intelligence algorithm, building outer wall damage type identification information and standard stored different building outer wall damage type repair process information are combined to carry out intelligent assessment on a damage repair scheme of the building outer wall, and meanwhile, building outer wall damage position coordinates are searched autonomously; intelligent data analysis of a building outer wall damage repair scheme and a damage position is realized, and intelligence and reliability of building maintenance operation are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of building maintenance management, and in particular to an artificial intelligence-based building use process maintenance management system and method. Background Art

[0002] Building maintenance refers to a series of work that involves regular inspection, repair, upkeep and management of buildings. As buildings age and their usage time increases, building maintenance becomes particularly important. Building maintenance is essential for the sustainable use and value retention of buildings. Regular inspections and repairs can ensure the structural safety of the building, the normal operation of the equipment, and the comfort of the environment. In addition, building maintenance can also help extend the service life of the building and reduce maintenance costs. Building maintenance work includes regular inspections, maintenance and upkeep, and equipment maintenance. Regular inspections: This means that the building maintenance service provider will regularly inspect the building to check the damage to the walls, ceilings, floors, doors, windows, stairs, and other parts, as well as the operating conditions of pipes, electrical appliances, HVAC equipment, etc. Repair and upkeep: This means that corresponding repair and upkeep work will be carried out based on the results of regular inspections. Repairs may include repairing damaged walls, replacing leaking water pipes, etc. Maintenance may include cleaning, painting, and replacing aging hardware accessories. Equipment maintenance: This means that the building maintenance service provider is also responsible for the maintenance of the equipment in the building, such as elevators, heating equipment, ventilation systems, etc. They will conduct regular inspections, upkeep and repairs to ensure the normal operation and safe use of the equipment. Equipment maintenance: means that building maintenance service providers are also responsible for the maintenance of equipment in the building, such as elevators, heating equipment, ventilation systems, etc. They will conduct regular inspections, maintenance and repairs to ensure the normal operation and safe use of the equipment. Among them, the maintenance and management of building exterior walls is very important to increase the service life of buildings and reduce the risk of falling objects from high altitudes; however, the existing maintenance of building exterior walls basically relies on manual regular high-altitude inspections, which cannot perform efficient and safe maintenance and inspections on building exterior walls, nor can they conduct scientific and intelligent analysis and processing of building exterior wall damage problems.

[0003] A Chinese invention patent application with publication number CN109716381A discloses a video-based integrated management method for building construction and maintenance, which produces a series of processes from the construction stage to completion and maintenance into a video and registers it on a building management server, so that building owners or occupants can confirm the video and realize transparent management of building construction; however, the above technical solution cannot provide owners with safe and reliable maintenance management technical services during the use of the building. Summary of the invention

[0004] 1. Technical issues to be solved

[0005] In order to solve the problem that the above-mentioned existing building exterior wall maintenance operations basically rely on manual regular high-altitude inspection operations, cannot carry out efficient and safe maintenance and inspection of building exterior walls, and cannot carry out scientific and intelligent analysis and processing of building exterior wall damage problems, the above purpose is to scientifically establish building exterior wall maintenance inspection position coordinate information, efficiently and safely perform building exterior wall maintenance operations, intelligently identify building exterior wall damage types, intelligently evaluate building exterior wall damage repair procedures, accurately locate building exterior wall damage locations, and timely and efficiently feedback abnormal building exterior wall maintenance results.

[0006] (II) Technical solution

[0007] The present invention is implemented by the following technical solution: a method for maintaining and managing a building during use, the method comprising the following steps:

[0008] S1. Collect the 3D model data of the building exterior wall;

[0009] S2, constructing and processing the spatial position coordinates of the building exterior wall maintenance inspection based on the building exterior wall three-dimensional model data to generate the building exterior wall maintenance inspection spatial position coordinate data;

[0010] S3, performing building exterior wall maintenance work based on the building exterior wall maintenance inspection space position coordinate data;

[0011] S4, collecting real-time inspection image data of building exterior walls and performing identification processing on the types of damage problems of building exterior walls with image data of different types of damage of building exterior walls, and generating identification data of the types of damage problems of building exterior walls; if the data does not exist, continue to perform steps S3 and S4 until the maintenance work of the building exterior walls is completed;

[0012] S5. When it exists, analyzing and processing the damage repair scheme of the building exterior wall according to the building exterior wall damage type identification data and the repair process data of different building exterior wall damage types to generate building exterior wall damage repair scheme analysis data;

[0013] S6, performing location processing of the building exterior wall damage position based on the building exterior wall damage type identification data and the building exterior wall maintenance inspection space position coordinate data to generate building exterior wall damage problem position coordinate data;

[0014] S7. Construct the data of abnormal problems in building exterior wall maintenance and execute the feedback and display operation of abnormal problems in building exterior wall maintenance; when the feedback and display operation of abnormal problems in building exterior wall maintenance is completed, continue to execute steps S3 and S4 until the building exterior wall maintenance operation is completed.

[0015] Preferably, the steps of collecting the three-dimensional model data of the building exterior wall are as follows:

[0016] S11. Use a 3D laser scanner mounted on a drone to collect the 3D solid model of the entire exterior wall of the target building online and generate 3D model data K of the exterior wall of the building.

