Energy-saving control method based on building group spatial form

Abstract

The invention discloses an energy-saving control method based on a building group spatial form, which relates to the technical field of energy-saving control, and comprises the following steps: constructing an airflow track analysis region based on a region with hot pressure difference direction change in a spatial pressure covering sensing matrix, and collecting continuous wind direction and wind speed data; judging whether the airflow direction is abnormal or not according to the change of the flow direction consistency entropy; performing spatial coincidence processing on the spatial pressure shielding sensing matrix and the airflow direction abnormal region, and generating an air outlet ventilation and heat extraction efficiency evaluation index based on the heat conduction rate in the coincidence region and the stability of an airflow output path; and on the basis of the ventilation and heat removal efficiency evaluation index of the exhaust outlet, an exhaust response condition matrix containing a space pressure shielding parameter, an airflow disturbance level and a heat retention level is constructed, and a ventilation control judgment sequence is generated. According to the method, the problem of air exhaust failure under pressure shielding interference is solved, ventilation efficiency accurate judgment and energy-saving regulation and control optimization are achieved, and the ventilation stability and the energy utilization efficiency of the building group are guaranteed.

Description

Technical Field

[0001] The present invention relates to the technical field of energy-saving control, and in particular to an energy-saving control method based on the spatial form of a building complex. Background Art

[0002] Energy-saving control based on the spatial form of a building complex refers to a method that analyzes and models the spatial morphological characteristics of individual buildings within the complex, such as their layout, orientation, height differences, shading, air duct accessibility, and sunlight conditions. This method then combines environmental and energy usage data to develop energy-saving control strategies tailored to the overall complex's operating environment. This control technology emphasizes the impact of the overall building complex on energy consumption distribution, rather than controlling individual buildings in isolation. Existing energy-saving control technologies based on the spatial form of building complexes are typically implemented through the following steps: First, spatial structural data of the building complex is acquired using methods such as BIM (Building Information Modeling), GIS (Geographic Information System), laser scanning, or remote sensing imagery, and a three-dimensional spatial model is constructed. Second, real-time environmental data collected by sensors (such as light intensity, temperature, wind speed, and occupant density) is combined to analyze the thermal environment, ventilation paths, and lighting conditions of each area within the building complex. Energy consumption prediction models and optimization algorithms are then used to evaluate the impact of different spatial forms on energy use (such as air conditioning systems, lighting systems, and shading systems). Finally, based on the analysis results, control strategies adapted to specific spatial conditions are generated, and intelligent equipment scheduling and energy consumption optimization control are implemented through building automation systems (BAS) or energy management systems (EMS). Overall, this approach achieves a closed-loop management system from "spatial cognition" to "strategy formulation" to "energy consumption execution," with the advantages of improving the energy efficiency of building complexes, reducing overall carbon emissions, and enhancing the level of intelligent operation and maintenance.

[0003] The existing technology has the following deficiencies: In the energy-saving control of building spatial form, natural ventilation is often achieved through high-rise vents based on the principle of thermal pressure differential to reduce air conditioning load and improve ventilation efficiency. However, when a building's roof is adjacent to a taller adjacent building mass, the spatial pressure-shading structure alters the original thermal pressure differential distribution, causing the direction of thermal airflow to reverse. Under certain meteorological conditions, this can lead to the abnormal phenomenon of hot air flowing inward from the outside. In this case, although the vents are open, they no longer effectively dissipate heat and instead exacerbate heat accumulation in the top floor. Existing energy-saving control technologies based on building spatial form fail to determine whether high-rise vents are still effective in ventilation and heat removal based on the airflow anomaly caused by the change in thermal pressure differential direction under the influence of pressure-shading. Their control logic still interprets the thermal pressure response as effective ventilation, resulting in the continuous erroneous opening of vents, increasing cooling loads, wasting energy, and further disrupting the air pressure balance within the building. This can lead to serious consequences such as temperature rise in the top floor, heat recirculation in the middle and lower floors, odor diffusion, and smoke backflow.

[0004] The above information disclosed in this Background section is only for enhancement of understanding of the background of the present disclosure and therefore it may contain information that does not form the prior art that is already known to a person of ordinary skill in the art. Summary of the Invention

[0005] The purpose of the present invention is to provide an energy-saving control method based on the spatial form of a building complex to solve the problems in the above-mentioned background technology.

[0006] In order to achieve the above-mentioned object, the present invention provides the following technical solution: an energy-saving control method based on the spatial form of a building complex, specifically comprising the following steps: S1. Obtain the three-dimensional spatial block structure information and real-time meteorological parameters of the building complex, establish a spatial shading perception matrix, extract the spatial closure angle relationship between the building blocks, and determine whether there is a shading interference condition that causes a change in the direction of the thermal pressure difference; S2. Based on the areas with thermal pressure difference direction changes in the spatial shading pressure perception matrix, an airflow trajectory analysis area is constructed, and continuous wind direction and wind speed data are collected. Changes in the flow direction consistency entropy value are used to determine whether there is an abnormal airflow direction. S3. Spatially overlap the spatial pressure sensing matrix with the area with abnormal airflow direction, and generate an exhaust vent ventilation and heat removal efficiency evaluation index based on the heat conduction rate and the stability of the airflow output path in the overlapped area; S4. Based on the exhaust vent ventilation and heat removal efficiency evaluation index, an exhaust response condition matrix including space shielding pressure parameters, airflow disturbance level, and heat retention level is constructed to generate a ventilation control decision sequence; S5. According to the control instructions in the ventilation control determination sequence, the exhaust vent opening state is dynamically regulated, and corresponding ventilation behaviors or alternative energy-saving strategies are triggered according to different states. At the same time, the ventilation control data is recorded in the spatial interference dataset.

[0007] Preferably, S1 is specifically: Obtaining the three-dimensional spatial block structure information and real-time meteorological parameters of the building complex, where the three-dimensional spatial block structure information includes the height data, boundary outline data, and facade orientation data of the building blocks, and the real-time meteorological parameters include the solar azimuth angle, solar altitude angle, and wind direction data; Based on the three-dimensional spatial block structure information of the building complex, a block ray projection model is constructed. A ray tracing algorithm is used to simulate the shadow superposition area of ​​each block under the current solar azimuth angle, and to form the spatial occlusion relationship between the blocks. The spatial closure angle relationship between building blocks is extracted based on the spatial occlusion relationship between the blocks. The angle integral method is used to calculate the sum of the visible angles of each block in multiple directions and generate a spatial closure angle distribution matrix. A coupling analysis is performed based on the spatial closure angle distribution matrix and the wind direction change trend in the real-time meteorological parameters to determine whether spatial shading causes a change in the direction of the thermal pressure difference. The judgment method is: when the spatial closure angle change rate is greater than the preset change threshold, and the wind direction and the maximum shading angle of the closed area are less than the set interference angle threshold, it is marked as a shading interference condition.

[0008] Preferably, S2 specifically includes the following steps: S201. Based on the areas where the thermal pressure difference direction changes in the spatial shading pressure perception matrix, taking the center of the thermal pressure difference gradient change in the area as the starting point, and combining the relative height difference and orientation relationship between building blocks, construct an airflow trajectory analysis area including multiple horizontal and vertical sections; S202: Arrange wind direction and wind speed collection units along the prevailing wind direction in the airflow trajectory analysis area to collect wind direction and wind speed data over multiple consecutive time periods. Map the data into an angle sequence on a unit circle based on the wind direction change trend, and establish a wind direction time series distribution model. S203. Calculate the flow direction consistency entropy value in each time period based on the wind direction time series distribution model. When the rate of change of the flow direction consistency entropy value in consecutive time periods is greater than the set disturbance judgment threshold, and the reversal angle of the dominant wind direction in the area exceeds the preset offset angle threshold, it is judged that there is an airflow direction anomaly in the airflow trajectory analysis area.

[0009] Preferably, in S203, the flow direction consistency entropy value in each time period is calculated based on the wind direction time series distribution model, specifically: The wind direction data in each time period in the wind direction time series distribution model is divided into a fixed number of wind direction intervals according to the unit angle; Count the frequency of wind direction data in each wind direction interval and form a probability distribution function after normalization; The Shannon entropy calculation formula is used to calculate the flow direction consistency entropy value based on the probability distribution function to characterize the degree of discreteness of the wind direction distribution within the time period.

