A method and system for reconstructing the temperature field of a large-span roof

Abstract

The application discloses a large-span roof temperature field reconstruction method and system, and particularly relates to the technical field of building structure health monitoring, and is used for solving the problems of compensation interruption and temperature field discontinuity caused by the time sequence asynchronism of environmental parameters and temperature data in the prior art; through synchronous collection of roof temperature sensors and micro-meteorological parameters and uniform time stamping, a compensation cache area is established to extract uncompensated data; a real-time / delay compensation area is dynamically divided according to the solar radiation intensity change rate; real-time area data is dynamically compensated to generate compensated data, and the delay area data is compensated asynchronously in the reconstruction gap; the compensated data is input into a reconstruction algorithm to generate a temperature field distribution; and the continuity is checked based on the historical and current temperature field gradient change quantity, so that the temperature field reconstruction process is realized without interruption, the time dimension continuity is ensured, and the real-time requirement is met through a resource optimization strategy.

Description

Technical Field

[0001] This invention relates to the field of building structural health monitoring technology, and more specifically, to a method and system for reconstructing the temperature field of a large-span roof. Background Technology

[0002] Temperature field monitoring of large-span roof structures (such as stadiums and airport terminals) relies on distributed sensor networks and dynamic environmental compensation techniques. Existing methods typically collect temperature data by deploying fiber optic sensors and infrared thermal imagers on the roof, and combine this with micro-weather stations to obtain parameters such as solar radiation and wind speed. Linear or nonlinear compensation formulas are then used to correct for the influence of environmental factors (such as changes in the solar incidence angle) on temperature measurement, thereby achieving dynamic modeling of the temperature field.

[0003] In existing technologies, the real-time updates of dynamic compensation parameters and environmental data are asynchronous. The acquisition frequency of micrometeorological parameters (such as radiation intensity and incident angle) and the data stream from temperature sensors inherently have a delay difference, causing the parameter updates of the compensation formula to be out of sync with the temperature data acquisition time. When environmental parameters change abruptly (such as a jump in radiation intensity caused by instantaneous cloud changes), the compensation calculation must interrupt the current reconstruction process to wait for the latest parameters, resulting in frequent interruptions in the temperature field generation process. This not only disrupts the continuity of the temperature field in the time dimension but also causes abrupt jumps in the reconstruction results, further affecting the stability of subsequent thermal stress analysis. Summary of the Invention

[0004] In order to overcome the above-mentioned defects of the prior art, the present invention provides a method and system for reconstructing the temperature field of a large-span roof to solve the problems mentioned in the background art.

[0005] To achieve the above objectives, the present invention provides the following technical solution:

[0006] A method for reconstructing the temperature field of a large-span roof includes the following steps:

[0007] S1. Synchronously collect data from distributed temperature sensors on the roof and micro-meteorological parameters, and add a unified timestamp to the temperature sensor data stream and the micro-meteorological parameter stream.

[0008] S2. Establish a compensation buffer area. When a new micro-meteorological parameter is detected, extract the uncompensated temperature sensor data within the preset time window.

[0009] S3. Based on the comparison results between the solar radiation intensity change rate in the new micro meteorological parameters and the preset threshold, the roof area is divided into a real-time compensation area and a delayed compensation area.

[0010] S4. Perform dynamic compensation on the uncompensated temperature sensor data in the real-time compensation zone to generate real-time compensated data. At the same time, mark the uncompensated temperature sensor data in the delayed compensation zone as the dataset to be compensated, and perform asynchronous compensation on the dataset to be compensated during the temperature field reconstruction interval.

[0011] S5. Input the real-time compensated data into the preset reconstruction algorithm to generate the roof temperature field distribution at the current moment;

[0012] S6. Based on the gradient change between the previously generated roof temperature field distribution and the currently generated roof temperature field distribution, perform time continuity verification on the reconstruction results.

[0013] Furthermore, data from distributed roof temperature sensors and micrometeorological parameters are collected simultaneously, and a unified timestamp is applied to both the temperature sensor data stream and the micrometeorological parameter stream, including:

[0014] The data stream from the first temperature sensor is acquired through a distributed optical fiber sensor, the data stream from the second temperature sensor is acquired through an infrared thermal imager, and the data stream from solar radiation intensity and solar incidence angle change is acquired through a micro weather station.

[0015] The data streams from the first temperature sensor of the distributed fiber optic sensor, the second temperature sensor of the infrared thermal imager, and the solar radiation intensity and solar incidence angle change data streams of the micro-weather station are all timestamped using a time synchronization server.

[0016] Furthermore, a compensation buffer is established. When a new micrometeorological parameter is detected, uncompensated temperature sensor data within a preset time window is extracted, including:

[0017] Configure a circular storage queue to cache the first and second temperature sensor data streams with a unified timestamp;

[0018] When the new solar radiation intensity data stream or the new solar incident angle change data stream arrives, a preset time window is calculated based on the timestamp corresponding to the new micro-meteorological parameters. All uncompensated first temperature sensor data streams and uncompensated second temperature sensor data streams within the corresponding time window are extracted as the dataset to be processed.

[0019] Furthermore, based on the comparison between the rate of change of solar radiation intensity in the new micro-meteorological parameters and the preset threshold, the roof area is divided into a real-time compensation zone and a delayed compensation zone, including:

[0020] Spatial coordinate distribution of the preloaded roof curvature abrupt change region;

[0021] Calculate the rate of change value of the new solar radiation intensity data stream;

[0022] When the rate of change exceeds the preset threshold, the entire coordinate range of the curvature abrupt change region will be included in the real-time compensation zone.

[0023] For regions without curvature abrupt changes, based on the comparison between the change amplitude of the solar incident angle data stream and the preset angle change threshold, sub-regions with change amplitudes exceeding the preset angle change threshold are classified into the real-time compensation zone, while the remaining regions are classified into the delay compensation zone.

[0024] Furthermore, dynamic compensation is performed on the uncompensated temperature sensor data in the real-time compensation zone to generate real-time compensated data, including:

[0025] Acquire the uncompensated first temperature sensor data stream and the uncompensated second temperature sensor data stream corresponding to the real-time compensation area;

[0026] Based on the timestamp difference between the new solar radiation intensity data stream, the new solar incidence angle change data stream and the uncompensated temperature sensor data stream, calculate the time-weighted compensation factor;

[0027] A dynamic compensation formula is used to perform batch compensation calculations on the uncompensated first temperature sensor data stream and the uncompensated second temperature sensor data stream respectively, generating real-time compensated data, including the real-time compensated first temperature sensor data stream and the real-time compensated second temperature sensor data stream.

[0028] Furthermore, the uncompensated temperature sensor data in the delay compensation region is labeled as the dataset to be compensated, including:

[0029] Store the uncompensated first temperature sensor data stream and the uncompensated second temperature sensor data stream corresponding to the delay compensation area into the compensation queue.

[0030] Furthermore, during the intervals between temperature field reconstructions, asynchronous compensation operations are performed on the dataset to be compensated, including:

[0031] During the refactoring task interval, data is extracted from the queue to be compensated, and the same dynamic compensation formula is used to perform compensation calculations.

[0032] Furthermore, the real-time compensated data is input into a preset reconstruction algorithm to generate the roof temperature field distribution at the current moment, including:

[0033] Receive the data stream from the first temperature sensor after real-time compensation and the data stream from the second temperature sensor after real-time compensation;

[0034] The radial basis interpolation algorithm based on adaptive kernel function is used to process the data streams of the first and second temperature sensors after real-time compensation.

[0035] During the interpolation calculation, the kernel function parameters are dynamically adjusted according to the roof curvature change rate, and the interpolation node density is increased preferentially in areas with high curvature changes.

[0036] Interpolation operations are performed using a parallel computing architecture to generate continuous temperature field distribution data covering the entire roof as the roof temperature field distribution at the current moment.