[0017] Preferably, the spatial position coordinate construction processing of the building exterior wall maintenance inspection is performed based on the building exterior wall three-dimensional model data, and the operation steps of generating the building exterior wall maintenance inspection spatial position coordinate data are as follows:

[0018] S21, taking the plane where the most prominent exterior wall structure of the target building on each direction of the exterior wall surface corresponding to the building exterior wall three-dimensional model data K is located as the building exterior wall reference surface, dividing the building exterior wall reference surface into grids according to squares with a side length of R, establishing a spatial rectangular coordinate system with the plane where the target building intersects the ground as the base of the spatial coordinate system, measuring the spatial coordinates of all the square grid center points in the spatial rectangular coordinate system and generating a building exterior wall maintenance inspection spatial position coordinate data set A = (a1, ..., a u ,…,a η ), u=1,2,3,…,eta; where a u represents u building exterior wall maintenance and inspection space position coordinate data, η represents the maximum number of building exterior wall maintenance and inspection space positions; the building exterior wall maintenance and inspection space position coordinate data includes the horizontal coordinate, vertical coordinate and vertical coordinate of the building exterior wall maintenance and inspection space position.

[0019] Preferably, the operation steps of performing building exterior wall maintenance work based on the building exterior wall maintenance inspection spatial position coordinate data are as follows:

[0020] S31, using a drone equipped with a cloud camera to detect the building exterior wall maintenance and inspection spatial position coordinate data set A according to the building exterior wall maintenance and inspection spatial position coordinate data set A. u Carry out maintenance and inspection work on the target building's exterior walls in an orderly manner according to the building's exterior wall maintenance and inspection space location number.

[0021] Preferably, the steps of collecting real-time inspection image data of building exterior walls and performing damage problem type identification processing on building exterior walls with image data of different building exterior wall damage types to generate building exterior wall damage type identification data are as follows:

[0022] S41, during the maintenance and inspection operation of the target building exterior wall, the cloud camera carried by the drone is used to simultaneously monitor the spatial position coordinate data a of the building exterior wall maintenance and inspection u The corresponding target building exterior wall specific maintenance inspection position performs building exterior wall real-time status image acquisition and generates building exterior wall real-time inspection image data set B u =(b u,1 ,…,b u,v ,…,b u,ι), v = 1, 2, 3, ..., ι; where B u represents the real-time inspection image data set of the building exterior wall corresponding to the u-th building exterior wall maintenance inspection space position; b u,v represents the vth building exterior wall real-time inspection image data collected at the uth building exterior wall maintenance inspection space position, ι represents the maximum number of collected building exterior wall real-time inspection images;

[0023] S42, establish different building exterior wall damage type image data set C = (c1, ..., c o ,…,c κ ), o=1,2,3,…,κ; where c o represents o different types of building exterior wall damage image data, κ represents the maximum number of building exterior wall damage types; the building exterior wall damage types include tile building exterior wall cracks, tile building exterior wall detachment, tile building exterior wall seepage, metal plate building exterior wall cracks, metal plate building exterior wall detachment, metal plate building exterior wall seepage, glass curtain wall building exterior wall cracks, glass curtain wall building exterior wall detachment, glass curtain wall building exterior wall seepage, composite board building exterior wall cracks, composite board building exterior wall detachment and composite board building exterior wall seepage;

[0024] S43, using the KD tree nearest neighbor search algorithm to search the building exterior wall real-time inspection image data set B u The real-time inspection image data b of the building exterior wall u,v The image data c of different building exterior wall damage types in the image data set C of different building exterior wall damage types o Perform image feature matching, and generate building exterior wall damage type identification data C′ based on the image feature matching results;

[0025] When b u,v With c o If the image feature matching is successful, it means that there is a building exterior wall damage problem at the u-th building exterior wall maintenance inspection space position, and the output building exterior wall damage type identification data C′ is the o-th building exterior wall damage type problem at the u-th building exterior wall maintenance inspection space position;

[0026] When b u,v With c o If the image feature matching is unsuccessful, it means that there is no building exterior wall damage problem at the u-th building exterior wall maintenance inspection space position, and the output building exterior wall damage type identification data C′ is non-existent. At this time, steps S31, S41, and S43 are continued in the order of the building exterior wall maintenance inspection space position numbering until the building exterior wall maintenance operation is completed.

[0027] Preferably, when the condition exists, the damage repair scheme of the building exterior wall is analyzed and processed according to the building exterior wall damage type identification data and the repair process data of different building exterior wall damage types, and the operation steps of generating the building exterior wall damage repair scheme analysis data are as follows:

[0028] S51. Establish a data set D = (d1, ..., d o ,…,d κ ), where d o Represents repair process data of different building exterior wall damage types corresponding to o types of building exterior wall damage; the repair process data of different building exterior wall damage types represent standard exterior wall repair step process data for different building exterior wall damage types;

[0029] S52, when the building exterior wall damage type identification data C' exists, compare the building exterior wall damage type identification data C' with the different building exterior wall damage type repair process data set D o According to the keyword matching of the building exterior wall damage type, the repair process data d of different building exterior wall damage types corresponding to the building exterior wall damage type identification data C' are searched out. o And construct analysis data for building exterior wall damage repair plan Execute and generate analysis data of building exterior wall damage repair plan The specific steps are as follows:

[0030] S521, initialization, defining the relevant structural parameters as vectors, and searching the data set D for the repair process data of different building exterior wall damage types in the search space. o In the κ-dimensional optimization problem, the process type search sand cat represents the 1×κ array of the problem solution, and each variable value q1,q2,…,q κ are all floating point numbers, and each variable value q1 to q κ represents a value between a lower bound and an upper bound in the search space of the repair process data set D of different building exterior wall damage types;