[0010] Preferably, S3 specifically includes the following steps: The spatial overlap processing is performed on the location areas with changes in the direction of thermal pressure difference and the areas with abnormal airflow direction in the spatial shielding pressure perception matrix. The two areas are voxel-decomposed using a three-dimensional Euclidean grid. Spatial alignment is performed based on the geometric center of each voxel in a unified coordinate system. The intersection area where the voxel coordinates coincide is extracted to construct a set of thermal-pressure disturbance coupling spatial units. Within a set of coupled thermal-pressure disturbance spatial units, the temperature gradient between adjacent units is calculated based on temperature sensor data, and the heat conduction rate of each unit is calculated using the Fourier heat conduction formula. An inter-unit airflow output path diagram is constructed based on the wind speed direction sequence, and the path continuity coefficient and velocity stability factor are extracted to construct a joint thermal-flow stability parameter set. The heat conduction rate and the airflow output path stability factor are normalized, and dynamic weight factors are assigned to highlight the local sensitivity of the shading area. The local efficiency value of each spatial unit is generated based on the weighted scoring model, and the local efficiency values ​​of all thermal-pressure disturbance-coupled spatial units are summarized to form the exhaust vent ventilation and heat exhaust efficiency evaluation index.

[0011] Preferably, S4 is specifically: S401. Normalize and segment the exhaust vent ventilation and heat removal efficiency evaluation index, extract the corresponding spatial shielding pressure parameters, airflow disturbance level, and heat retention level based on the spatial unit position correlation, and assign them to three-dimensional label vectors to form an initial response factor set. S402. Establish a multidimensional matrix mapping structure based on the combination relationship of the three types of grade indicators in the initial response factor set, and construct an exhaust response condition matrix including space shielding pressure parameters, airflow disturbance level, and heat retention level. Each element in the matrix corresponds to a unique grade combination configuration. S403. Relying on the matching rules between each level combination in the exhaust response condition matrix and the exhaust control behavior, a ventilation control determination sequence is generated according to the preset control threshold mapping logic. The ventilation behavior control instructions output by the ventilation control determination sequence are used for subsequent exhaust control execution.

[0012] Preferably, S401 specifically includes: The ventilation and heat removal efficiency evaluation index of the exhaust vent is normalized to the minimum and maximum values ​​within the set upper and lower limits, and divided into multiple discrete grade segments based on the equal spacing method, and mapped to each spatial unit in the building complex; The spatial shielding parameters are calculated based on the relative height, closure angle, and azimuth angle between the spatial unit and the source of the shielding. The airflow disturbance level is generated by combining the wind direction reversal probability and the rate of change of the flow direction consistency entropy value. The heat retention level is calculated by combining the heat conduction rate and the duration of heat accumulation. The spatial shielding parameters, airflow disturbance level and heat retention level are assigned to the respective dimensional components of the three-dimensional label vector, and are combined and stored according to the spatial unit coordinate index to form an initial response factor set.

[0013] Preferably, S402 specifically includes: The spatial shielding parameter level set, airflow disturbance level set, and heat retention level set are mapped into independent one-dimensional integer index axes, and an axial index system for constructing a three-dimensional matrix mapping structure is defined. The spatial shielding parameter level, airflow disturbance level and heat retention level of each response factor in the initial response factor set are input into the three-dimensional matrix structure as three-dimensional coordinates, and the corresponding coordinate unit is assigned a unique identification value for the response configuration index; Traverse all three-dimensional coordinate combinations to construct an exhaust response condition matrix, and explicitly map each cell element in the exhaust response condition matrix to a unique combination of spatial shielding parameter level, airflow disturbance level, and heat retention level for the subsequent generation and retrieval of ventilation control judgment sequences.

[0014] Preferably, S5 is specifically: The exhaust control execution unit is driven according to the control instructions in the ventilation control judgment sequence. The opening angle and opening duration of the exhaust vents are dynamically adjusted using the opening adjustment mechanism. An exhaust vent opening state mapping model covering the building space area is established to represent the ventilation capacity boundary of each exhaust vent in real time. Based on the exhaust vent opening state mapping model and the status field of the ventilation control judgment sequence, a linkage analysis is performed to trigger the ventilation behavior instructions that match the current state. When the current state is marked as a high risk level of thermal pressure disturbance in the judgment sequence, an alternative energy-saving strategy is switched to execute, including switching to a central exhaust system, activating a heat exchange recovery mechanism, or delaying the exhaust response strategy. Key data such as the execution time of ventilation behavior or alternative energy-saving strategies, changes in exhaust vent opening, response type, and ventilation control judgment sequence number are written into the spatial interference dataset in a unified structure format, and the interference type index is marked with the exhaust vent spatial coordinates as the primary key for subsequent traceability modeling and optimization learning of ventilation control behavior.

[0015] In the above technical solution, the technical effects and advantages provided by the present invention are: 1. By introducing a spatial pressure-shading perception matrix and an airflow trajectory analysis mechanism, the present invention can accurately identify changes in the direction of thermal pressure differences and airflow reversals under pressure-shading interference conditions, effectively avoiding the problem of traditional exhaust control systems misjudging ventilation status. Through the combined identification method of the wind direction time-series distribution model and the flow direction consistency entropy, the system can monitor and determine in real time whether there is a risk of heat exhaust failure at high-rise exhaust vents in the building complex, ensuring that ventilation control behavior is based on the actual state of thermal pressure and airflow, improving the accuracy and dynamics of the thermal response mechanism, preventing phenomena such as increased cooling load, local heat accumulation, and airflow backflow caused by hot air backflow, and ensuring the thermal stability and air quality safety of the indoor environment.

[0016] 2. The present invention further constructs an exhaust response condition matrix that integrates spatial shielding parameters, airflow disturbance levels, and heat retention levels, forming a ventilation control judgment logic based on multi-parameter collaboration, which significantly improves the accuracy and adaptability of exhaust regulation. The system uses the exhaust outlet ventilation and heat exhaust efficiency evaluation index as a key indicator. On the basis of dynamic perception of local ventilation effectiveness, it links the alternative energy-saving strategy execution mechanism with the update and storage of the spatial interference data set to achieve closed-loop management and intelligent optimization of ventilation behavior. Compared with the existing control method that relies solely on a single parameter such as thermal pressure or wind speed, this method has the ability to predict abnormal airflow conditions and adaptively match exhaust strategies. It can be widely used in ventilation energy efficiency management and energy-saving control scenarios in high-density urban buildings. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, a brief introduction to the drawings required for use in the embodiments will be given below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For ordinary technicians in this field, other drawings can also be obtained based on these drawings.

[0018] Figure 1 It is a flow chart of the energy-saving control method based on the spatial form of building complexes of the present invention. DETAILED DESCRIPTION

[0019] Example embodiments will now be described more fully with reference to the accompanying drawings. However, example embodiments can be implemented in many forms and should not be construed as limited to the examples set forth herein; rather, these example embodiments are provided so that the description of this disclosure will be thorough and complete and will fully convey the concepts of the example embodiments to those skilled in the art.

[0020] The present invention provides Figure 1 The energy-saving control method based on the spatial form of a building complex shown in the figure specifically includes the following steps: S1. Obtain the three-dimensional spatial block structure information and real-time meteorological parameters of the building complex, establish a spatial shading perception matrix, extract the spatial closure angle relationship between the building blocks, and determine whether there is a shading interference condition that causes a change in the direction of the thermal pressure difference; In this embodiment, S1 is specifically: Obtaining the three-dimensional spatial block structure information and real-time meteorological parameters of the building complex, where the three-dimensional spatial block structure information includes the height data, boundary outline data, and facade orientation data of the building blocks, and the real-time meteorological parameters include the solar azimuth angle, solar altitude angle, and wind direction data; The process of acquiring three-dimensional spatial structural information and real-time meteorological parameters for a building complex involves the dual collection and processing of building spatial form and environmental dynamics. Building elevation data can be obtained through drone LiDAR scanning or vertical elevation information extracted from a Building Information Modeling (BIM) platform. Boundary contour data can be extracted using remote sensing imagery, real-world modeling, or existing two-dimensional vector layers within an urban GIS platform. Combined with measurements of building facade orientation, a three-dimensional point cloud model can be used to calculate the orientation angle of each facade relative to geographic north. Real-time meteorological parameters are collected using environmental sensors deployed at the building complex's roofs and at key ventilation nodes. Solar azimuth and altitude angles are dynamically acquired using time, geographic coordinates, and astronomical calculation models (such as the sun's trajectory formula). Wind direction data is acquired and calibrated using multiple wind vanes. All of this spatial structural information and meteorological parameters are converted to a standard spatial data format and used as input for the subsequent construction of a spatial shading perception matrix. This collection method not only ensures the accuracy of the spatial morphological data of the building complex, but also realizes the dynamic coupling of environmental status and spatial structure, providing reliable data support for subsequent energy-saving control.