[0037] Furthermore, based on the gradient change between the previously generated roof temperature field distribution and the currently generated roof temperature field distribution, the temporal continuity of the reconstruction results is verified, including:

[0038] Extract historical temperature gradient distribution data from the previously generated roof temperature field distribution;

[0039] Calculate the temperature gradient change at the same spatial coordinates in the currently generated roof temperature field distribution;

[0040] Compare the temperature gradient change with a preset gradient threshold;

[0041] When the temperature gradient change exceeds the preset gradient threshold, locate the coordinates of the corresponding roof curvature abrupt change area.

[0042] A local recompensation command is generated based on the positioning results. The local recompensation command includes the spatial coordinate range that needs to be recompensated.

[0043] On the other hand, the present invention provides a temperature field reconstruction system for a large-span roof, comprising the following modules:

[0044] The synchronous acquisition module is used to synchronously acquire data from the roof distributed temperature sensor and micro-meteorological parameters, and to add a unified timestamp to the temperature sensor data stream and the micro-meteorological parameter stream.

[0045] The cache extraction module is used to establish a compensation cache area. When a new micro-meteorological parameter is detected, it extracts the uncompensated temperature sensor data within a preset time window.

[0046] The zoning decision module is used to divide the roof area into a real-time compensation zone and a delayed compensation zone based on the comparison results between the rate of change of solar radiation intensity in the new micro-meteorological parameters and the preset threshold.

[0047] The dynamic compensation module is used to perform dynamic compensation on the uncompensated temperature sensor data in the real-time compensation zone to generate real-time compensated data. At the same time, the uncompensated temperature sensor data in the delayed compensation zone is marked as the dataset to be compensated, and asynchronous compensation is performed on the dataset to be compensated during the temperature field reconstruction interval.

[0048] The field reconstruction module is used to input the real-time compensated data into a preset reconstruction algorithm to generate the roof temperature field distribution at the current moment.

[0049] The verification and positioning module is used to verify the temporal continuity of the reconstruction results based on the gradient change between the previously generated roof temperature field distribution and the currently generated roof temperature field distribution.

[0050] Compared with the prior art, the present invention has the following beneficial effects:

[0051] 1. By eliminating time-series deviations in multi-source data through synchronous acquisition and a unified timestamp mechanism, and combining a dynamic compensation buffer and a real-time / delay partitioning strategy, only local compensation calculations need to be interrupted when environmental parameters change abruptly, ensuring the continuous operation of the temperature field reconstruction process. Based on the continuity verification mechanism of the gradient changes between the previous and current temperature fields, the jump regions are accurately identified and local recompensation is triggered, fundamentally avoiding the temperature field step distortion caused by global waiting in traditional methods. This scheme enables the temperature field of large-span roofs to maintain high continuity in the time dimension, providing a stable and reliable data foundation for subsequent thermal stress analysis.

[0052] 2. A zoning decision-making mechanism driven by the rate of change of radiation intensity is adopted to prioritize the processing of highly sensitive areas (curvature change zones) and implement delay compensation for low-impact areas, effectively reducing the real-time computing load; asynchronous compensation operations are performed during the temperature field reconstruction intervals, and resource conflicts between compensation calculation and reconstruction processes are avoided by reusing resources during idle periods; combined with a GPU parallel-accelerated temperature field reconstruction algorithm, the entire process time is compressed to the minute level while ensuring spatial resolution, meeting the stringent real-time requirements of large-span roof safety monitoring. Attached Figure Description

[0053] Figure 1 This is a flowchart of a method for reconstructing the temperature field of a large-span roof according to the present invention;

[0054] Figure 2 This is a schematic diagram of the temperature field reconstruction system for a large-span roof according to the present invention. Detailed Implementation

[0055] 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 of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0056] Example 1: Figure 1 This invention provides a method for reconstructing the temperature field of a large-span roof, which includes the following steps:

[0057] S1. Synchronously collect data from distributed temperature sensors on the roof and micro-meteorological parameters, and add a unified timestamp to the temperature sensor data stream and the micro-meteorological parameter stream.

[0058] S2. Establish a compensation buffer area. When a new micro-meteorological parameter is detected, extract the uncompensated temperature sensor data within the preset time window.

[0059] S3. Based on the comparison results between the solar radiation intensity change rate in the new micro meteorological parameters and the preset threshold, the roof area is divided into a real-time compensation area and a delayed compensation area.

[0060] S4. Perform dynamic compensation on the uncompensated temperature sensor data in the real-time compensation zone to generate real-time compensated data. At the same time, mark the uncompensated temperature sensor data in the delayed compensation zone as the dataset to be compensated, and perform asynchronous compensation on the dataset to be compensated during the temperature field reconstruction interval.

[0061] S5. Input the real-time compensated data into the preset reconstruction algorithm to generate the roof temperature field distribution at the current moment;

[0062] S6. Based on the gradient change between the previously generated roof temperature field distribution and the currently generated roof temperature field distribution, perform time continuity verification on the reconstruction results.

[0063] The first temperature sensor data stream is acquired through distributed optical fiber sensors. These distributed optical fiber sensors are deployed at specific locations on the surface of the roof steel structure grid, specifically at the intersection of the main beam and secondary beam of the roof. The sensor spacing is controlled within a range of 2 meters (e.g., 2 meters, 3 meters, etc., which can be adjusted according to the specific roof span). The sensor works based on the Raman scattering effect to measure temperature. The specific process is as follows: the pulsed laser transmitted inside the optical fiber interacts with the thermal vibration of the material molecules to generate backscattered light. The intensity of the scattered light signal is directly proportional to the temperature change. This scattered light is converted into a voltage signal by a photoelectric converter, and then sampled by an analog-to-digital converter to form discrete temperature values. The sampling frequency is set to a specific value (e.g., once per second or once every 2 seconds). Each sampling point corresponds to a pre-set three-dimensional spatial coordinate position on the roof, thus forming a continuous spatially distributed roof temperature data sequence, which is the first temperature sensor data stream.

[0064] The infrared thermal imager is installed at a fixed location on the surrounding high-rise building or on a mobile platform of a drone. The imaging spectrum is selected in the long-wave infrared range (e.g., 8-14 micrometer wavelength range), the single pixel spatial coverage size is set to a specific resolution (e.g., 0.5 m × 0.5 m or 1 m × 1 m), and the acquisition frequency is set to a specific interval (e.g., one frame of thermal image is acquired every 2 seconds or every 5 seconds). During thermal image processing, the infrared radiation intensity values ​​of each pixel area are segmented, and the radiation values ​​are converted into the corresponding steel structure surface temperature values ​​using a blackbody radiation correction curve. The temperature values ​​of each pixel are correlated with latitude and longitude coordinates to generate spatial array data, i.e., the second temperature sensor data stream.

[0065] The micro-weather station is installed in an unobstructed area at the highest point of the roof. The radiation sensor uses a thermopile-type total radiation meter, and the sensing element includes a constantan thermocouple stack structure. It generates a voltage signal by measuring the electromotive force caused by the temperature difference due to solar radiation. The voltage signal is collected at a specific sampling interval (e.g., 0.5 seconds or 1 second). The voltage value is converted into a radiation power value per unit area (watts per square meter) using a calibration formula, thus forming a solar radiation intensity data stream. The change in solar incidence angle is measured using a dual-axis electronic inclinometer. The inclinometer is fixed on a vertical mounting base. The gravity vector direction is calculated in real time using a built-in three-axis accelerometer and a three-axis gyroscope. The change in angle between the gravity vector and the roof normal vector is calculated and recorded at a fixed period (e.g., 0.5 seconds or 1 second). This change is processed by a specific normalization function (e.g., arctangent function) to form a solar incidence angle change data stream.

[0066] The data streams from the first temperature sensor of the distributed fiber optic sensor, the second temperature sensor of the infrared thermal imager, and the solar radiation intensity and solar incidence angle change data streams of the micro-weather station are all timestamped. In specific implementation, a high-precision time synchronization server is deployed. This server uses the GPS satellite signal as the reference time source. The receiver continuously receives Coordinated Universal Time (UTC) signals sent by multiple positioning satellites (e.g., 4 or more). The local clock offset is corrected by a time calibration algorithm (e.g., least squares estimation algorithm). The internally configured precision time base device (e.g., rubidium atomic clock) controls the clock error within a specific accuracy range (e.g., less than 1 microsecond).