[0031] S522, Searching for prey, the final and main parameters controlling the transition between exploration and exploitation phases are when When the process type search sand cat searches the search space of the different building exterior wall damage type repair process data set D for the different building exterior wall damage type repair process data d that matches the building exterior wall damage type identification data C′. oThe process type search sand cat search process relies on the release of low-frequency noise to search for the different building exterior wall damage type repair process data d that matches the building exterior wall damage type identification data C'. o , assuming that the sensitivity range of the process type search sand cat ω is from 0 to 2kHz, ξ represents the auditory characteristics inspired by the process type search sand cat, assuming its value is 2, t is the current iteration number, T is the maximum iteration number, rand(0,1) represents a random number with a value of 0 to 1; in Represents the sensitivity vector; each process type searches for the sand cat according to the best candidate position J best and the current position J and its sensitivity range ω to update its own position, that is, to update the different building exterior wall damage type repair process data d matching the building exterior wall damage type identification data C' in the search space of the different building exterior wall damage type repair process data set D o The position calculation formula is as follows: J(t+1)=ω×(J best (t)-rand(0,1)×J best (t)), where J(t+1) represents the current position of the sand cat individual in the search space of the repair process data set D for the different building exterior wall damage types in the t+1th iteration process type search, and J best (t) represents the t-times iteration process type search sand cat searches for the different building exterior wall damage type repair process data set D in the search space of the different building exterior wall damage type repair process data set D that best matches the building exterior wall damage type identification data C′. o The best candidate position for

[0032] S523, attack prey, when When the process type search sand cat is used, it searches in the search space of the repair process data set D of the different building exterior wall damage types, and uses the best candidate position J best Generate a random position with the current position J, that is, randomly search for the different building exterior wall damage type repair process data set D that matches the building exterior wall damage type identification data C' in the search space of the different building exterior wall damage type repair process data set D o Assuming that the sensitivity range of the process type search sand cat is a circle, the roulette method is used to randomly select an angle ψ for each process type search sand cat, and the random position calculation formula is:

[0033] Perform the repair process data d of the different building exterior wall damage types that match the building exterior wall damage type identification data C o A random position search of represents the sensitivity vector, cos(ψ) represents the cosine value of the angle ψ, and J rand represents the random position of the process type search sand cat in the search space of the different building exterior wall damage type repair process data set D, and the random position makes the process type search sand cat approach and attack the prey, that is, the different building exterior wall damage type repair process data d matching the building exterior wall damage type identification data C' is searched in the search space of the different building exterior wall damage type repair process data set D. o ;

[0034] S524, when the search algorithm reaches the maximum number of iterations T, output the repair process data d of the different building exterior wall damage types that best matches the building exterior wall damage type identification data C′ o ;

[0035] S525, the repair process data of different building exterior wall damage types output in step S524 are o The corresponding building exterior wall damage type repair process data is constructed as building exterior wall damage repair plan analysis data

[0036] 7. A building maintenance management method according to claim 6, characterized in that: based on the building exterior wall damage type identification data and the building exterior wall maintenance inspection space position coordinate data, the building exterior wall damage position location processing is performed to generate the building exterior wall damage problem position coordinate data in the following steps:

[0037] S61, using a depth-first search algorithm, according to the building exterior wall damage type identification data C′ corresponding to the building exterior wall maintenance inspection spatial position number when it exists, search for the building exterior wall maintenance inspection spatial position coordinate data set A in the search space to find the building exterior wall maintenance inspection spatial position coordinate data a corresponding to the building exterior wall damage problem. u And constructed as building exterior wall damage problem location coordinate data f, the building exterior wall damage problem location coordinate data includes the horizontal coordinate, the vertical coordinate and the vertical coordinate of the building exterior wall damage problem location.

[0038] Preferably, the steps of constructing the building exterior wall maintenance abnormality problem data and performing the building exterior wall maintenance abnormality problem feedback display operation, when the building exterior wall maintenance abnormality problem feedback display operation is completed, continuing to perform steps S3 and S4 until the building exterior wall maintenance operation is completed are as follows:

[0039] S71, the building exterior wall damage type identification data C', the building exterior wall damage repair plan analysis data The data is combined with the building exterior wall damage problem location coordinate data f to construct the building exterior wall maintenance abnormality problem data M, where

[0040] S72. Push and feed back the abnormal problem data M of the building exterior wall maintenance to the building maintenance management platform through the Internet of Things communication network and display it on a display screen. When the feedback and display operation of the abnormal problem of the building exterior wall maintenance is completed, continue to execute S31, S41, and S43 in the order of the spatial position number of the building exterior wall maintenance inspection until the building exterior wall maintenance operation is completed.

[0041] An artificial intelligence-based building maintenance management system for implementing a building maintenance management method, the system comprising a building maintenance control module, a building maintenance problem identification and processing module, and a building maintenance problem output module;

[0042] The building maintenance control module includes a building exterior wall three-dimensional model acquisition unit, a building exterior wall maintenance inspection position construction unit, and a building exterior wall maintenance operation execution unit;

[0043] The building exterior wall three-dimensional model acquisition unit collects the building exterior wall three-dimensional model data by means of a three-dimensional laser scanner mounted on an unmanned aerial vehicle; the building exterior wall maintenance inspection position construction unit performs spatial position coordinate construction processing for building exterior wall maintenance inspection based on the building exterior wall three-dimensional model data to generate building exterior wall maintenance inspection spatial position coordinate data; the building exterior wall maintenance operation execution unit executes the building exterior wall maintenance operation based on the building exterior wall maintenance inspection spatial position coordinate data;

[0044] The building maintenance problem identification and processing module includes a building exterior wall real-time inspection image acquisition unit, a building exterior wall damage type image storage unit, a building exterior wall damage type identification unit, a building exterior wall damage type repair process storage unit, a building exterior wall damage repair process analysis unit, and a building exterior wall damage location information search unit;

