Testing Methods for Thermal Performance of Building Wall Materials Based on Multimodal Sensing

Abstract

This invention relates to the field of building material testing technology, and discloses a method for testing the thermal performance of building wall materials based on multimodal sensing. The method involves installing multimodal sensors on the building wall to collect thermal performance data. Simultaneously, the building wall is divided into multiple test sections, and historical thermal performance data for each section is acquired. Thermal performance parameters are calculated based on the historical thermal performance data and the real-time collected wall thermal performance data to determine the thermal performance level of each test section, and a test plan is then developed based on this level. During the test, environmental interference is detected, real-time thermal data is analyzed, and the test process is corrected based on the analysis results. Finally, the thermal performance indicators of the building wall material are calculated. This method can collect data from multiple dimensions, achieve accurate monitoring of the thermal performance of different sections of the wall, effectively cope with environmental interference, improve the accuracy of test results, and provide reliable thermal performance data support for building-related work.

Description

Technical Field

[0001] This invention relates to the field of building material testing technology, specifically a method for testing the thermal performance of building wall materials based on multimodal sensing. Background Technology

[0002] In the development of the construction industry, the thermal performance of wall materials has a direct and crucial impact on building energy consumption and indoor thermal comfort. Accurately understanding the thermal performance of wall materials is an indispensable part of building energy-saving design, energy-saving renovation of existing buildings, and wall material research and development. Currently, various testing methods exist for the thermal performance of building wall materials, commonly including the heat flow meter method and the protective heat chamber method.

[0003] The heat flux meter method typically involves placing heat flux meters and temperature sensors on the wall surface to calculate thermal performance parameters by collecting heat flux density and temperature differences. However, this method has significant limitations in practical applications. It requires a high degree of stability in the testing environment; fluctuations in ambient temperature, wind speed, and other factors can easily lead to significant deviations in the collected data, making it difficult to accurately reflect the thermal performance of the wall during actual use. Furthermore, this method can mostly only perform overall wall testing, failing to accurately capture the differences in thermal performance between different sections of the wall. When localized thermal bridging or uneven material distribution exists in the wall, it is difficult to accurately identify the problem, posing significant challenges to subsequent building energy efficiency optimization and troubleshooting.

[0004] While the protective thermal chamber method offers improved accuracy compared to the heat flow meter method, enabling relatively accurate measurement of wall thermal performance in a laboratory environment, it typically requires cutting wall samples for testing within specialized laboratory equipment. This method cannot provide in-situ testing of building walls' thermal performance. Since the wall's installation method, surrounding environment, and connection to other building components all influence its thermal performance, laboratory test results often differ from the wall's actual thermal performance in real-world applications, failing to accurately reflect its performance in real-world usage scenarios. Furthermore, the protective thermal chamber method involves bulky equipment, complex testing procedures, long testing cycles, and relatively high costs, making it unsuitable for large-scale, rapid testing of building wall thermal performance. Summary of the Invention

[0005] The purpose of this invention is to provide a testing method for the thermal performance of building wall materials based on multimodal sensing, so as to solve the problems mentioned in the background art.

[0006] To achieve the above objectives, the present invention provides a method for testing the thermal performance of building wall materials based on multimodal sensing, the method comprising:

[0007] Multimodal sensors are installed on the building walls to collect thermal performance data; the building walls are divided into multiple test sections; historical thermal performance data of each test section is acquired; thermal performance parameters are calculated based on the historical thermal performance data and the real-time collected thermal performance data; the thermal performance level of each test section is determined; a test plan is formulated based on the thermal performance level; environmental interference is detected during the test; real-time thermal performance data is analyzed; the test process is corrected based on the analysis results; and the thermal performance indicators of the building wall materials are calculated.

[0008] Preferably, the collection of wall thermal performance data includes:

[0009] The multimodal sensor is used to acquire surface temperature distribution data and heat flux density data of the wall.

[0010] The process of dividing the building wall into multiple test sections includes:

[0011] The test sections are divided according to the structural characteristics of the wall, and each test section corresponds to an independent thermal performance evaluation unit.

[0012] Preferably, obtaining the historical thermal data for each test section includes:

[0013] Extract historical temperature change records and thermal resistance values ​​for each test segment from the storage unit;

[0014] The calculation of thermal performance parameters based on the historical thermal data and the real-time collected wall thermal performance data includes:

[0015] By combining the historical temperature change records and wall surface temperature distribution data, thermal deformation calculations are performed to generate thermal strain parameters.

[0016] Preferably, determining the thermal performance level of each test section includes:

[0017] Calculate the coefficient of performance based on the aforementioned thermal strain parameters;

[0018] The thermal performance coefficient is compared with a preset threshold to determine the thermal performance level of each test section as the first performance level, the second performance level, or the third performance level.

[0019] Preferably, the step of developing a test plan based on the thermal performance level includes:

[0020] The thermal performance level is used as the primary selection factor, and the distance between the test section and the sensor reference point is used as the secondary selection factor.

[0021] If multiple test sections have the same thermal performance level, the test section closest to the sensor reference point shall be selected for testing.

[0022] Preferably, detecting environmental interference during the test includes:

[0023] Obtain data on the center location and interference range of environmental interference sources;

[0024] Calculate the real-time distance between the real-time test location and the center location of the environmental interference source;

[0025] An interference confirmation signal is generated when the real-time distance is less than or equal to the interference range.