[0067] Each data acquisition terminal device is equipped with a Network Time Protocol (NTP) communication module. Distributed fiber optic sensors are connected via fiber optic Ethernet, and infrared thermal imagers and micro-weather stations are accessed via wireless communication modules. All terminal devices send clock synchronization request data packets to the time synchronization server at specific time intervals (e.g., every minute).

[0068] The time synchronization server uses a bidirectional time transmission mechanism to respond. Specifically, it records the arrival time of the request packet and the sending time of the response packet. The client device records the time of receiving the response packet. It uses the four time values ​​of request sending time, request arrival time, response sending time and response receiving time to calculate the transmission delay of the communication path and corrects the local clock deviation through a time compensation algorithm.

[0069] After time synchronization is completed, each acquisition terminal adds a time tag in a specific format (e.g., 64-bit UTC time code) when generating data each time. The time tag accuracy reaches a specific level (e.g., microsecond level). The time tag of the first temperature sensor data stream is embedded in a specific field of the message header of the fiber optic demodulator output data. The time tag of the second temperature sensor data stream is written to the metadata area of ​​the thermal image file. The time tag of the solar radiation intensity data stream is placed in the status word area of ​​the serial communication protocol data packet. The time tag of the solar incidence angle change data stream is stored in the index tag area of ​​the circular buffer.

[0070] After all data streams converge to the central data processing unit, the time stamp information of various data streams is extracted, the difference between the time stamps of different data streams is compared, and when the time deviation is detected to exceed a specific threshold (e.g., 50 milliseconds), the time series interpolation algorithm is activated to generate transition data points to achieve timeline alignment. Finally, the set of data streams with strictly matched timestamps, including the first temperature sensor data stream, the second temperature sensor data stream, the solar radiation intensity data stream, and the solar incidence angle change data stream, is output.

[0071] The time synchronization quality verification adopts a spatial sampling method, randomly selecting some roof measurement points (e.g., sampling points of 5% of the roof area), and comparing the difference between the recording time of the fiber optic sensor and the exposure time of the infrared thermal imager at that location. When the time difference exceeds a specific alarm threshold (e.g., 5 milliseconds), the abnormal handling process is triggered.

[0072] In areas with weak signal coverage, such as the roof edge, fiber optic time synchronization signal repeaters are deployed. These repeaters receive the optical synchronization signal transmitted from the backbone fiber and regenerate the clock signal locally, distributing it to surrounding terminal devices via optical switching equipment. The resulting unified timestamp data set is stored in a time-series database system. This database is indexed along a millisecond-level time axis to support fast retrieval, providing a strictly time-aligned foundational data set for subsequent data processing stages.

[0073] A circular storage queue is configured to cache the first and second temperature sensor data streams with unified timestamps. The specific implementation process is as follows: A contiguous address region is allocated in the central processing unit's memory space as the carrier of the circular storage queue. This storage area manages data according to a first-in, first-out (FIFO) rule, and its capacity is determined based on the temperature sensor sampling frequency and the maximum allowable delay (e.g., with a sampling interval of 2 seconds, the capacity is designed to store 30 minutes of data). The data writing mechanism is as follows: When the first and second temperature sensor data streams with unified timestamps are received, the parser extracts the timestamp and spatial coordinate information from the data packets and stores the data packets sequentially into contiguous memory blocks of the circular storage queue. The size of the memory block matches the number of bytes in the temperature data packet (e.g., a single temperature data packet contains coordinate values ​​and temperature values, totaling 64 bytes). The circular storage queue is equipped with both write and read pointers. The write pointer always points to the latest data storage location. When the queue is full, it automatically overwrites the oldest historical data, with the overwrite rule based on a preset data validity period (e.g., a default validity period of 15 minutes). The first temperature sensor data stream and the second temperature sensor data stream are stored in a partitioned queue. The first temperature sensor data stream is stored at the front address of the queue (e.g., the address range from the start bit 0000 to the end bit 7FFF), and the second temperature sensor data stream is stored at the back address (e.g., the address range from the start bit 8000 to the end bit FFFF). During storage, a status flag bit is added to indicate the data compensation status (uncompensated is marked with status code 00, and compensated is marked with status code 01).

[0074] The implementation of detecting the arrival of new solar radiation intensity data streams or new solar incidence angle change data streams includes: deploying an interrupt triggering mechanism at the micrometeorological data receiving port, which continuously monitors the input status of the data port. Upon detecting the start frame signal of a new solar radiation intensity data stream (e.g., Ethernet frame synchronization header AA) or the start identifier of a new solar incidence angle change data stream (e.g., serial port header 55), an interrupt service routine is immediately activated. The interrupt service routine first parses the timestamp field (e.g., the 64-bit UTC timecode defined in the previous steps) in the new data packet, and simultaneously verifies the checksum of the data packet to ensure integrity. For the new solar radiation intensity data stream, its timestamp value and the sequence of newly arrived radiation values ​​are extracted; for the new solar incidence angle change data stream, its timestamp value and the sequence of newly arrived angle changes are extracted. Upon detection of new data arrival, an event log is generated, recording the event type (radiation update or angle update) and the arrival time accurate to microseconds.

[0075] The operation procedure for calculating the preset time window based on the timestamp of the new micro-meteorological parameter is as follows: Take the timestamp of the newly arrived micro-meteorological parameter as the reference time point. Set the preset time window length as the window duration (the window duration value is determined based on the thermal conduction response rate; for example, if the thermal response time constant of steel is 30 seconds, the window duration is adjustable within the range of 15 to 60 seconds). Calculate the starting time point of the time window by subtracting the window duration from the reference time point. Scan the storage area in the circular storage queue, and perform index matching on the data streams of the first and second temperature sensors respectively: traverse the timestamp field of each data block in the queue, filter data blocks whose timestamps are greater than or equal to the starting time point of the time window and less than or equal to the reference time point, and simultaneously verify that the status flag is in an uncompensated state (status code 00). The extracted data blocks form a temporary data set. If multiple time point data exist at the same spatial location (such as the overlapping area of ​​infrared and fiber optics), retain the latest valid data.

[0076] The process of extracting uncompensated temperature sensor data within the corresponding time window as the dataset to be processed includes: spatial integration of the selected first and second temperature sensor data streams. Through a unified coordinate system transformation (geographic coordinate system X, Y, and Z axes), the first temperature sensor data stream from the fiber optic sensor is mapped to a 3D grid according to the installation path coordinates, and the second temperature sensor data stream from the infrared thermal imager is projected onto the same grid according to pixel latitude and longitude. The grid size is set according to the roof accuracy requirements (e.g., 0.5m x 0.5m grid). After coordinate transformation, data fusion is performed: within the same grid cell, if both the first and second temperature sensor data streams exist, the average value is taken as the temperature value of that cell (or weighted according to sensor accuracy weights); single data source cells directly use the original value. Finally, a spatiotemporally aligned dataset to be processed is generated. The data record format of this dataset includes the spatial grid index number, the original temperature value, the compensation status flag, and the source sensor type identifier.

[0077] The preset time window length is derived from the thermal model: a transient heat conduction equation is constructed based on the specific heat capacity of the roof panel material (e.g., 502 joules per kilogram Kelvin for steel), thermal conductivity (50 watts per meter Kelvin), and ambient wind speed. The temperature response lag time constant is determined through numerical simulation. The preset window duration must satisfy the condition that half the lag time constant is less than the window duration and the window duration is less than twice the lag time constant (e.g., the lag time constant for a steel roof is approximately 30 seconds, so the window duration is between 15 and 60 seconds). In engineering applications, material characteristic parameters are input through the parameter calibration interface, and simulation calculations are performed to generate recommended window duration values. A linear buffer expansion strategy is used for time window boundary handling: when the reference time point is close to the oldest data time in the queue (e.g., the queue storage start time), if the time window start time point is less than the storage start time, the window range is adaptively adjusted to the distance from the storage start time point to the reference time point to avoid out-of-range indexing; simultaneously, a system alarm is triggered to indicate data overwrite risk, with the alarm threshold set to the number of times the window duration exceeds the limit per hour (e.g., 5 times). The dataset to be processed is ultimately stored in a dedicated buffer with a time window identifier attached, awaiting the compensation operation. The buffer structure uses a two-dimensional matrix for storage, with row records corresponding to spatial grid indices and column records arranged in time series to display temperature values, enabling fast access in both spatiotemporal dimensions.

