Intelligent monitoring method and system for public safety of intelligent building based on digital twinning

Abstract

The invention relates to the technical field of intelligent building management, and discloses an intelligent public safety supervision method and system for an intelligent building based on digital twinning, and the system comprises an intelligent building safety risk monitoring management module, an intelligent building safety risk digital model building module, and an intelligent building safety risk output module. Based on building supervision object entity model information, building supervision object entity model real-time state sampling coordinate information and building supervision object real-time physical state parameters, building supervision object real-time three-dimensional data scene intelligent building with fire protection risks is carried out. Refined visual monitoring of the spatial feature change trend, the time feature change trend and the physical state feature change trend of a building supervision object is realized, and reliable supervision of intelligent building public safety based on digital twinning is realized.

Description

Technical Field

[0001] The present invention relates to the technical field of smart building management, and specifically to a method and system for intelligent public safety supervision of smart buildings based on digital twins. Background Art

[0002] Intelligent building management is the comprehensive management of building structures, building systems, building services, building management, and their combination, making buildings safe, convenient, efficient, and energy-efficient. Intelligent building management is a cutting-edge, interdisciplinary discipline involving automation, building technology, and more and more new technologies are being applied in smart buildings. With the increase in building personnel turnover and installed equipment, public fire safety management in smart buildings has become a crucial issue. Existing intelligent building safety management systems cannot intelligently identify and locate firefighting targets that pose safety risks, nor can they implement scenario-based digital supervision of firefighting supervision targets that pose safety risks, reducing the public safety of smart buildings.

[0003] A Chinese invention patent with announcement number CN116703252B discloses a SaaS-based smart building information management method. By collecting smoke alarm information from smart buildings, the obtained power supply information and environmental status information are generated into a status assessment index. Smoke alarms that generate abnormal alarm operation signals are analyzed, and the status of the smoke alarm is determined based on the analysis results. The management area where the smoke alarm is located is analyzed, and the regional data information and equipment association information are generated into a level assessment coefficient to determine the different maintenance priorities of pre-repair smoke alarms. However, the above technical solution cannot perform dynamic digital supervision of supervision objects that pose safety risks. Summary of the Invention

[0004] (1) Technical problems solved

[0005] In order to solve the above-mentioned problems that the existing intelligent building safety management cannot intelligently identify and locate fire objects with safety risks, nor can it realize scenario-based digital supervision of fire supervision objects with safety risks, the problem of public safety in smart buildings is reduced, and the above-mentioned purposes of intelligently analyzing the fire risks of building fire supervision objects, accurately locating building fire supervision objects, dynamically constructing the digital status model of building fire supervision objects, and visually and dynamically feedback the digital status of building fire supervision objects are achieved.

[0006] (2) Technical solution

[0007] The present invention is implemented through the following technical solution: a method for intelligent public safety supervision of smart buildings based on digital twins, the method comprising the following steps:

[0008] S1. Collect real-time status image data of building supervision objects;

[0009] S2. Performing real-time fire risk analysis on the target building supervision object based on the real-time state image data of the building supervision object and the fire risk image data of different types of buildings to generate fire risk analysis data for the building supervision object; and ending the safety supervision operation on the target building supervision object when no risk exists;

[0010] S3. When there is a risk, collect the spatial location coordinate data of the building supervision object;

[0011] S4. Collecting and processing a spatial three-dimensional entity model of the target building supervision object based on the spatial position coordinate data of the building supervision object to generate entity model data of the building supervision object;

[0012] S5. Performing real-time physical state sampling position construction processing of the spatial three-dimensional physical model of the target building supervision object based on the building supervision object physical model data, generating real-time state sampling coordinate data of the building supervision object physical model, and performing real-time physical state parameter collection processing of the target building supervision object to generate real-time physical state data of the building supervision object;

[0013] S6. Based on the building supervision object physical model data, the real-time state sampling coordinate data of the building supervision object physical model, and the real-time physical state data of the building supervision object, a real-time data scene construction process of the fire risk of the target building supervision object is performed to generate real-time fire risk supervision digital scene data of the building supervision object;

[0014] S7. Construct smart building safety supervision result data and perform smart building safety supervision feedback operations.

[0015] Preferably, the steps for collecting real-time status image data of building supervision objects are as follows:

[0016] S11. Use a mobile robot equipped with a cloud camera to capture real-time status image information of the real scene of the building fire monitoring object online, and generate real-time status image data P of the building supervision object. The building fire monitoring objects include fire passages, building office rooms, power supply and distribution facilities, safety alarm equipment, fire emergency lighting equipment, fire hydrant equipment, water fire extinguishing equipment, foam fire extinguishing equipment, gas fire extinguishing equipment and access elevators.

[0017] Preferably, a real-time fire risk analysis process of the target building supervision object is performed based on the real-time state image data of the building supervision object and the image data of fire risk of different types of buildings to generate fire risk analysis data of the building supervision object; when no risk exists, the operation steps for ending the safety supervision operation of the target building supervision object are as follows:

[0018] S21. Establish the image data matrix O of different types of fire risk in buildings = (o1,…,o x ,…,o θ ), x=1,2,3,…,θ; where o x represents the image data of different building fire risk types corresponding to the θth building fire risk type, where θ represents the maximum number of building fire risk types; building fire risk types include damaged power lines, improper use of open flames, illegal construction work, debris piled in corridors, occupation of fire escapes, stacking of flammable and explosive items, and illegal charging; the image data of different building fire risk types represents standard image information set for different building fire risk types;