[0026] Preferably, the analysis of real-time thermal data includes:

[0027] Capture abnormal temperature regions in the real-time thermal data;

[0028] By integrating the temperature anomaly region and heat flux density data through multimodal data fusion technology, a thermal anomaly signal is generated.

[0029] Preferably, the test process based on the analysis results includes:

[0030] The test error is determined based on the aforementioned thermal anomaly signal;

[0031] Based on the adaptive adjustment algorithm, a test adjustment signal is generated to correct the test plan.

[0032] Preferably, the calculation of the thermal performance indicators of the building wall material includes:

[0033] Based on the corrected test process and the thermal strain parameters, thermal permeability calculation is performed to generate thermal performance index values.

[0034] Preferably, the thermal permeability calculation includes:

[0035] Based on the aforementioned thermal performance index values ​​and historical thermal data, the thermal performance evaluation units for each test section are updated.

[0036] Compared with the prior art, the beneficial effects of the present invention are:

[0037] By installing multimodal sensors on building walls to collect thermal performance data, the limitations of traditional single sensors can be overcome. This allows for the acquisition of multi-dimensional parameters related to thermal performance, such as temperature, heat flux, and humidity, resulting in more comprehensive and abundant data that more accurately reflects the thermal state of the wall. The application of multimodal sensors reduces data deviations caused by the inherent characteristics of single sensors or external factors, improving data reliability from the data acquisition source and providing a higher-quality data foundation for subsequent calculations of thermal performance parameters.

[0038] Dividing building walls into multiple test sections and acquiring historical thermal data for each section changes the traditional testing method, which often tests the entire wall. This allows for precise monitoring and analysis of the thermal performance of different sections of the wall. By comparing historical thermal data with real-time data from different test sections, differences in thermal performance between different sections of the wall can be clearly identified. This allows for the timely detection of sections with abnormal thermal performance, such as areas with localized thermal bridging or material aging leading to decreased thermal performance. This provides a clear direction for subsequent targeted optimization measures or maintenance and renovation, avoiding the drawbacks of traditional overall testing that struggles to pinpoint localized problems.

[0039] Thermal performance parameters are calculated based on historical thermal data and real-time collected wall thermal performance data. This method fully integrates the long-term performance information contained in historical data with the current performance status reflected by real-time data, making the calculated thermal performance parameters more comprehensive and accurate. Historical data can reflect the changes in the thermal performance of the wall under different seasons and environmental conditions. Combining it with real-time data allows for a more accurate assessment of the current thermal performance of the wall, reducing calculation errors that may result from relying solely on real-time or historical data, and ensuring that the thermal performance parameters better reflect the actual performance of the wall.

[0040] Determining the thermal performance level of each test section and developing a test plan based on that level allows for differentiated and targeted testing. For sections with good and stable thermal performance, the testing frequency and duration can be adjusted appropriately to reduce unnecessary consumption of testing resources. Conversely, for sections with low thermal performance, performance fluctuations, or anomalies, the testing frequency can be increased, the testing time extended, and the focus shifted to monitoring changes in thermal performance. This allows for the rational allocation of testing resources, improved testing efficiency, and ensures thorough monitoring of the thermal performance of critical sections.

[0041] This method detects environmental interference during testing and corrects the testing process based on the analysis results, effectively solving the problem of traditional testing methods being easily affected by environmental factors. External factors such as ambient temperature, wind speed, and sunlight can all interfere with the test results of wall thermal performance. This method detects these environmental interference factors in real time, analyzes their impact on real-time thermal data, and then takes corresponding corrective measures to eliminate or reduce the adverse effects of environmental interference on the test results. This ensures that the test results can still accurately reflect the thermal performance of the wall material itself under complex and changing environmental conditions, improving the reliability and stability of the test results.

[0042] By analyzing real-time thermal data and correcting the testing process, and then calculating the thermal performance indicators of building wall materials, a complete closed-loop testing process is formed. The analysis of real-time thermal data can promptly identify anomalies during the testing process. Combined with environmental interference detection results for process correction, the testing steps can be continuously optimized, reducing testing errors. The final calculated thermal performance indicators of building wall materials more accurately and objectively reflect their thermal performance levels, providing a more reliable basis for building energy-saving design, energy-saving renovation of existing buildings, and quality assessment of wall materials. This helps promote further development in the energy-saving field of the construction industry and meets the demand for precise thermal performance data in intelligent buildings. Attached Figure Description

[0043] Figure 1 This is a schematic diagram illustrating the working principle of the testing method for the thermal performance of building wall materials based on multimodal sensing as described in this invention.

[0044] Figure 2 Flowchart for determining thermal performance rating;

[0045] Figure 3 A flowchart for detecting environmental interference;

[0046] Figure 4 A flowchart for analyzing real-time thermal data. Detailed Implementation

[0047] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0048] Please see Figure 1 This invention provides a method for testing the thermal performance of building wall materials based on multimodal sensing, and its overall implementation scheme is as follows:

[0049] A multimodal sensor array is deployed on the surface of the building wall to collect thermal performance data. The entire wall is divided into multiple independent test sections based on its structural characteristics. Historical thermal performance data for each test section is retrieved from the data storage system. Thermal performance parameters are obtained by fusing historical data with real-time collected thermal performance data. The thermal performance level of each test section is determined based on the parameter calculation results. Differentiated test plans are developed based on the performance levels. Environmental interference factors are monitored in real-time during test execution. Multi-dimensional analysis is performed on the collected real-time thermal performance data. The test process is dynamically adjusted based on the analysis results. Finally, the comprehensive thermal performance index of the building wall material is calculated.