[0045] The real-time inspection image acquisition unit of the building exterior wall collects real-time inspection image data of the building exterior wall through a cloud lens carried by an unmanned aerial vehicle; the image storage unit of different building exterior wall damage types is used to store image data of different building exterior wall damage types; the building exterior wall damage type identification unit performs damage problem type identification processing on the building exterior wall based on the real-time inspection image data of the building exterior wall and the image data of different building exterior wall damage types, and generates building exterior wall damage type identification data; the repair process storage unit of different building exterior wall damage types is used to store repair process data of different building exterior wall damage types; the building exterior wall damage repair process analysis unit performs damage repair solution analysis processing on the building exterior wall based on the building exterior wall damage type identification data combined with the repair process data of different building exterior wall damage types based on an artificial intelligence algorithm, and generates building exterior wall damage repair solution analysis data; the building exterior wall damage position information search unit performs building exterior wall damage position positioning processing based on the building exterior wall damage type identification data and the building exterior wall maintenance inspection space position coordinate data, and generates building exterior wall damage problem position coordinate data;

[0046] The building maintenance problem output module includes a building exterior wall maintenance abnormality problem information collection unit and a building exterior wall maintenance abnormality problem feedback display unit;

[0047] The building exterior wall maintenance abnormality problem information collection unit is used to construct building exterior wall maintenance abnormality problem data; the building exterior wall maintenance abnormality problem feedback display unit performs building exterior wall maintenance abnormality problem feedback display operations based on the building exterior wall maintenance abnormality problem data.

[0048] (III) Beneficial effects

[0049] The present invention provides a building maintenance management system and method based on artificial intelligence, which has the following beneficial effects:

[0050] 1. Through the building exterior wall three-dimensional model acquisition unit, a drone-mounted three-dimensional laser scanner is used to realize accurate and efficient acquisition of the exterior wall three-dimensional model features of the target building maintenance object; the building exterior wall maintenance inspection position construction unit divides the model plane grid of the building exterior wall three-dimensional model to scientifically and accurately establish the position coordinates of the building exterior wall maintenance operation, ensuring that the building exterior wall maintenance operation fully covers the entire building exterior wall and improves the quality of the building maintenance operation; the building exterior wall maintenance operation execution unit executes the building exterior wall maintenance operation based on the building exterior wall maintenance inspection space position coordinate information combined with the drone-mounted cloud camera, thereby improving the efficiency and safety of the building maintenance operation.

[0051] 2. Accurately collect the exterior wall status images of the real-time inspection position of the building exterior wall through the cloud camera carried by the drone, and combine the intelligent search algorithm with the scientific preset image information of different types of building exterior wall damage to intelligently identify the types of damage problems of the building exterior wall, and realize accurate analysis of the building exterior wall damage problems; based on the artificial intelligence algorithm, combined with the building exterior wall damage type identification information and the standard stored information on the repair process of different building exterior wall damage types, intelligent evaluation of the building exterior wall damage repair plan is carried out, and at the same time, the coordinates of the building exterior wall damage location are autonomously searched to realize the intelligent data analysis of the building exterior wall damage repair plan and the damage location, thereby improving the intelligence and reliability of building maintenance operations.

[0052] 3. Through the online dynamic collection of information on damage problems, repair plans and specific locations of damage to building exterior walls, and the scientific construction of information on abnormal maintenance problems of building exterior walls, efficient and accurate collection of building maintenance operation problems can be achieved; the information on abnormal maintenance problems of building exterior walls is combined with the Internet of Things communication network to achieve instant transmission to the building maintenance management platform, and is instantly visualized and output through a display screen, thereby achieving efficient and intuitive feedback of building maintenance operation result information. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] Figure 1 A schematic diagram of a module of a building use process maintenance management system based on artificial intelligence provided by the present invention;

[0054] Figure 2 The present invention provides a flow chart of a method for maintaining and managing a building during use. DETAILED DESCRIPTION

[0055] 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.

[0056] The embodiments of the building use process maintenance management system and method based on artificial intelligence are as follows:

[0057] Embodiment 1:

[0058] See also Figure 1 - Figure 2 A method for maintaining and managing a building during use, the method comprising the following steps:

[0059] S1. Collect the 3D model data of the building exterior wall;

[0060] S2. Perform spatial position coordinate construction processing on the building exterior wall maintenance inspection based on the building exterior wall three-dimensional model data to generate the building exterior wall maintenance inspection spatial position coordinate data;

[0061] S3, performing building exterior wall maintenance work based on the building exterior wall maintenance inspection spatial position coordinate data;

[0062] S4, collecting real-time inspection image data of building exterior walls and performing identification processing on the types of damage problems of building exterior walls with image data of different types of damage of building exterior walls, and generating identification data of the types of damage problems of building exterior walls; if the data does not exist, continue to perform steps S3 and S4 until the maintenance work of the building exterior walls is completed;

[0063] S5. When it exists, analyzing and processing the damage repair scheme of the building exterior wall according to the building exterior wall damage type identification data and the repair process data of different building exterior wall damage types, and generating building exterior wall damage repair scheme analysis data;

[0064] S6. Based on the building exterior wall damage type identification data and the building exterior wall maintenance inspection spatial position coordinate data, the building exterior wall damage position location processing is performed to generate the building exterior wall damage problem position coordinate data;

[0065] S7. Construct the data of abnormal problems in building exterior wall maintenance and execute the feedback and display operation of abnormal problems in building exterior wall maintenance; when the feedback and display operation of abnormal problems in building exterior wall maintenance is completed, continue to execute steps S3 and S4 until the building exterior wall maintenance operation is completed.

[0066] For further information, see Figure 1 - Figure 2 ,The steps for collecting the 3D model data of the building exterior wall are as follows:

[0067] S11. Use a 3D laser scanner mounted on a drone to collect the 3D solid model of the entire exterior wall of the target building online and generate 3D model data K of the exterior wall of the building.