[0019] S22, compare the real-time state image data P of the building supervision object with the different types of fire risk image data o in the building different types of fire risk image data matrix O. x Perform building fire risk type character matching and generate building supervision object fire risk analysis data P based on the building fire risk type character matching results fengxian , execute to generate the fire risk analysis data P of the building supervision object fengxian The specific steps are as follows:

[0020] S221, initialization, update the maximum number of iterations T, update and generate the individual positions of the white shark population for fire risk analysis, and the calculation formula for the individual positions of the white shark population for fire risk analysis is as follows: Among them E i,θ represents the position of the fire risk analysis white shark individual i in the search space of the building fire risk image data matrix O with spatial dimension θ, l and They represent the upper and lower limits of the search space of the image data matrix O of different types of fire risk in the building, and Ω represents a random number between (0, 1);

[0021] S222, speed update stage, the fire risk analysis white shark individual searches for the building different types of fire risk image data O that matches the real-time state image data P of the building supervision object in the search space of the building different types of fire risk image data matrix O x Prey moves to perceive different types of fire risk image data of the building x Position, and update its own speed. The calculation formula for updating its own speed is as follows: where N i,t+1 N represents the speed of the fire risk analysis white shark individual i in the search space of the building fire risk image data matrix O after t+1 iterations, i,trepresents the speed of the fire risk analysis white shark individual i in the search space of the building fire risk image data matrix O after t iterations, E best,t It represents the optimal position of the fire risk analysis white shark individual in the search space of the image data matrix O of different types of fire risk in the building after t+1 iterations, E i,t represents the position of the fire risk analysis white shark individual i in the search space of the fire risk image data matrix O of different types of fire risk of the building after t iterations; It represents the fire risk analysis of individual white shark i in the search space of different types of fire risk image data matrix O of the building after t iterations and the speed N i,t corresponding positions; represents the algorithm shrinkage coefficient, y1 and y2 represent E best,t and The control coefficients of , Λ1 and Λ2 both represent random numbers between (0,1);

[0022] S223, position update stage, the fire risk analysis white shark individual moves in the search space of the building different types of fire risk image data matrix O towards the building different types of fire risk image data o that best matches or sub-optimally matches the real-time state image data P of the building supervision object x The prey moves to update its own position in the search space of the image data matrix O of different types of fire risk in the building. The calculation formula for the individual position update of the fire risk analysis white shark is as follows: where E′ i,t+1 The fire risk analysis of the white shark individual i in the position update phase of the search for prey after t+1 iterations is the position of the search space of the image data matrix O of different types of fire risks of the building, Represents a bitwise operator, is a logical vector, ∫ and ε are both binary vectors, Π represents the attraction coefficient of the white shark individual approaching the prey, ◇ represents the wave frequency of the white shark individual movement product;

[0023] The fire risk analysis white shark individual moves toward the optimal fire risk analysis white shark individual position in the search space of the building different types of fire risk image data matrix O to approach the building different types of fire risk image data o that best matches the real-time status image data P of the building supervision object x The location of prey, fire risk analysis, and the calculation formula for the updated location of white shark individuals are as follows: E″ i,t+1 =E best,t +Γ1×Z×sgn(Γ2-0.5),Γ3<Γ, where E″ i,t+1The position of the fire risk analysis white shark individual i in the search space of the building different types of fire risk image data matrix O after t+1 iterations in the position update phase is represented. Γ1, Γ2, and Γ3 all represent random numbers between (0, 1). Z represents the distance between the fire risk analysis white shark individual and the prey. In the search space of the building different types of fire risk image data matrix O, the building different types of fire risk image data o that matches the real-time status image data P of the building supervision object are searched. x , sgn represents the sign return function, Γ represents the olfactory and visual parameters of individual white sharks approaching the optimal prey for fire risk analysis;

[0024] S224, in the school behavior stage, the fire risk analysis white shark population retains the optimal fire risk analysis white shark individual position in the search space of the building different types of fire risk image data matrix O in the position update stage through feeding behavior, and updates other fire risk analysis white shark individual positions based on the optimal fire risk analysis white shark individual position to obtain the building different types of fire risk image data O that matches the real-time status image data P of the building supervision object in the search space of the building different types of fire risk image data matrix O. x ,The calculation formula for the update of the individual position of white shark in fire risk analysis is as follows: E″′ it+1 =(E' i,t+1 +E″ i,t+1 )*2Ω, where E″′ i,t+1 It represents the position of the white shark individual i in the search space of the building fire risk image data matrix O after t+1 iterations in the fish school behavior stage. The building fire risk image data matrix O is searched for the building fire risk image data o that matches the real-time state image data P of the building supervision object in the search space. x ;

[0025] S225: When the maximum number of iterations is met, output the real-time state image data P of the building supervision object and the image data o of different types of fire risk of the building. x Perform building fire risk type character matching results and generate building supervision object fire risk analysis data P fengxian ;

[0026] When P and o x If the building fire risk type character matching is successful, it means that the target building monitoring object has fire danger, then the fire risk analysis data P of the building supervision object is output. fengxian There are risks;

[0027] When P and ox If the building fire risk type character is not matched successfully, it means that the target building monitoring object does not have fire danger, then the fire risk analysis data P of the building supervision object is output. fengxian In order to eliminate the risk, the security supervision operation of the target building monitoring object is terminated directly at this time.

[0028] Preferably, when there is a risk, the steps for collecting spatial position coordinate data of a building supervision object are as follows:

[0029] S31, when the fire risk analysis data P of the building supervision object fengxian In order to prevent risks, a mobile robot equipped with a position sensor is used to collect the spatial position coordinate information of the building fire monitoring object online, and generate the building supervision object spatial position coordinate data J, which includes the longitude, latitude and altitude of the building fire monitoring object.