[0050] Example 1: In the implementation of the multimodal sensing-based method for testing the thermal performance of building wall materials, the multimodal sensor deployment adopts a combination of an infrared thermal imager and a heat flux sensor. The infrared thermal imager is mounted on an adjustable pan-tilt unit, which is kept parallel to the wall surface via a fixed base. The installation position must ensure that the lens's field of view covers the entire area to be measured. The infrared thermal imager operates in the long-wave infrared range, with temperature measurement accuracy controlled within a reasonable range. The sampling frequency is set to acquire one complete temperature distribution image per minute. Each frame of the temperature distribution image contains temperature data from millions of pixels, with each pixel corresponding to the temperature value at a specific location on the wall surface. The data format is a floating-point matrix. The heat flux sensor is mounted as a patch, directly attached to the wall surface. The sensor's sensing surface is completely in contact with the wall surface, with thermally conductive silicone grease applied in between to reduce contact thermal resistance. The heat flux sensor acquires heat flux density data five times per second, and the data includes heat flux intensity values ​​and direction vectors. In the multimodal sensor deployment phase, besides the combination of infrared thermal imagers and heat flux sensors, the sensor array density is determined based on the wall area and structural complexity. For walls with large areas or complex structures such as numerous door and window openings or structural columns, the number of sensors is appropriately increased to ensure that at least two sensors cover each test section. This avoids data acquisition blind spots due to insufficient sensor coverage and further ensures the spatial continuity of wall surface temperature distribution and heat flux density data, laying the foundation for subsequent segmented testing and accurate data analysis. Simultaneously, during sensor installation, appropriate installation methods are selected based on the characteristics of different wall materials. For example, for rigid wall materials such as concrete, expansion screws are used to assist in fixing the sensor base; for lightweight wall materials such as insulation boards, special adhesives are used to ensure a tight fit between the sensor and the wall surface, reducing the impact of improper installation on data acquisition accuracy.

[0051] The test sections are divided based on the structural features of the wall. A high-resolution visible light camera first acquires a panoramic image of the wall, which is then transmitted to an image processing system for edge detection and feature recognition. The system automatically identifies significant feature lines on the wall, such as material joints, edges of door and window openings, and boundaries of structural columns, dividing the wall surface into several regular rectangular regions based on these feature lines. Each rectangular region is assigned a unique number, generated in a left-to-right, top-to-bottom order. Each rectangular region serves as an independent thermal performance evaluation unit, and a mapping relationship is established between the unit number and the data acquisition channels of the multimodal sensor. This mapping relationship is stored in a configuration file, which contains the geometric center coordinates, area data, and boundary coordinate set for each evaluation unit.

[0052] During data acquisition, the infrared thermal imager transmits temperature distribution data to the central processing unit in real time via a gigabit Ethernet interface. The temperature distribution data is stored in a matrix format, where the row and column indices correspond to the two-dimensional coordinates of the wall surface, and the matrix element values ​​represent the measured temperature at that location. The heat flux sensor transmits heat flux density data to the data acquisition card via an RS-485 bus. The acquisition card performs analog-to-digital conversion on the data before uploading it to the central processing unit. The heat flux density data is stored in a time-series format, with each data point including a timestamp, heat flux intensity value, and heat flux direction angle. The central processing unit establishes an independent data buffer for each test segment, using a circular buffer structure to store real-time data from the most recent several hours.

[0053] Temperature distribution data processing includes background temperature compensation and radiation correction. Background temperature compensation is achieved using a reference blackbody placed near the sensor, while radiation correction considers the effects of environmental radiation reflection and atmospheric transmittance. Heat flux density data processing includes signal filtering and unit conversion. Filtering uses a digital filter to eliminate high-frequency noise, and unit conversion converts the raw voltage signal to standard heat flux density units. The processed data is stored in a distributed database system, with the database structured into tables based on test section numbers. Temperature data for each test section is stored as a time-series matrix, while heat flux data is stored as a multi-dimensional time series.

[0054] The geometric information of the test sections is stored in a spatial database, which records the vertex coordinates, center point coordinates, and area value of each rectangular region. The sensor reference point coordinates are obtained through GPS measurements, achieving centimeter-level accuracy. The reference point coordinates and the center point coordinates of each test section are jointly stored in the positioning database, which establishes a spatial index to support rapid distance lookups. Distance calculation employs a planar projection algorithm, which considers the planar characteristics of the wall surface, simplifying the three-dimensional distance problem into a two-dimensional planar distance calculation.

[0055] Calibration of multimodal sensors is performed regularly, with the calibration cycle dynamically adjusted based on equipment operating status. Infrared thermal imager calibration uses a standard temperature source, collecting calibration data at different temperature points to establish a temperature-voltage correspondence table. Heat flux sensor calibration employs a standard heat flux source, and the calibration process includes zero-point calibration and range calibration. Calibration data is stored in the equipment archive database for online correction during real-time data acquisition. The data acquisition software automatically adjusts acquisition parameters based on the calibration data to ensure the accuracy of measurement data.