[0021] Based on the three-dimensional spatial block structure information of the building complex, a block ray projection model is constructed. A ray tracing algorithm is used to simulate the shadow superposition area of ​​each block under the current solar azimuth angle, and to form the spatial occlusion relationship between the blocks. Constructing a block ray projection model is a key process for quantitatively modeling spatial occlusion relationships based on the three-dimensional spatial block structure information of a building complex. This model is based on the geometric boundaries, facade orientation, and height information of each building block. By setting the solar azimuth and solar altitude angles at a specific time and geographic location, the direction vector of the incident light is determined. The ray tracing algorithm is a spatial lighting simulation algorithm derived from computer graphics. Its basic principle is to emit a large number of rays from the direction of the light source and calculate the intersection position of the rays with the surface of the object in three-dimensional space to determine whether the light is blocked, reflected, or penetrated. In this method, the ray tracing algorithm is used to simulate the projection process of sunlight on the surface of building blocks, thereby identifying which blocks will cast shadows on other blocks at the current solar azimuth. This simulation process can use GPU-based accelerated rendering calculation methods to improve the efficiency of spatial occlusion analysis.

[0022] Constructing a volumetric ray projection model first requires discretizing the three-dimensional model of the building complex into a geometric structure composed of polygonal patches and introducing sunlight into a three-dimensional coordinate system in the form of a vector array. Each ray is tested for intersection with the surface of the volume. If the ray cannot continue to propagate to the target point after entering a volume, the target point is considered to be blocked. By traversing the ray blocking relationships between multiple volumes, a layer of shadow overlay areas between buildings is generated. In this layer, the spatial blocking relationships between volumes at the current solar azimuth angle are calculated based on the distribution of blocked areas of different volumes in their respective time periods and directions. The resulting spatial blocking relationship includes parameters such as the blocking building, the blocked building, the blocking angle, and the blocking time period, providing a quantitative basis for the subsequent extraction of spatial closure angle relationships.

[0023] The spatial closure angle relationship between building blocks is extracted based on the spatial occlusion relationship between the blocks. The angle integral method is used to calculate the sum of the visible angles of each block in multiple directions and generate a spatial closure angle distribution matrix. To extract the spatial enclosure angle relationship between building blocks based on the spatial occlusion relationship between them, a 3D model of the building complex is used to construct a full-range radiation viewing angle field with the building block as the observation center. This field is then divided into several equally spaced directions (e.g., 10° sectors). For each direction, the view is determined to determine whether it is blocked by other blocks and the angle between the blocked building and the observing building is measured. The angle integration method is a calculation method used to quantify the intensity of multi-directional occlusion. Its core concept is to integrate the angle values ​​in each direction on a unit circle or sphere to calculate the sum of the angles in the visible direction. For example, for a low-rise building surrounded by three high-rise buildings, if 360° is divided into 36 sectors, 20 of which are blocked, and the average angle of each blocked direction is 40°, the total visible angle is: (360° × 36 - 20 × 40°) ÷ 36 = 133.3°, indicating a high degree of spatial enclosure for the building. After normalizing the total visible angles of all building blocks, a spatial closed angle distribution matrix is ​​formed. Each element in this matrix represents the degree of openness of the corresponding block in a given orientation, which helps to determine whether the building is in a "closed recessed area" or an "open edge area", providing a spatial structural basis for subsequent thermal pressure ventilation assessment.

[0024] A coupling analysis is performed based on the spatial closure angle distribution matrix and the wind direction change trend in the real-time meteorological parameters to determine whether spatial shading causes a change in the direction of the thermal pressure difference. The judgment method is: when the spatial closure angle change rate is greater than the preset change threshold, and the wind direction and the maximum shading angle of the closed area are less than the set interference angle threshold, it is marked as a shading interference condition.

[0025] The purpose of coupling the spatial enclosure angle distribution matrix with wind direction trends derived from real-time meteorological parameters is to identify areas within a building complex where the direction of thermal pressure differentials has shifted abnormally due to spatial shading. This analysis first calculates the rate of change of each building's enclosure angle in the dominant ventilation direction based on the spatial enclosure angle distribution matrix generated continuously over a historical period. This rate of change represents the magnitude of the change in the block's enclosure degree per unit time and is used to identify the dynamic impact of structural shading. Simultaneously, the wind direction trend derived from real-time meteorological parameters constructs a wind direction trajectory curve using high-frequency wind direction acquisition points, extracting the deviation range of the principal wind direction axis within the analysis period. By coupling these two data types, the author identifies whether the principal wind direction axis is aligned with the maximum shading angle of the target building. If the rate of change of the enclosure angle exceeds a preset threshold (e.g., 15%) and the angle between the wind direction deviation and the maximum shading angle is less than an interference angle threshold (e.g., 25°), it is inferred that the shading structure in this area may have blocked the original thermal pressure-driven channel, causing ventilation direction reversal or failure. This judgment method integrates the evolution of spatial structure and the actual deviation of airflow, and can accurately screen out areas with shielding and pressure interference characteristics, providing a spatial basis for subsequent exhaust efficiency judgment.

[0026] S2. Based on the areas with thermal pressure difference direction changes in the spatial shading pressure perception matrix, an airflow trajectory analysis area is constructed, and continuous wind direction and wind speed data are collected. Changes in the flow direction consistency entropy value are used to determine whether there is an abnormal airflow direction. In this embodiment, S2 specifically includes the following steps: S201. Based on the areas where the thermal pressure difference direction changes in the spatial shading pressure perception matrix, taking the center of the thermal pressure difference gradient change in the area as the starting point, and combining the relative height difference and orientation relationship between building blocks, construct an airflow trajectory analysis area including multiple horizontal and vertical sections; To construct an airflow trajectory analysis area encompassing multiple horizontal and vertical sections, it is necessary to first spatially locate the areas marked as experiencing directional changes in thermal pressure differences within the spatial shading perception matrix. The location with the most significant thermal pressure gradient change is then extracted as the initial calculation center point. This center point is determined by calculating the maximum value of the derivatives in each direction of the thermal pressure gradient field, reflecting the core area where spatial morphological changes disrupt the thermal pressure conduction path. Furthermore, the relative height differences between building blocks and the orientation of the facades are combined to identify the ventilation channels where airflow is most likely to flow and deflect. To construct the analysis area, a set of parallel horizontal sections is first developed along the prevailing wind direction axis in a three-dimensional coordinate system. The vertical spacing between each section is adaptively set based on the inter-building gap height, wind speed attenuation coefficient, and shading severity. Several vertical sections are then inserted at the center point perpendicular to the prevailing wind direction to capture wind deflection or turbulent swirls. Each section and cross section forms a regular grid analysis volume, mapping multidimensional meteorological data such as wind speed and direction through grid point interpolation. This provides spatially decoupled modeling capabilities for subsequent flow field trend analysis and anomaly detection. The entire construction process can be integrated with computational fluid dynamics (CFD) modeling tools and GIS platforms, enabling the computation and visualization of high-resolution spatial airflow trajectories.

[0027] S202: Arrange wind direction and wind speed collection units along the prevailing wind direction in the airflow trajectory analysis area to collect wind direction and wind speed data over multiple consecutive time periods. Map the data into an angle sequence on a unit circle based on the wind direction change trend, and establish a wind direction time series distribution model. Wind direction and speed collection units are arranged along the prevailing wind direction within the airflow trajectory analysis area. High-precision ultrasonic anemometers and multi-point wind direction sensors are evenly distributed along the centerline and boundaries of the analysis area to capture microclimate disturbance characteristics at different heights and locations. The prevailing wind direction is determined through regression analysis based on the mean wind direction over the past 72 hours. Deviations are corrected for building shading characteristics to ensure that the sampling direction aligns with the main airflow path. After collection, wind direction and speed data are segmented and aggregated at a uniform temporal resolution. Wind direction data for each time period are converted to angles relative to true north and mapped to vector points on the unit circle in a two-dimensional polar coordinate system. This vector point set is then organized into a wind direction angle sequence, forming a continuous angle trajectory in the temporal dimension. To develop a temporal distribution model for wind direction, a weighted sliding window technique is used to locally smooth the angle sequence and extract the prevailing wind direction trend. Wind speed data are then superimposed as a weighting term to enhance sensitivity to wind field fluctuations. The resulting wind direction temporal distribution model uses time as the horizontal axis and angle as the vertical axis, along with a wind speed weighted heat map. This serves as the foundational data structure for subsequent flow direction consistency entropy calculations. This model can be modeled and visualized using the SciPy and Matplotlib libraries in MATLAB or Python, offering high resolution and dynamic scalability.

[0028] S203. Calculate the flow direction consistency entropy value in each time period based on the wind direction time series distribution model. When the rate of change of the flow direction consistency entropy value in consecutive time periods is greater than the set disturbance judgment threshold, and the reversal angle of the dominant wind direction in the area exceeds the preset offset angle threshold, it is judged that there is an airflow direction anomaly in the airflow trajectory analysis area.