[0030] Preferably, the steps of collecting and processing the spatial three-dimensional entity model of the target building supervision object based on the spatial position coordinate data of the building supervision object and generating the entity model data of the building supervision object are as follows:

[0031] S41. Use a three-dimensional laser scanner mounted on an unmanned aerial vehicle to fly to the spatial position of the target building supervision object according to the spatial position coordinate data J of the building supervision object, and scan and model the spatial three-dimensional entity model of the target building supervision object to generate building supervision object entity model data M.

[0032] Preferably, based on the building supervision object entity model data, a real-time physical state sampling position construction process of the spatial three-dimensional entity model of the target building supervision object is performed, the real-time state sampling coordinate data of the building supervision object entity model is generated, and the real-time physical state parameter collection process of the target building supervision object is performed. The operation steps for generating the real-time physical state data of the building supervision object are as follows:

[0033] S51. Use the MarchingCubes meshing algorithm to perform triangular meshing of equal area ψ on the surface of the building supervision object entity model corresponding to the building supervision object entity model data M, establish a spherical coordinate system with the earth as the base surface, measure the spatial coordinate parameters of the center point of the triangular mesh on the surface of the building supervision object entity model, and generate a real-time state sampling coordinate data set of the building supervision object entity model. where d k represents the real-time state sampling coordinate data of the building supervision object entity model corresponding to the k-th building supervision object real-time state sampling point, Indicates the maximum number of real-time status sampling points of a building supervision object; the real-time status sampling coordinate data of the building supervision object entity model includes the longitude, latitude and altitude of the real-time status sampling points of the building supervision object;

[0034] S52, using a state monitoring sensor carried by a drone based on the real-time state sampling coordinate data set D of the building supervision object entity model real-time state sampling coordinate data d1 to d2. Orderly execute the physical state parameters of the real-time state sampling points of the building supervision objects and generate the real-time physical state data set of the building supervision objects where d' k The real-time physical status data of the building supervision object corresponding to the k-th real-time status sampling point of the building supervision object is represented. The real-time physical status data of the building supervision object includes the temperature, humidity and smoke concentration of the real-time status sampling point of the building supervision object; the status monitoring sensor includes a temperature monitoring sensor, a humidity monitoring sensor and a smoke monitoring sensor.

[0035] Preferably, based on the building supervision object entity model data, the real-time state sampling coordinate data of the building supervision object entity model, and the real-time physical state data of the building supervision object, the real-time data scene construction processing of the fire risk of the target building supervision object is performed to generate the real-time supervision digital scene data of the fire risk of the building supervision object as follows:

[0036] S61, the real-time physical status data d' of the building supervision object in the real-time physical status data set D' of the building supervision object k According to the real-time state sampling coordinate data set D of the building supervision object entity model, the real-time state sampling coordinate data d of the building supervision object entity model is obtained. k The corresponding real-time status sampling point coordinates of the building supervision object are mapped in order to the building supervision object entity model surface corresponding to the building supervision object entity model data M, and the building supervision object fire risk real-time supervision digital scene data G is generated. The building supervision object fire risk real-time supervision digital scene data represents the building supervision object fire real-time digital twin scene information composed of the building supervision object spatial three-dimensional entity model, the building supervision object real-time status sampling point coordinates and the real-time physical state parameters of the building supervision object real-time status sampling points.

[0037] Preferably, the steps of constructing the smart building safety supervision result data and performing the smart building safety supervision feedback operation are as follows:

[0038] S71, the fire risk analysis data P of the building supervision object fengxian, the spatial position coordinate data J of the building supervision object, the real-time fire risk supervision digital scene data G of the building supervision object are combined to construct the smart building safety supervision result data R, where R=(P fengxian ,J,G);

[0039] S72. Transmit the smart building safety supervision result data R online to the smart building supervision platform through the Internet of Things communication network, and output it through the display screen to perform the smart building safety supervision feedback operation.

[0040] A digital twin-based smart building public safety intelligent supervision system is used to implement the digital twin-based smart building public safety intelligent supervision method. The system includes a smart building safety risk monitoring and management module, a smart building safety risk digital model establishment module, and a smart building safety risk output module;

[0041] The smart building safety risk monitoring and management module includes a real-time status acquisition unit for building supervision objects, a storage unit for images of different types of fire risk in buildings, a fire risk analysis unit for building supervision objects, and a building supervision object positioning unit;

[0042] The real-time state acquisition unit for building supervision objects collects real-time state image data of building supervision objects through a mobile robot equipped with a cloud camera; the different types of building fire risk image storage unit is used to store different types of building fire risk image data; the building supervision object fire risk analysis unit performs real-time fire risk analysis of the target building supervision object based on the real-time state image data of the building supervision object and the different types of building fire risk image data to generate building supervision object fire risk analysis data; the building supervision object positioning unit collects spatial position coordinate data of the building supervision object through a mobile robot equipped with a position sensor;

[0043] The smart building safety risk digital model establishment module includes a building supervision object physical model acquisition unit, a building supervision object physical model real-time state sampling position establishment unit, a building supervision object real-time state parameter acquisition unit, and a smart building fire risk digital scene establishment unit;