[0056] Quality monitoring is implemented during data acquisition, with monitoring indicators including data integrity, data validity, and data consistency. Data integrity checks whether the acquired data packet contains all required fields; data validity checks whether the measured values ​​are within a reasonable range; and data consistency checks whether there are logical conflicts in the synchronized data collected by different sensors. When a data anomaly is detected, the system automatically records the anomaly event and triggers a data re-acquisition mechanism. The re-acquisition mechanism adopts different strategies based on the anomaly type: for transient anomalies, immediate re-acquisition is used; for persistent anomalies, the sensor operating parameters are adjusted before re-acquisition.

[0057] Data transmission employs a redundancy check mechanism, adding a checksum to each data packet for verification by the receiving end. Data packets failing verification require retransmission from the sending end. If the number of transmission failures exceeds a threshold, a fault diagnosis program is initiated. This program monitors network connectivity, sensor operational status, and data acquisition card status, recording the diagnostic results in the system log. The system log records all operational events and data flow status in real time, with log files rotated by date and retention periods meeting relevant requirements.

[0058] A comprehensive data traceability mechanism was established throughout the implementation process, recording the operation time and results at every stage from data acquisition to storage. Data traceability is achieved through unique identifiers; each data packet carries the acquisition time, sensor number, and test section number. This information remains linked throughout data processing, ensuring the traceability of the data source. Data storage employs a multi-layered backup strategy, with real-time data simultaneously stored in memory cache, disk storage, and network storage. Data at different layers undergoes periodic consistency checks.

[0059] Example 2: See Figure 2The storage unit employs a distributed database architecture for managing historical thermal data. This database system consists of multiple nodes, each responsible for storing historical data for a specific test section. In managing historical thermal data, the distributed database stores past temperature change records and historical thermal resistance values ​​by test section, and also labels and categorizes historical data based on the season and weather type (e.g., sunny, cloudy, rainy) during the test. This allows for more accurate filtering of historical data similar to the current test environment conditions when extracting historical data for fusion calculations with real-time data, reducing calculation deviations in thermal performance parameters caused by significant differences in environmental conditions. Past temperature change records are stored in time-series format, containing extreme temperature data, daily average temperature fluctuation curves, and temperature trend data collected during historical tests for each test section. Extreme temperature data includes the daily highest and lowest temperatures and their timestamps, while the daily average temperature fluctuation curve is generated from 24-hour continuous sampling data after smoothing. Historical thermal resistance values ​​are derived from calculations based on previous test records and stored as a set of equivalent thermal resistance values ​​for each test section. The set contains thermal resistance value records under different seasons and weather conditions.

[0060] During data extraction, the system retrieves data from the corresponding partition in the database based on the test segment number. The extraction time range is set to the most recent twelve complete test cycles, with each test cycle containing all data records from the start to the end of the test. The data query statement includes time range filtering conditions, test segment number matching conditions, and data integrity verification conditions. The extracted data package contains a temperature time series array and a thermal resistance value array, with each array element bearing a precise timestamp and data quality identifier.

[0061] Before calculating thermal performance parameters, the system performs data preprocessing. Preprocessing includes two main steps: time axis alignment and spatial coordinate matching. Time axis alignment unifies the sampling frequency of historical temperature data to the sampling frequency of real-time data using an interpolation algorithm. Spatial coordinate matching unifies the coordinate system of the pixel grid of historical temperature data with that of real-time temperature distribution data. During alignment, a nearest neighbor interpolation algorithm is used to handle pixel position deviations, ensuring that the historical data and real-time data of each pixel have a consistent spatial correspondence.

[0062] Thermal deformation calculations are performed using a differential algorithm, which calculates the pixel differences between the real-time temperature data matrix and the historical baseline temperature data matrix. The baseline temperature data is selected from historical test records under the same environmental conditions, including parameters such as ambient temperature, humidity, and solar radiation intensity. During the thermal deformation calculation, in addition to combining past temperature change records and wall surface temperature distribution data, the aging degree of the wall material is also considered. By linking the construction time and past maintenance records of each test section of the wall in the database, a material aging correction factor is appropriately introduced for walls with longer service lives, making the generated thermal strain parameters more closely reflect the thermal deformation under actual use conditions. The thermal strain parameter matrix generated by the differential calculation contains the strain value of each pixel, reflecting the deformation characteristics of the material due to temperature changes. In addition to the basic coefficient of linear expansion, the thermal strain parameters also include derived parameters such as the rate of change of temperature gradient and the uniformity of thermal strain distribution.

[0063] The coefficient of performance (COP) is calculated using a multiple regression model. The model input parameters include the values ​​of various indices in the thermal strain parameter matrix. The model weights are dynamically configured based on the wall material type, with different weight sets corresponding to different material types. These weights are stored in a material property database containing the physical property parameters of common wall materials. The regression model outputs a comprehensive performance score, which is the COP.