[0068] The steps for constructing and processing the spatial position coordinates of the building exterior wall maintenance inspection based on the building exterior wall three-dimensional model data and generating the spatial position coordinates of the building exterior wall maintenance inspection data are as follows:

[0069] S21. The plane where the most prominent exterior wall structure of the target building in all directions corresponding to the building exterior wall three-dimensional model data K is located is used as the building exterior wall reference surface, and the building exterior wall reference surface is gridded according to squares with a side length of R. A spatial rectangular coordinate system is established with the plane where the target building intersects the ground as the base of the spatial coordinate system, and the spatial coordinates of all the square grid center points in the spatial rectangular coordinate system are measured to generate the building exterior wall maintenance inspection spatial position coordinate data set A = (a1,…,a u ,…,aη ), u=1,2,3,…,eta; where a u represents u building exterior wall maintenance and inspection space position coordinate data, η represents the maximum number of building exterior wall maintenance and inspection space positions; the building exterior wall maintenance and inspection space position coordinate data includes the horizontal coordinate, vertical coordinate and vertical coordinate of the building exterior wall maintenance and inspection space position.

[0070] The steps for performing building exterior wall maintenance work based on the building exterior wall maintenance inspection spatial position coordinate data are as follows:

[0071] S31, using a drone equipped with a cloud camera to detect the building exterior wall maintenance and inspection spatial position coordinate data set A based on the building exterior wall maintenance and inspection spatial position coordinate data a u Carry out maintenance and inspection work on the target building's exterior walls in an orderly manner according to the building's exterior wall maintenance and inspection space location number.

[0072] Through the building exterior wall three-dimensional model acquisition unit, a drone-mounted three-dimensional laser scanner is used to achieve accurate and efficient acquisition of the exterior wall three-dimensional model features of the target building maintenance object; the building exterior wall maintenance inspection position construction unit divides the building exterior wall three-dimensional model into a model plane grid to scientifically and accurately establish the position coordinates of the building exterior wall maintenance operation, ensuring that the building exterior wall maintenance operation fully covers the entire building exterior wall and improves the quality of the building maintenance operation; the building exterior wall maintenance operation execution unit executes the building exterior wall maintenance operation based on the building exterior wall maintenance inspection space position coordinate information combined with the drone-mounted cloud camera, thereby improving the efficiency and safety of the building maintenance operation.

[0073] For further information, see Figure 1 - Figure 2 , collect real-time inspection image data of building exterior walls and perform damage problem type identification processing on building exterior walls with image data of different building exterior wall damage types. The operation steps for generating building exterior wall damage type identification data are as follows:

[0074] S41. During the maintenance and inspection of the target building's exterior wall, the cloud camera carried by the drone is used to simultaneously obtain the spatial position coordinate data of the building's exterior wall maintenance and inspection. u The corresponding target building exterior wall specific maintenance inspection position performs building exterior wall real-time status image acquisition and generates building exterior wall real-time inspection image data set B u =(b u,1 ,…,b u,v ,…,b u,ι ), v = 1, 2, 3, ..., ι; where B u represents the real-time inspection image data set of the building exterior wall corresponding to the u-th building exterior wall maintenance inspection space position; b u,vrepresents the vth building exterior wall real-time inspection image data collected at the uth building exterior wall maintenance inspection space position, ι represents the maximum number of collected building exterior wall real-time inspection images;

[0075] S42, establish different building exterior wall damage type image data set C = (c1, ..., c o ,…,c κ ), o=1,2,3,…,κ; where c o represents o different types of building exterior wall damage image data, κ represents the maximum number of building exterior wall damage types; building exterior wall damage types include tile building exterior wall cracks, tile building exterior wall detachment, tile building exterior wall seepage, metal plate building exterior wall cracks, metal plate building exterior wall detachment, metal plate building exterior wall seepage, glass curtain wall building exterior wall cracks, glass curtain wall building exterior wall detachment, glass curtain wall building exterior wall seepage, composite board building exterior wall cracks, composite board building exterior wall detachment and composite board building exterior wall seepage;

[0076] S43, using the KD tree nearest neighbor search algorithm to collect the building exterior wall real-time inspection image data set B u Real-time inspection image data of building exterior walls u,v The image data c of different building exterior wall damage types in the image data set C o Perform image feature matching, and generate building exterior wall damage type identification data C′ based on the image feature matching results;

[0077] When b u,v With c o If the image feature matching is successful, it means that there is a building exterior wall damage problem at the u-th building exterior wall maintenance inspection space position, and the output building exterior wall damage type identification data C′ is the o-th building exterior wall damage type problem at the u-th building exterior wall maintenance inspection space position;

[0078] When b u,v With c o If the image feature matching is unsuccessful, it means that there is no building exterior wall damage problem at the u-th building exterior wall maintenance inspection space position, and the output building exterior wall damage type identification data C′ is non-existent. At this time, steps S31, S41, and S43 are continued in the order of the building exterior wall maintenance inspection space position numbering until the building exterior wall maintenance operation is completed.

