Building external wall thermal performance detection method based on data driving

Abstract

The invention discloses a building external wall thermal performance detection method based on data driving, and relates to the technical field of building detection, and the method comprises the following steps: S1, a data acquisition unit monitors infrared data and structural data of a building external wall; s2, the edge calculation unit analyzes thermal parameters of the building wall; s3, the central processing unit carries out regional anomaly intelligent diagnosis and thermal performance optimization inversion on the building wall; s4, displaying the abnormal area of the wall body by a data output unit, obtaining a maintenance suggestion, and generating an optimal construction scheme of the outer wall of the building by combining the optimal thermal parameters; through multi-source data fusion and three-dimensional infrared modeling, thermotechnical defects are accurately positioned, full-process automation is achieved, manual intervention is reduced, detection efficiency and precision are improved, through data driving and model fusion, full-life-cycle management of the thermal performance of the building outer wall is achieved, the detection industry is promoted to be transformed to intelligentization and datamation, and the detection efficiency is improved. And energy-saving reconstruction and sustainable development of the building are realized.

Description

Technical Field

[0001] The present invention relates to the technical field of building detection, and in particular to a data-driven building exterior wall thermal performance detection method. Background Art

[0002] A large number of old buildings around the world have degraded exterior wall insulation performance, such as insulation layer shedding, thermal bridge effect, etc., which require detection to locate defective areas and guide targeted renovation. In addition, construction defects such as uneven laying of insulation materials and poor sealing may lead to hidden heat losses, and thermal performance acceptance must be carried out at the completion stage. Thermal defects are often accompanied by problems such as hollowing and water seepage, which may cause structural safety hazards. Regular inspections are required to implement preventive maintenance to ensure building safety.

[0003] However, traditional detection methods rely on manual experience, are inefficient and prone to missing details, such as thermal bridges and cracks. Existing systems generally detect through local temperature data, which is difficult to reflect the overall thermal performance of the wall, lacks three-dimensional spatial correlation, lacks intelligent diagnosis capabilities, and has high latency due to large data processing volume. It is impossible to guide on-site maintenance in real time, resulting in unscientific construction plans, and it is difficult to combine real-time thermal parameters for dynamic adjustment, which can easily lead to excessive construction or incomplete repairs, resulting in high renovation costs. In view of the above technical defects, a solution is now proposed. Summary of the invention

[0004] The purpose of the present invention is to solve the problems that traditional detection methods rely on manual experience, are inefficient and easily miss details, such as thermal bridges and cracks, and that existing systems generally perform detection through local temperature data, which makes it difficult to reflect the overall thermal performance of the wall, lack three-dimensional spatial correlation, lack intelligent diagnostic capabilities, and cause high latency due to large data processing volume, making it impossible to guide on-site maintenance in real time, resulting in unscientific construction plans, and difficult to combine with real-time thermal parameters for dynamic adjustment, which easily leads to excessive construction or incomplete repairs, resulting in high renovation costs.

[0005] In order to achieve the above object, the present invention adopts the following technical solutions: The data-driven building exterior wall thermal performance detection method includes the following steps: S1, the data acquisition unit monitors the infrared data and structural data of the building exterior wall; S2, the edge computing unit analyzes the thermal parameters of the building wall: by constructing a three-dimensional infrared building model, the thermal parameters of the building wall are analyzed, and the distribution of the thermal performance of the building exterior wall is displayed; S3, the central processing unit performs intelligent diagnosis of regional abnormalities and thermal performance optimization inversion of building walls: regional planning and comparative marking are performed through the thermal performance distribution of the building's exterior walls, the abnormal wall areas are identified and output, and m types of construction plans for the building's exterior walls are set to make the predicted temperature approach the measured temperature, build a parameter optimization objective function for thermal performance, and output the optimal thermal parameters of the building's exterior wall area; S4, the data output unit displays the abnormal area of ​​the wall and obtains maintenance suggestions, and then combines the optimal thermal parameters to generate the optimal construction plan for the building exterior wall.

[0006] Further, the data acquisition unit includes an infrared thermal imager and a three-dimensional laser scanner; The infrared data of the building's exterior walls are obtained through an infrared thermal imager, and the structural data is obtained through a 3D laser scanner; By combining the infrared data of the building's exterior wall with the structural data, a three-dimensional infrared building model is constructed; Infrared data include the internal and external surface temperatures of building walls, and indoor and outdoor air temperatures; Structural data includes the 3D location, thickness, and thermal conductivity of building walls.