[0044] The building supervision object entity model acquisition unit performs spatial three-dimensional entity model acquisition processing of the target building supervision object based on the spatial position coordinate data of the building supervision object in combination with the three-dimensional laser scanner carried by the drone, and generates building supervision object entity model data; the building supervision object entity model real-time state sampling position establishment unit performs real-time physical state sampling position construction processing of the spatial three-dimensional entity model of the target building supervision object based on the building supervision object entity model data, and generates real-time state sampling coordinate data of the building supervision object entity model; the building supervision object real-time state parameter acquisition unit performs real-time physical state parameter acquisition processing of the target building supervision object based on the real-time state sampling coordinate data of the building supervision object entity model in combination with the state monitoring sensor carried by the drone, and generates real-time physical state data of the building supervision object; the smart building fire risk digital scene establishment unit performs real-time data scene construction processing of the fire risk of the target building supervision object based on the building supervision object entity model data, the real-time state sampling coordinate data of the building supervision object entity model, and the real-time physical state data of the building supervision object, and generates real-time fire risk supervision digital scene data of the building supervision object;

[0045] The smart building safety risk output module includes a smart building safety supervision result construction unit and a smart building safety supervision result feedback unit;

[0046] The smart building safety supervision result construction unit constructs smart building safety supervision result data based on the fire risk analysis result information of the building supervision object and the real-time fire risk supervision digital scene information of the building supervision object; the smart building safety supervision result feedback unit transmits the smart building safety supervision result data online to the smart building supervision platform through the Internet of Things communication network, and outputs it through the display screen to execute the smart building safety supervision feedback operation.

[0047] (3) Beneficial effects

[0048] The present invention provides a method and system for intelligent public safety supervision of smart buildings based on digital twins. It has the following beneficial effects:

[0049] 1. Accurately collect real-time status image information of building supervision objects through mobile robots equipped with cloud cameras, providing data support for scientific identification of fire risks of building supervision objects; conduct real-time fire risk intelligent analysis of target building supervision objects based on real-time status image information of building supervision objects combined with artificial intelligence recognition algorithms and different types of fire risk image data of buildings based on big data storage, realizing intelligent and efficient fire supervision of building supervision objects; dynamically collect spatial location information of building supervision objects with fire risks through mobile robots equipped with position sensors, realizing accurate judgment and positioning of fire risks of building supervision objects, and improving the efficiency and quality of public safety supervision of smart buildings.

[0050] 2. By combining the spatial position coordinate information of the building supervision object with the three-dimensional laser scanner carried by the drone, efficient and accurate physical modeling of the three-dimensional entity model of the building supervision object space is achieved. The spatial entity grid division algorithm is used to comprehensively and accurately establish the real-time physical state parameter sampling points of the building supervision object, and the real-time physical state parameters of the building supervision object are dynamically collected in combination with the state monitoring sensor carried by the drone, so as to realize the scientific and efficient collection of the entity model and physical state parameters of the building supervision object; improve the reliability of the public safety supervision of smart buildings; based on the entity model information of the building supervision object, the real-time state sampling coordinate information of the entity model of the building supervision object, and the real-time physical state parameters of the building supervision object, the real-time three-dimensional data scene of the building supervision object with fire risks is intelligently constructed, so as to realize the refined and intuitive monitoring of the spatial feature change trend, temporal feature change trend and physical state feature change trend of the building supervision object, and realize reliable supervision of the public safety of smart buildings based on digital twins.

[0051] 3. By efficiently and accurately constructing smart building safety supervision result information based on the fire risk analysis result information of building supervision objects and the real-time fire risk supervision digital scene information of building supervision objects, the security of public safety supervision of intelligent buildings is improved; the smart building safety supervision result information is transmitted online to the smart building supervision platform through the Internet of Things communication network, and visually output with the help of the display screen, so as to realize the visual integrated display of the spatial feature change trend, temporal feature change trend and physical state feature change trend of the building supervision objects, realize remote intelligent feedback on the real status of building supervision objects with fire risks, and improve the effect of public safety supervision of intelligent buildings. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] Figure 1 A schematic diagram of the modules of the digital twin-based smart building public safety intelligent supervision system provided by the present invention;

[0053] Figure 2 Flowchart of the intelligent public safety supervision method for smart buildings based on digital twins provided by the present invention. DETAILED DESCRIPTION

[0054] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. 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 making creative efforts are within the scope of protection of the present invention.

[0055] The embodiments of the digital twin-based smart building public safety intelligent supervision method and system are as follows:

[0056] Example 1:

[0057] See also Figure 1-Figure 2 , a smart building public safety intelligent supervision method based on digital twins, the method includes the following steps:

[0058] S1. Collect real-time status image data of building supervision objects;

[0059] S2. Performing real-time fire risk analysis and processing on the target building supervision object based on the real-time status image data of the building supervision object and the image data of different types of fire risk of the building, generating fire risk analysis data for the building supervision object; when no risk exists, ending the safety supervision operation of the target building supervision object;

[0060] S3. When there is a risk, collect the spatial location coordinate data of the building supervision object;

[0061] S4. Collecting and processing a spatial three-dimensional entity model of the target building supervision object based on the spatial position coordinate data of the building supervision object to generate entity model data of the building supervision object;

[0062] S5. Based on the building supervision object entity model data, construct and process the real-time physical state sampling positions of the spatial three-dimensional entity model of the target building supervision object, generate real-time state sampling coordinate data of the building supervision object entity model, and perform real-time physical state parameter collection and processing of the target building supervision object to generate real-time physical state data of the building supervision object;

[0063] S6. Based on the building supervision object entity model data, the real-time state sampling coordinate data of the building supervision object entity model, and the real-time physical state data of the building supervision object, a real-time data scenario construction and processing of the fire risk of the target building supervision object is performed to generate real-time fire risk supervision digital scenario data of the building supervision object;

[0064] S7. Construct smart building safety supervision result data and perform smart building safety supervision feedback operations.