[0079] When it exists, the damage repair scheme of the building exterior wall is analyzed and processed according to the building exterior wall damage type identification data and the repair process data of different building exterior wall damage types. The operation steps for generating the building exterior wall damage repair scheme analysis data are as follows:

[0080] S51. Establish a data set D = (d1, ..., d o ,…,d κ ), where d o Represents the repair process data of different building exterior wall damage types corresponding to o types of building exterior wall damage; the repair process data of different building exterior wall damage types represents the standard exterior wall repair step process data for different building exterior wall damage types;

[0081] S52, when the building exterior wall damage type identification data C' exists, the building exterior wall damage type identification data C' is combined with the different building exterior wall damage type repair process data d in the different building exterior wall damage type repair process data set D. o According to the keyword matching of the building exterior wall damage type, search for the repair process data d of different building exterior wall damage types corresponding to the building exterior wall damage type identification data C′ o And construct analysis data for building exterior wall damage repair plan Execute and generate analysis data of building exterior wall damage repair plan The specific steps are as follows:

[0082] S521, initialization, defining the relevant structural parameters as vectors, searching for different building exterior wall damage type repair process data sets D in the different building exterior wall damage type repair process data sets D o In the κ-dimensional optimization problem, the process type search sand cat represents the 1×κ array of the problem solution, and each variable value q1,q2,…,q κ are all floating point numbers, and each variable value q1 to q κ It means that it is between the lower bound and the upper bound in the search space of the data set D of the repair process of different building exterior wall damage types;

[0083] S522, Searching for prey, the final and main parameters controlling the transition between exploration and exploitation phases are when When the process type search sand cat searches for different building exterior wall damage type repair process data sets D in the search space to find different building exterior wall damage type repair process data d that match the building exterior wall damage type identification data C'. o The process of searching for process types relies on the release of low-frequency noise to search for different building exterior wall damage type repair process data d that match the building exterior wall damage type identification data C′. o , assuming that the sensitivity range of the process type search sand cat ω is from 0 to 2kHz, ξ represents the auditory characteristics inspired by the process type search sand cat, assuming its value is 2, t is the current iteration number, T is the maximum iteration number, rand(0,1) represents a random number with a value of 0 to 1; in Represents the sensitivity vector; each process type searches for the sand cat according to the best candidate position J best Update its own position with the current position J and its sensitivity range ω, that is, update the different building exterior wall damage type repair process data d matching the building exterior wall damage type identification data C′ in the search space of the different building exterior wall damage type repair process data set D o The position calculation formula is as follows: J(t+1)=ω×(J best (t)-rand(0,1)×J best (t)), where J(t+1) represents the current position of the sand cat individual in the search space of the repair process data set D for different building exterior wall damage types in the t+1th iteration process type search, and J best (t) indicates the t-iteration process type search. Sandcat searches for the different building exterior wall damage type repair process data d that best matches the building exterior wall damage type identification data C' in the search space of different building exterior wall damage type repair process data set D. o The best candidate position for

[0084] S523, attack prey, when When the process type search sand cat is used, it searches in the search space of the repair process data set D of different building exterior wall damage types, and uses the best candidate position J best Generate a random position with the current position J, that is, randomly search for different building exterior wall damage type repair process data d matching the building exterior wall damage type identification data C' in the search space of different building exterior wall damage type repair process data set D o Assuming that the sensitivity range of the process type search sand cat is a circle, the roulette method is used to randomly select an angle ψ for each process type search sand cat, and the random position calculation formula is: Perform repair process data d of different building exterior wall damage types that match the building exterior wall damage type identification data C' o A random position search of represents the sensitivity vector, cos(ψ) represents the cosine value of the angle ψ, and J rand represents the random position of the process type search sand cat in the search space of different building exterior wall damage type repair process data set D. The random position makes the process type search sand cat approach and attack the prey, that is, search for different building exterior wall damage type repair process data d that matches the building exterior wall damage type identification data C′ in the search space of different building exterior wall damage type repair process data set D. o ;

[0085] S524, when the search algorithm reaches the maximum number of iterations T, output the repair process data d of different building exterior wall damage types that best matches the building exterior wall damage type identification data C′ o ;

[0086] S525, the repair process data of different building exterior wall damage types outputted from step S524 are o The corresponding building exterior wall damage type repair process data is constructed as building exterior wall damage repair plan analysis data

[0087] 7. A building maintenance management method according to claim 6, characterized in that: based on the building exterior wall damage type identification data and the building exterior wall maintenance inspection space position coordinate data, the building exterior wall damage position location processing is performed, and the operation steps of generating the building exterior wall damage problem position coordinate data are as follows:

[0088] S61, using a depth-first search algorithm, according to the building exterior wall damage type identification data C' corresponding to the building exterior wall maintenance inspection spatial position number when it exists, searching for the building exterior wall maintenance inspection spatial position coordinate data set A in the search space to find the building exterior wall maintenance inspection spatial position coordinate data a corresponding to the building exterior wall damage problem. u And it is constructed as the coordinate data f of the location of the building exterior wall damage problem, and the coordinate data f of the location of the building exterior wall damage problem includes the horizontal coordinate, the vertical coordinate and the vertical coordinate of the location of the building exterior wall damage problem.

[0089] Through the cooperation of the building exterior wall real-time inspection image acquisition unit and the building exterior wall damage type identification unit, the drone equipped with a cloud lens is used to accurately collect the exterior wall status image of the real-time inspection position of the building exterior wall, and the intelligent search algorithm and the scientifically preset image information of different building exterior wall damage types are combined to intelligently identify the damage problem type of the building exterior wall, so as to achieve accurate analysis of the building exterior wall damage problem; the building exterior wall damage repair process analysis unit and the building exterior wall damage position information search unit cooperate with each other, based on the artificial intelligence algorithm combined with the building exterior wall damage type identification information and the standard stored different building exterior wall damage type repair process information to intelligently evaluate the building exterior wall damage repair plan, and at the same time, the building exterior wall damage position coordinates are autonomously searched to realize the intelligent data analysis of the building exterior wall damage repair plan and the damage position, thereby improving the intelligence and reliability of building maintenance operations.