[0007] Furthermore, the thermal parameters of the building wall are analyzed through the heat transfer model. The specific process is as follows: The building walls are planned into N regional walls through the three-dimensional infrared building model, and any regional wall is marked as P; Set the regional wall P to include n layers of wall materials, mark the total wall thickness as L, mark any layer of the wall as i, mark the thickness of the wall i as di, and mark the thermal conductivity of the wall i as ki; The inner surface temperature of the area wall P is marked as , the outer surface temperature is marked as ; The indoor air temperature of the area wall P is marked as , the outdoor air temperature is marked as ; Obtain indoor heat flux density through heat flux meter , Outdoor heat flux ; The inner surface convection heat transfer coefficient is marked as , and the external surface convection heat transfer coefficient is marked as ; Convection heat transfer coefficient through the inner surface , External surface convection heat transfer coefficient , the thickness di of the wall i and the thermal conductivity ki are combined to obtain the total thermal resistance R of the regional wall P, and then the heat transfer coefficient U of the regional wall P is obtained; Thermal parameters include heat transfer coefficient and total thermal resistance.

[0008] Furthermore, the specific process of intelligent diagnosis of regional abnormalities of building walls is as follows: The thermal performance distribution of the building's exterior wall is displayed through a three-dimensional infrared building model, and then regional planning and comparison marking are carried out. The standard threshold of the total thermal resistance of the regional wall is set and marked as Qr, and the standard threshold of the heat transfer coefficient of the regional wall is set and marked as Qu; When the total thermal resistance R of the regional wall P is higher than the standard threshold value Qr, and the heat transfer coefficient U is lower than the standard threshold value Qu, the thermal performance of the regional wall P is judged to be normal; otherwise, the thermal performance of the regional wall P is judged to be abnormal, thereby identifying and outputting the abnormal wall area; It then conducts intelligent diagnosis on abnormal areas of the wall to determine insulation damage, thermal bridges, cracks and poor sealing.

[0009] Furthermore, the specific process of the thermal performance optimization inversion of the building wall is as follows: The steady-state heat transfer and transient heat transfer of the wall are analyzed by the surface heat transfer equation and the time-series heat conduction equation respectively. The specific process is as follows: Through the heat transfer coefficient U, indoor air temperature and outdoor air temperature Combined with the heat flux density of steady-state heat transfer of wall P, ; The heat flux Substitute into the surface heat transfer equation to obtain the predicted internal and external surface temperatures; mark is the predicted inner surface temperature, is the predicted external surface temperature; The inner surface boundary space point q is marked as 0, the outer surface boundary space point q is marked as L, any time node is marked as t, and the wall surface temperature corresponding to the space point q and time node t is marked as T (q, t); By the density of the wall material , specific heat capacity c, thermal conductivity k to obtain thermal diffusivity ; Establish the time-series heat conduction equation and set the boundary conditions of the internal and external surfaces; Then, by setting the parameter optimization objective function of the thermal performance, the thermal performance optimization inversion is performed, the predicted temperature is compared with the measured temperature, and the optimal thermal parameters of the building's exterior wall area are obtained.

[0010] Furthermore, the parameter optimization objective function of thermal performance is set, and the specific process of obtaining the optimal thermal parameters of the building exterior wall area is as follows: The thermal conductivity coefficient ki of wall i and the convection heat transfer coefficient of the inner surface , External surface convection heat transfer coefficient , density of wall materials and specific heat capacity c integrated as the target parameter vector ; By minimizing the mean square error between the predicted temperature and the actual temperature, the target parameter vector Perform inverse optimization, set m construction schemes for the building's exterior wall, mark any construction scheme for the building's exterior wall as j, and build a parameter optimization objective function for thermal performance; By setting m kinds of construction schemes for the building exterior wall and calculating the minimum value of the objective function, the target parameter vector obtained thereby is marked as the optimal thermal parameter, and the construction scheme obtained thereby is marked as the optimal construction scheme for the building exterior wall.