[0078] In practice, the preset time window length is determined based on the thermophysical properties of the roof panel material (such as the density, specific heat capacity, and thermal conductivity of steel) and the ambient wind speed. The transient heat conduction equation is a general physical equation used in building thermal engineering to calculate the temperature change of materials over time. It describes the relationship between the rate of heat transfer within the material and the material's own properties (density, specific heat capacity, and thermal conductivity) and the ambient heat input (solar radiation). The temperature response lag time constant is determined using the following standard laboratory method: a standard specimen of the same material as the roof is selected, and a step increase in solar radiation intensity (e.g., from 200 watts per square meter to 800 watts per square meter) is simulated in a controlled environment chamber. A high-precision temperature sensor is used to record the time required for the specimen surface temperature to reach approximately 63.2% of the new equilibrium state. For example, for a 0.6 mm thick Q345B steel plate, this time is measured to be approximately 28 seconds under windless conditions. The preset time window length is then taken as 1 to 1.5 times this measured time value (e.g., 28 to 42 seconds). In engineering applications, this time constant can be directly input into the system configuration interface, and the system will automatically set the recommended window length.

[0079] In practice, the transient heat transfer process is achieved through standard thermal tests:

[0080] Material thermal response test: A standard specimen of the same material as the roof (e.g., 0.6mm thick Q345B steel plate) was selected and subjected to a fixed radiation intensity of 800W / m² in an environmental chamber. 2After heating until the temperature stabilizes, the radiation is suddenly reduced to 200 W / m². 2 The temperature drop curve was recorded using an infrared thermal imager (sampling rate 1Hz).

[0081] Lag time determination: When the temperature drops to 63.2% of the initial temperature difference (e.g., from 45℃ to 37.6℃), the time required is recorded as the lag time constant (typical value for steel: 28 seconds).

[0082] Explanation of the relationship: The rate of heat transfer is as follows: the higher the density / specific heat capacity, the slower the temperature change (e.g., steel is 3 times slower than concrete); the higher the thermal conductivity, the faster the transfer of environmental radiation (aluminum is 2 times faster than steel).

[0083] System configuration: After entering the material parameters (density 7850 kg / m³) in the software interface... 3 (Specific heat capacity 502 J / kg·K), the system automatically calculates the recommended time window (28-42 seconds).

[0084] The implementation process for preloading the spatial coordinate distribution of roof curvature abrupt change regions includes: extracting the roof's three-dimensional geometric model data from the Building Information Model (BIM). The model format is an industrial basic class standard file, containing the non-uniform rational B-spline mathematical expression of all curved surfaces on the roof. Gaussian curvature values ​​are calculated for each surface patch. The specific algorithm is as follows: take the principal curvature radius value at any point on the surface; the Gaussian curvature is equal to the product of the two principal curvature values. When the absolute value of the Gaussian curvature at a point exceeds a set threshold (e.g., 0.05 per meter), that point is marked as a curvature abrupt change point. All curvature abrupt change points are projected onto a two-dimensional plane coordinate system (using the architectural plane projection coordinate system) to generate a density distribution cloud map. Cluster analysis algorithms are used to identify abrupt change point clusters. The cluster radius is set to a specific value (e.g., 2 meters). The boundary of each cluster region is calculated using a convex hull algorithm, and the final output is a set of polygon vertex coordinates. For example, in stadium engineering (such as a 250-meter span cable-membrane structure), 15 curvature abrupt change regions were identified, including ridge and valley regions. Each region was stored as an independent set of coordinate ranges (such as a sequence of polygon vertices) with a coordinate accuracy of a specific value (such as 0.01 meters). The coordinate set of curvature abrupt change regions was stored in a spatial database, and an index mapping relationship was established with the roof grid.

[0085] The procedure for calculating the rate of change value of a new solar radiation intensity data stream is as follows: Take the two most recent arriving solar radiation intensity data streams, which are the radiation values ​​R1 corresponding to time label T1 and R2 corresponding to time label T2 (T2>T1). Calculate the time difference ΔT = T2 - T1 (in seconds) and the change in radiation value ΔR = R2 - R1 (in watts per square meter). The rate of change value is defined as the derivative of the change in radiation with respect to time, specifically calculated as ΔR divided by ΔT (in watts per square meter per second). When the new data stream is arriving for the first time (i.e., no historical data), the rate of change value is set to a specific initial value (e.g., 0). In engineering implementation, the rate of change calculation results are stored in a dynamic array, with the array index corresponding to the micro-meteorological data time label. To prevent noise interference, a minimum time difference threshold is set (e.g., 0.5 seconds). When ΔT is less than this threshold, the time window is automatically expanded to include the first three consecutive data points, and the linear regression slope is used as the rate of change value. Output the data sequence of time labels and corresponding rate of change values ​​for subsequent zoning decisions.

[0086] When the rate of change exceeds a preset threshold, the method for including the entire coordinate range of the curvature abrupt change region into the real-time compensation zone is as follows: The preset threshold is determined through material thermal sensitivity analysis. Specifically, the transient heat conduction differential equation of the roof panel material (e.g., aluminum-magnesium-manganese alloy) is established, and the boundary condition that the critical rate of change of radiation causes the surface temperature rise rate to exceed a specific value (e.g., 2 degrees Celsius per minute) is solved. Under typical operating conditions, the preset threshold is set to a specific value (e.g., 100 watts per square meter per second). The decision-making process is as follows: the current rate of change value is compared with the preset threshold in real time. When the rate of change value is detected to be greater than the preset threshold, the coordinate ranges of all curvature abrupt change regions are retrieved from the spatial database. The roof grid cells corresponding to these coordinate ranges are marked as real-time compensation status (status code 01), and a set of region identifiers is generated. In large airport roof applications (e.g., an area of ​​30,000 square meters), when clouds move rapidly at midday in summer, the measured rate of radiation change reaches a specific value (e.g., 120 watts per square meter per second), triggering the system to complete the status marking of 12 curvature change zones (total area of ​​approximately 5,000 square meters) within a specific time (e.g., within 200 milliseconds).

[0087] Methods for determining the critical rate of change of radiation threshold:

[0088] Engineering measurements:

[0089] A radiometer (Apogee SP-510) and an infrared thermal imager (FLIR A655sc) were installed in a typical area of ​​the roof; when the radiation intensity was monitored to change by ΔR (in W / m²) within Δt seconds, the following was observed. 2 When ΔR / Δt ≥ 100 W / m, record the temperature rise rate ΔT / Δt (°C / min); statistical analysis of 100 sets of data shows that: when ΔR / Δt ≥ 100 W / m 2At 1000 ℃, ΔT / Δt > 2℃ / min in 90% of cases.

[0090] Software verification:

[0091] Use the ANSYS Thermal module to create a roof panel model (material parameters are the same as in section 0035); set up transient analysis: apply solar radiation heat flux to the surface, and insulate the bottom surface; parametrically scan the rate of change of radiation (50-150W / m). 2 ·s), output the temperature rise rate curve; when the slope of the curve is greater than 2℃ / min, mark the corresponding radiation change rate as the critical value.

[0092] Equation Description: The differential equation is encapsulated in the ANSYS solver. Its physical essence is "Solar radiation energy absorbed by the material = internal heat accumulation + surface heat loss". The system only needs to call the software interface (command: / SOLVE), and no manual processing of the equation is required.

