Cold and hot channel simulation analysis method and system based on BIM and CFD

Abstract

An embodiment of the present invention discloses a cold and hot channel simulation analysis method and system based on BIM and CFD, the method comprising: obtaining a building structure data set of a target building area, generating a BIM building model based on the building structure data set, the BIM building model including equipment distribution information and spatial topological relationships of the target building area; performing channel division processing on the BIM building model to generate an initial layout plan for the cold and hot channels, performing fluid dynamics analysis based on the initial layout plan to simulate air flow states and obtain fluid distribution data for the cold and hot channels; coupling the BIM building model with the fluid distribution data for analysis to generate air leakage state prediction results and temperature distribution prediction results for the cold and hot channels; generating a layout optimization strategy based on the air leakage state prediction results and the temperature distribution prediction results, and adjusting the channel layout information in the initial layout plan according to the layout optimization strategy to obtain a target layout plan.

Description

Technical Field

[0001] The embodiments of the present invention relate to the field of computer technology, and in particular to a cold and hot channel simulation analysis method and system based on BIM and CFD. Background Art

[0002] With the development of the construction industry, BIM (Building Information Modeling) and CFD (Computational Fluid Dynamics) technologies have been widely used in architectural design and analysis. BIM integrates all aspects of a building's information to create a three-dimensional digital model, visually presenting the building's structure and equipment layout. CFD, on the other hand, specializes in numerical simulation and analysis of fluid flow phenomena. The emergence of these two technologies has provided a more scientific and precise approach to architectural design, allowing designers to evaluate building performance from different perspectives.

[0003] However, in existing technologies, while BIM and CFD each play their own roles, there's a lack of effective integration and synergy between the two. When addressing issues like hot and cold aisle layout within a building, analysis often relies on either technology in isolation. BIM alone makes it difficult to deeply analyze fluid-related issues like air flow, while relying solely on CFD lacks comprehensive information about the building's structure and equipment. This disconnected analysis approach results in an incomplete understanding of the complex internal conditions of a building, making it difficult to develop a scientifically sound strategy for optimizing hot and cold aisle layouts. Summary of the Invention

[0004] The embodiments of the present invention provide a hot and cold aisle simulation analysis method and system based on BIM and CFD, which are used to effectively improve the rationality of hot and cold aisle layout, optimize air flow and temperature distribution inside the building, and improve the overall functionality and comfort of the building.

[0005] In a first aspect, an embodiment of the present invention provides a hot and cold channel simulation and analysis method based on BIM and CFD, which is applied to a hot and cold channel simulation and analysis system, the method comprising: obtaining a building structure data set of a target building area, generating a BIM building model based on the building structure data set, the BIM building model including equipment distribution information and spatial topological relationships of the target building area; performing channel division processing on the BIM building model to generate an initial layout plan for hot and cold channels, performing fluid dynamics analysis based on the initial layout plan to simulate air flow states, and obtaining fluid distribution data of the hot and cold channels; coupling the BIM building model with the fluid distribution data for analysis to generate air leakage state prediction results and temperature distribution prediction results of the hot and cold channels; generating a layout optimization strategy based on the air leakage state prediction results and the temperature distribution prediction results, and adjusting the channel layout information in the initial layout plan according to the layout optimization strategy to obtain a target layout plan.

[0006] In a second aspect, an embodiment of the present invention provides a cold and hot channel simulation analysis system, comprising:

[0007] processor;

[0008] a storage device having a computer program stored thereon,

[0009] When the computer program is executed by the processor, the processor implements any of the cold and hot channel simulation analysis methods based on BIM and CFD.

[0010] An embodiment of the present invention provides a readable storage medium, on which a program or instruction is stored. When the program or instruction is executed by a processor, the steps of the hot and cold channel simulation analysis method based on BIM and CFD are implemented.

[0011] It can be seen that the embodiments of the present invention have the following beneficial effects: first, by acquiring a set of building structure data, a BIM building model containing equipment distribution information and spatial topological relationships is generated, and the BIM building model is divided into channels and combined with fluid dynamics analysis to simulate the air flow state and obtain fluid distribution data; secondly, the BIM building model is coupled with the fluid distribution data for analysis to predict the air leakage state and temperature distribution, and through cross-domain data fusion analysis, strong support is provided for accurately grasping the conditions of hot and cold channels; then, a layout optimization strategy is generated based on the prediction results and the channel layout is adjusted to obtain the target layout plan, which effectively improves the rationality of the hot and cold channel layout, optimizes the air flow and temperature distribution inside the building, improves the overall functionality and comfort of the building, and realizes a more efficient and scientific building layout planning method. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] Figure 1 A flow chart of a hot and cold channel simulation analysis method based on BIM and CFD provided in an embodiment of the present invention.

[0013] Figure 2 A schematic diagram of the basic structure of a hot and cold channel simulation analysis system provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0014] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the embodiments of the present invention are further described in detail below with reference to the accompanying drawings and specific implementation methods.

[0015] See also Figure 1 As shown in FIG, this figure is a flow chart of a cold and hot channel simulation analysis method based on BIM and CFD provided by an embodiment of the present invention, which can be applied to the cold and hot channel simulation analysis system. Figure 1 As shown, the method includes steps 110 to 140.

[0016] Step 110: Acquire a building structure data set of a target building area, and generate a BIM building model based on the building structure data set, wherein the BIM building model includes equipment distribution information and spatial topology relationships of the target building area.