[0064] The performance level determination process compares the calculated coefficient of performance (COP) with preset threshold ranges. These threshold ranges are divided into three distinct areas: the first performance level corresponds to a COP value between 0 and 0.3; the second performance level corresponds to a COP value between 0.3 and 0.7; and the third performance level corresponds to a COP value between 0.7 and 1.0. The comparison algorithm employs an interval inclusion detection method to determine which threshold range the COP value falls into. The determination result generates a performance level identifier, which is represented by a numerical code.

[0065] The performance level determination results are written to a dedicated data table in the test section attribute database. This table records the test section's number, test time, thermal performance coefficient value, and performance level code. Simultaneously, the system associates the performance level information with the corresponding area in the 3D wall model. The 3D wall model is constructed using Building Information Modeling (BIM) technology, and each test section has a corresponding geometric entity within the model. The association operation is achieved through spatial location matching, assigning the performance level data to the attribute fields of the corresponding geometric entity.

[0066] The visualization process employs a color-coding scheme, with different performance levels corresponding to different display colors. The first performance level area is displayed in green, the second in yellow, and the third in red. Color rendering is accelerated through a graphics processing unit, updating the color distribution on the wall model surface in real time. The visualization system supports multi-angle viewing and detailed zooming, allowing operators to observe the detailed performance characteristics of each test section.

[0067] The entire implementation process establishes a complete data pipeline, forming a closed-loop process from historical data extraction to performance level visualization. Operation logs are recorded at each stage of data processing, including processing time, data volume, and result status. The data pipeline incorporates multiple quality checkpoints to verify the validity of intermediate data processing results. When data anomalies are detected, a reprocessing mechanism is activated. Depending on the anomaly type, the reprocessing mechanism selects to re-execute a single processing step or the entire processing flow to ensure the reliability of the final result.

[0068] The system periodically archives historical data. Archived data is compressed, encrypted, and then transferred to long-term storage media. The archiving strategy is based on data timestamps; data exceeding a certain period is automatically archived. Archived data retains all original information and processing records, supporting data backtracking and analysis when needed. The data management interface provides historical data query and retrieval functions, allowing users to search for desired data records by specifying time ranges, test segment numbers, and other criteria.

[0069] Example 3: See Figure 3 The test plan is formulated based on two key factors: thermal performance level and spatial distance. The system sorts all test sections in descending order of thermal performance level, with higher level values ​​indicating higher test priority. When multiple test sections have the same thermal performance level, the spatial distance between the center point of these sections and the sensor reference point is calculated, and the section with the closest distance is selected for priority testing. Taking the testing of the east exterior wall of a building as an example, the system divides the wall into 12 test sections; the thermal performance level and distance data are shown in Table 1.

[0070] Table 1: Priority sorting table for test sections.

[0071]

[0072] Distance calculations utilize coordinate data obtained from the Global Positioning System (GPS). After Gauss-Kruger projection coordinate transformation, the planar distance between the center point of each test section and the sensor reference point is calculated. The sensor reference point is located at the geometric center of the wall, and its coordinates are obtained through differential GPS measurements, with planar coordinate accuracy controlled to the centimeter level. The planar characteristics of the wall surface are considered during distance calculations; the three-dimensional spatial distance is projected onto a two-dimensional plane for calculation, with the projection plane parallel to the wall surface.

[0073] Environmental interference detection is achieved through a distributed monitoring network deployed at the test site. This network consists of eight ultrasonic sensors and six electromagnetic field sensors, evenly distributed around the test building. In addition to the ultrasonic and electromagnetic field sensors, temperature and wind speed sensors are added to this distributed monitoring network to collect real-time data on temperature changes and wind speed in the test environment, providing supplementary information for environmental interference assessment. When the ambient temperature fluctuates significantly within a short period or the wind speed exceeds a set threshold, an interference warning signal is generated, further refining the environmental interference detection system and enhancing its ability to identify complex environmental interference. The ultrasonic sensors cover a 50-meter radius area, sampling at 10 times per second, and can detect ultrasonic interference generated by mechanical vibrations, personnel activity, etc. The electromagnetic field sensors operate in the 50Hz-2.4GHz frequency band, with sensitivity reaching the microtesla level, and can monitor electromagnetic interference generated by power equipment, wireless communication equipment, etc.

[0074] The interference monitoring network collects environmental data in real time and transmits it to the central processing unit via a dedicated data bus. Data processing algorithms identify interference signal characteristics and determine the center coordinates of the interference source. Location calculation employs a multi-point positioning algorithm, using the time difference and intensity difference of signals received from multiple sensors to calculate the spatial coordinates of the interference source. Interference range data is calculated using a signal attenuation model that considers the influence of obstacles along the signal propagation path, providing the three-dimensional spatial distribution range of the interference field.

[0075] Real-time test location is obtained through a BeiDou / GPS dual-mode positioning module installed on the mobile test equipment. The positioning data is updated once per second, and the positioning accuracy reaches sub-meter level. The positioning module outputs latitude and longitude coordinates, which are converted into a plane coordinate system consistent with the interference monitoring network through a coordinate transformation algorithm.