[0029] By calculating the flow direction consistency entropy value for each time period based on the wind direction time-series distribution model and determining whether the rate of change in the entropy value within consecutive time periods exceeds a set disturbance detection threshold, the system aims to quantify the stability of the airflow direction and capture sudden disturbances. The flow direction consistency entropy value essentially measures the concentration of wind direction distribution within a given time period. Lower entropy values ​​indicate more consistent wind direction, while higher entropy values ​​indicate more dispersed or turbulent wind direction. When the rate of change in the entropy value for consecutive time periods increases significantly, it typically indicates that the airflow in the area is experiencing rapid fluctuations in wind direction due to pressure reflection, turbulent recirculation, or other physical disturbances. In this case, if the prevailing wind direction reverses within the analysis area and this reversal angle exceeds a preset deviation angle threshold, it further confirms that the airflow behavior has deviated from the original ventilation design path. The disturbance detection threshold is obtained by statistically modeling entropy value change samples under historical typical scenarios. Empirical percentiles are typically used to set the boundaries of the stability interval to ensure high-sensitivity detection. The preset deviation angle threshold is determined based on the building geometry, the maximum angle between the prevailing wind direction and the available ventilation path. This ensures that any deviation from the normal wind direction triggers an anomaly flag, effectively identifying the risk of backflow and ventilation failure caused by shading pressure. The combined use of these two thresholds enables accurate identification of airflow anomalies, improving the energy-saving control system's responsiveness to disturbances in complex building spaces.

[0030] In this embodiment, in S203, the flow direction consistency entropy value in each time period is calculated based on the wind direction time series distribution model, specifically: The wind direction data in each time period in the wind direction time series distribution model is divided into a fixed number of wind direction intervals according to the unit angle; In the wind direction time-series distribution model, wind direction data within each time period is divided into a fixed number of wind direction intervals per unit angle. This is done to discretize the wind direction distribution state and facilitate subsequent calculation of its statistical distribution characteristics. This can be achieved by dividing a 360-degree circle into a number of equally spaced angular segments. Common configurations include 12 30-degree intervals, 24 15-degree intervals, or 36 10-degree intervals for higher precision. The specific number of intervals can be adaptively set based on the terrain complexity surrounding the building complex and the frequency of wind direction fluctuations. Within each time period, collected wind direction data is mapped to the corresponding wind direction interval according to its corresponding angle and counted. For example, if a wind direction data value is 73 degrees and the interval is 10 degrees, it is classified as falling between 70 and 80 degrees. This conversion of the original continuous wind direction data into a segmented count structure not only helps reduce the interference of high-frequency noise on entropy value determination but also improves the accuracy and stability of the flow direction consistency measurement. This process can be implemented using discretization functions in programming languages, such as the digitize function in the Python NumPy library or the discretize method in Matlab. Combined with batch data processing, this allows for efficient real-time computation. This operation, the foundation for entropy calculation, determines the precision of subsequent probability distribution function construction and the rationality of classification.

[0031] Count the frequency of wind direction data in each wind direction interval and form a probability distribution function after normalization; Counting the frequency of wind direction data within each wind direction interval and normalizing it is a key step in constructing a probability distribution function. The goal is to convert the discretized wind direction data into a statistically meaningful probability model. This is achieved by first counting all wind direction data according to the defined angle intervals, recording the number of occurrences within each interval to form an original frequency distribution vector. Each element in this frequency vector is then divided by the total number of wind direction data samples within that interval to convert the frequency to the corresponding relative frequency, thus obtaining the probability value for each wind direction interval. This process can be expressed mathematically as follows: if the number of samples within a certain interval is ni and the total number of samples is N, then the probability of that interval is pi = ni / N. For example, in a set of 36 wind direction intervals with 10-degree intervals, if the ninth interval (80-90 degrees) has 45 occurrences of wind direction data, and the total number of samples for that interval is 360, then the probability of that interval is 45 / 360 = 0.125. Combining all interval probability values ​​forms a complete probability distribution function, which forms a probabilistic model of wind direction distribution within that time period. This probability distribution function serves as the input for subsequent calculations of the flow direction consistency entropy value. It accurately reflects the degree of dispersion and concentration of current wind direction changes and is commonly used in wind field disturbance identification, ventilation path determination, and abnormal airflow behavior detection. This calculation process can be implemented using the value_counts(normalize=True) function in Pandas in Python or the frequency vector normalization operation in NumPy.

[0032] The Shannon entropy calculation formula is used to calculate the flow direction consistency entropy value based on the probability distribution function to characterize the degree of discreteness of the wind direction distribution within the time period.

[0033] The Shannon entropy formula is used to calculate the flow direction consistency entropy based on the probability distribution function. This formula aims to quantitatively characterize the degree of dispersion of wind direction data within a specific time period. Shannon entropy, originating from information theory, was proposed by Claude Shannon as a key tool for measuring information uncertainty. Its calculation formula is: H = -∑(pi × log²(pi)), where pi represents the probability value of the i-th wind direction interval in the wind direction probability distribution function. The closer each pi is to 1, the more concentrated the wind direction data is in a certain direction, and the lower the overall entropy. Conversely, if all pi are close to a uniform distribution, the wind direction is more dispersed and the entropy is larger. For example, if the wind direction within a certain time period is concentrated at 90° and the probability distribution is {0, 0, 1, 0, ..., 0}, the entropy is 0. If the distribution is {0.1, 0.1, ..., 0.1}, the entropy is log²(n), where n is the number of wind direction intervals. By calculating the entropy of the wind direction probability distribution function for each time period, we can quantify the stability and consistency of the current wind direction, facilitating the subsequent identification of sudden disturbances or anomalies such as pressure and backflow. This process can be implemented using the scipy.stats.entropy function in Python or custom coding. The input is an array of wind direction probability distributions, and the output is the entropy value of the flow direction consistency corresponding to that time period. This entropy value serves as a key parameter in dynamic ventilation control and can be used to perceive the airflow field state within a building space in real time.

[0034] S3. Spatially overlap the spatial pressure sensing matrix with the area with abnormal airflow direction, and generate an exhaust vent ventilation and heat removal efficiency evaluation index based on the heat conduction rate and the stability of the airflow output path in the overlapped area; In this embodiment, S3 specifically includes the following steps: The spatial overlap processing is performed on the location areas with changes in the direction of thermal pressure difference and the areas with abnormal airflow direction in the spatial shielding pressure perception matrix. The two areas are voxel-decomposed using a three-dimensional Euclidean grid. Spatial alignment is performed based on the geometric center of each voxel in a unified coordinate system. The intersection area where the voxel coordinates coincide is extracted to construct a set of thermal-pressure disturbance coupling spatial units. A three-dimensional Euclidean grid is a voxelized structure that is regularly divided in three-dimensional Euclidean space. It is used to discretize continuous spatial regions into cubic units of uniform size and indexable positions. Each voxel can be precisely located by the three-dimensional coordinates of its geometric center, facilitating spatial registration and coincidence calculations. In spatial coincidence processing, the three-dimensional Euclidean grid can consistently structure spatial data from two independent sources so that they can be compared at a unified spatial scale. Voxel decomposition maps continuous blocks into a collection of small cubes, with each cubic voxel recording attribute information in the space, such as shading pressure intensity, thermal pressure gradient, and airflow disturbance level. By comparing the spatial indexes of two grid voxels, their overlapping areas in physical space can be efficiently identified, thereby achieving joint modeling of multi-source spatial data.

[0035] To spatially coincide locations where thermal pressure differentials change in the spatial pressure perception matrix and areas with abnormal airflow direction are present, the two types of areas are first modeled in three dimensions and unified in coordinate system. For example, the boundary surfaces of the corresponding blocks in the spatial pressure perception matrix can be extracted from the building complex BIM model and discretized into a three-dimensional Euclidean grid consisting of 1m³ voxels. Similarly, the areas with abnormal airflow trajectories generated by wind direction disturbances are converted into voxel representations. Within the unified coordinate system, the center point of each voxel is used as the spatial registration reference point, and the coordinate matching relationship of all voxels in the x, y, and z dimensions is calculated. When the geometric center position of a voxel in the two areas is exactly the same, the voxel is considered to be spatially coincident. The set of all voxels that meet this condition is defined as the thermal-pressure disturbance coupling spatial unit set, and the voxels in the set are numbered, which serves as the input for subsequent thermal-fluid stability analysis. For example, in the northwest corner of the top floor of a building, if the pressure shading analysis results show that this is the center of thermal pressure inversion, and the wind direction data also indicates the existence of abnormal backflow, the overlapping area in this area will be identified as the key interference area after voxel registration.