[0017] In an embodiment of the present invention, the computer room area of ​​a large data center is taken as an example of the target building area. First, the building structure data set of the computer room area is obtained through existing building data acquisition tools. The above data covers the structural information of the walls, floor slabs, beams, etc. of the computer room, as well as the location information of various equipment such as server cabinets, air-conditioning units, etc. Then, BIM modeling software is used to start generating a BIM building model based on the above building structure data set. During the modeling process, the data can be parsed and processed by the BIM modeling software, and the location information of the equipment can be accurately integrated into the model. At the same time, the spatial topological relationship between the equipment can be constructed, such as determining which devices are connected by channels and the direction of the channels. The BIM building model thus generated completely contains the equipment distribution information and spatial topological relationship of the computer room area.

[0018] As an optional embodiment, generating a BIM building model based on the building structure data set in step 110 further includes:

[0019] Step 111: Perform geometric analysis on the building structure data set to extract the structural contour features and equipment location coordinates of the target building area.

[0020] In the application scenario of the data center computer room, the collected building structure data set can be analyzed one by one through geometric analysis processing. For the walls, floor slabs and other structural parts of the computer room, the structural contour features, such as the length, height, angle and other information of the wall, can be extracted based on the geometric description in the data, thereby determining the overall geometric shape of the computer room. For the extraction of equipment location coordinates, equipment-related information can be filtered out from the data to accurately determine the specific coordinate position of each server cabinet, air conditioning unit and other equipment in the computer room space. For example, the coordinates of server cabinet A in the computer room coordinate system are (x1, y1, z1), and the coordinates of air conditioning unit B are (x2, y2, z2), etc. The above coordinate information accurately locates the position of the equipment in the computer room.

[0021] Step 112: Generate a spatial grid model of the target building area based on the structural contour features, and mark the device identifier corresponding to the device position coordinates in the spatial grid model.

[0022] When generating a spatial grid model, the space of the computer room is divided into small grid cells using a grid generation algorithm based on the previously extracted structural contour features. Taking the data center computer room as an example, the size of the above-mentioned grid cells can be set according to the actual situation and the requirements of analysis accuracy, such as being set to a cube grid cell with a side length of ds meters. Then, in the spatial grid model, based on the previously extracted device location coordinates, the identifier corresponding to each device is marked on the corresponding grid cell position. For example, the identifier corresponding to server cabinet A is "S1", so "S1" is marked on the grid cell where it is located. The identifier corresponding to air-conditioning unit B is "K1", so "K1" is also marked on the corresponding grid cell. In this way, the position of each device can be identified in the spatial grid model.

[0023] Step 113: Determine a heat dissipation information set of the target building area according to the device attribute data associated with the device identifier, wherein the heat dissipation information set includes device power data, heat dissipation efficiency data, and heat radiation range data.

[0024] In this embodiment of the present invention, each device has its corresponding device attribute data stored in a pre-set database. Once a device identifier is obtained, the database query is used to correlate the identifier with the corresponding device attribute data. Taking a server cabinet as an example, the device power data for cabinet S1 can be queried to determine P1 watts, the heat dissipation efficiency data is E1 (the ratio of heat dissipated per unit time to power consumed by the device), and the heat radiation range data can be determined through thermal imaging analysis and other technical means as a spherical area with a radius of R1 meters centered on the cabinet. Similarly, for air conditioner unit K1, its corresponding device power data P2, heat dissipation efficiency data E2, and heat radiation range data (such as the shape and range centered on the air conditioner outlet) can also be obtained. Together, these data constitute the heat dissipation information set for the target building area.

[0025] Step 114: Fusing the spatial grid model with the heat dissipation information set to generate a BIM building model including the device distribution information and the spatial topological relationship, where the spatial topological relationship is used to describe the relative positions and channel connectivity between devices.

[0026] During the data fusion process, a data fusion algorithm is used to match device identifiers in the spatial grid model with those in the cooling information set. For example, using server cabinet S1, the data fusion algorithm determines the location of the label "S1" in the spatial grid model. This location is then correlated with the cooling information for server cabinet S1 in the cooling information set (e.g., power P1, cooling efficiency E1, and heat radiation radius R1). This process is repeated for all equipment in the entire computer room, completing the data fusion between the spatial grid model and the cooling information set. The resulting BIM building model not only includes device distribution information (device locations in the spatial grid model), but also describes the relative positions and channel connectivity between devices using the previously constructed spatial topology. For example, server cabinet S1 is connected to adjacent server cabinet S2 via a channel, and channel width and length information are also included in the model, providing comprehensive data support for subsequent hot and cold aisle analysis.

[0027] Step 120: Perform channel division processing on the BIM building model to generate an initial layout plan for hot and cold channels, perform fluid dynamics analysis based on the initial layout plan to simulate air flow conditions, and obtain fluid distribution data for the hot and cold channels.

[0028] In the data center computer room of an embodiment of the present invention, when performing channel division processing on the BIM building model, channel planning is first performed based on the functional layout of the computer room and the heat dissipation requirements of the equipment. For example, considering the heat dissipation direction of the server cabinet and the air supply direction of the air conditioning unit, the space in the computer room is divided into different areas, and it is determined which areas are used as cold channels and which areas are used as hot channels. Through the channel division algorithm, an initial layout plan for the cold and hot channels is generated. The plan clearly defines the location, direction, and width of the cold and hot channels. For example, the cold channel is arranged along one side of the computer room with a width of D1 meters, and the hot channel is on the other side with a width of D2 meters. Then, based on this initial layout plan, fluid dynamics analysis software is used to simulate the air flow state. The software will set corresponding boundary conditions and parameters based on the channel information in the initial layout plan and the equipment distribution in the BIM building model, perform simulation calculations, and finally obtain the fluid distribution data of the cold and hot channels. The above data reflects the flow of air in the cold and hot channels.