[0076] Real-time distance calculation uses the planar Euclidean distance formula to calculate the straight-line distance between the current position of the test equipment and the center point of the interference source. In the spatial distance calculation stage of the test plan, in addition to using the planar Euclidean distance formula to calculate the distance between the center point of the test section and the sensor reference point, the distance calculation must avoid protruding components on the wall surface (such as window sills, waistlines, etc.). If the line connecting the center point of the test section and the sensor reference point passes through a protruding component, a broken line distance calculation method is used, that is, the shortest path distance from the center point of the test section around the protruding component to the sensor reference point, ensuring that the distance calculation result accurately reflects the actual movement path length of the test equipment, providing a more accurate spatial distance basis for optimizing the test sequence. During the calculation process, the system acquires the dynamic position coordinates of the test equipment and the updated position data of the interference source in real time. When the calculated real-time distance value is less than or equal to the radius of the interference range, the system generates a digital interference confirmation signal.

[0077] The interference confirmation signal uses a standard data format and includes fields such as interference type code, interference source strength, interference duration, and affected area. The interference type code is classified according to international standards, including categories such as mechanical vibration, electromagnetic radiation, and thermal radiation. The interference source strength is represented by a normalized numerical value, ranging from zero to one. The interference duration records the time from the occurrence of the interference to the present. The affected area field records the three-dimensional spatial dimensions of the interference field.

[0078] When an interference confirmation signal is generated, the system automatically triggers the test process adjustment mechanism. Adjustment measures include pausing data acquisition, reducing sensor sensitivity, and activating shielding devices. For brief interference, the system records the interference period and resumes testing after the interference disappears; for persistent interference, the system recalculates the test path to avoid the interference-affected area.

[0079] During testing, the system continuously monitors environmental interference, updating the interference source database every second. The database records information on all detected interference sources, including their occurrence time, duration, and intensity variation curves. This data is used for interference removal during subsequent test quality assessment and data analysis.

[0080] The entire test plan execution process incorporates a dynamic adjustment mechanism. Every five minutes, the system reassesses the priority ranking of test sections, updates the thermal performance ratings based on the latest collected data, and adjusts the test order accordingly. Simultaneously, the system updates interference source information in real time and dynamically optimizes the test path to ensure the test process is conducted under optimal conditions. All adjustment operations are recorded in the system log, including detailed information such as adjustment time, reason for adjustment, and parameter changes before and after the adjustment, forming a complete traceability record of the test process.

[0081] Example 4: See Figure 4In the implementation of the multimodal sensing-based method for testing the thermal performance of building wall materials, a sliding time window mechanism is used for real-time thermal data analysis. Each time window contains temperature data from sixty consecutive sampling points, and the window width is dynamically adjusted according to the sampling frequency. In the sliding time window mechanism for real-time thermal data analysis, in addition to dynamically adjusting the window width according to the sampling frequency, the window sliding step size is set according to the following rules: for test phases with relatively gentle temperature changes, a larger sliding step size is used to reduce data processing volume and improve analysis efficiency; for test phases with frequent temperature fluctuations, a smaller sliding step size is used to increase data sampling density, ensuring that subtle temperature anomalies can be captured in a timely manner. Temperature anomaly detection is achieved through an improved statistical outlier identification algorithm. This algorithm calculates the deviation of each sampling point from the average temperature within the window and establishes a dynamic threshold model based on historical data. When the temperature value of a sampling point exceeds the threshold range, the point is marked as a potential anomaly.

[0082] The multimodal data fusion technology employs an improved Kalman filter algorithm, which spatiotemporally matches the spatial coordinates of temperature anomaly regions with synchronously acquired heat flux density data. The fusion process first performs spatial interpolation on the heat flux density data to ensure its spatial resolution matches that of the temperature distribution data. In addition to spatial interpolation of the heat flux density data, the temperature anomaly region data undergoes smoothing preprocessing to remove isolated anomalous temperature points caused by sensor transient errors. This avoids interference from isolated anomalies in the calculation results of the heat flux-temperature coupling model, ensuring that the generated thermal anomaly signal more accurately reflects the true thermal anomaly situation of the wall. By establishing a heat flux-temperature coupling model, the correlation index between heat flux and temperature at each spatial location is calculated.

[0083]

[0084] in: Represents the correlation coefficient of thermal anomalies. Represents the number of sampling points. It is the measured value of heat flux density at the i-th sampling point. This represents the average heat flux density within the window. It is the standard deviation of the heat flux density data. Let i be the temperature measurement value at the i-th sampling point. This represents the average temperature within the window. This is the standard deviation of the temperature data. When... When the value is lower than the set threshold, the system determines that there is a thermal anomaly in the area and generates a thermal anomaly signal containing the boundary coordinates of the anomaly area and the anomaly intensity level.

[0085] The test error determination process establishes an error propagation model based on thermal anomaly signals. This model maps anomaly intensity values ​​to confidence deviation values ​​of the test data, and the mapping relationship is achieved through a pre-established calibration curve. The calibration curve is established based on a large amount of experimental data, including the measurement error range corresponding to different anomaly intensities. The confidence deviation value is expressed on a percentage scale; the larger the value, the lower the reliability of the test data.