[0093] The partitioning operation for non-curvature abrupt change regions includes: first, excluding the coordinate range of curvature abrupt change regions from the roof global mesh; the remaining region is defined as a non-curvature abrupt change region. The current solar incidence angle change data stream is acquired, containing time-series angle change values ​​(in degrees). The change amplitude is calculated using a sliding window standard deviation algorithm: a dataset of incidence angle changes within a specific time window (e.g., 5 minutes) is taken, and the statistical standard deviation of this dataset is calculated as the change amplitude indicator. The preset angle change threshold is determined based on solar trajectory simulation. Specifically, a time-varying model of the solar altitude angle / azimuth angle for this geographical location is established, the theoretical maximum angular velocity value is calculated, and a specific proportion (e.g., 50%) is taken as the threshold benchmark. In engineering applications, the preset angle change threshold is set to a specific value (e.g., 0.3 degrees per second).

[0094] In practice, the preset angular change threshold (e.g., 0.3 degrees per second) is determined based on the project location's latitude and longitude and the date. The solar altitude angle / azimuth time-varying model refers to internationally accepted astronomical calculation methods (such as the ISO 19115 standard or the NRELSOLPOS algorithm). The theoretical maximum angular velocity value is obtained through the following executable steps: using standard astronomical calculation software, input the precise latitude and longitude of the project location and the target date (e.g., the summer solstice), calculate the numerical sequence of the solar azimuth angle per second from sunrise to sunset; calculate the absolute value of the angular change between adjacent seconds of data, and the maximum value in the sequence is the theoretical maximum angular velocity value. For example, the calculated maximum rate of change of the azimuth angle at noon on the summer solstice in Guangzhou is approximately 0.25 degrees per second. The preset angular change threshold is taken as 1.2 times this value (i.e., 0.3 degrees per second) as a safety margin.

[0095] The partitioning decision logic is as follows: Traverse each grid cell in the non-curvature abrupt region, extract the incident angle change data corresponding to that cell (matched according to the cell's geographical location), and calculate the change amplitude. When the change amplitude is greater than a preset angle change threshold, the grid cell is assigned to the real-time compensation zone; otherwise, it is assigned to the delayed compensation zone. In gently curved areas (e.g., the central roof), when a sudden change in wind speed causes the solar incident angle to change by a specific value (e.g., 0.5 degrees per second), the system automatically marks that area as a real-time compensation zone. The partitioning results are stored as a compensation state matrix, where rows correspond to grid row numbers, columns correspond to grid column numbers, and element values ​​of 0 represent delayed compensation zones and 1 represent real-time compensation zones. A bidirectional index is established between the state matrix and the roof spatial coordinates to support fast region lookups. Finally, the system outputs the coordinate sets of the real-time compensation zone and the delayed compensation zone for subsequent compensation operations.

[0096] The specific implementation process for obtaining the uncompensated first temperature sensor data stream and uncompensated second temperature sensor data stream corresponding to the real-time compensation zone is as follows: Based on the set of coordinate ranges of the real-time compensation zone generated in the previous step, data records with matching spatial locations are retrieved from the circular storage queue. Spatial matching adopts a grid index query mechanism, based on the three-dimensional grid coordinate system divided by the roof, to extract all data entries whose grid index numbers are within the real-time compensation zone. For each matching entry, its data type identifier code (first temperature sensor data stream identifier code or second temperature sensor data stream identifier code) is identified, and the uncompensated first temperature sensor data stream and uncompensated second temperature sensor data stream are separated. Uncompensated status verification is achieved by checking the status flag bit (status code 00). If a data block has been marked as compensated (status code 01), it is skipped. Integrity verification is performed during data extraction. When data loss at a specific grid point is detected (e.g., due to packet loss during transmission), a temporary data block is generated by spatial proximity interpolation. The temporary data block is marked as pending verification and a special identifier is attached.

[0097] Based on the timestamp differences between the new solar radiation intensity data stream, the new solar incidence angle change data stream, and the uncompensated temperature sensor data stream, the time-weighted compensation factor is calculated using the following operations: Let the timestamp of the uncompensated temperature data be Ttemp, the timestamp of the newly arrived solar radiation intensity data stream be Trad, and the timestamp of the solar incidence angle change data stream be Tang. The radiation time difference is calculated as |Ttemp - Trad|, and the incidence angle time difference is calculated as |Ttemp - Tang|. The calculation model for the time-weighted compensation factor is as follows: the radiation compensation weight equals 1 minus the quotient of the radiation time difference divided by the preset maximum effective time difference; the incidence angle compensation weight equals 1 minus the quotient of the incidence angle time difference divided by the preset maximum effective time difference. The maximum effective time difference is determined through thermodynamic analysis, specifically based on the material's thermal relaxation time constant (e.g., 30 seconds for steel). When the time difference exceeds the maximum effective time difference, the corresponding weight is reset to zero. The weight calculation results are stored in vector form, with the vector dimension matching the number of temperature data points. Each data point is associated with two weight values ​​(radiation weight and incident angle weight).

[0098] The dynamic compensation formula is used to perform batch compensation calculations on the uncompensated data streams of the first and second temperature sensors, respectively, as follows: The core structure of the dynamic compensation formula includes a temperature reference term and an environmental disturbance term. The temperature reference term corresponds to the original uncompensated temperature value, and the environmental disturbance term is composed of a linear superposition of a radiation compensation component and an incident angle compensation component. The specific calculation process is as follows: the compensated temperature value equals the original temperature value plus an adjustment amount. This adjustment amount equals the radiation compensation weight multiplied by the new solar radiation intensity value multiplied by the solar radiation coefficient, plus the incident angle compensation weight multiplied by the change in the new solar incident angle multiplied by the incident angle coefficient. The solar radiation coefficient is calibrated through material photothermal absorption experiments (e.g., 0.65 W / m² / degree Celsius for aluminum plates), and the incident angle coefficient is obtained through an angle-heat flux lookup table. During batch processing, three parallel calculations are performed for each uncompensated temperature data point: First, the radiation compensation component value is calculated as the radiation compensation weight at that point multiplied by the new solar radiation intensity value multiplied by the solar radiation coefficient. Second, the incident angle compensation component value is calculated as the incident angle compensation weight at that point multiplied by the new solar incident angle change value multiplied by the incident angle coefficient. Third, the final compensated temperature value is calculated as the original temperature value plus the radiation compensation component value plus the incident angle compensation component value. The compensation calculation results are stored as a new data block, which inherits the original spatial coordinate information and updates the status flag to compensated status (status code 01). An anomaly detection mechanism is set in batch processing. When the compensation value exceeds the material's allowable temperature range (e.g., -40 to 200 degrees Celsius for steel), boundary truncation is automatically triggered and an anomaly event log is recorded.

[0099] In practice, the calculation of radiation compensation weight and incident angle compensation weight relies solely on time difference data: the weight value equals 1 minus the difference between the temperature data timestamp and the corresponding meteorological parameter timestamp, divided by the temperature response lag time constant (see definition in section 0034). If the time difference exceeds this time constant, the weight is directly set to 0. For example, if the temperature data timestamp is 10:00:00, the solar radiation data timestamp is 10:00:05, and the time constant is 28 seconds, then the radiation compensation weight is 1 - (5 / 28) ≈ 0.82.

[0100] The operation of storing the uncompensated first temperature sensor data stream and the uncompensated second temperature sensor data stream corresponding to the delay compensation zone into the compensation queue includes: filtering relevant entries in the circular storage queue based on the coordinate range set of the delay compensation zone. The data extraction rules are the same as for the real-time compensation zone, but a priority control strategy is added. Priorities are arranged in reverse chronological order (the one with the latest time tag has the highest priority). The queue storage structure adopts a two-level index design, with the first-level index being the time segment and the second-level index being the spatial grid number. Data compression is performed during storage, and redundant spatial coordinate information is deleted (only the starting coordinates are retained for the same grid point).