[0090] For further information, see Figure 1 - Figure 2 , construct the building exterior wall maintenance abnormality problem data and perform the building exterior wall maintenance abnormality problem feedback display operation. When the building exterior wall maintenance abnormality problem feedback display operation is completed, continue to perform steps S3 and S4 until the building exterior wall maintenance operation is completed. The operation steps are as follows:

[0091] S71, building exterior wall damage type identification data C′, building exterior wall damage repair plan analysis data The data is combined with the building exterior wall damage problem location coordinate data f to construct the building exterior wall maintenance abnormality problem data M, where

[0092] S72. Push the abnormal problem data M of building exterior wall maintenance to the building maintenance management platform through the Internet of Things communication network and display it on a display screen. When the feedback and display operation of the abnormal problem of building exterior wall maintenance is completed, continue to execute S31, S41, and S43 in the order of the spatial position number of the building exterior wall maintenance inspection until the building exterior wall maintenance operation is completed.

[0093] Through the building exterior wall maintenance abnormality problem information collection unit, the damage problems, repair plans and specific damage location information of the building exterior wall are collected online and dynamically, and the building exterior wall maintenance abnormality problem information is scientifically constructed to achieve efficient and accurate collection of building maintenance operation problems; the building exterior wall maintenance abnormality problem feedback and display unit combines the building exterior wall maintenance abnormality problem information with the Internet of Things communication network to achieve instant transmission to the building maintenance management platform, and instantly visualizes the output through the display screen, to achieve efficient and intuitive feedback of building maintenance operation result information.

[0094] Embodiment 2:

[0095] See also Figure 1 - Figure 2 , a building use process maintenance management system based on artificial intelligence, used to implement a building use process maintenance management method, the system includes a building maintenance control module, a building maintenance problem identification and processing module, and a building maintenance problem output module;

[0096] The building maintenance control module includes a building exterior wall three-dimensional model acquisition unit, a building exterior wall maintenance inspection position construction unit, and a building exterior wall maintenance operation execution unit;

[0097] A building exterior wall three-dimensional model acquisition unit, which collects building exterior wall three-dimensional model data through a drone equipped with a three-dimensional laser scanner; a building exterior wall maintenance inspection position construction unit, which constructs and processes the spatial position coordinates of the building exterior wall maintenance inspection based on the building exterior wall three-dimensional model data, and generates the building exterior wall maintenance inspection spatial position coordinate data; a building exterior wall maintenance operation execution unit, which executes the building exterior wall maintenance operation based on the building exterior wall maintenance inspection spatial position coordinate data;

[0098] The building maintenance problem identification and processing module includes a building exterior wall real-time inspection image acquisition unit, a building exterior wall damage type image storage unit, a building exterior wall damage type identification unit, a building exterior wall damage type repair process storage unit, a building exterior wall damage repair process analysis unit, and a building exterior wall damage location information search unit;

[0099] A real-time inspection image acquisition unit for building exterior walls, which collects real-time inspection image data of building exterior walls through a cloud camera mounted on a drone; an image storage unit for different types of damage to building exterior walls, which is used to store image data of different types of damage to building exterior walls; a building exterior wall damage type identification unit, which identifies and processes the types of damage problems of building exterior walls based on the real-time inspection image data of building exterior walls and image data of different types of damage to building exterior walls, and generates identification data of building exterior wall damage types; a storage unit for repair procedures for different types of damage to building exterior walls, which is used to store repair procedure data of different types of damage to building exterior walls; a building exterior wall damage repair procedure analysis unit, which analyzes and processes damage repair plans for building exterior walls based on the building exterior wall damage type identification data combined with the repair procedure data of different types of damage to building exterior walls based on an artificial intelligence algorithm, and generates analysis data of building exterior wall damage repair plans; a building exterior wall damage position information search unit, which locates and processes the building exterior wall damage position based on the building exterior wall damage type identification data and the building exterior wall maintenance inspection space position coordinate data, and generates building exterior wall damage problem position coordinate data;

[0100] The building maintenance problem output module includes a building exterior wall maintenance abnormal problem information collection unit and a building exterior wall maintenance abnormal problem feedback display unit;

[0101] The building exterior wall maintenance abnormality problem information collection unit is used to construct the building exterior wall maintenance abnormality problem data; the building exterior wall maintenance abnormality problem feedback display unit performs the building exterior wall maintenance abnormality problem feedback display operation based on the building exterior wall maintenance abnormality problem data.

[0102] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for maintenance management during the use of a building, characterized in that: The method comprises the following steps: S1. Collect the 3D model data of the building exterior wall; S2, constructing and processing the spatial position coordinates of the building exterior wall maintenance inspection based on the building exterior wall three-dimensional model data to generate the building exterior wall maintenance inspection spatial position coordinate data; S3, performing building exterior wall maintenance work based on the building exterior wall maintenance inspection space position coordinate data; S4, collecting real-time inspection image data of building exterior walls and performing identification processing on the types of damage problems of building exterior walls with image data of different types of damage of building exterior walls, and generating identification data of the types of damage problems of building exterior walls; if the data does not exist, continue to perform steps S3 and S4 until the maintenance work of the building exterior walls is completed; S5. When it exists, analyzing and processing the damage repair scheme of the building exterior wall according to the building exterior wall damage type identification data and the repair process data of different building exterior wall damage types to generate building exterior wall damage repair scheme analysis data; S6, performing location processing of the building exterior wall damage position based on the building exterior wall damage type identification data and the building exterior wall maintenance inspection space position coordinate data to generate building exterior wall damage problem position coordinate data; S7. Construct the data of abnormal problems in building exterior wall maintenance and execute the feedback and display operation of abnormal problems in building exterior wall maintenance; when the feedback and display operation of abnormal problems in building exterior wall maintenance is completed, continue to execute steps S3 and S4 until the building exterior wall maintenance operation is completed.