[0029] The step 120 of performing fluid dynamics analysis based on the initial layout plan to simulate air flow conditions and obtain fluid distribution data of the hot and cold aisles includes:

[0030] Step 121: Determine the fluid inlet boundary conditions and the fluid outlet boundary conditions according to the channel division information in the initial layout plan, wherein the boundary conditions include the inlet flow rate, the outlet pressure and the initial value of the temperature.

[0031] In an embodiment of the present invention, the boundary conditions are determined based on the channel division information in the previously generated initial layout plan of the hot and cold channels. For the fluid inlet boundary conditions, the air outlet of the air-conditioning unit is the fluid inlet. By analyzing the performance parameters of the air-conditioning unit and the design requirements of the channel, the inlet flow rate is determined to be V1 meters per second. The determination of this flow rate should take into account the air supply capacity of the air-conditioning unit and the carrying capacity of the channel. The outlet pressure is determined to be P0 Pascal based on factors such as the ventilation system design of the machine room and the atmospheric pressure. The initial temperature value is set according to the ambient temperature of the machine room, for example, it is set to T0 degrees Celsius. For the fluid outlet boundary conditions, the outlet pressure, temperature and other parameters are also determined according to the discharge direction of the hot channel and the design of the ventilation system. For example, the outlet pressure is Pa1 Pascal, and the initial temperature value is T1 degrees Celsius. The above boundary conditions provide accurate initial parameters for subsequent fluid dynamics simulations.

[0032] Step 122: setting obstacle constraint features based on the equipment distribution information in the BIM building model, wherein the obstacle constraint features include equipment geometric shape features and surface roughness features.

[0033] In the data center computer room, the BIM building model includes the distribution information of various equipment such as server cabinets and air-conditioning units. Based on the above information, obstacle constraint features are set. For server cabinets, their geometric shape characteristics can be described as a cuboid with a length of L1 meter, a width of W1 meter, and a height of H1 meter. The surface roughness characteristics can be determined as the roughness coefficient r1 through experimental measurement or reference to relevant standards. For air-conditioning units, their geometric shapes may be more complex, such as irregular box shapes. Their accurate geometric shape characteristics are obtained through technologies such as three-dimensional modeling, and the surface roughness characteristics are also determined as the roughness coefficient r2 accordingly. The above obstacle constraint features are used in fluid dynamics simulations to simulate the obstructions and influences that air encounters when flowing through the equipment, making the simulation results closer to the actual situation.

[0034] Step 123: Determine the air velocity distribution and pressure gradient distribution in the hot and cold channels by combining the obstacle constraint characteristics and the target fluid dynamics control model.

[0035] In an embodiment of the present invention, a preset target fluid dynamics control model is used to integrate previously set obstacle constraint features into the model for calculation. The model simulates the air flow in the hot and cold channels based on the basic principles of fluid mechanics, such as the continuity equation and momentum equation, combined with the geometric shape and surface roughness characteristics of the obstacle, as well as the previously determined boundary conditions. Through iterative calculations and numerical solutions, the air velocity distribution and pressure gradient distribution at different positions in the hot and cold channels are ultimately determined. For example, at one position in the cold channel, the air velocity is calculated to be Vx meters per second, and at another position in the hot channel, the pressure gradient is Gx Pascals per meter. The above distribution information provides detailed data for understanding the flow characteristics of the air in the channel.

[0036] Step 124: Generate the fluid distribution data according to the air velocity distribution and the pressure gradient distribution, wherein the fluid distribution data includes a streamline trajectory set, a vortex region position, and a turbulence intensity index.

[0037] Optionally, fluid distribution data is generated using a relevant data generation algorithm based on the air velocity distribution and pressure gradient distribution calculated previously. For the streamline trajectory set, the algorithm will draw the trajectory of the air flow in the space of the hot and cold channels according to the direction and magnitude of the air velocity, forming a streamline trajectory set. For example, in the cold channel, by analyzing the air velocity at different positions, a series of streamlines representing the direction of air flow are drawn, and the above streamlines constitute the streamline trajectory set. For determining the position of the vortex area, the algorithm will identify the area where vortices are formed in the air flow based on the changes in flow velocity and pressure, and determine its specific position in the hot and cold channels. For the turbulence intensity index, the algorithm will calculate the turbulence intensity index at different positions based on the fluctuation of flow velocity and the relevant turbulence model. The above index reflects the degree of turbulence of the air flow. Finally, the above streamline trajectory set, vortex area position and turbulence intensity index are integrated together to form the fluid distribution data of the hot and cold channels.

[0038] Step 130: performing coupling analysis on the BIM building model and the fluid distribution data to generate air leakage state prediction results and temperature distribution prediction results of the hot and cold channels.

[0039] In an embodiment of the present invention, when coupling and analyzing a BIM building model with fluid distribution data, a corresponding coupling analysis algorithm is utilized. This algorithm correlates and integrates the equipment distribution and heat dissipation information in the BIM building model with the air flow information in the fluid distribution data. By analyzing the heat dissipation of the equipment and the impact of air flow on heat dissipation, predictions of air leakage and temperature distribution in hot and cold aisles are generated. For example, air leakage can be predicted by analyzing whether air flow within hot and cold aisles causes air to leak from the cold aisle to the hot aisle, or vice versa. Furthermore, the temperature distribution of different aisle regions is predicted based on factors such as the equipment's heat dissipation power, the amount of heat removed by air flow, and heat conduction within the aisle.