[0101] During the reconstruction task interval, the asynchronous compensation operation for the dataset to be compensated is implemented as follows: The system continuously monitors the reconstruction process status of the temperature field. When a signal indicating completion of the reconstruction algorithm is detected (e.g., the GPU kernel returns a result code), an idle time slot timer is started. The minimum available duration of the idle time slot is determined based on the reconstruction cycle (e.g., if the reconstruction takes 20 seconds, the minimum idle time is set to 5 milliseconds). The asynchronous compensation scheduler checks the length of the queue to be compensated. If the queue is not empty, data block groups are extracted in priority order (the amount extracted at one time is determined by the length of the idle time slot). After extraction, the same dynamic compensation formula as real-time compensation is immediately called for calculation. The compensation result is stored in an independent storage area (distinct from the real-time compensation result) and marked as delayed compensation (status code 10). The compensation process implements a resource limiting strategy: dynamically allocating computing resources to no more than a specific proportion (e.g., 30%) of the total system resources to prevent impact on real-time reconstruction performance.

[0102] The specific operations for receiving the real-time compensated first and second temperature sensor data streams include: reading data blocks marked as compensation complete from the compensation result storage area. These data blocks contain the real-time compensated first and second temperature sensor data streams generated in the previous steps. The receiving end is configured with a dual-buffer mechanism: while the reconstruction algorithm processes the current data, the next batch of data is simultaneously written to the preparatory buffer, achieving pipelined operation. The data transmission interface uses a direct memory access engine, which automatically identifies the data block header identifier (first or second temperature sensor data stream identifier) ​​and stores it in different memory partitions according to data type. The memory partitions are configured as high-speed cache memory areas with access speeds reaching a specific level (e.g., 6GB / s bandwidth). During reception, data integrity checks are performed, calculating the cyclic redundancy check value for each data block and comparing it with the header checksum. If an error is detected, the source end is immediately requested to retransmit the missing data block. For rooftops with extremely large spans (e.g., exceeding 300 meters), distributed receiving nodes are set up, with nodes synchronously transmitting status via a fiber optic data bus.

[0103] The implementation process of the radial basis function interpolation algorithm based on adaptive kernel functions includes: the core of the algorithm adopts an improved radial basis function model, which includes two parts: a basic function and an adaptive adjustment mechanism. The basic function is a Gaussian kernel function, whose characteristic parameters include two adjustable variables: support radius and shape parameter. The interpolation calculation principle is: each sensor temperature data point is regarded as the center point of the spatial basis function, and a continuous temperature field is constructed by superimposing basis functions. The specific implementation steps are divided into three stages: initialization, matrix construction, and equation solving. In the initialization stage, the roof spatial grid structure definition is loaded (see step S2 for grid generation), and the mapping relationship between grid points and sensor data points is established. In the matrix construction stage, the interpolation system matrix is ​​created. The matrix element calculation rule is: the distance value between grid points and sensor points is input into the kernel function to generate influence coefficients. In the equation solving stage, the preprocessed conjugate gradient method is used to solve large sparse linear equation systems. The iteration termination condition is set to the residual norm being less than a specific threshold (e.g., 10 to the power of -6). Algorithm optimization strategies include: excluding areas with low temperature differences during initialization (areas with temperature differences less than a set value, such as 1 degree Celsius, are directly assigned nearest neighbor values), and using a spatial tree index to accelerate distance retrieval. In the airport roof application (500,000 grid points), the algorithm initialization phase time is controlled within a specific time (e.g., 700 milliseconds).

[0104] The method for dynamically adjusting kernel function parameters based on the roof curvature change rate during interpolation calculation is as follows: Curvature data for the current processing area is retrieved from a pre-stored curvature distribution database (see the S3 step example). The kernel function parameter adjustment strategy is: the larger the curvature change rate, the smaller the support radius value, and the larger the shape parameter value. Specific conversion rules are determined through experimental calibration curves: a mapping table between the curvature change rate and the optimal parameter set is established. For every specific increase in the curvature change rate (e.g., 0.1 per meter), the support radius is reduced by a specific percentage (e.g., 5%), and the shape parameter is increased by a specific percentage (e.g., 8%). The dynamic parameter loading mechanism uses a region grouping method, dividing the roof into multiple processing blocks (approximately 100×100 grid size) based on curvature characteristics. Each block uses a unified parameter set. A transition buffer is set during parameter switching to prevent step errors caused by abrupt boundary changes.

[0105] In practice, the curvature distribution database is constructed during the system deployment phase according to the following process: Three-dimensional surface data of the roof are extracted from the project's BIM model; standard curvature calculation tools in computer-aided design software (such as Rhino's "CurvatureAnalysis" command) are used to calculate the Gaussian curvature values ​​of densely sampled points on the roof (e.g., one point every 0.5 meters); the coordinates and curvature values ​​of points with absolute curvature values ​​greater than a set threshold (e.g., 0.05) are stored in the spatial database. For example, the ridge area of ​​a stadium roof is automatically identified as a high-curvature area and its coordinate boundaries are stored.

[0106] Curvature distribution database construction process:

[0107] Data input: Export the roof BIM model (.rvt format) from Revit and convert it into a .3dm file that can be read by Rhino;

[0108] Automated processing: Execute the script in Rhino:

[0109]

[0110]

[0111] Region aggregation: Using Grasshopper's "Cluster Points" component (cluster radius 2m), 15 mutation regions were generated;

[0112] Boundary generation: Execute the "Convex Hull" command for each point set to output polygon vertices (e.g., ridge area: [(12.3,45.6),(12.5,45.8)...]);

[0113] Database storage: Vertex coordinates are stored in the PostGIS database, and an R-tree spatial index is created.

[0114] The procedure for prioritizing increased interpolation node density in high curvature variation regions is as follows: High curvature variation regions are defined as those where the rate of curvature change exceeds a set threshold (e.g., 0.15 degrees Celsius per meter). The node densification strategy employs a quadtree subdivision algorithm: first, the boundary of the high curvature region is detected, and then recursive subdivision is performed within the boundary. Specifically, the temperature gradient value in the original mesh cell is calculated. If the gradient value exceeds a set temperature gradient threshold (e.g., 3 degrees Celsius per meter) and is located in a high curvature region, the cell is subdivided into four sub-cells. The temperature value of newly added nodes is calculated using a cubic spline interpolation algorithm, with the spline control points selected from the surrounding 9 original nodes. The densification depth is linearly controlled by the rate of curvature change; for every specific increase in the rate of curvature change (e.g., 0.05 degrees Celsius per meter), the recursive depth increases by one level (the maximum depth is limited to a specific value, e.g., 3 levels).

[0115] The implementation of interpolation operations using a parallel computing architecture includes: a CPU-GPU heterogeneous computing model, where the CPU handles logic control and data preprocessing, and the GPU performs the core computational tasks. The computational task decomposition strategy employs spatial domain decomposition: the roof is divided into multiple computational unit blocks (e.g., 64 blocks), each assigned to an independent GPU core. Two levels of parallelism are implemented within the GPU core: first-level block-level parallelism (coarse-grained), and second-level grid-point parallelism (fine-grained). Memory management uses zero-copy technology, directly transferring data from the host fixed memory to the GPU's video memory. The core computing kernel is configured with 2048 concurrent threads, each responsible for interpolation calculations of a specific number (e.g., 32) grid points. A paged memory barrier synchronization mechanism ensures data consistency at block boundaries. Under typical hardware configurations (e.g., NVIDIA V100 GPU), the computation time for 100,000 interpolation points is reduced to a specific time (e.g., 80 milliseconds), achieving a speedup of a specific level (e.g., 11x).

[0116] Generating continuous temperature field distribution data covering the entire roof involves: the final output data format is designed as a four-dimensional array structure, with the first three dimensions representing spatial coordinate indices (X, Y, and Z axis grid indices), and the fourth dimension storing the temperature values. Spatial continuity is achieved using cubic spline surface fitting, with the interpolation results used as control points to generate a continuous surface. Data validation is performed by randomly selecting a specific percentage (e.g., 5%) of the roof surface and comparing the interpolated temperature with the real-time infrared scan temperature. If the deviation exceeds a specific threshold (e.g., ±1.5 degrees Celsius), abnormal areas are automatically marked and a local recalculation is triggered. The resulting data is output to a temperature field database, which supports spatial range queries and time series backtracking. The final generated roof temperature field distribution data format conforms to the ISO geospatial data exchange standard, achieving seamless integration with building information systems.