[0011] Furthermore, the specific process of intelligently diagnosing abnormal areas of the wall and determining insulation damage, thermal bridges, cracks and poor sealing is as follows: By calculating the thermal resistance change rate of the regional wall And set the threshold to , when the thermal resistance change rate Exceeding the set threshold When , it is determined that the insulation layer of the wall is damaged; By measuring the temperature difference between the area wall P and the surrounding walls And set the threshold to , when the temperature difference Exceeding the set threshold When , it is determined that there is a thermal bridge in the wall; A grayscale image of the wall P in the area is obtained by a three-dimensional laser scanner, and a Gaussian filter is used to remove noise on the grayscale image of the wall, so as to compare and determine whether there are cracks in the wall; By analyzing the temperature change rate of the wall indoors and comparing it with the temperature change rate of the normal area wall in the sealed state, the temperature change rate difference is obtained. And set the threshold to , when the temperature change rate difference Threshold exceeded When the wall is not tightly sealed, it is determined that the wall is not tightly sealed.

[0012] Furthermore, the specific process of using Gaussian filtering to remove noise and compare to determine whether there are cracks in the wall is as follows: Set the wall grayscale image to I (x, y), where x is the horizontal coordinate of the image and y is the vertical coordinate of the image; The Gaussian filter function is marked as G(x, y, σ), where σ is the standard deviation of the grayscale value of the image pixel; After filtering the wall grayscale image I(x, y) by the Gaussian filter function G(x, y, σ), the denoised image Iqz(x, y) is obtained, and then the gradient amplitude M(x, y) and the gradient direction γ(x, y) are obtained; and then Mnms(x, y) is obtained after non-maximum suppression; Set the high threshold of Mnms(x,y) and low threshold , and then compare the threshold: When Mnms(x,y)>high threshold , then the pixel is marked as a strong edge point; When Mnms(x, y)<low threshold , then the pixel is marked as a non-edge point; When the low threshold ≤Mnms(x,y)≤highthreshold , the pixel point is marked as a crack edge, and it is determined that there is a crack in the wall.

[0013] Furthermore, corresponding maintenance suggestions are preset in advance for each diagnosis result, and the corresponding maintenance suggestions are matched by displaying the abnormal area of ​​the wall and the intelligent diagnosis result; By actually inspecting the walls in the abnormal area, the intelligent diagnosis results can be verified; When the verification diagnosis is correct, the wall maintenance management is carried out according to the maintenance suggestions output by the system; conversely, when the verification diagnosis is wrong, professional technicians are prompted to carry out wall maintenance management according to the actual situation.

[0014] In summary, due to the adoption of the above technical solution, the beneficial effects of the present invention are: The present invention realizes spatial mapping of temperature, thickness, and material properties through multi-source data fusion and three-dimensional infrared modeling, thereby accurately locating thermal defects. The entire process is automated, from data collection, analysis, diagnosis to optimization, reducing human intervention and achieving both improved detection efficiency and accuracy. Through data-driven and model fusion, the shortcomings of traditional methods in efficiency, accuracy, and real-time performance are overcome, and the full life cycle management of the thermal performance of building exterior walls is achieved, which promotes the transformation of the detection industry towards intelligence and dataization, and realizes energy-saving transformation and sustainable development of buildings.

[0015] Among them, the present invention combines an infrared thermal imager with a three-dimensional laser scanner to construct a three-dimensional infrared building model, supports local high-precision re-testing, ensures the detection rate of abnormal defects in key areas, and completes the preliminary analysis of the thermal parameters of the building wall through the edge end, generates a thermal performance distribution map in real time, and improves the speed of data processing and response; The present invention uses a cloud processor to plan and compare the regional markings of the building's exterior walls, so as to quickly locate abnormal areas of the wall, and diagnose and identify problems such as insulation layer damage and thermal bridges, so as to generate a customized construction plan based on the optimal thermal parameters, avoid material waste, reduce energy-saving renovation costs, extend the life of the building, and reduce energy consumption losses caused by thermal defects; The present invention integrates the steady-state heat flux equation and the transient time-series heat conduction equation to perform model fusion, thereby improving the comprehensiveness of data processing, and then constructs a parameter optimization objective function to minimize the mean square error, inverts and outputs the optimal thermal parameters, and improves the inversion accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 A schematic diagram showing the overall solution of the present invention is shown; Figure 2 A schematic diagram of the framework of the system modules of the present invention is shown. DETAILED DESCRIPTION

[0017] 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.