[0040] In a preferred embodiment, step 130 includes:

[0041] Step 131: extracting heat source features from the equipment distribution information in the BIM building model to obtain heat source intensity features corresponding to each device.

[0042] In data center computer rooms, heat source characteristics are extracted from the equipment distribution information in the BIM building model. Taking server cabinets as an example, heat source intensity characteristics are calculated by analyzing the equipment power and heat dissipation efficiency data. For example, if server cabinet S1 has a power of P1 watt and a heat dissipation efficiency of E1, its heat source intensity characteristics can be calculated by multiplying the power and heat dissipation efficiency, i.e., the heat source intensity is P1 × E1 watt. For other equipment, such as air conditioning units, a similar method is used to calculate the corresponding heat source intensity characteristics based on their device attribute data. These heat source intensity characteristics reflect the amount of heat dissipated by each device per unit time.

[0043] Step 132: Determine the spatial overlap region between the air flow path and the heat source intensity feature based on the streamline trajectory set in the fluid distribution data, wherein the spatial overlap region represents the coverage range of the air flowing through the heat source.

[0044] In an embodiment of the present invention, a spatial overlap region is determined based on a set of streamline trajectories in the fluid distribution data. For example, a set of streamline trajectories shows that air flows along a certain path from the air outlet of an air conditioning unit. When this streamline trajectory intersects with the heat source region of server cabinet S1, the spatial overlap region between the air flow path and the heat source intensity characteristics is determined. By analyzing the specific direction of the streamline trajectory and the range of the heat source, the shape and size of the spatial overlap region can be accurately determined. This spatial overlap region represents the coverage area of ​​air flowing through a heat source such as server cabinet S1, which is of great significance for understanding the range of influence of air on equipment heat dissipation.

[0045] Step 133: Perform thermodynamic equilibrium analysis in the spatial overlap region to obtain heat exchange state characteristics of the hot and cold channels. The heat exchange state characteristics are used to quantify the contribution of air flow to equipment heat dissipation.

[0046] This step performs a thermodynamic equilibrium analysis within the defined spatial overlap region. Using thermodynamic principles and relevant heat exchange models, factors such as air velocity, temperature, and the heat source intensity of the equipment are considered. For example, assume that air flows through the spatial overlap region at a certain velocity Vx, the air temperature is Tx, and the heat source intensity of server cabinet S1 is Q1. Using heat exchange formulas and the law of conservation of energy, the amount of heat exchanged between the air and the equipment within this spatial overlap region is calculated. Based on the heat exchange amount and relevant parameters of the air and equipment, the heat exchange state characteristics of the hot and cold aisles are derived. This characteristic can be expressed as a quantitative indicator, such as the heat exchange efficiency η, which reflects the contribution of air flow to equipment heat dissipation.

[0047] Step 134: Generate an air leakage state prediction model based on the heat exchange state characteristics and the turbulence intensity index, and calculate the air leakage state prediction results of the cold and hot channels using the air leakage state prediction model.

[0048] In an embodiment of the present invention, a leakage state prediction model is generated based on the heat exchange state characteristics and the turbulence intensity index. First, a large amount of historical engineering data is collected. The data includes different heat exchange efficiencies, turbulence intensities, and corresponding leakage rate values. Then, a multivariate linear regression analysis is performed on the data, and the regression analysis algorithm is used to determine the first regression coefficient k1 between the heat exchange state characteristics and the leakage rate, and the second regression coefficient k2 between the turbulence intensity index and the leakage rate. Next, the heat exchange state characteristics are multiplied by the first regression coefficient to obtain the first prediction component, that is, the first prediction component = heat exchange state characteristics × k1. The turbulence intensity index is multiplied by the second regression coefficient to obtain the second prediction component, that is, the second prediction component = turbulence intensity index × k2. Finally, the first prediction component and the second prediction component are superimposed to generate the leakage state prediction results of the hot and cold channels, that is, the leakage state prediction result = first prediction component + second prediction component.

[0049] Exemplarily, generating an air leakage state prediction model based on the heat exchange state characteristics and the turbulence intensity index includes:

[0050] Step 1340: Obtain quantitative information on air leakage status from historical engineering data, where the quantitative information on air leakage status includes leakage rate values ​​corresponding to different heat exchange efficiencies and turbulence intensities; perform a multivariate linear regression analysis on the quantitative information on air leakage status to determine a first regression coefficient between the heat exchange status characteristic and the air leakage rate, and a second regression coefficient between the turbulence intensity index and the air leakage rate; multiply the heat exchange status characteristic by the first regression coefficient to obtain a first prediction component, and multiply the turbulence intensity index by the second regression coefficient to obtain a second prediction component; and perform a superposition operation on the first prediction component and the second prediction component to generate a leakage status prediction result for the hot and cold channels.

[0051] In practice, quantitative information on air leakage status is extracted from historical engineering databases. For example, a set of data is found where, when the heat exchange efficiency is E1, the turbulence intensity is TF1, and the corresponding air leakage rate is L1; when the heat exchange efficiency is E2, the turbulence intensity is TF2, and the corresponding air leakage rate is L2, and so on. This data is input into multivariate linear regression analysis software. The software fits and calculates the data to determine the first regression coefficient k1 between the heat exchange state characteristics and the air leakage rate, and the second regression coefficient k2 between the turbulence intensity index and the air leakage rate. The heat exchange state characteristics of the current hot and cold channels are E, and the turbulence intensity index is TF. The first prediction component is calculated as E×k1, and the second prediction component is TF×k2. Superimposing the two yields the air leakage status prediction result of E×k1+TF×k2.