[0117] The specific steps for extracting historical temperature gradient distribution data from the previously generated roof temperature field distribution are as follows: Retrieve the roof temperature field distribution dataset with the time label Tprev from the historical temperature field database. This dataset contains the 3D coordinates and temperature values ​​of all grid points on the roof. The temperature gradient is calculated using the central difference method: for any grid point P(x,y), its X-direction temperature gradient value is the temperature of the adjacent point to the right minus the temperature of the adjacent point to the left divided by twice the X-direction grid spacing; the Y-direction temperature gradient value is the temperature of the adjacent point above minus the temperature of the adjacent point below divided by twice the Y-direction grid spacing; the combined gradient value is the Euclidean norm of the X-direction and Y-direction gradients. The gradient calculation result for each grid point is stored in a two-dimensional gradient matrix, with the matrix row and column indices corresponding to the roof grid coordinate system. Data integrity verification is performed during data extraction: when a boundary point is found to be missing adjacent points, a virtual mirror point algorithm is automatically activated for boundary processing. The virtual point temperature value is generated by linear extrapolation from the three nearest actual points. The historical temperature gradient distribution data storage structure includes three key fields: spatial coordinate index, gradient value, and calculation time label.

[0118] The process for calculating the temperature gradient change at the same spatial coordinates in the currently generated roof temperature field distribution is as follows: First, align the time series by taking the previous generation time Tprev and the current generation time Tcurrent, ensuring the time interval meets preset requirements (e.g., 1-5 minutes). Spatial coordinate alignment is achieved through grid index matching: the current temperature field distribution data is mapped according to the same roof grid coordinate system. The temperature gradient change is defined as: the current temperature gradient value minus the previous temperature gradient value. Specifically, the change equals Gcurrent minus Gprev, where Gcurrent is the value in the current gradient matrix, and Gprev is the value of the same grid index in the historical gradient matrix. The result of the change calculation forms a new matrix whose spatial dimensions are completely consistent with the roof grid, with each element representing the gradient change at the corresponding grid point. In gradient abrupt change regions (such as the ridge), an additional secondary gradient change is calculated: the rate of change of the gradient change, which is stored as an auxiliary judgment factor in an additional data layer.

[0119] The comparison of temperature gradient change with a preset gradient threshold involves the following steps: The preset gradient threshold is derived from a material thermal stress model, specifically by establishing a mechanical relationship between the thermal expansion coefficient (e.g., 11 × 10⁻⁶ per degree Celsius), elastic modulus (210 GPa), and temperature gradient of the roof steel. The critical temperature gradient value corresponding to the critical thermal stress is determined through numerical simulation (e.g., the gradient threshold corresponding to 10 MPa stress in a finite element model), and the final threshold is set as a specific percentage (e.g., 80%) of the critical value. The comparison process employs matrix parallel computation technology: the gradient change matrix and the constant value threshold matrix are compared element-wise, and the comparison result generates a Boolean label matrix. A true value in the label matrix indicates exceeding the threshold, and a false value indicates normal operation. In engineering applications, the preset gradient threshold is dynamically adjusted according to the ambient temperature range: when the ambient temperature is below a specific value (e.g., 10 degrees Celsius), the threshold is reduced by a specific percentage (e.g., 20%) to address the risk of increased steel brittleness.

[0120] In practice, the preset gradient threshold (e.g., 2.15 Kelvin per meter) is determined based on the material's mechanical properties. The mechanical formula refers to the standard method used in structural engineering to calculate thermal stress: thermal stress equals the elastic modulus multiplied by the coefficient of thermal expansion multiplied by the temperature difference. The critical temperature gradient value is determined through the following steps: First, set the allowable safe thermal stress value for the material (e.g., 10 MPa for steel); based on the material parameters (elastic modulus 210 GPa, coefficient of thermal expansion 0.000011 Kelvin), calculate the maximum allowable local temperature difference; then divide by the temperature sensor spacing (e.g., 2 meters) to obtain the critical gradient value. For example: maximum allowable temperature difference = 10 MPa / (210 GPa × 0.000011) ≈ 4.3 Kelvin, critical gradient = 4.3 Kelvin / 2 meters = 2.15 Kelvin per meter. The preset gradient threshold is taken as 80% of this critical value (i.e., 1.72 Kelvin per meter).

[0121] When the temperature gradient change exceeds a preset gradient threshold, locating the coordinates of the corresponding roof curvature abrupt change region involves: first, filtering all ground truth elements in the Boolean marker matrix to obtain a set of grid coordinates exceeding the threshold. Coordinate space association uses a surface nearest neighbor algorithm: using each threshold point as the center, a curvature abrupt change region within a specific radius (e.g., 2 meters) is searched. Curvature abrupt change region coordinate data is retrieved from a pre-loaded spatial database (see step S3), which stores the vertex coordinate sequence of the polygons representing roof curvature abrupt change regions. The location method is as follows: calculate the minimum distance from each threshold point to each curvature abrupt change region polygon; if the minimum distance is less than a preset relevant distance threshold (e.g., 0.5 meters), the point is determined to belong to the curvature abrupt change region. The relevant distance threshold is set according to the thermal conductivity of the roof material: the higher the thermal conductivity of the material, the higher the relevant distance threshold. The location result is output as a list of region codes, with each code corresponding to the center coordinates and boundary range of a curvature abrupt change region.

[0122] The local recompensation command generated based on the positioning results includes: a machine-readable command data structure consisting of a command header, payload, and checksum. The payload defines the spatial coordinate range to be recompensated: for each located curvature abrupt change region, its polygon vertex coordinate sequence is obtained, and the outer rectangular boundary is calculated. The rectangular boundary is defined as: minimum X-coordinate, maximum X-coordinate, minimum Y-coordinate, and maximum Y-coordinate. This coordinate range is encapsulated as a spatial range object and appended to the command payload. Immediately after command generation, it is sent to a distributed message queue, where a routing mechanism automatically allocates it to the corresponding processing node based on the spatial coordinates.

[0123] The command priority setting strategy is as follows: regions with larger gradient changes have higher priority, with priority levels ranging from 0 to 3 (level 0 being the highest). The priority value is appended to the command header as metadata. During continuous operation, the system maintains a command execution status tracking table and triggers a timeout retransmission mechanism for commands that do not respond in a timely manner. The timeout threshold is set to a specific time (e.g., 30 seconds), and the maximum number of retries is limited to 3. The local recompensation command is ultimately delivered for compensation execution, providing precise spatial positioning guidance for the reacquisition and processing of local area temperature data. The effectiveness of the commands is verified through infrared imaging; test results show that the command positioning accuracy reaches a specific level (e.g., spatial deviation less than 0.3 meters).

[0124] Example 2: Figure 2 A schematic diagram of a temperature field reconstruction system for a large-span roof according to the present invention is provided. The temperature field reconstruction system for a large-span roof includes the following modules:

[0125] The synchronous acquisition module is used to synchronously acquire data from the roof distributed temperature sensor and micro-meteorological parameters, and to add a unified timestamp to the temperature sensor data stream and the micro-meteorological parameter stream.

[0126] The cache extraction module is used to establish a compensation cache area. When a new micro-meteorological parameter is detected, it extracts the uncompensated temperature sensor data within a preset time window.

[0127] The zoning decision module is used to divide the roof area into a real-time compensation zone and a delayed compensation zone based on the comparison results between the rate of change of solar radiation intensity in the new micro-meteorological parameters and the preset threshold.

[0128] The dynamic compensation module is used to perform dynamic compensation on the uncompensated temperature sensor data in the real-time compensation zone to generate real-time compensated data. At the same time, the uncompensated temperature sensor data in the delayed compensation zone is marked as the dataset to be compensated, and asynchronous compensation is performed on the dataset to be compensated during the temperature field reconstruction interval.