[0065] For further information, see Figure 1-Figure 2,The steps for collecting real-time status image data of building ,supervision objects are as follows:

[0066] S11. Use a mobile robot equipped with a cloud camera to capture real-time status image information of the real scene of the building fire monitoring object online, and generate real-time status image data P of the building supervision object. The building fire monitoring objects include fire passages, building office rooms, power supply and distribution facilities, safety alarm equipment, fire emergency lighting equipment, fire hydrant equipment, water fire extinguishing equipment, foam fire extinguishing equipment, gas fire extinguishing equipment and access elevators.

[0067] Based on the real-time status image data of the building supervision object and the image data of different types of fire risk of the building, real-time fire risk analysis and processing of the target building supervision object are performed to generate fire risk analysis data of the building supervision object; when there is no risk, the operation steps for ending the safety supervision operation of the target building supervision object are as follows:

[0068] S21. Establish the image data matrix O of different types of fire risk in buildings = (o1,…,o x ,…,o θ ), x=1,2,3,…,θ; where o x represents the image data of different types of building fire risks corresponding to the θth building fire risk type, where θ represents the maximum number of building fire risk types. Building fire risk types include damaged power lines, improper use of open flames, illegal construction work, debris piled up in corridors, occupation of fire escapes, stacking of flammable and explosive items, and illegal charging. The image data of different types of building fire risks represents the standard image information set for different building fire risk types.

[0069] S22, compare the real-time state image data P of the building supervision object with the different types of fire risk image data o in the building different types of fire risk image data matrix O. x Perform building fire risk type character matching and generate building supervision object fire risk analysis data P based on the building fire risk type character matching results fengxian , execute and generate fire risk analysis data P of building supervision objects fengxian The specific steps are as follows:

[0070] S221, initialization, update the maximum number of iterations T, update and generate the individual positions of the white shark population for fire risk analysis, and the calculation formula for the individual positions of the white shark population for fire risk analysis is as follows: Among them E i,θ represents the position of the fire risk analysis white shark individual i in the search space of the building fire risk image data matrix O with a spatial dimension of θ, l and They represent the upper and lower limits of the search space of the image data matrix O of different types of fire risk in buildings, and Ω represents a random number between (0,1);

[0071] S222, speed update stage, the fire risk analysis white shark individual searches for different types of building fire risk image data o that matches the real-time state image data P of the building supervision object in the search space of different types of building fire risk image data matrix O x Prey movement to perceive different types of fire risk image data of buildings x Position, and update its own speed. The calculation formula for updating its own speed is as follows: where N i,t+1 N represents the speed of fire risk analysis white shark individual i in the search space of building fire risk image data matrix O after t+1 iterations, i,t represents the speed of fire risk analysis white shark individual i in the search space of building fire risk image data matrix O after t iterations, E best,t It represents the optimal position of the fire risk analysis white shark individual in the search space of the building fire risk image data matrix O after t+1 iterations, E i,t represents the position of the fire risk analysis white shark individual i in the search space of the fire risk image data matrix O of different types of building fire risks after t iterations; It represents the fire risk analysis of white shark individual i in the search space of different types of building fire risk image data matrix O after t iterations and the speed N i,t corresponding positions; represents the algorithm shrinkage coefficient, y1 and y2 represent E best,t and The control coefficients of , Λ1 and Λ2 both represent random numbers between (0,1);

[0072] S223, position update stage, the fire risk analysis white shark individual moves in the search space of the building different types of fire risk image data matrix O towards the building different types of fire risk image data o that is optimally matched or suboptimally matched with the real-time state image data P of the building supervision object x The prey moves to update its own position in the search space of the image data matrix O of different types of building fire risks. The calculation formula for the individual position update of the fire risk analysis white shark is as follows: where E′ i,t+1 The fire risk analysis of the white shark individual i in the search space of the image data matrix O of different types of building fire risk after t+1 iterations in the position update phase is represented. Represents a bitwise operator, is a logical vector, ∫ and ε are both binary vectors, Π represents the attraction coefficient of the white shark individual approaching the prey, ◇ represents the wave frequency of the white shark individual movement product;

[0073] The fire risk analysis white shark individual moves towards the optimal fire risk analysis white shark individual position in the search space of the building different types of fire risk image data matrix O to approach the building different types of fire risk image data o that best matches the real-time status image data P of the building supervision object. x The location of prey, fire risk analysis, and the calculation formula for the updated location of white shark individuals are as follows: E″ i,t+1 =E best,t +Γ1×Z×sgn(Γ2-0.5),Γ3<Γ, where E″ i,t+1 represents the position of the fire risk analysis white shark individual i in the search space of the building different types of fire risk image data matrix O after t+1 iterations in the position update phase, Γ1, Γ2, and Γ3 all represent random numbers between (0, 1), and Z represents the distance between the fire risk analysis white shark individual and the prey, that is, the building different types of fire risk image data o that matches the real-time status image data P of the building supervision object is searched in the search space of the building different types of fire risk image data matrix O. x ; sgn represents the sign return function, Γ represents the olfactory and visual parameters of white shark individuals close to the optimal prey for fire risk analysis;

[0074] S224, in the school behavior stage, the fire risk analysis white shark population retains the optimal fire risk analysis white shark individual position in the search space of the building different types of fire risk image data matrix O in the position update stage through feeding behavior, and updates the positions of other fire risk analysis white shark individuals based on the optimal fire risk analysis white shark individual position to obtain the building different types of fire risk image data o that matches the real-time status image data P of the building supervision object in the search space of the building different types of fire risk image data matrix O. x ,The calculation formula for the update of the individual position of white shark in fire risk analysis is as follows: E″′ i,t+1 =(E' i,t+1 +E″ i,t+1 ) / 2Ω, where E″′ i,t+1 It represents the position of the white shark individual i in the search space of the building fire risk image data matrix O after t+1 iterations in the fish school behavior stage. That is, the building fire risk image data o that matches the real-time status image data P of the building supervision object is searched out in the search space of the building fire risk image data matrix O. x ;