[0018] Embodiment 1: like Figure 1-2 As shown, the data-driven building exterior wall thermal performance detection method includes the following steps: S1, the data acquisition unit monitors the infrared data and structural data of the building exterior wall; The data acquisition unit includes an infrared thermal imager and a 3D laser scanner; The infrared data of the building's exterior walls are obtained through an infrared thermal imager, and the structural data is obtained through a 3D laser scanner; By combining the infrared data of the building's exterior wall with the structural data, a three-dimensional infrared building model is constructed; Infrared data include the internal and external surface temperatures of building walls, and indoor and outdoor air temperatures; Structural data includes the three-dimensional location, thickness, and thermal conductivity of building walls; By combining infrared thermal imagers with 3D laser scanners, a 3D infrared building model is constructed to support local high-precision re-testing and ensure the detection rate of abnormal defects in key areas. In this solution, "data-driven" runs through the entire process of data collection, model construction, parameter optimization, and decision generation, especially the data analysis part of edge computing and central processing unit.

[0019] S2, the edge computing unit analyzes the thermal parameters of the building wall: by constructing a three-dimensional infrared building model, the thermal parameters of the building wall are analyzed, and the distribution of the thermal performance of the building exterior wall is displayed; The thermal parameters of the building wall are analyzed by the heat transfer model. The specific process is as follows: The building walls are planned into N regional walls through the three-dimensional infrared building model, and any regional wall is marked as P; Set the regional wall P to include n layers of wall materials, mark the total wall thickness as L, mark any layer of the wall as i, mark the thickness of the wall i as di, and mark the thermal conductivity of the wall i as ki; The inner surface temperature of the area wall P is marked as , the outer surface temperature is marked as ; The indoor air temperature of the area wall P is marked as , the outdoor air temperature is marked as ; Obtain indoor heat flux density through heat flux meter , Outdoor heat flux ; The inner surface convection heat transfer coefficient is marked as : ; The external surface convection heat transfer coefficient is labeled : ; Convection heat transfer coefficient through the inner surface , External surface convection heat transfer coefficient , the thickness di of the wall i and the thermal conductivity ki to obtain the total thermal resistance R of the regional wall P: ; Then the heat transfer coefficient U of the regional wall P is obtained: ; Thermal parameters include heat transfer coefficient and total thermal resistance; The edge computing port is used to complete the preliminary analysis of the thermal parameters of the building walls, and the thermal performance distribution map is generated in real time, which improves the speed of data processing and response.

[0020] S3, the central processing unit performs intelligent diagnosis of regional abnormalities and thermal performance optimization inversion of building walls: regional planning and comparative marking are carried out through the thermal performance distribution of the building's exterior walls, the abnormal wall areas are identified and output, and by setting m types of construction plans for the building's exterior walls, the predicted temperature is continuously approached to the measured temperature, the parameter optimization objective function of the thermal performance is established, and the optimal thermal parameters of the building's exterior wall area are output.

[0021] S3-1, the specific process of intelligent diagnosis of regional abnormalities of building walls is as follows: The thermal performance distribution of the building's exterior wall is displayed through a three-dimensional infrared building model, and then regional planning and comparison marking are carried out. The standard threshold of the total thermal resistance of the regional wall is set and marked as Qr, and the standard threshold of the heat transfer coefficient of the regional wall is set and marked as Qu; When the total thermal resistance R of the regional wall P is higher than the standard threshold value Qr, and the heat transfer coefficient U is lower than the standard threshold value Qu, the thermal performance of the regional wall P is judged to be normal; otherwise, the thermal performance of the regional wall P is judged to be abnormal, thereby realizing intelligent diagnosis of regional abnormalities, identifying and outputting abnormal wall areas; Intelligent diagnosis of abnormal wall areas to determine insulation damage, thermal bridges, cracks and poor sealing; Through regional planning and comparative marking of building exterior walls, abnormal areas of the wall can be quickly located, and problems such as insulation damage and thermal bridges can be diagnosed and identified, so as to generate customized construction plans based on optimal thermal parameters, avoid material waste, reduce energy-saving renovation costs, extend the life of the building, and reduce energy consumption losses caused by thermal defects.