[0052] Step 135: Generate the temperature distribution prediction result according to the heat conduction relationship between the air flow path and the heat source intensity characteristic, wherein the temperature distribution prediction result reflects the temperature gradient change of different channel areas.

[0053] In an embodiment of the present invention, a temperature distribution prediction result is generated based on the heat conduction relationship between the air flow path and the heat source intensity characteristics. The influence of factors such as the air flow velocity, heat source intensity, and channel material on heat conduction is considered. For example, in a cold channel, air flows from the air conditioning unit to the server cabinet at a certain speed, and the heat emitted by the server cabinet is transferred to the surrounding space through the flow of air and heat conduction. Using the heat conduction equation and the relevant heat transfer model, combined with the air flow path and heat source intensity characteristics, the temperature changes at different locations are calculated. By calculating different areas of the entire hot and cold channels, a temperature distribution prediction result is generated. The result can be represented by a temperature distribution matrix. Each element in the matrix corresponds to the temperature value of a different channel area, thereby clearly reflecting the temperature gradient changes in different channel areas.

[0054] Step 140: Generate a layout optimization strategy based on the air leakage state prediction result and the temperature distribution prediction result, and adjust the channel layout information in the initial layout plan according to the layout optimization strategy to obtain a target layout plan.

[0055] In data center computer rooms, a layout optimization strategy is generated based on the air leakage status prediction results and the temperature distribution prediction results. If the air leakage status prediction results indicate a high air leakage rate in one area, and the temperature distribution prediction results indicate uneven temperatures in certain areas, a corresponding optimization strategy will be necessary. For example, to address air leakage issues, one can consider adjusting the sealing of the aisles; to address uneven temperatures, one can adjust the aisle width or the placement of equipment. Based on these analysis results, an optimization algorithm is used to generate a layout optimization strategy containing parameter adjustment instructions. This strategy is then used to adjust the aisle layout information in the initial layout plan, such as aisle width and aisle inclination angle, to ultimately obtain the target layout plan to improve the performance of hot and cold aisles.

[0056] In an optional embodiment, generating a layout optimization strategy based on the air leakage state prediction result and the temperature distribution prediction result includes:

[0057] Step 141: Identify the air leakage potential area in the initial layout plan according to the air leakage state prediction result, and extract the channel boundary geometry vector of the air leakage potential area.

[0058] In an embodiment of the present invention, an analysis is performed based on the air leakage status prediction results. If the air leakage status prediction results indicate that the air leakage rate in one area exceeds a preset standard, then that area is a potential air leakage area. For example, at the junction of hot and cold aisles, the air leakage status prediction results indicate a high air leakage rate, thus determining that this area is a potential air leakage area. Then, using a geometric analysis algorithm, the channel boundary geometric vector of this potential air leakage area is extracted. For this potential air leakage area at the junction of hot and cold aisles, the channel boundary may be irregular in shape, but it can be approximated as a polygon using mathematical methods. For example, the channel boundary is divided into multiple line segments, each of which can be represented by a vector. For example, if the channel boundary is composed of three points PA, PB, and PC connected in sequence, then the line segment PAPB can be represented as the vector PAPB (its coordinates are the coordinates of point PB minus the coordinates of point PA), and the line segment PBPC can be represented as the vector PBPC (its coordinates are the coordinates of point PC minus the coordinates of point PB). Together, these vectors constitute the channel boundary geometric vector of the potential air leakage area. These vectors contain information such as the direction and length of the channel boundary, providing an important data foundation for subsequent layout optimization.

[0059] Step 142: determining the temperature abnormality area in the hot and cold channels based on the temperature distribution prediction result, and calculating the thermal radiation superposition range of adjacent devices according to the spatial coordinates of the temperature abnormality area.

[0060] In a data center computer room, analysis is performed based on the predicted temperature distribution results. The results are presented in matrix form. By comparing and analyzing the temperature values ​​represented by each element in the matrix, if the temperature of a region deviates significantly from the overall average temperature, exceeding the preset normal range, that region is identified as a temperature anomaly. For example, if the temperature in a corner of a hot aisle is significantly higher than other areas, that corner is considered a temperature anomaly region. Then, based on the spatial coordinates of the temperature anomaly region, the devices located in and around that region are identified. For example, if server cabinets S3 and S4 are located within that region, a spatial geometry algorithm is used to calculate the overlapping thermal radiation ranges of adjacent devices using the previously determined device thermal radiation range data (e.g., the thermal radiation range of server cabinet S3 is a spherical region with a radius of R3 at its center, and the thermal radiation range of S4 is a spherical region with a radius of R4 at its center) and their spatial coordinates. To calculate the overlapping thermal radiation ranges of two spherical regions, the distance d between the two sphere centers (i.e., the device locations) is first calculated. Then, based on geometric relationships, the intersection of the two spheres is determined. If d is less than R3+R4, the two spheres intersect. The shape and range of the intersection are determined through geometric calculations (such as using geometric models such as spherical caps and spherical segments). This intersection is the overlapping range of thermal radiation from adjacent devices. Determining this range helps understand the degree of thermal interference between devices and provides a basis for optimizing layout.

[0061] Step 143: Obtain a multi-objective optimization function including minimizing the air leakage rate index, maximizing the temperature uniformity index, and minimizing the equipment thermal interference index.