[0075] S225: When the maximum number of iterations is met, output the real-time state image data P of the building supervision object and the image data o of different types of fire risk of the building x Perform building fire risk type character matching results and generate building supervision object fire risk analysis data P fengxian ;

[0076] When P and o x If the building fire risk type character matching is successful, it means that the target building monitoring object has fire danger, and the building supervision object fire risk analysis data P is output. fengxian There are risks;

[0077] When P and o x If the building fire risk type character is not matched successfully, it means that the target building monitoring object does not have fire danger, then the building supervision object fire risk analysis data P is output. fengxian In order to eliminate the risk, the security supervision operation of the target building monitoring object is terminated directly at this time.

[0078] When there is a risk, the steps for collecting spatial location coordinate data of building supervision objects are as follows:

[0079] S31. Fire risk analysis data of building supervision objects P fengxian In order to prevent risks, the spatial position coordinate information of the building fire monitoring object is collected online by using a mobile robot equipped with a position sensor, and the spatial position coordinate data J of the building supervision object is generated. The spatial position coordinate data of the building supervision object includes the longitude, latitude and altitude of the building fire monitoring object.

[0080] Through the real-time status collection unit of building supervision objects, a mobile robot equipped with a cloud camera is used to accurately collect real-time and true status image information of building supervision objects, providing data support for the scientific identification of fire risks of building supervision objects; the fire risk analysis unit of building supervision objects performs real-time fire risk intelligent analysis of target building supervision objects based on the real-time status image information of building supervision objects combined with artificial intelligence recognition algorithms and different types of building fire risk image data stored based on big data, realizing intelligent and efficient fire supervision of building supervision objects; the building supervision object positioning unit dynamically collects spatial position information of building supervision objects with fire risks through mobile robots equipped with position sensors, realizing accurate judgment and positioning of fire risks of building supervision objects, and improving the efficiency and quality of public safety supervision of intelligent buildings.

[0081] For further information, see Figure 1-Figure 2 , based on the spatial position coordinate data of the building supervision object, the spatial three-dimensional entity model of the target building supervision object is collected and processed, and the operation steps for generating the building supervision object entity model data are as follows:

[0082] S41. Use a drone equipped with a three-dimensional laser scanner to fly to the spatial location of the target building supervision object based on the spatial position coordinate data J of the building supervision object, and scan and model the spatial three-dimensional entity model of the target building supervision object to generate the building supervision object entity model data M.

[0083] The steps for constructing and processing the real-time physical state sampling positions of the spatial three-dimensional physical model of the target building supervision object based on the building supervision object physical model data, generating the real-time state sampling coordinate data of the building supervision object physical model, and performing the real-time physical state parameter collection and processing of the target building supervision object to generate the real-time physical state data of the building supervision object are as follows:

[0084] S51. Use the MarchingCubes meshing algorithm to perform triangular meshing of equal area ψ on the surface of the building supervision object entity model corresponding to the building supervision object entity model data M, establish a spherical coordinate system with the earth as the base surface, measure the spatial coordinate parameters of the center point of the triangular mesh on the surface of the building supervision object entity model, and generate a real-time state sampling coordinate data set of the building supervision object entity model. where d k represents the real-time state sampling coordinate data of the building supervision object entity model corresponding to the k-th building supervision object real-time state sampling point, Indicates the maximum number of real-time status sampling points of a building supervision object; the real-time status sampling coordinate data of the building supervision object entity model includes the longitude, latitude and altitude of the real-time status sampling points of the building supervision object;

[0085] S52, using the drone equipped with a status monitoring sensor to monitor the real-time status sampling coordinate data d1 to d2 of the building supervision object entity model in the real-time status sampling coordinate data set D of the building supervision object entity model. Orderly execute the physical state parameters of the real-time state sampling points of the building supervision objects and generate the real-time physical state data set of the building supervision objects where d' k It represents the real-time physical status data of the building supervision object corresponding to the k-th building supervision object real-time status sampling point. The real-time physical status data of the building supervision object includes the temperature, humidity and smoke concentration of the building supervision object real-time status sampling point; the status monitoring sensor includes a temperature monitoring sensor, a humidity monitoring sensor and a smoke monitoring sensor.

[0086] Based on the building supervision object entity model data, the building supervision object entity model real-time state sampling coordinate data, and the building supervision object real-time physical state data, the real-time data scene construction and processing of the fire risk of the target building supervision object is performed to generate the building supervision object fire risk real-time supervision digital scene data as follows:

[0087] S61, the real-time physical state data d' of the building supervision object in the real-time physical state data set D' k According to the real-time state sampling coordinate data set D of the building supervision object entity model, the real-time state sampling coordinate data d of the building supervision object entity model is obtained. k The corresponding real-time status sampling point coordinates of the building supervision object are mapped in order to the building supervision object entity model surface corresponding to the building supervision object entity model data M, and the building supervision object fire risk real-time supervision digital scene data G is generated. The building supervision object fire risk real-time supervision digital scene data represents the building supervision object fire real-time digital twin scene information composed of the building supervision object spatial three-dimensional entity model, the building supervision object real-time status sampling point coordinates and the real-time physical state parameters of the building supervision object real-time status sampling points.