[0022] S3-2, the specific process of the thermal performance optimization inversion of the building wall is: The steady-state heat transfer and transient heat transfer of the wall are analyzed by the surface heat transfer equation and the time-series heat conduction equation respectively. The specific process is as follows: Through the heat transfer coefficient U, indoor air temperature and outdoor air temperature Combined with the heat flux density of steady-state heat transfer of wall P, : ; The heat flux Substituting into the surface heat transfer equation, we obtain the predicted inner and outer surface temperatures: ; in, is the predicted inner surface temperature, is the predicted external surface temperature; The inner surface boundary space point q is marked as 0, the outer surface boundary space point q is marked as L, any time node is marked as t, and the wall surface temperature corresponding to the space point q and time node t is marked as T (q, t); Establish the time series heat conduction equation: ; in, is the thermal diffusivity, and , is the density of the wall material, c is the specific heat capacity of the wall material, and k is the thermal conductivity of the wall material; Then set the inner surface boundary conditions as: ; The boundary conditions on the outer surface are: ; Then, by setting the parameter optimization objective function of the thermal performance, the thermal performance optimization inversion is performed, the predicted temperature is compared with the measured temperature, and the optimal thermal parameters of the building's exterior wall area are obtained.

[0023] S3-3, setting the parameter optimization objective function of thermal performance, and obtaining the specific process of the optimal thermal parameters of the building exterior wall area is: The thermal conductivity coefficient ki of wall i and the convection heat transfer coefficient of the inner surface , External surface convection heat transfer coefficient , density of wall materials and specific heat capacity c integrated as the target parameter vector : ; By minimizing the mean square error between the predicted temperature and the actual temperature, the target parameter vector Perform inverse optimization, set the construction schemes of the building exterior wall to have m types, mark any construction scheme of the building exterior wall as j, and build the parameter optimization objective function of thermal performance: ; By integrating the steady-state heat flux equation with the transient time-series heat conduction equation, the model is integrated to improve the comprehensiveness of data processing, and then a parameter optimization objective function is built to minimize the mean square error, and the optimal thermal parameters are inverted and output, thereby improving the inversion accuracy.

[0024] S4, the data output unit displays the abnormal area of ​​the wall and obtains maintenance suggestions, and then combines the optimal thermal parameters to generate the optimal construction plan for the building exterior wall.

[0025] By setting m kinds of construction schemes for the building exterior wall and calculating the minimum value of the objective function, the target parameter vector obtained thereby is marked as the optimal thermal parameter, and the construction scheme obtained thereby is marked as the optimal construction scheme for the building exterior wall.

[0026] In summary, the present invention realizes spatial mapping of temperature, thickness, and material properties through multi-source data fusion and three-dimensional infrared modeling, so as to accurately locate thermal defects, and the entire process is automated, from data collection, analysis, diagnosis to optimization, reducing manual intervention, and achieving a double improvement in detection efficiency and accuracy. Through data-driven and model fusion, the shortcomings of traditional methods in efficiency, accuracy, and real-time performance are overcome, and the full life cycle management of the thermal performance of building exterior walls is realized, which promotes the transformation of the detection industry towards intelligence and dataization, and realizes energy-saving transformation and sustainable development of buildings.

[0027] The present invention combines an infrared thermal imager with a three-dimensional laser scanner to construct a three-dimensional infrared building model, supports local high-precision retesting, ensures the detection rate of abnormal defects in key areas, and completes the preliminary analysis of the thermal parameters of the building wall through the edge port, generates a thermal performance distribution map in real time, and improves the speed of data processing and response; The present invention uses regional planning and comparative marking of building exterior walls to quickly locate abnormal areas of the wall, and diagnose and identify problems such as insulation layer damage and thermal bridges, thereby generating a customized construction plan based on optimal thermal parameters, avoiding material waste, reducing energy-saving renovation costs, extending the life of the building, and reducing energy consumption losses caused by thermal defects. The present invention integrates the steady-state heat flux equation and the transient time-series heat conduction equation to perform model fusion, thereby improving the comprehensiveness of data processing, and then constructs a parameter optimization objective function to minimize the mean square error, inverts and outputs the optimal thermal parameters, and improves the inversion accuracy.

[0028] The size of the interval and threshold is set to facilitate comparison. The size of the threshold depends on the amount of sample data and the number of bases set by technical personnel in this field for each group of sample data; as long as it does not affect the proportional relationship between the parameter and the quantized value.