[0086] The adaptive adjustment algorithm is implemented using fuzzy control logic, which includes multiple input and output variables. Input variables include parameters such as confidence deviation, anomaly duration, and anomaly area ratio. Output variables include sampling frequency adjustment, test duration correction coefficient, and sensor gain adjustment parameters. The fuzzy control rule base contains a series of empirical rules based on expert knowledge and historical debugging data. In addition to existing rules based on confidence deviation and anomaly duration, a control rule based on the location of the anomaly area is added to the fuzzy control rule base of the adaptive adjustment algorithm: if the thermal anomaly area is located in a critical load-bearing part of the wall or a node prone to thermal bridging, more aggressive adjustment measures (such as significantly increasing the sampling frequency and extending the test duration) are prioritized to ensure more detailed monitoring and analysis of thermal anomalies in critical areas.

[0087] The control rules adopt an if-then statement format, for example: if the confidence deviation value is large and the anomaly duration is long, the sampling frequency is significantly reduced. The fuzzy inference process uses the Mamdani inference method, which converts the fuzzy output into precise control commands through defuzzification. These commands are issued in the form of digital control signals, containing specific parameter adjustment values ​​and effective time points.

[0088] The test signal adjustment employs a gradual adjustment strategy to avoid the impact of sudden parameter changes on the testing process. Sampling frequency adjustment is achieved by changing the clock frequency of the data acquisition card, with the adjustment step dynamically determined based on the current frequency value. Test duration correction is achieved by recalculating the remaining test time, taking into account the degree of anomaly impact and test schedule factors. Sensor gain adjustment is achieved by changing amplifier circuit parameters, and the adjustment process includes calibration and verification steps.

[0089] Throughout the adjustment process, the system monitors the adjustment effect in real time and optimizes the adjustment parameters through a feedback mechanism. The effect monitoring indicators include data quality index, signal stability coefficient, and measurement consistency index. These indicators are calculated every second and compared with the baseline values ​​before adjustment. If the adjustment effect does not meet the expected goals, the system initiates a secondary adjustment process, using different control rules for further optimization.

[0090] All adjustment operations are recorded in the system log, including detailed information such as adjustment time, adjustment parameters, adjustment reason, and adjustment effect. The log data is stored in a structured format, supporting subsequent querying and analysis. The system also establishes an adjustment effect evaluation mechanism, continuously optimizing fuzzy control rules through machine learning algorithms to improve the accuracy and efficiency of adaptive adjustments. All parameter changes involved in the adjustment process are displayed in real time through a visual interface, facilitating operator monitoring and understanding of the system status.

[0091] Example 5: In the implementation of the multimodal sensing-based method for testing the thermal performance of building wall materials, the thermal permeability calculation employs a numerical solution method to handle unsteady-state heat conduction problems. At the start of the calculation, the system first establishes a physical model of the wall material, which includes basic parameters such as thermal conductivity, specific heat capacity, and density. In the physical model construction phase for thermal permeability calculation, in addition to the basic parameters such as thermal conductivity, specific heat capacity, and density, information on the wall's structural layers is also incorporated. If the wall is composed of layers of different materials, the thickness of each layer and the interfacial contact thermal resistance parameters are clearly defined in the model. A layered calculation approach is used to handle unsteady-state heat conduction problems, making the calculation results more consistent with the multi-layered structural characteristics of actual walls. The corrected test process data is used as boundary condition input, including the wall surface temperature change history and heat flow boundary conditions. Thermal strain parameters are used as material property adjustment variables in the calculation, reflecting the influence of temperature changes on the material's thermophysical properties.

[0092] The computational domain is discretized into a fine-grained mesh system, with each mesh cell assigned a corresponding material property value. The time step is dynamically adjusted according to the rate of temperature change, using a smaller step size when temperature changes drastically and a larger step size when the change is gradual. During the solution process, the temperature and heat flux values ​​of each mesh cell at each time step are recorded, and these data are used for subsequent calculations of thermal performance indicators.

[0093] The thermal performance indicators are derived from the temperature and heat flow field data obtained through analysis and calculation. The system calculates the heat storage and heat transfer capacity of the wall over a certain time period, and integrates this data into thermal performance indicators. The indicator values ​​include multiple dimensions, including parameters such as thermal inertia, thermal response rate, and heat decay factor. Each parameter is normalized, ultimately generating a comprehensive performance score within the range of zero to one.

[0094] Historical thermal performance data updates employ an incremental learning mechanism. Newly calculated thermal performance index values ​​are weighted and fused with historical values ​​stored in the database, with weighting coefficients determined based on the data's timestamp. In this incremental learning mechanism, in addition to determining weighting coefficients based on data timestamps, the weights are also adjusted based on the similarity of the historical data to the testing environment. Historical data with high similarity to current testing environment conditions have their weight increased, while those with less similarity have their weight decreased, ensuring that the fused historical data provides a more effective reference for current thermal performance index calculations. Recent data has a higher weight, while older data has a lower weight, with the weight allocation using an exponential decay function. The fused new values ​​replace the original values ​​in the database, while historical version records are retained.

[0095] The update operation of the thermal performance evaluation unit is performed at both the data and model levels. At the data level, the system writes the updated thermal performance index values ​​into a dedicated field in the evaluation unit's attribute database. The database records the timestamp of each update, the value before the update, the value after the update, and the data source on which the update was based. At the model level, the system retrains the thermal performance prediction model using the updated dataset. The training process employs a supervised learning algorithm, with input features including environmental parameters, material parameters, and historical performance data, and the output being the predicted thermal performance index.