[0062] In the data center room layout optimization scenario, a multi-objective optimization function is required to comprehensively consider multiple performance indicators to optimize the hot and cold aisle layout. Minimizing the air leakage rate metric aims to reduce air leakage between the hot and cold aisles and improve energy efficiency. The air leakage rate can be defined as a function related to the predicted air leakage status. For example, let the air leakage rate be L, which can be the ratio of the total leakage flow rate at the junction of the hot and cold aisles and other potential leakage areas to the total air supply volume. Layout optimization can be used to minimize L. Maximizing the temperature uniformity metric requires minimizing temperature differences between areas within the hot and cold aisles to ensure stable equipment operation. Temperature uniformity can be measured using the temperature standard deviation σ, with smaller σ indicating more uniform temperature. Layout parameters can be adjusted to minimize σ. Minimizing the device thermal interference metric aims to reduce the mutual influence of thermal radiation between adjacent devices and ensure efficient heat dissipation. Device thermal interference can be defined as the sum of the heat within the overlapping range of thermal radiation from adjacent devices, denoted as H. H can be minimized by optimizing device spacing and layout. These three indicators are combined to construct a multi-objective optimization function F, for example, F = q1 × L + q2 × σ + q3 × H, where q1, q2, and q3 are weight coefficients, respectively indicating the importance of each indicator in the overall optimization. They are set according to actual needs, such as q1 = 0.3, q2 = 0.3, and q3 = 0.4. By adjusting the above weights, the optimization priorities of different indicators can be balanced, thereby achieving comprehensive optimization of the hot and cold channel layout.

[0063] Step 144: adjusting the channel boundary geometric vector and the thermal radiation superposition range through an iterative optimization algorithm so that the multi-objective optimization function reaches a preset convergence condition and generates the layout optimization strategy.

[0064] During the data center room layout optimization process, an iterative optimization algorithm is used to adjust the channel boundary geometry vectors and the thermal radiation superposition range. First, relevant parameters are initialized. For example, the channel boundary geometry vectors are set to initial values ​​(these initial values ​​can be determined based on the channel boundary information in the initial layout plan), and the thermal radiation superposition range is also set to initial values ​​based on previously calculated results. Then, the value of the multi-objective optimization function F is calculated at the current parameter values ​​(i.e., the current channel boundary geometry vectors and thermal radiation superposition range). Next, an iterative optimization algorithm (such as a gradient descent algorithm) is used to calculate the gradient vector of the multi-objective optimization function F at the current parameters. The gradient vector indicates the direction of the fastest increase in the function value, while the opposite direction indicates the direction of the fastest decrease in the function value, which is the direction of parameter adjustment. Following this gradient vector's direction, the channel boundary geometry vectors and thermal radiation superposition range are gradually updated at a predetermined step size. After each update, the value of the multi-objective optimization function F is recalculated and checked to see if it meets the preset convergence criteria. The preset convergence criteria can be that the change in the multi-objective optimization function F is less than a preset threshold, such as ε. If the change in F is less than ε, the algorithm is considered to have converged. The layout adjustment plan corresponding to the parameter values ​​obtained at this time (i.e., the updated channel boundary geometry vector and thermal radiation superposition range) is the generated layout optimization strategy. If the convergence condition is not met, the next round of iterative calculation is continued, and the parameters are continuously adjusted until the convergence condition is met.

[0065] In detail, step 144 may include:

[0066] Step 1440: Initialize the channel boundary geometry vector as a first initial feature parameter set, and the thermal radiation superposition range as a second initial feature parameter set; calculate the gradient vector of the multi-objective optimization function at the first initial feature parameter set and the second initial feature parameter set, wherein the gradient vector indicates the parameter adjustment direction; gradually update the first initial feature parameter set and the second initial feature parameter set along the direction of the gradient vector until the change of the multi-objective optimization function is less than a set threshold; generate a layout optimization strategy containing parameter adjustment instructions based on the updated first initial feature parameter set and the second initial feature parameter set.

[0067] In actual optimization of a data center room, the channel boundary geometry vectors are set as the first initial feature parameter set, Set1. For example, Set1 contains the initial coordinate values ​​of multiple channel boundary vectors, determined based on the initial layout plan. The thermal radiation overlap range is set as the second initial feature parameter set, Set2. Set2 may contain parameters such as the initial shape and size of the thermal radiation overlap range of each device. Next, using specialized optimization software or a programmatic algorithm, the gradient vector G of the multi-objective optimization function F at Set1 and Set2 is calculated. This gradient vector G is a multidimensional vector, with each dimension corresponding to the direction of change of a parameter. For example, for a vector coordinate parameter x in the channel boundary geometry vectors, the corresponding dimension value in the gradient vector G represents the degree and direction of the impact of a change in x on the multi-objective optimization function F. Next, Set1 and Set2 are updated in the opposite direction of the gradient vector G, with a certain step size α (e.g., α = 0.1). For example, for one of the vector coordinate parameters x in Set1, the updated value of x is x-α×Gx (Gx is the dimension value corresponding to x in the gradient vector G). Similar update operations are performed on all parameters in Set1 and Set2 to obtain new parameter sets Set1' and Set2'. The values ​​of the multi-objective optimization function F at Set1' and Set2' are calculated and compared with the F value calculated in the previous round to calculate the change ΔF. If ΔF is less than the set threshold (such as the set threshold is 0.01), the algorithm is considered to have converged. At this time, a layout optimization strategy containing parameter adjustment instructions is generated based on Set1' and Set2'. For example, based on the channel boundary geometry vector information in Set1', instructions for adjusting the channel width and tilt angle are generated. Based on the thermal radiation superposition range information in Set2', instructions for adjusting the device spacing, etc. are generated. The above instructions constitute the layout optimization strategy.

[0068] In a preferred embodiment, adjusting the channel layout information in the initial layout solution according to the layout optimization strategy to obtain a target layout solution includes:

[0069] Step 145: Modify the channel width and channel inclination angle of the initial layout solution according to the parameter adjustment instruction in the layout optimization strategy to obtain a modified channel layout solution.