[0129] The field reconstruction module is used to input the real-time compensated data into a preset reconstruction algorithm to generate the roof temperature field distribution at the current moment.

[0130] The verification and positioning module is used to verify the temporal continuity of the reconstruction results based on the gradient change between the previously generated roof temperature field distribution and the currently generated roof temperature field distribution.

[0131] The calculations involved in the embodiments are all dimensionless numerical calculations, and the preset parameters and thresholds in the calculations are set by those skilled in the art according to the actual situation.

[0132] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, in the form of a computer program product.

[0133] Those skilled in the art will recognize that the modules and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and inventive constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0134] In addition, the functional modules in the various embodiments of this application can be integrated into one processing module, or each module can exist physically separately, or two or more modules can be integrated into one module.

[0135] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of modules is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple modules or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or modules may be electrical, mechanical, or other forms.

[0136] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0137] In conclusion, the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for reconstructing temperature field of a large-span roof, characterized in that, The method comprises the following steps: S1, synchronously collecting roof distributed temperature sensor data and microclimate parameters, and marking the temperature sensor data stream and the microclimate parameter stream with a unified timestamp; S2, establishing a compensation buffer area, and extracting uncompensated temperature sensor data within a preset time window when detecting arrival of new microclimate parameters; S3, dividing the roof area into a real-time compensation area and a delayed compensation area according to a comparison result of a solar radiation intensity change rate in the new microclimate parameters and a preset threshold value; S4, performing a dynamic compensation operation on the uncompensated temperature sensor data in the real-time compensation area to generate real-time compensated data, and marking the uncompensated temperature sensor data in the delayed compensation area as a to-be-compensated data set, and performing an asynchronous compensation operation on the to-be-compensated data set during a temperature field reconstruction gap period; S5, inputting the real-time compensated data into a preset reconstruction algorithm to generate a current roof temperature field distribution; S6, performing time continuity verification on the reconstruction result based on a gradient change amount of a previously generated roof temperature field distribution and a currently generated roof temperature field distribution.

2. The temperature field reconstruction method for long-span roofs according to claim 1, characterized in that, Synchronously collecting roof distributed temperature sensor data and microclimate parameters, and marking the temperature sensor data stream and the microclimate parameter stream with a unified timestamp, comprising: Obtaining a first temperature sensor data stream through a distributed optical fiber sensor, obtaining a second temperature sensor data stream through an infrared thermal imager, and obtaining a solar radiation intensity data stream and a solar incident angle change amount data stream through a microclimate station; Marking the first temperature sensor data stream of the distributed optical fiber sensor, the second temperature sensor data stream of the infrared thermal imager, and the solar radiation intensity data stream and the solar incident angle change amount data stream of the microclimate station with a unified timestamp through a time synchronization server.

3. The method of claim 2, wherein, Establishing a compensation buffer area, and extracting uncompensated temperature sensor data within a preset time window when detecting arrival of new microclimate parameters, comprising: Configuring a ring-shaped storage queue to cache the first temperature sensor data stream and the second temperature sensor data stream with a unified timestamp; When detecting arrival of new solar radiation intensity data stream or new solar incident angle change amount data stream, calculating a preset time window based on a corresponding timestamp of the corresponding new microclimate parameter, and extracting all uncompensated first temperature sensor data stream and uncompensated second temperature sensor data stream within the corresponding time window as a to-be-processed data set.

4. The method of claim 3, wherein, Dividing the roof area into a real-time compensation area and a delayed compensation area according to a comparison result of a solar radiation intensity change rate in the new microclimate parameters and a preset threshold value, comprising: Preloading spatial coordinate distribution of a roof curvature mutation area; Calculating a change rate value of the new solar radiation intensity data stream; When the change rate value exceeds the preset threshold value, all coordinate ranges of the curvature mutation area are divided into the real-time compensation area; For a non-curvature mutation area, according to a comparison result of a change amplitude of the solar incident angle change amount data stream and a preset angle change threshold value, a sub-area with a change amplitude exceeding the preset angle change threshold value is divided into the real-time compensation area, and the remaining area is divided into the delayed compensation area.

5. The method of claim 4, wherein, Performing a dynamic compensation operation on the uncompensated temperature sensor data in the real-time compensation area to generate real-time compensated data, comprising: acquire the uncompensated first temperature sensor data stream and the uncompensated second temperature sensor data stream corresponding to the real-time compensation zone; calculate the time-weighted compensation factor based on the time stamp difference between the new solar radiation intensity data stream and the new solar incident angle change data stream and the uncompensated temperature sensor data stream; perform batch compensation operation on the uncompensated first temperature sensor data stream and the uncompensated second temperature sensor data stream respectively by using the dynamic compensation formula to generate real-time compensated data, including real-time compensated first temperature sensor data stream and real-time compensated second temperature sensor data stream.

6. The temperature field reconstruction method for long-span roofs according to claim 5, characterized in that, mark the uncompensated temperature sensor data of the delay compensation zone as a to-be-compensated data set, including: store the uncompensated first temperature sensor data stream and the uncompensated second temperature sensor data stream corresponding to the delay compensation zone into the to-be-compensated queue.

7. The method according to claim 6, wherein, perform asynchronous compensation operation on the to-be-compensated data set during the temperature field reconstruction gap period, including: extract data from the to-be-compensated queue during the reconstruction task interval period and perform compensation operation by using the same dynamic compensation formula.

8. The temperature field reconstruction method for long-span roofs according to claim 7, characterized in that, input the real-time compensated data into the preset reconstruction algorithm to generate the roof temperature field distribution at the current time, including: receive the real-time compensated first temperature sensor data stream and the real-time compensated second temperature sensor data stream; process the real-time compensated first temperature sensor data stream and the real-time compensated second temperature sensor data stream based on the adaptive kernel function radial basis interpolation algorithm; during the interpolation calculation process, dynamically adjust the kernel function parameters according to the roof curvature change rate to preferentially increase the interpolation node density in the high curvature change area; perform interpolation operation through a parallel computing architecture to generate continuous temperature field distribution data covering the entire roof as the roof temperature field distribution at the current time.

9. The temperature field reconstruction method for long-span roofs according to claim 8, characterized in that, perform time continuity verification on the reconstruction result based on the gradient change amount of the previously generated roof temperature field distribution and the currently generated roof temperature field distribution, including: extract historical temperature gradient distribution data from the previously generated roof temperature field distribution; calculate the temperature gradient change amount of the same spatial coordinate point in the currently generated roof temperature field distribution; compare the temperature gradient change amount with the preset gradient threshold value; when the temperature gradient change amount exceeds the preset gradient threshold value, locate the corresponding roof curvature sudden change area coordinates; generate a local re-compensation instruction according to the positioning result, and the local re-compensation instruction contains the spatial coordinate range that needs to be re-compensated.

10. A temperature field reconstruction system for long-span roofs for implementing the method of any one of claims 1 to 9, characterized by include the following modules: a synchronous acquisition module for synchronously acquiring roof distributed temperature sensor data and microclimate parameters, and applying a unified time stamp to the temperature sensor data stream and the microclimate parameter stream; a cache extraction module for establishing a compensation cache area, and extracting uncompensated temperature sensor data within a preset time window when new microclimate parameters are detected; a partition decision module for dividing the roof area into a real-time compensation zone and a delay compensation zone according to the comparison result of the solar radiation intensity change rate in the new microclimate parameters and the preset threshold value; The dynamic compensation module is configured to perform a dynamic compensation operation on the uncompensated temperature sensor data in the real-time compensation area to generate real-time compensated data, mark the uncompensated temperature sensor data in the delay compensation area as a to-be-compensated data set, and perform an asynchronous compensation operation on the to-be-compensated data set during a temperature field reconstruction gap period; The field reconstruction module is configured to input the real-time compensated data into a preset reconstruction algorithm to generate a roof temperature field distribution at a current time point; The verification and positioning module is configured to perform a time continuity verification on the reconstruction result based on a gradient change amount of the roof temperature field distribution generated last time and the roof temperature field distribution generated currently.

Images