[0036] Within a set of coupled thermal-pressure disturbance spatial units, the temperature gradient between adjacent units is calculated based on temperature sensor data, and the heat conduction rate of each unit is calculated using the Fourier heat conduction formula. An inter-unit airflow output path diagram is constructed based on the wind speed direction sequence, and the path continuity coefficient and velocity stability factor are extracted to construct a joint thermal-flow stability parameter set. In a collection of coupled thermal-pressure-disturbance spatial units, the heat conduction rate is a key parameter for measuring the ability of a region to transfer heat energy within it. Calculating the heat conduction rate between adjacent voxels requires first obtaining temperature sensor data. The temperature gradient is obtained by dividing the temperature difference between two voxels by the distance between their centers. The Fourier heat conduction equation, q = -k × A × (dT / dx), describes steady-state heat transfer: where q represents the heat conduction rate, k is the thermal conductivity of the material, A is the area of ​​the heat conduction interface, and dT / dx is the temperature gradient. This equation, based on the principle that heat naturally diffuses from high-temperature to low-temperature areas, is applicable to computational scenarios involving heat transfer between building blocks. An airflow output path map is constructed using a sequence of wind speed directions. Its core objective is to trace a coherent path of dominant wind direction between voxels. The path continuity coefficient indicates whether wind direction remains consistent between adjacent voxels, while the velocity stability factor assesses the magnitude of wind speed fluctuations. The joint thermal-fluid stability parameter set, a set of metrics that integrates heat conduction performance and airflow continuity characteristics, is used to evaluate the effectiveness of exhaust paths.

[0037] In the analysis of thermal-pressure disturbances in building complexes, real-time temperature data within the thermal-pressure coupling region is first collected using temperature sensors placed within each voxel. The temperature difference between two adjacent voxels is divided by the distance between their geometric centers to obtain a temperature gradient, which is then substituted into the Fourier heat conduction formula to calculate the heat conduction rate. For example, if the center temperature difference between two voxels is 4°C, the distance between them is 1 meter, and the thermal conductivity is 1.7 W / (m·K), the value for concrete, then the heat flux density is q = -1.7 × A × 4. Wind speed sensors are also used to obtain the wind speed and direction within each voxel. A voxel connectivity graph is constructed, with each edge connecting two voxels with a dominant wind direction transmission relationship, and the wind speed trend is recorded. Paths with continuous wind direction and minimal wind speed fluctuations are assigned higher path continuity coefficients and speed stability factors. For example, a voxel in the top floor corridor of a building, if the wind direction is consistently directed toward the exhaust vent and heat is continuously conducted to that point, it scores high in the heat-flow stability parameter set, indicating an efficient ventilation and heat removal path.

[0038] The heat conduction rate and the airflow output path stability factor are normalized, and dynamic weight factors are assigned to highlight the local sensitivity of the shading area. The local efficiency value of each spatial unit is generated based on the weighted scoring model, and the local efficiency values ​​of all thermal-pressure disturbance-coupled spatial units are summarized to form the exhaust vent ventilation and heat exhaust efficiency evaluation index.

[0039] The heat transfer rate and the airflow output path stability factor measure the heat transfer efficiency and ventilation path reliability of a spatial unit, respectively. To achieve uniform evaluation, these two metrics need to be normalized. A common approach is linear normalization, where the raw values ​​are converted to the [0, 1] range using a maximum-minimum normalization method. A dynamic weighting factor is used to assign higher weights to specific areas during the evaluation process. For example, areas with strong shading pressure should prioritize the airflow stability factor, while areas with significant wind direction fluctuations should prioritize the heat transfer rate. Dynamic weighting factors can be calculated based on spatial location, meteorological disturbance intensity, or historical disturbance frequency. A weighted scoring model can be a linear weighted sum model, such as: local performance value = α × normalized heat transfer rate + β × normalized airflow stability factor, where α and β are dynamically adjusted weighting parameters adaptively determined by contextual thermal disturbance sensitivity. This model integrates the combined impact of these two parameters on ventilation performance, ensuring accurate evaluation in complex shading pressure environments.

[0040] For each thermally coupled spatial unit in an area with significant shading interference, the heat transfer rate and airflow path stability factor are first normalized. For example, if a unit has a heat transfer rate of 0.85 (high among all units) and a wind speed stability factor of 0.60, the normalized values ​​are 0.90 and 0.65, respectively. Based on the historical frequency of thermal disturbances and the shading intensity in the area, weights α = 0.4 and β = 0.6 are assigned, reflecting the area's greater sensitivity to airflow continuity. The final local efficiency value is calculated as 0.4 × 0.90 + 0.6 × 0.65 = 0.75. This process is repeated for all thermally coupled spatial units, and their local efficiency values ​​are weighted averaged or summed to form the exhaust outlet ventilation and heat removal efficiency evaluation index for that area. If the final evaluation index is significantly below the normal threshold, the exhaust duct is deemed inoperable due to shading and airflow reversal, and the ventilation strategy should be adjusted. This approach enables intelligent assessment of complex shading interference areas, preventing ventilation failures from being misinterpreted as successful ventilation.

[0041] S4. Based on the exhaust vent ventilation and heat removal efficiency evaluation index, an exhaust response condition matrix including space shielding pressure parameters, airflow disturbance level, and heat retention level is constructed to generate a ventilation control decision sequence; In this embodiment, S4 is specifically: S401. Normalize and segment the exhaust vent ventilation and heat removal efficiency evaluation index, extract the corresponding spatial shielding pressure parameters, airflow disturbance level, and heat retention level based on the spatial unit position correlation, and assign them to three-dimensional label vectors to form an initial response factor set. S402. Establish a multidimensional matrix mapping structure based on the combination relationship of the three types of grade indicators in the initial response factor set, and construct an exhaust response condition matrix including space shielding pressure parameters, airflow disturbance level, and heat retention level. Each element in the matrix corresponds to a unique grade combination configuration. S403. Relying on the matching rules between each level combination in the exhaust response condition matrix and the exhaust control behavior, a ventilation control determination sequence is generated according to the preset control threshold mapping logic. The ventilation behavior control instructions output by the ventilation control determination sequence are used for subsequent exhaust control execution.

[0042] The generation of ventilation control decision sequences is based on the exhaust response condition matrix. A mapping mechanism is formed by establishing matching rules between each level combination and the corresponding exhaust control behavior. Matching rules are defined through expert system modeling, historical control data training, or scenario feedback based on simulated environments to ensure the selection of the most appropriate ventilation response strategy under different combinations of thermal and wind field disturbances. For example, a combination of high spatial shading parameter levels, severe airflow disturbance levels, and persistently high heat retention levels will be mapped to "closing exhaust vents and activating alternative heat exhaust channels." However, in scenarios with mild shading effects but moderate heat retention levels, the "opening of some exhaust vents to enhance local heat exhaust" may be triggered. This mapping relationship is encoded in the form of a rule table or conditional logic tree. The system can quickly match and output control behaviors within the exhaust response condition matrix based on real-time monitoring data tags.

[0043] The preset control threshold mapping logic is used to refine the dynamic response relationship between level combinations and exhaust behaviors. This logic not only defines which behavior should be triggered under which combination conditions, but also allows the setting of progressive strategies and fault-tolerant mechanisms for triggering thresholds. For example, when the airflow disturbance level fluctuates critically between medium and high, it can be configured to trigger a strategy of completely closing the exhaust vents only after the heat retention level remains above a certain duration, so as to avoid frequent control switching caused by short-cycle fluctuations. This logic is implemented in the control system in the form of state transfer functions, condition matching trees, or fuzzy logic mapping, which not only ensures the stability of the response, but also improves the system's adaptability to complex dynamic environments. The resulting ventilation control decision sequence is a timed set of control instructions that supports linkage execution, priority scheduling, and feedback closed-loop correction, and is used to subsequently drive the exhaust device to complete partition response, energy-saving switching, or alternative ventilation strategy execution.

[0044] In this embodiment, S401 specifically includes: The ventilation and heat removal efficiency evaluation index of the exhaust vent is normalized to the minimum and maximum values ​​within the set upper and lower limits, and divided into multiple discrete grade segments based on the equal spacing method, and mapped to each spatial unit in the building complex; The exhaust vent ventilation and heat removal efficiency evaluation index is normalized between the minimum and maximum limits within a set range. This process aims to normalize the raw index value to a uniform numerical range for subsequent grading and comparison. This process typically constructs a normalized mapping range based on the historical maximum and minimum values ​​of the actual evaluation index, ensuring uniform comparability of evaluation values ​​across different spatial units. After normalization, the numerical range is graded using the equal-interval method. This discretization method evenly divides the normalized continuous numerical range into several sub-ranges. Its core concept is to divide the index space into several graded segments at a fixed interval, with each segment corresponding to a distinct discrete level. For example, when the normalized result is limited to between 0 and 1, the equal-interval method can divide it into five grade segments, corresponding to the five grade labels: very low, low, medium, high, and very high. Each spatial unit is automatically classified into the corresponding grade segment based on its normalized value. Through this division process, each spatial unit in the building complex can be assigned a clear discrete level, thereby achieving a structural expression of the exhaust vent ventilation and heat exhaust efficiency evaluation index at the spatial scale, laying the foundation for the subsequent construction of the ventilation control matrix.