[0088] Through the mutual cooperation between the building supervision object entity model acquisition unit, the building supervision object entity model real-time state sampling position establishment unit, and the building supervision object real-time state parameter acquisition unit, based on the spatial position coordinate information of the building supervision object combined with the drone-mounted three-dimensional laser scanner, efficient and accurate physical modeling of the building supervision object's spatial three-dimensional entity model is achieved. The spatial entity meshing algorithm is used to comprehensively and accurately establish the real-time physical state parameter sampling points of the building supervision object, and combined with the drone-mounted state monitoring sensor to dynamically collect the real-time physical state parameters of the building supervision object, thereby achieving scientific and efficient collection of the entity model and physical state parameters of the building supervision object; improving the reliability of smart building public safety supervision; the smart building fire risk digital scene establishment unit, based on the building supervision object entity model information, the building supervision object entity model real-time state sampling coordinate information, and the building supervision object real-time physical state parameters, intelligently builds real-time three-dimensional data scenes for building supervision objects with fire risks, realizing refined and intuitive monitoring of the spatial feature change trends, temporal feature change trends, and physical state feature change trends of the building supervision object, and realizing reliable supervision of smart building public safety based on digital twins.

[0089] For further information, see Figure 1-Figure 2 The steps to construct smart building safety supervision result data and perform smart building safety supervision feedback operations are as follows:

[0090] S71. Fire risk analysis data of building supervision objects P fengxian , the spatial position coordinate data J of the building supervision object, the real-time fire risk supervision digital scene data G of the building supervision object are combined to construct the smart building safety supervision result data R, where R = (P fengxian ,J,G);

[0091] S72. The smart building safety supervision result data R is transmitted online to the smart building supervision platform through the Internet of Things communication network, and is output through the display screen to perform the smart building safety supervision feedback operation.

[0092] Through the smart building safety supervision result construction unit, the smart building safety supervision result information is constructed efficiently and accurately based on the fire risk analysis result information of the building supervision object and the real-time fire risk supervision digital scene information of the building supervision object, thereby improving the security of the public safety supervision of smart buildings; the smart building safety supervision result feedback unit transmits the smart building safety supervision result information online to the smart building supervision platform through the Internet of Things communication network, and cooperates with the display screen for visual and intuitive output, thereby realizing the visual integrated display of the spatial feature change trend, time feature change trend and physical state feature change trend of the building supervision object, realizing remote intelligent feedback of the real status of the building supervision object with fire risk, and improving the effect of public safety supervision of smart buildings.

[0093] Example 2:

[0094] See also Figure 1-Figure 2 , a smart building public safety intelligent supervision system based on digital twins is used to implement a smart building public safety intelligent supervision method based on digital twins. The system includes a smart building safety risk monitoring and management module, a smart building safety risk digital model establishment module, and a smart building safety risk output module;

[0095] The smart building safety risk monitoring and management module includes a real-time status acquisition unit for building supervision objects, a storage unit for images of different types of fire risk in buildings, a fire risk analysis unit for building supervision objects, and a positioning unit for building supervision objects;

[0096] The real-time status acquisition unit for building supervision objects uses a mobile robot equipped with a cloud camera to collect real-time status image data of building supervision objects; the building different types of fire risk image storage unit is used to store different types of fire risk image data of buildings; the building supervision object fire risk analysis unit performs real-time fire risk analysis and processing of target building supervision objects based on the real-time status image data of building supervision objects and different types of fire risk image data of buildings, and generates building supervision object fire risk analysis data; the building supervision object positioning unit uses a mobile robot equipped with a position sensor to collect spatial position coordinate data of building supervision objects;

[0097] The smart building safety risk digital model establishment module includes a building supervision object physical model acquisition unit, a building supervision object physical model real-time state sampling position establishment unit, a building supervision object real-time state parameter acquisition unit, and a smart building fire risk digital scenario establishment unit;

[0098] The building supervision object entity model acquisition unit acquires and processes the spatial three-dimensional entity model of the target building supervision object based on the spatial position coordinate data of the building supervision object in combination with the three-dimensional laser scanner carried by the drone, and generates the building supervision object entity model data; the building supervision object entity model real-time state sampling position establishment unit constructs and processes the real-time physical state sampling position of the spatial three-dimensional entity model of the target building supervision object based on the building supervision object entity model data, and generates the building supervision object entity model real-time state sampling coordinate data; the building supervision object real-time state parameter acquisition unit acquires and processes the real-time physical state parameters of the target building supervision object based on the real-time state sampling coordinate data of the building supervision object entity model in combination with the state monitoring sensor carried by the drone, and generates the building supervision object real-time physical state data; the smart building fire risk digital scene establishment unit constructs and processes the fire risk real-time data scene of the target building supervision object based on the building supervision object entity model data, the building supervision object entity model real-time state sampling coordinate data, and the building supervision object real-time physical state data, and generates the building supervision object fire risk real-time supervision digital scene data;

[0099] The smart building safety risk output module includes a smart building safety supervision result construction unit and a smart building safety supervision result feedback unit;

[0100] The smart building safety supervision result construction unit constructs the smart building safety supervision result data based on the fire risk analysis result information of the building supervision object and the real-time fire risk supervision digital scene information of the building supervision object; the smart building safety supervision result feedback unit transmits the smart building safety supervision result data online to the smart building supervision platform through the Internet of Things communication network, and outputs it through the display screen to execute the smart building safety supervision feedback operation.