2. A building maintenance management method according to claim 1, characterized in that: The S1 comprises the following steps: S11. Use a 3D laser scanner mounted on a drone to collect the 3D solid model of the entire exterior wall of the target building online and generate 3D model data K of the exterior wall of the building.

3. A building maintenance management method according to claim 2, characterized in that: The S2 comprises the following steps: S21, taking the plane where the most prominent external wall structure of the target building's external wall surface in all directions corresponding to K as the building's external wall reference plane, dividing the building's external wall reference plane into grids according to squares with a side length of R, establishing a spatial rectangular coordinate system with the plane where the target building intersects the ground as the base of the spatial coordinate system, measuring the spatial coordinates of all the square grid center points in the spatial rectangular coordinate system, and generating a set of spatial position coordinate data for building external wall maintenance inspection A=(a1,…,a u ,…,a η ), u=1,2,3,…,eta; where a u represents u building exterior wall maintenance inspection space position coordinate data, and η represents the maximum number of building exterior wall maintenance inspection space positions.

4. A building maintenance management method according to claim 3, characterized in that: The S3 comprises the following steps: S31, using a drone equipped with a cloud camera according to the a in A u Carry out maintenance and inspection work on the target building's exterior walls in an orderly manner according to the building's exterior wall maintenance and inspection space location number.

5. A building maintenance management method according to claim 4, characterized in that: The S4 comprises the following steps: S41, during the maintenance and inspection operation on the exterior wall of the target building, the cloud camera carried by the drone is used to simultaneously monitor the a u The corresponding target building exterior wall specific maintenance inspection position performs building exterior wall real-time status image acquisition and generates building exterior wall real-time inspection image data set B u =(b u,1 ,…,b u,v ,…,b u,ι ), v = 1, 2, 3, ..., ι; where B u represents the real-time inspection image data set of the building exterior wall corresponding to the u-th building exterior wall maintenance inspection space position; b u,v represents the vth building exterior wall real-time inspection image data collected at the uth building exterior wall maintenance inspection space position, ι represents the maximum number of collected building exterior wall real-time inspection images; S42, establish different building exterior wall damage type image data set C = (c1, ..., c o ,…,c κ ), o=1,2,3,…,κ; where c o represents o different types of building exterior wall damage image data, κ represents the maximum number of building exterior wall damage types; S43, using the KD tree nearest neighbor search algorithm to u b u,v The same as the C in c o Perform image feature matching, and generate building exterior wall damage type identification data C′ based on the image feature matching results; When b u,v With c o If the image feature matching is successful, the output building exterior wall damage type identification data C' is the problem of the oth building exterior wall damage type existing in the uth building exterior wall maintenance inspection space position; When b u,v With c o If the image feature matching is unsuccessful, the output building exterior wall damage type identification data C' is non-existent. At this time, steps S31, S41, and S43 are continued in the order of the building exterior wall maintenance inspection space position number until the building exterior wall maintenance operation is completed.

6. A building maintenance management method according to claim 5, characterized in that: The S5 comprises the following steps: S51. Establish a data set D = (d1, ..., d o ,…,d κ ), where d o Represents the repair process data of different building exterior wall damage types corresponding to o types of building exterior wall damage; S52, when the C' is present, the C' and the D o According to the keyword matching of the building exterior wall damage type, search for the d corresponding to the C' o And construct analysis data for building exterior wall damage repair plan Execute and generate analysis data of building exterior wall damage repair plan The specific steps are as follows: S521, initialization, defining the relevant structural parameters as vectors, the d o In the κ-dimensional optimization problem, the process type search sand cat represents the 1×κ array of the problem solution, and each variable value q1,q2,…,q κ are all floating point numbers, and each variable value q1 to q κ represents a value between the lower bound and the upper bound in the D search space; S522, Searching for prey, the final and main parameters controlling the transition between exploration and exploitation phases are when When the process type search cat searches for the d matching the C' in the D search space, o , the process type search sand cat search process relies on the release of low frequency noise to search for the d that matches the C' o , that is, update the d matching the C' in the D search space o location; S523, attack prey, when When the process type search sand cat is searched in the D search space, the best candidate position J is used best Generate a random position with the current position J, that is, randomly search for the d that matches the C' in the D search space o Assuming that the sensitivity range of the process type search sand cat is a circle, the roulette method is used to randomly select an angle ψ for each process type search sand cat, and the d matching the C' is calculated according to the random position calculation formula. o Random position search of ; S524, when the search algorithm reaches the maximum number of iterations T, output the d that best matches the C' o ; S525, the output of step S524 is o The corresponding building exterior wall damage type repair process data is constructed as building exterior wall damage repair plan analysis data 7. A building maintenance management method according to claim 6, characterized in that: The S6 comprises the following steps: S61, using a depth-first search algorithm, searching in the A search space for the a corresponding to the building exterior wall damage problem according to the building exterior wall maintenance inspection space position number corresponding to the C′ when it exists. u And it is constructed as the coordinate data f of the location of the building exterior wall damage problem.

8. A building maintenance management method according to claim 7, characterized in that: The S7 comprises the following steps: S71, the C′, the The data is combined with f to construct the building exterior wall maintenance abnormality problem data M, where S72. Push and feed back the abnormal problem data M of the building exterior wall maintenance to the building maintenance management platform through the Internet of Things communication network and display it on a display screen. When the feedback and display operation of the abnormal problem of the building exterior wall maintenance is completed, continue to execute S31, S41, and S43 in the order of the spatial position number of the building exterior wall maintenance inspection until the building exterior wall maintenance operation is completed.

9. An artificial intelligence-based building maintenance management system for implementing a building maintenance management method according to any one of claims 1 to 8, characterized in that: The system includes a building maintenance control module, a building maintenance problem identification and processing module, and a building maintenance problem output module.

Images