[0096] After model training is complete, the system uses the new model to re-predict the thermal performance levels of all test sections. The predicted results are compared with the existing levels, and if significant differences are found, a level adjustment process is triggered. The level adjustment is based on the degree of agreement between the predicted results and the actual measured data, and the adjustment range is determined by predefined rules.

[0097] The updated evaluation unit generates new thermal performance level labels, which include a level code and a confidence index. The confidence index reflects the reliability of the prediction results and is calculated based on model prediction error and consistency with historical data. The new labels are written to the evaluation unit attribute database and re-associated with the 3D wall model.

[0098] The test plan reset calculation is automatically initiated after the evaluation unit update is completed. The reset process maintains the principle of prioritizing thermal performance level as the primary selection factor, but uses the updated level data for priority ranking. The system recalculates the test priority of all test sections and generates a new test sequence. The new test plan considers the latest performance level distribution and test resource constraints, optimizing test paths and test time allocation.

[0099] The entire update process forms a complete closed-loop control system. The system periodically collects new test data, updates thermal performance indicators, retrains the predictive model, adjusts performance levels, and optimizes the test plan. This closed-loop design enables the system to adapt to the changing thermal performance of wall materials over time, maintaining the accuracy and reliability of test results. Each closed-loop run generates a detailed operation log, recording the entire process from data acquisition to plan update, supporting subsequent auditing and optimization analysis.

[0100] The system also establishes a version control mechanism to track all changes to data and models. Each update generates a new version number, and the version information includes metadata such as update time, changes, and the person responsible for the changes. The version history supports rewinding to any historical state, facilitating comparison of performance trends over different periods. This design provides technical support for long-term monitoring of the evolution of building wall thermal performance.

[0101] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

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

Claims

1. A method for testing the thermal performance of building wall materials based on multimodal sensing, characterized in that, include: Multimodal sensors are installed on the building walls to collect data on the thermal performance of the walls; The building walls were divided into multiple test sections; Acquire historical thermal performance data for each test section; calculate thermal performance parameters based on the historical thermal performance data and real-time collected wall thermal performance data; determine the thermal performance level of each test section; and formulate a test plan based on the thermal performance level. Detect environmental interference during the testing process; Analyze real-time thermal data; The testing process was corrected based on the analysis results; Calculate the thermal performance indicators of building wall materials; The collected wall thermal performance data includes: The multimodal sensor is used to acquire surface temperature distribution data and heat flux density data of the wall. The process of dividing the building wall into multiple test sections includes: The test sections are divided according to the structural characteristics of the wall, and each test section corresponds to an independent thermal performance evaluation unit.

2. The method for testing the thermal performance of building wall materials based on multimodal sensing according to claim 1, characterized in that, The acquisition of historical thermal data for each test section includes: Extract historical temperature change records and thermal resistance values ​​for each test segment from the storage unit; The calculation of thermal performance parameters based on the historical thermal data and the real-time collected wall thermal performance data includes: By combining the historical temperature change records and wall surface temperature distribution data, thermal deformation calculations are performed to generate thermal strain parameters.

3. The method for testing the thermal performance of building wall materials based on multimodal sensing according to claim 2, characterized in that, The determination of the thermal performance level of each test section includes: Calculate the coefficient of performance based on the aforementioned thermal strain parameters; The thermal performance coefficient is compared with a preset threshold to determine the thermal performance level of each test section as the first performance level, the second performance level, or the third performance level.

4. The method for testing the thermal performance of building wall materials based on multimodal sensing according to claim 3, characterized in that, The process of developing a test plan based on the thermal performance level includes: The thermal performance level is used as the primary selection factor, and the distance between the test section and the sensor reference point is used as the secondary selection factor. If multiple test sections have the same thermal performance level, the test section closest to the sensor reference point shall be selected for testing.

5. The method for testing the thermal performance of building wall materials based on multimodal sensing according to claim 4, characterized in that, The detection of environmental interference during the testing process includes: Obtain data on the center location and interference range of environmental interference sources; Calculate the real-time distance between the real-time test location and the center location of the environmental interference source; An interference confirmation signal is generated when the real-time distance is less than or equal to the interference range.

6. The method for testing the thermal performance of building wall materials based on multimodal sensing according to claim 5, characterized in that, The real-time thermal data analyzed includes: Capture abnormal temperature regions in the real-time thermal data; By integrating the temperature anomaly region and heat flux density data through multimodal data fusion technology, a thermal anomaly signal is generated.

7. The method for testing the thermal performance of building wall materials based on multimodal sensing according to claim 6, characterized in that, The calibration test process based on the analysis results includes: The test error is determined based on the aforementioned thermal anomaly signal; Based on the adaptive adjustment algorithm, a test adjustment signal is generated to correct the test plan.

8. The method for testing the thermal performance of building wall materials based on multimodal sensing according to claim 7, characterized in that, The thermal performance indicators for calculating building wall materials include: Based on the corrected test process and the thermal strain parameters, thermal permeability calculation is performed to generate thermal performance index values.

9. The method for testing the thermal performance of building wall materials based on multimodal sensing according to claim 8, characterized in that, The thermal permeability calculation includes: Based on the aforementioned thermal performance index values ​​and historical thermal data, the thermal performance evaluation units for each test section are updated.

Images