[0070] In an embodiment of the present invention, operations are performed according to the parameter adjustment instructions in the layout optimization strategy. For example, the instructions in the layout optimization strategy require that the width of the cold channel be increased from the original 2 meters to 2.5 meters, the width of the hot channel be reduced from 1.5 meters to 1.2 meters, and the inclination angle of one of the channels be adjusted from 30 degrees to 45 degrees. According to the above instructions, the initial layout plan is modified. During the modification process, professional layout design software is used, and the software can adjust the geometric shape of the channel according to the input parameters. For the modification of the channel width, the software will expand or shrink the boundary of the corresponding channel according to the instructions. For the adjustment of the channel inclination angle, the software will use the selected point as the rotation center and rotate the channel according to the angle value of the instruction, thereby obtaining a modified channel layout plan, which meets the requirements of the layout optimization strategy in terms of channel width and inclination angle, and lays the foundation for further optimizing the performance of hot and cold channels.

[0071] Step 146: Re-determine the device spacing feature in the modified channel layout scheme to obtain a channel layout division scheme, where the device spacing feature is used to constrain the heat radiation ranges between adjacent devices to have no overlapping areas.

[0072] After obtaining the revised aisle layout, the device spacing characteristics are redefined. Since the aisle layout has changed, the relative positions of the devices have also changed, necessitating a reconsideration of device spacing to prevent overlap in the heat radiation ranges of adjacent devices. Taking server cabinets as an example, by analyzing the heat radiation range of each server cabinet (e.g., a spherical area with a radius of R centered on the cabinet), combined with the revised aisle layout, a spatial geometry algorithm is used to determine the appropriate device spacing. Under the new aisle layout, the minimum spacing between two adjacent server cabinets should ensure that their heat radiation ranges do not overlap. The device spacing is determined by calculating the boundary distance between the heat radiation ranges of the two cabinets and considering factors such as space utilization and air flow within the aisle. The device spacing characteristics are redefined for all devices in the entire computer room using this method, ultimately resulting in a channel layout partitioning scheme. This scheme not only accounts for changes in the aisle layout but also effectively avoids thermal interference between adjacent devices through reasonable device spacing constraints, thereby improving the heat dissipation efficiency of the computer room.

[0073] Step 147: Perform fluid dynamics verification analysis on the channel layout division scheme to obtain a verification result. If the verification result meets the preset air leakage tolerance threshold and temperature uniformity threshold, generate the target layout scheme; otherwise, re-execute the coupling analysis steps until the verification conditions are met.

[0074] In an embodiment of the present invention, a fluid dynamics validation analysis is performed on the channel layout scheme. First, the channel layout scheme is converted into the boundary condition input information required for fluid dynamics analysis. For example, the new channel dimensions (such as the modified channel width and length) and the new device location information are accurately input into the fluid dynamics analysis software. Next, the meshing parameters for the fluid dynamics simulation are set based on the channel geometry complexity. If the channel geometry is complex, a finer meshing may be required, such as setting the mesh cell side length to 0.2 meters. If the channel geometry is relatively simple, the mesh cell size can be appropriately increased, such as to 0.5 meters. Based on these meshing parameters, a fluid dynamics simulation is performed in the fluid dynamics analysis software. During the simulation, the software calculates the air flow within the channel based on the set boundary conditions and meshing, and outputs leakage status verification data and temperature distribution verification data. The leakage status verification data is compared and analyzed with a preset leakage tolerance threshold. For example, the preset leakage tolerance threshold is 5%. If the leakage status verification data indicates that the actual leakage rate is less than 5%, the leakage meets the requirements. At the same time, a consistency analysis is performed on the temperature distribution verification data and the preset temperature uniformity threshold. For example, if the preset temperature uniformity threshold requires a temperature standard deviation of no more than 3 degrees Celsius, then if the temperature standard deviation calculated from the temperature distribution verification data is less than 3 degrees Celsius, the temperature uniformity requirement is met. If the air leakage status verification data is less than the air leakage tolerance threshold and the temperature distribution verification data meets the temperature uniformity threshold, the verification result is confirmed to meet the preset conditions and the target layout plan is generated. Otherwise, it indicates that there are still problems with the current channel layout division plan, and the coupling analysis step needs to be repeated. That is, the BIM building model and fluid distribution data are coupled and analyzed again, and the layout plan needs to be adjusted until the verification conditions are met.

[0075] In an exemplary embodiment, performing a fluid dynamics verification analysis on the channel layout division scheme to obtain a verification result includes:

[0076] Step 1470: Convert the target layout scheme into boundary condition input information required for fluid dynamics analysis, the boundary condition input information including updated channel size and equipment location information; set the meshing parameters of fluid dynamics simulation based on the channel geometry complexity; perform fluid dynamics simulation based on the meshing parameters and the meshing parameters, and output air leakage status verification data and temperature distribution verification data; compare and analyze the air leakage status verification data with the air leakage tolerance threshold, and perform consistency analysis on the temperature distribution verification data and the temperature uniformity threshold; if the air leakage status verification data is less than the air leakage tolerance threshold and the temperature distribution verification data meets the temperature uniformity threshold, confirm that the verification result meets the preset air leakage tolerance threshold and temperature uniformity threshold.