[0029] The above formulas are all dimensionless and numerical calculations. The formula is a formula obtained by collecting a large amount of data and performing software simulation to obtain the most recent real situation. The preset parameters in the formula are set by technicians in this field according to actual conditions. The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can make equivalent replacements or changes according to the technical scheme and inventive concept of the present invention within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. A data-driven building exterior wall thermal performance detection method, characterized in that: The following steps are involved: S1, the data acquisition unit monitors the infrared data and structural data of the building exterior wall; S2, the edge computing unit analyzes the thermal parameters of the building wall: by constructing a three-dimensional infrared building model, the thermal parameters of the building wall are analyzed, and the distribution of the thermal performance of the building exterior wall is displayed; S3, the central processing unit performs intelligent diagnosis of regional abnormalities and thermal performance optimization inversion of building walls: regional planning and comparative marking are performed through the thermal performance distribution of the building's exterior walls, the abnormal wall areas are identified and output, and m types of construction plans for the building's exterior walls are set to make the predicted temperature approach the measured temperature, build a parameter optimization objective function for thermal performance, and output the optimal thermal parameters of the building's exterior wall area; S4, the data output unit displays the abnormal area of ​​the wall and obtains maintenance suggestions, and then combines the optimal thermal parameters to generate the optimal construction plan for the building exterior wall.

2. The data-driven building exterior wall thermal performance detection method according to claim 1 is characterized in that: The data acquisition unit includes an infrared thermal imager and a 3D laser scanner; The infrared data of the building's exterior walls are obtained through an infrared thermal imager, and the structural data is obtained through a 3D laser scanner; By combining the infrared data of the building's exterior wall with the structural data, a three-dimensional infrared building model is constructed; Infrared data include the internal and external surface temperatures of building walls, and indoor and outdoor air temperatures; Structural data includes the 3D location, thickness, and thermal conductivity of building walls.

3. The data-driven building exterior wall thermal performance detection method according to claim 2 is characterized in that: The thermal parameters of the building wall are analyzed by the heat transfer model. The specific process is as follows: The building walls are planned into N regional walls through the three-dimensional infrared building model, and any regional wall is marked as P; Set the regional wall P to include n layers of wall materials, mark the total wall thickness as L, mark any layer of the wall as i, mark the thickness of the wall i as di, and mark the thermal conductivity of the wall i as ki; The inner surface temperature of the area wall P is marked as , the outer surface temperature is marked as ; The indoor air temperature of the area wall P is marked as , the outdoor air temperature is marked as ; Obtain indoor heat flux density through heat flux meter , Outdoor heat flux ; The inner surface convection heat transfer coefficient is marked as , and the external surface convection heat transfer coefficient is marked as ; Heat transfer coefficient by inner surface convection , External surface convection heat transfer coefficient , the thickness di of the wall i and the thermal conductivity ki are combined to obtain the total thermal resistance R of the regional wall P, and then the heat transfer coefficient U of the regional wall P is obtained; Thermal parameters include heat transfer coefficient and total thermal resistance.

4. The data-driven building exterior wall thermal performance detection method according to claim 3 is characterized in that: The specific process of intelligent diagnosis of regional abnormalities of building walls is as follows: The thermal performance distribution of the building's exterior wall is displayed through a three-dimensional infrared building model, and then regional planning and comparison marking are carried out. The standard threshold of the total thermal resistance of the regional wall is set and marked as Qr, and the standard threshold of the heat transfer coefficient of the regional wall is set and marked as Qu; When the total thermal resistance R of the regional wall P is higher than the standard threshold value Qr, and the heat transfer coefficient U is lower than the standard threshold value Qu, the thermal performance of the regional wall P is judged to be normal; otherwise, the thermal performance of the regional wall P is judged to be abnormal, thereby identifying and outputting the abnormal wall area; It then conducts intelligent diagnosis on abnormal areas of the wall to determine insulation damage, thermal bridges, cracks and poor sealing.