[0045] The spatial shielding parameters are calculated based on the relative height, closure angle, and azimuth angle between the spatial unit and the source of the shielding. The airflow disturbance level is generated by combining the wind direction reversal probability and the rate of change of the flow direction consistency entropy value. The heat retention level is calculated by combining the heat conduction rate and the duration of heat accumulation. Spatial shading parameters are calculated by analyzing the three-dimensional spatial relationship between each spatial unit and its corresponding shading source. Relative height measures the vertical shading strength of the unit's overhead, closure angle assesses the degree of line of sight between two units in planar projection, and azimuth angle reflects the spatial orientation of the shading source relative to the unit's main facade. Quantifying these three dimensions enables the construction of a multi-factor weighted model to express the overall shading degree. The airflow disturbance level is determined by the wind direction reversal probability and the rate of change of the flow direction consistency entropy within the airflow trajectory analysis area. The wind direction reversal probability represents the frequency of a 180-degree reversal of wind direction per unit time, while the rate of change of the flow direction consistency entropy measures the tendency of disturbance in the local airflow direction stability. Disturbance levels are categorized into high, medium, and low levels based on established disturbance classification criteria. The heat retention level is a composite evaluation based on the heat conduction rate and the duration of heat accumulation within each spatial unit. Spatial units with low heat conduction rates and prolonged heat accumulation are designated as areas of high heat retention. Taking a unit at the top of a high-rise building as an example, if it is blocked by a taller block above, the relative height is positive, the closure angle is close to 90 degrees, and the azimuth angle is less than 30 degrees, it is judged that its shading pressure is significant. At the same time, if the wind direction in this unit has frequently reversed and the flow entropy value has fluctuated significantly in the past two hours, the disturbance level is judged to be high. If the temperature is retained for a long time without an obvious diffusion path, its heat retention level is also high, and it is identified as a high-risk ventilation control point in the spatial response analysis.

[0046] The spatial shielding parameters, airflow disturbance level and heat retention level are assigned to the respective dimensional components of the three-dimensional label vector, and are combined and stored according to the spatial unit coordinate index to form an initial response factor set.

[0047] After mapping the three indicators of spatial obstruction, airflow disturbance, and heat retention into discrete numerical labels, they can be assigned to the three independent dimensional components of a three-dimensional label vector. For example, the spatial obstruction parameter can be divided into five levels (0 to 4), the airflow disturbance level into three levels (0 to 2), and the heat retention level into four levels (0 to 3), thus forming a three-dimensional vector structure with unique combinational meaning. Each spatial unit within the building complex is stored using its three-dimensional coordinate index within the unified spatial grid system as the primary key, and the corresponding three-dimensional label vector is bound to it, thus forming a structured data set. The entire process of constructing the initial response factor set is based on spatial mapping and label aggregation mechanisms, and sparse tensor or multidimensional array structures can be used for data organization and rapid retrieval. Each response factor in the set is indexed by spatial coordinates and has a three-dimensional vector as its value, ensuring spatial uniqueness and the ability to distinguish between different levels. For example, a spatial unit numbered A-12-5 at the top of a building has a spatial shading parameter of 3, an airflow disturbance level of 2, and a heat retention level of 1. Its label vector is [3, 2, 1]. This is bound to the spatial coordinates and written into the initial response factor set, which is then used to construct the exhaust response condition matrix and ventilation control decision sequence. This ensures that the environmental state of each spatial point is uniformly modeled and accurately participates in the ventilation control logic.

[0048] In this embodiment, S402 specifically includes: The spatial shielding parameter level set, airflow disturbance level set, and heat retention level set are mapped into independent one-dimensional integer index axes, and an axial index system for constructing a three-dimensional matrix mapping structure is defined. The spatial obstruction parameter level set, the airflow disturbance level set, and the heat retention level set each contain multiple discrete level labels. For example, the spatial obstruction parameter may consist of five levels, the airflow disturbance level may consist of three levels, and the heat retention level may consist of four levels. To uniquely identify each level combination in the data structure, each level set must be mapped to a one-dimensional integer index axis, where each level corresponds to a unique integer index number. For example, spatial obstruction parameter levels 0 to 4 are mapped to positions 0 to 4 on the index axis X, airflow disturbance levels 0 to 2 are mapped to positions 0 to 2 on the index axis Y, and heat retention levels 0 to 3 are mapped to positions 0 to 3 on the index axis Z. The three indicators occupy the three coordinate axes of the three-dimensional matrix, respectively. This method creates a three-dimensional matrix mapping structure in which each three-dimensional coordinate point represents a combination of spatial obstruction parameter level, airflow disturbance level, and heat retention level, and has a unique index address. This indexing system not only facilitates the subsequent rapid matching of ventilation control rules but also provides a data structure foundation for rapid retrieval of response configurations and automatic mapping of level combinations. This mapping structure can be implemented based on a three-dimensional array, a sparse matrix, or a hash map. It has the characteristics of clear structure, efficient operation, and strong scalability, and is suitable for complex control scenarios under dynamic changes in building environments.

[0049] The spatial shielding parameter level, airflow disturbance level and heat retention level of each response factor in the initial response factor set are input into the three-dimensional matrix structure as three-dimensional coordinates, and the corresponding coordinate unit is assigned a unique identification value for the response configuration index; Each set of initial response factors consists of three dimensions, corresponding to the spatial obstruction parameter level, airflow disturbance level, and heat retention level. The level values ​​for each dimension are mapped to integer indexes. These three index values ​​are treated as three-dimensional coordinate points and input into a predefined three-dimensional matrix structure, precisely locating a unique location within the matrix. This location represents a specific response factor combination, and its unique coordinates are used to precisely associate and retrieve the corresponding ventilation control rule. Within this coordinate cell, a response configuration identifier is assigned to encode the control action label for the current level combination. This identifier can be a preset ventilation command code, control mode index, or priority factor. This static mapping between spatial level combinations and action responses creates an efficient data indexing mechanism, enabling rapid level identification and control strategy linkage under complex environmental parameter fluctuations. This three-dimensional matrix structure can be implemented as a multidimensional array. In actual deployment, a key-value mapping table is used to improve access efficiency. An index caching mechanism also enhances the real-time and stability of control responses. The entire process, leveraging a multidimensional tag structure and matrix mapping mechanism, enhances the ventilation control system's ability to precisely respond to building obstruction interference and its thermal effects.

[0050] Traverse all three-dimensional coordinate combinations to construct an exhaust response condition matrix, and explicitly map each cell element in the exhaust response condition matrix to a unique combination of spatial shielding parameter level, airflow disturbance level, and heat retention level for the subsequent generation and retrieval of ventilation control judgment sequences.

[0051] The process of constructing the exhaust response condition matrix requires traversing all possible three-dimensional combinations of spatial shielding parameter levels, airflow disturbance levels, and heat retention levels, and mapping each set of level combinations to a unique cell position in the matrix structure. Each cell represents a combination configuration of a specific environmental state. The three-dimensional coordinate axis is formed by the Cartesian product of the spatial shielding parameter level set, the airflow disturbance level set, and the heat retention level set. The traversal process is completed in a nested loop manner to ensure that all possible combinations are enumerated and mapped. In the matrix, each cell element is precisely bound to a unique level combination through the index value corresponding to its three-dimensional coordinate. This binding relationship is implemented in a table-driven manner during the initialization phase, and each combination is associated with a specific ventilation control strategy or instruction through the response rule configuration file. This matrix structure is used as a core index library during operation. When the actual monitoring value generates a new level label combination, the system can quickly locate the corresponding matrix cell and then call the preset control response. The entire matrix structure is implemented with sparse storage or three-dimensional array structure, combined with bitmap identification and cache mechanism, which can significantly improve the control logic matching efficiency and system response speed. At the same time, it has good scalability, which facilitates the subsequent introduction of more control factors or level granularity adjustment.

[0052] S5. According to the control instructions in the ventilation control determination sequence, the exhaust vent opening state is dynamically regulated, and corresponding ventilation behaviors or alternative energy-saving strategies are triggered according to different states. At the same time, the ventilation control data is recorded in the spatial interference dataset.