[0101] While 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 these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. The intelligent supervision method for public safety of smart buildings based on digital twins is characterized by: The method comprises the following steps: S1. Collect real-time status image data of building supervision objects; S2. Perform real-time fire risk analysis on the target building supervision object to generate fire risk analysis data for the building supervision object; when no risk exists, terminate the safety supervision operation on the target building supervision object; S3. When there is a risk, collect the spatial location coordinate data of the building supervision object; S4. Collect and process the spatial three-dimensional entity model of the target building supervision object to generate entity model data of the building supervision object; S5. Perform real-time physical state sampling position construction processing of the spatial three-dimensional entity model of the target building supervision object to generate real-time state sampling coordinate data of the entity model of the building supervision object and perform real-time physical state parameter collection processing of the target building supervision object to generate real-time physical state data of the building supervision object; S6. Build and process the real-time data scenario of the fire risk of the target building supervision object to generate real-time digital scenario data of the fire risk of the building supervision object; S7. Construct smart building safety supervision result data and perform smart building safety supervision feedback operations.

2. The method for intelligent public safety supervision of smart buildings based on digital twins according to claim 1 is characterized by: Said S1 comprises the following steps: S11. Use a mobile robot equipped with a cloud camera to shoot real-time state image information of a building fire monitoring object online, and generate real-time state image data P of the building supervision object.

3. The method for intelligent public safety supervision of smart buildings based on digital twins according to claim 2 is characterized by: The S2 comprises the following steps: S21. Establish the image data matrix O of different types of fire risk in buildings = (o1,…,o x ,…,o θ ), x=1,2,3,…,θ; where o x represents the image data of different types of building fire risk corresponding to the θth building fire risk type, and θ represents the maximum number of building fire risk types; S22, the P and the o in the O x Perform building fire risk type character matching and generate building supervision object fire risk analysis data P based on the building fire risk type character matching results fengxian , execute to generate the fire risk analysis data P of the building supervision object fengxian The specific steps are as follows: S221, initialization, updating the maximum number of iterations T, and updating the individual positions of the white shark population for fire risk analysis; S222, speed update stage, fire risk analysis white shark individual searches for the O that matches the P in the search space of the O x The prey moves to sense the o x Position, and update its own speed; S223, position update stage, fire risk analysis white shark individual in the search space of O by moving towards the o that is the best match or suboptimal match with P x The prey moves its position to update its position in the search space of O; the fire risk analysis white shark individual moves towards the optimal fire risk analysis white shark individual position in the search space of O to get close to the o that best matches P. x the location of the prey; S224, in the school behavior stage, the white shark population retains the optimal individual position of the white shark in the search space of O in the position update stage through feeding behavior, and updates the positions of other white sharks according to the optimal individual position of the white shark to obtain the o that matches P in the search space of O. x ; S225. When the maximum number of iterations is met, output the P and the o x Perform building fire risk type character matching results and generate building supervision object fire risk analysis data P fengxian ; When P and o x If the building fire risk type character matching is successful, the P fengxian There are risks; When P and o x If the building fire risk type character is not matched successfully, the P fengxian In order to eliminate the risk, the security supervision operation of the target building monitoring object is terminated directly at this time.

4. The method for intelligent public safety supervision of smart buildings based on digital twins according to claim 3 is characterized by: The S3 includes the following steps: S31, when the P fengxian In order to avoid risks, the spatial position coordinate information of building fire monitoring objects is collected online by using a mobile robot equipped with a position sensor, and the spatial position coordinate data J of the building supervision objects is generated.

5. The method for intelligent public safety supervision of smart buildings based on digital twins according to claim 4 is characterized by: The S4 comprises the following steps: S41. Use a drone equipped with a three-dimensional laser scanner to fly to the spatial location of the target building supervision object according to the J, and perform scanning and modeling processing on the spatial three-dimensional entity model of the target building supervision object to generate building supervision object entity model data M.

6. The method for intelligent public safety supervision of smart buildings based on digital twins according to claim 5 is characterized by: The S5 comprises the following steps: S51. Use the MarchingCubes meshing algorithm to perform triangular meshing of equal area ψ on the surface of the building supervision object entity model corresponding to M, establish a spherical coordinate system with the earth as the base surface, measure the spatial coordinate parameters of the center point of the triangular mesh on the surface of the building supervision object entity model, and generate a real-time state sampling coordinate data set of the building supervision object entity model. where d k represents the real-time state sampling coordinate data of the building supervision object entity model corresponding to the k-th building supervision object real-time state sampling point, Indicates the maximum number of real-time status sampling points of a building supervision object; S52, using a drone equipped with a status monitoring sensor based on the data from d1 to d2 in D. Orderly execute the physical state parameters of the real-time state sampling points of the building supervision objects and generate the real-time physical state data set of the building supervision objects where d' k Represents the real-time physical status data of the building supervision object corresponding to the k-th building supervision object real-time status sampling point.

7. The method for intelligent public safety supervision of smart buildings based on digital twins according to claim 6 is characterized by: The S6 comprises the following steps: S61, the d' in the D' k Follow the instructions in D k The corresponding real-time status sampling point coordinates of the building supervision object are mapped in order to the surface of the building supervision object entity model corresponding to the M, and the real-time supervision digital scene data G of the fire risk of the building supervision object is generated.

8. The method for intelligent public safety supervision of smart buildings based on digital twins according to claim 7 is characterized by: The S7 comprises the following steps: S71, the P fengxian , the J and the G are combined to construct the smart building safety supervision result data R; S72. Transmit the R online to the smart building supervision platform through the Internet of Things communication network, and output it through the display screen to perform the smart building safety supervision feedback operation.

9. A digital twin-based intelligent building public safety intelligent supervision system, used to implement the digital twin-based intelligent building public safety intelligent supervision method according to any one of claims 1 to 8, characterized in that: The system includes a smart building safety risk monitoring and management module, a smart building safety risk digital model establishment module, and a smart building safety risk output module.

Images