5. The data-driven building exterior wall thermal performance detection method according to claim 3 is characterized in that: The specific process of thermal performance optimization inversion of building walls is as follows: The steady-state heat transfer and transient heat transfer of the wall are analyzed by the surface heat transfer equation and the time-series heat conduction equation respectively. The specific process is as follows: Through the heat transfer coefficient U, indoor air temperature and outdoor air temperature Combined with the heat flux density of steady-state heat transfer of wall P, ; The heat flux Substitute into the surface heat transfer equation to obtain the predicted internal and external surface temperatures; mark is the predicted inner surface temperature, is the predicted external surface temperature; The inner surface boundary space point q is marked as 0, the outer surface boundary space point q is marked as L, any time node is marked as t, and the wall surface temperature corresponding to the space point q and time node t is marked as T (q, t); By the density of the wall material , specific heat capacity c, thermal conductivity k to obtain thermal diffusivity ; Establish the time-series heat conduction equation and set the boundary conditions of the internal and external surfaces; Then, by setting the parameter optimization objective function of the thermal performance, the thermal performance optimization inversion is performed, the predicted temperature is compared with the measured temperature, and the optimal thermal parameters of the building's exterior wall area are obtained.

6. The data-driven building exterior wall thermal performance detection method according to claim 5 is characterized in that: The specific process of setting the parameter optimization objective function of thermal performance and obtaining the optimal thermal parameters of the building exterior wall area is as follows: The thermal conductivity coefficient ki of wall i and the convection heat transfer coefficient of the inner surface , External surface convection heat transfer coefficient , density of wall materials and specific heat capacity c integrated as the target parameter vector ; By minimizing the mean square error between the predicted temperature and the actual temperature, the target parameter vector Perform inverse optimization, set m construction schemes for the building's exterior wall, mark any construction scheme for the building's exterior wall as j, and build a parameter optimization objective function for thermal performance; By setting m kinds of construction schemes for the building exterior wall and calculating the minimum value of the objective function, the target parameter vector obtained thereby is marked as the optimal thermal parameter, and the construction scheme obtained thereby is marked as the optimal construction scheme for the building exterior wall.

7. The data-driven building exterior wall thermal performance detection method according to claim 4 is characterized in that: The specific process of intelligently diagnosing abnormal areas of the wall and determining insulation damage, thermal bridges, cracks and poor sealing is as follows: By calculating the thermal resistance change rate of the regional wall And set the threshold to , when the thermal resistance change rate Exceeding the set threshold When , it is determined that the insulation layer of the wall is damaged; By measuring the temperature difference between the area wall P and the surrounding walls And set the threshold to , when the temperature difference Exceeding the set threshold When , it is determined that there is a thermal bridge in the wall; A grayscale image of the wall P in the area is obtained by a three-dimensional laser scanner, and a Gaussian filter is used to remove noise on the grayscale image of the wall, so as to compare and determine whether there are cracks in the wall; By analyzing the temperature change rate of the wall indoors and comparing it with the temperature change rate of the normal area wall in the sealed state, the temperature change rate difference is obtained. And set the threshold to , when the temperature change rate difference Threshold exceeded When the wall is not tightly sealed, it is determined that the wall is not tightly sealed.

8. The data-driven building exterior wall thermal performance detection method according to claim 7 is characterized in that: The specific process of using Gaussian filtering to remove noise and compare to determine whether there are cracks in the wall is as follows: Set the wall grayscale image to I (x, y), where x is the horizontal coordinate of the image and y is the vertical coordinate of the image; The Gaussian filter function is marked as G(x, y, σ), where σ is the standard deviation of the grayscale value of the image pixel; After filtering the wall grayscale image I(x, y) by the Gaussian filter function G(x, y, σ), the denoised image Iqz(x, y) is obtained, and then the gradient amplitude M(x, y) and the gradient direction γ(x, y) are obtained; and then Mnms(x, y) is obtained after non-maximum suppression; Set the high threshold of Mnms(x,y) and low threshold , and then compare the thresholds: When Mnms(x, y)>high threshold , then the pixel is marked as a strong edge point; When Mnms(x, y)<low threshold , then the pixel is marked as a non-edge point; When the low threshold ≤Mnms(x,y)≤highthreshold , the pixel point is marked as a crack edge, and it is determined that there is a crack in the wall.

9. The data-driven building exterior wall thermal performance detection method according to claim 8, characterized in that: Preset corresponding maintenance suggestions for each diagnosis result in advance, and match corresponding maintenance suggestions by displaying abnormal wall areas and intelligent diagnosis results; By actually inspecting the walls in the abnormal area, the intelligent diagnosis results can be verified; When the verification diagnosis is correct, the wall maintenance management is carried out according to the maintenance suggestions output by the system; conversely, when the verification diagnosis is wrong, professional technicians are prompted to carry out wall maintenance management according to the actual situation.

Images