[0053] In this embodiment, S5 is specifically: The exhaust control execution unit is driven according to the control instructions in the ventilation control judgment sequence. The opening angle and opening duration of the exhaust vents are dynamically adjusted using the opening adjustment mechanism. An exhaust vent opening state mapping model covering the building space area is established to represent the ventilation capacity boundary of each exhaust vent in real time. The control instructions in the ventilation control decision sequence establish a communication channel with the exhaust control execution unit through the logic parsing module, driving the exhaust control execution unit to electrically open and close the exhaust vents. During this process, the opening adjustment mechanism finely adjusts the opening angle and duration of the exhaust vents based on the exhaust intensity level and duration parameters contained in the control instructions. To uniformly manage the operating status of multiple exhaust vents, the exhaust equipment is spatially annotated within the 3D building model. Combining the current actual opening status and time information, an exhaust vent opening status mapping model covering the entire building space is constructed. This model dynamically tracks the operating status boundaries of each exhaust vent along the timeline, thereby reflecting the distribution of ventilation capacity at different locations in the building within a specific timeframe. For example, in a high-rise building block, if the control instructions determine low heat removal efficiency, the corresponding exhaust vent will be displayed as "low opening + short duration" in the mapping model, triggering subsequent intervention by the alternative strategy response module.

[0054] The exhaust control execution unit consists of electric actuators connected to the building automation control system. It receives precise numerical commands from the control sequence to open and close the exhaust vents. The opening adjustment mechanism is based on an exhaust level parameter, typically divided into multiple discrete levels, such as 0%, 25%, 50%, 75%, and 100%, each corresponding to a set of ventilation capacity estimates. The opening duration is synchronously managed by the system's timing module, ensuring that exhaust operations at different locations do not interfere with each other and achieve coordinated heat removal. The exhaust vent opening state mapping model is constructed based on a building information model (BIM) or GIS spatial information system. The mapping data structure includes the exhaust vent's spatial location, opening percentage, opening and closing time, and the corresponding control sequence number to support subsequent visualization analysis, predictive modeling, and energy strategy optimization. This structure enables dynamic representation of ventilation capacity boundaries across time and space, providing a critical foundation for control efficiency and strategy adaptation.

[0055] Based on the exhaust vent opening state mapping model and the status field of the ventilation control judgment sequence, a linkage analysis is performed to trigger the ventilation behavior instructions that match the current state. When the current state is marked as a high risk level of thermal pressure disturbance in the judgment sequence, an alternative energy-saving strategy is switched to execute, including switching to a central exhaust system, activating a heat exchange recovery mechanism, or delaying the exhaust response strategy. A linkage parsing mechanism is established between the exhaust vent opening state mapping model and the ventilation control decision sequence to achieve dynamic control logic response and strategy adaptation. This mechanism matches the current exhaust vent opening state and opening / closing period in the mapping model with the corresponding state field values ​​in the decision sequence, such as ventilation intensity level, airflow disturbance level, and heat retention level. If the analysis identifies a high thermal pressure disturbance risk level, the system triggers the strategy diversion module, terminating the output of regular ventilation behavior instructions and instead implementing a backup ventilation strategy or energy-saving mechanism. Specifically, programmable logic can be used to define conditional trigger statements in the control system, such as "If the state is marked as high risk and ventilation is ineffective for three consecutive cycles, activate the heat recovery device." For example, in a specific implementation, if strong obstruction pressure and airflow reversal occur within a continuous monitoring period in the south area of ​​a high-rise building, the exhaust vents in this area are marked as having low ventilation efficiency in the mapping model and high risk in the decision sequence. At this point, the system automatically activates the central exhaust unit or a delayed opening strategy to alleviate the ventilation load.

[0056] The exhaust vent opening state mapping model is a multi-dimensional data structure that records the current spatial position, opening angle, opening duration, and opening and closing timestamps of each exhaust vent, and is used to reflect the exhaust behavior boundaries of each area of ​​the building in real time. The state field in the ventilation control judgment sequence includes the three-dimensional label vector combination corresponding to the exhaust response condition matrix and its matching strategy weight. During the linkage analysis process, the state field is matched with the current data item of the mapping model to drive the execution module to output specific ventilation behavior. When the thermal pressure disturbance is judged to be a high-risk level, the alternative energy-saving strategy module will select a strategy path based on the energy-saving priority, including activating the central exhaust system to break the ventilation blind spot of the pressure-blocking dead corner, activating the heat exchange device to recover the energy in the high-temperature exhaust, or suspending the exhaust behavior through the dynamic scheduling algorithm to avoid further reduction of local heat exhaust efficiency. This design improves the system's intelligent adaptability under complex aerodynamic disturbances.

[0057] Key data such as the execution time of ventilation behavior or alternative energy-saving strategies, changes in exhaust vent opening, response type, and ventilation control judgment sequence number are written into the spatial interference dataset in a unified structure format, and the interference type index is marked with the exhaust vent spatial coordinates as the primary key for subsequent traceability modeling and optimization learning of ventilation control behavior.

[0058] The execution of ventilation behavior or alternative energy-saving strategies involves several key variables, including the execution time period, dynamic changes in exhaust vent opening, the type of response triggered, and the ventilation control decision sequence number underlying the trigger. Recording these data items uniformly in a spatial interference dataset requires standardized processing using a structured format. For example, a field-oriented data table structure is employed, with each record indexed by the exhaust vent's three-dimensional spatial coordinates as the primary key, and supplementary fields identifying the specific interference type and response mechanism. This approach facilitates rapid identification of the historical behavior trajectory of ventilation response areas during subsequent system evaluations, enabling causal tracing and optimized modeling of ventilation behavior. For example, during a sustained high heat pressure disturbance in the north area of ​​a building, system records indicate that the exhaust vents in this area implemented a delayed exhaust and heat exchange mechanism during a certain period. This ventilation control decision sequence number differed from that of adjacent areas. Through the structured records in the spatial interference dataset, the system can track the effectiveness of this specific response pattern, providing a reference model for future response strategies under similar disturbance scenarios.

[0059] The spatial interference dataset uses an indexing method with spatial coordinates as the primary key to achieve spatial management of exhaust control behavior. The execution time record reflects the timeliness of the response strategy, the change in the exhaust vent opening reflects the physical behavior process of the exhaust control unit, the response type is used to distinguish the switching logic between conventional ventilation and alternative energy-saving mechanisms, and the ventilation control judgment sequence number is used as a logical basis to associate the matching relationship between the upstream judgment mechanism and the downstream behavior response. The application of a unified structural format ensures that the dataset can be directly read and processed by machine learning modules or optimization algorithms, providing a data foundation for the intelligent evolution of ventilation control systems. Through this mechanism, the system can not only identify the control behavior patterns under different shading pressure and airflow disturbance scenarios, but also train prediction models based on historical data accumulation, and adjust future exhaust behavior strategies in advance to avoid thermal pressure interference risks.

[0060] The above embodiments can be implemented in whole or in part via software, hardware, firmware, or any other combination. When implemented using software, the above embodiments can be implemented in whole or in part in the form of a computer program product. A computer program product comprises one or more computer instructions or computer programs. When the computer instructions or computer program are loaded or executed on a computer, the processes or functions according to the embodiments of the present application are fully or partially generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. Computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired or wireless means (e.g., infrared, wireless, microwave, etc.). A computer-readable storage medium can be any available medium accessible by a computer or a data storage device such as a server or data center that contains a collection of one or more available media. Available media can be magnetic media (e.g., floppy disks, hard disks, magnetic tapes), optical media (e.g., DVDs), or semiconductor media. Semiconductor media can be solid-state drives.

[0061] It should be understood that in the various embodiments of the present application, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0062] Those skilled in the art will appreciate that the units and algorithm steps of each example 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 performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel 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.

[0063] In the several embodiments provided in this application, it should be understood that the disclosed systems and methods can be implemented in other ways. For example, the embodiments described above are merely illustrative. For example, the division of units is merely a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interface, device or unit, which can be electrical, mechanical or other forms.

[0064] Units described as separate components may or may not be physically separate, and components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of this embodiment according to actual needs.

[0065] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0066] The above are only specific embodiments of the present application, but the scope of protection of this application is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in this application should be included in the scope of protection of this application. Therefore, the scope of protection of this application should be based on the scope of protection of the claims.

Claims

1. An energy-saving control method based on the spatial form of a building complex, characterized in that: The specific steps include: S1. Obtain the three-dimensional spatial block structure information and real-time meteorological parameters of the building complex, establish a spatial shading perception matrix, extract the spatial closure angle relationship between the building blocks, and determine whether there is a shading interference condition that causes a change in the direction of the thermal pressure difference; S2. Based on the areas with thermal pressure difference direction changes in the spatial shading pressure perception matrix, an airflow trajectory analysis area is constructed, and continuous wind direction and wind speed data are collected. Changes in the flow direction consistency entropy value are used to determine whether there is an abnormal airflow direction. S3. Spatially overlap the spatial pressure sensing matrix with the area with abnormal airflow direction, and generate an exhaust vent ventilation and heat removal efficiency evaluation index based on the heat conduction rate and the stability of the airflow output path in the overlapped area; S4. Based on the exhaust vent ventilation and heat removal efficiency evaluation index, an exhaust response condition matrix including space shielding pressure parameters, airflow disturbance level, and heat retention level is constructed to generate a ventilation control decision sequence; S5. According to the control instructions in the ventilation control determination sequence, the exhaust vent opening state is dynamically regulated, and corresponding ventilation behaviors or alternative energy-saving strategies are triggered according to different states. At the same time, the ventilation control data is recorded in the spatial interference dataset.

2. The energy-saving control method based on the spatial form of building complexes according to claim 1 is characterized in that: S1 is specifically: Obtaining the three-dimensional spatial block structure information and real-time meteorological parameters of the building complex, where the three-dimensional spatial block structure information includes the height data, boundary outline data, and facade orientation data of the building blocks, and the real-time meteorological parameters include the solar azimuth angle, solar altitude angle, and wind direction data; Based on the three-dimensional spatial block structure information of the building complex, a block ray projection model is constructed. A ray tracing algorithm is used to simulate the shadow superposition area of ​​each block under the current solar azimuth angle, and to form the spatial occlusion relationship between the blocks. The spatial closure angle relationship between building blocks is extracted based on the spatial occlusion relationship between the blocks. The angle integral method is used to calculate the sum of the visible angles of each block in multiple directions and generate a spatial closure angle distribution matrix. A coupling analysis is performed based on the spatial closure angle distribution matrix and the wind direction change trend in the real-time meteorological parameters to determine whether spatial shading causes a change in the direction of the thermal pressure difference. The judgment method is: when the spatial closure angle change rate is greater than the preset change threshold, and the wind direction and the maximum shading angle of the closed area are less than the set interference angle threshold, it is marked as a shading interference condition.

3. The energy-saving control method based on the spatial form of building complexes according to claim 1 is characterized in that: S2 specifically includes the following steps: S201. Based on the areas where the thermal pressure difference direction changes in the spatial shading pressure perception matrix, taking the center of the thermal pressure difference gradient change in the area as the starting point, and combining the relative height difference and orientation relationship between building blocks, construct an airflow trajectory analysis area including multiple horizontal and vertical sections; S202: Arrange wind direction and wind speed collection units along the prevailing wind direction in the airflow trajectory analysis area to collect wind direction and wind speed data over multiple consecutive time periods. Map the data into an angle sequence on a unit circle based on the wind direction change trend, and establish a wind direction time series distribution model. S203. Calculate the flow direction consistency entropy value in each time period based on the wind direction time series distribution model. When the rate of change of the flow direction consistency entropy value in consecutive time periods is greater than the set disturbance judgment threshold, and the reversal angle of the dominant wind direction in the area exceeds the preset offset angle threshold, it is judged that there is an airflow direction anomaly in the airflow trajectory analysis area.

4. The energy-saving control method based on the spatial form of building complexes according to claim 3 is characterized in that: In S203, the flow direction consistency entropy value in each time period is calculated based on the wind direction time series distribution model, specifically: The wind direction data in each time period in the wind direction time series distribution model is divided into a fixed number of wind direction intervals according to the unit angle; Count the frequency of wind direction data in each wind direction interval and form a probability distribution function after normalization; The Shannon entropy calculation formula is used to calculate the flow direction consistency entropy value based on the probability distribution function to characterize the degree of discreteness of the wind direction distribution within the time period.

5. The energy-saving control method based on the spatial form of building complexes according to claim 1 is characterized in that: S3 specifically includes the following steps: The spatial overlap processing is performed on the location areas with changes in the direction of thermal pressure difference and the areas with abnormal airflow direction in the spatial shielding pressure perception matrix. The two areas are voxel-decomposed using a three-dimensional Euclidean grid. Spatial alignment is performed based on the geometric center of each voxel in a unified coordinate system. The intersection area where the voxel coordinates coincide is extracted to construct a set of thermal-pressure disturbance coupling spatial units. Within a set of coupled thermal-pressure disturbance spatial units, the temperature gradient between adjacent units is calculated based on temperature sensor data, and the heat conduction rate of each unit is calculated using the Fourier heat conduction formula. An inter-unit airflow output path diagram is constructed based on the wind speed direction sequence, and the path continuity coefficient and velocity stability factor are extracted to construct a joint thermal-flow stability parameter set. The heat conduction rate and the airflow output path stability factor are normalized, and dynamic weight factors are assigned to highlight the local sensitivity of the shading area. The local efficiency value of each spatial unit is generated based on the weighted scoring model, and the local efficiency values ​​of all thermal-pressure disturbance-coupled spatial units are summarized to form the exhaust vent ventilation and heat exhaust efficiency evaluation index.

6. The energy-saving control method based on the spatial form of building complexes according to claim 1 is characterized in that: S4 is specifically: S401. Normalize and segment the exhaust vent ventilation and heat removal efficiency evaluation index, extract the corresponding spatial shielding pressure parameters, airflow disturbance level, and heat retention level based on the spatial unit position correlation, and assign them to three-dimensional label vectors to form an initial response factor set. S402. Establish a multidimensional matrix mapping structure based on the combination relationship of the three types of grade indicators in the initial response factor set, and construct an exhaust response condition matrix including space shielding pressure parameters, airflow disturbance level, and heat retention level. Each element in the matrix corresponds to a unique grade combination configuration. S403. Relying on the matching rules between each level combination in the exhaust response condition matrix and the exhaust control behavior, a ventilation control determination sequence is generated according to the preset control threshold mapping logic. The ventilation behavior control instructions output by the ventilation control determination sequence are used for subsequent exhaust control execution.

7. The energy-saving control method based on the spatial form of building complexes according to claim 6 is characterized in that: S401 is specifically: The ventilation and heat removal efficiency evaluation index of the exhaust vent is normalized to the minimum and maximum values ​​within the set upper and lower limits, and divided into multiple discrete grade segments based on the equal spacing method, and mapped to each spatial unit in the building complex; The spatial shielding parameters are calculated based on the relative height, closure angle, and azimuth angle between the spatial unit and the source of the shielding. The airflow disturbance level is generated by combining the wind direction reversal probability and the rate of change of the flow direction consistency entropy value. The heat retention level is calculated by combining the heat conduction rate and the duration of heat accumulation. The spatial shielding parameters, airflow disturbance level and heat retention level are assigned to the respective dimensional components of the three-dimensional label vector, and are combined and stored according to the spatial unit coordinate index to form an initial response factor set.

8. The energy-saving control method based on the spatial form of building complexes according to claim 6 is characterized in that: S402 is specifically: The spatial shielding parameter level set, airflow disturbance level set, and heat retention level set are mapped into independent one-dimensional integer index axes, and an axial index system for constructing a three-dimensional matrix mapping structure is defined. The spatial shielding parameter level, airflow disturbance level and heat retention level of each response factor in the initial response factor set are input into the three-dimensional matrix structure as three-dimensional coordinates, and the corresponding coordinate unit is assigned a unique identification value for the response configuration index; Traverse all three-dimensional coordinate combinations to construct an exhaust response condition matrix, and explicitly map each cell element in the exhaust response condition matrix to a unique combination of spatial shielding parameter level, airflow disturbance level, and heat retention level for the subsequent generation and retrieval of ventilation control judgment sequences.

9. The energy-saving control method based on the spatial form of building complexes according to claim 1, characterized in that: S5 is specifically: The exhaust control execution unit is driven according to the control instructions in the ventilation control judgment sequence. The opening angle and opening duration of the exhaust vents are dynamically adjusted using the opening adjustment mechanism. An exhaust vent opening state mapping model covering the building space area is established to represent the ventilation capacity boundary of each exhaust vent in real time. Based on the exhaust vent opening state mapping model and the status field of the ventilation control judgment sequence, a linkage analysis is performed to trigger the ventilation behavior instructions that match the current state. When the current state is marked as a high risk level of thermal pressure disturbance in the judgment sequence, an alternative energy-saving strategy is switched to execute, including switching to a central exhaust system, activating a heat exchange recovery mechanism, or delaying the exhaust response strategy. Key data such as the execution time of ventilation behavior or alternative energy-saving strategies, changes in exhaust vent opening, response type, and ventilation control judgment sequence number are written into the spatial interference dataset in a unified structure format, and the interference type index is marked with the exhaust vent spatial coordinates as the primary key for subsequent traceability modeling and optimization learning of ventilation control behavior.

Images