[0077] During the actual verification of the data center room, the target layout aisle dimensions (e.g., changing the cold aisle width to 2.5 meters and the hot aisle width to 1.2 meters) and equipment location information (e.g., moving server cabinet S1 to a new coordinate location) were accurately formatted into the required format for the fluid dynamics analysis software and used as boundary condition input. Meshing parameters were determined based on the complexity of the aisles. For example, if the aisle contained numerous corners and obstacles, a finer mesh was created, with a cell side length of 0.2 meters. These meshing parameters were input into the fluid dynamics analysis software, and the simulation was started. After a series of calculations, the software outputs leakage verification data (e.g., an actual leakage rate of 3%) and temperature distribution verification data (e.g., a calculated temperature standard deviation of 2°C). The leakage verification data was compared with the preset leakage tolerance threshold (set to 5%). If 3% was less than 5%, the leakage level met the requirements. The temperature distribution verification data was then analyzed for consistency with the preset temperature uniformity threshold (requiring a temperature standard deviation of no more than 3°C). If 2°C was less than 3°C, the temperature uniformity also met the requirements. Since the air leakage status verification data is less than the air leakage tolerance threshold and the temperature distribution verification data meets the temperature uniformity threshold, the verification result is confirmed to meet the preset air leakage tolerance threshold and temperature uniformity threshold. At this time, the target layout plan has passed the verification and can be applied to the actual computer room construction or optimization.

[0078] As an independently implementable technical solution, after adjusting the channel layout information in the initial layout solution according to the layout optimization strategy to obtain the target layout solution, it also includes: deploying a temperature sensor array and an air flow velocity sensor array based on the target layout solution to collect the actual temperature distribution data and actual air flow velocity data of the hot and cold channels in real time; performing residual calculation on the actual temperature distribution data and the temperature distribution prediction result to generate a temperature prediction error distribution map, and extracting abnormal monitoring areas where the temperature deviation exceeds a preset threshold based on the error distribution map; performing dynamic path matching analysis based on the actual air flow velocity data and the streamline trajectory set in the fluid distribution data to identify channel sections where the flow velocity matching is lower than the preset matching degree; generating feedback optimization instructions based on the abnormal monitoring area and the channel section, and adjusting the equipment spacing characteristics and channel inclination angle in the target layout solution to generate a first optimized layout solution.

[0079] In the data center computer room, sensor arrays are deployed based on the target layout plan. Temperature sensor arrays and airflow velocity sensor arrays are installed at key locations in the hot and cold aisles. For example, a temperature sensor is installed every 2 meters in the cold aisle and every 3 meters in the hot aisle. Airflow velocity sensors are also installed at different heights in the aisles to comprehensively collect actual temperature distribution data and actual air flow velocity data. After collecting this data, a residual is calculated between the actual temperature distribution data and the previously generated temperature distribution prediction results. For example, at a location in the cold aisle, the actual temperature is Tactual, and the temperature at that location in the temperature distribution prediction result is TPredicted. The residual is Tactual - TPredicted. By performing this calculation for each location in the hot and cold aisles, a temperature prediction error distribution map is generated. A preset threshold, such as 5 degrees Celsius, is set. Based on the error distribution map, areas with temperature deviations exceeding 5 degrees Celsius are identified as abnormal monitoring areas. The actual air flow velocity data is dynamically matched with the streamline trajectory set in the fluid distribution data. For example, the theoretical air flow rate and direction of one of the channel sections are determined based on the set of streamline trajectories, and the actual air flow rate data is compared with it to calculate the flow rate matching degree. If the preset matching degree is 80%, when the flow rate matching degree of one of the channel sections is lower than 80%, the channel section is identified as abnormal. Based on the above abnormal monitoring areas and channel sections, feedback optimization instructions are generated. For example, if it is found that the temperature in one of the abnormal monitoring areas is too high due to the superposition of heat radiation caused by the small equipment spacing, an instruction to increase the equipment spacing is generated; if the flow rate matching degree of one of the channel sections is low because the channel inclination angle is unreasonable, an instruction to adjust the channel inclination angle is generated. According to the above instructions, the equipment spacing characteristics and channel inclination angles in the target layout plan are adjusted to generate a first optimized layout plan, thereby further optimizing the performance of the hot and cold channels.

[0080] As an independently implementable technical solution, after adjusting the channel layout information in the initial layout solution according to the layout optimization strategy to obtain the target layout solution, it also includes: calculating the thermal radiation superposition influence factor of each device based on the device distribution information in the target layout solution, the thermal radiation superposition influence factor is determined by the product of the overlapping area of ​​the thermal radiation range of adjacent devices and the device heat dissipation power; constructing a thermal environment energy efficiency evaluation matrix based on the thermal radiation superposition influence factor and the temperature distribution prediction result, each element in the matrix represents the unit area heat dissipation efficiency and energy consumption ratio of the corresponding channel area; dividing the thermal environment energy efficiency evaluation matrix into regions based on a preset energy efficiency threshold, and screening out inefficient channel areas with energy efficiency values ​​lower than the energy efficiency threshold; adjusting the equipment installation direction or adding auxiliary heat dissipation devices according to the spatial coordinates of the inefficient channel area to generate a second optimized layout solution.

[0081] In this embodiment of the present invention, the heat radiation superposition impact factor is calculated based on the device distribution information in the target layout plan. Taking server cabinets as an example, the adjacent devices of each server cabinet are first determined. For example, server cabinet S1's adjacent devices are S2 and S3. Based on the previously determined device heat radiation range data, the overlapping areas of the heat radiation ranges of S1, S2, and S3 are calculated. For example, through spatial geometry calculations, the overlapping areas of the heat radiation ranges of S1 and S2 are A12, and the overlapping areas of the heat radiation ranges of S1 and S3 are A13. Given that the heat dissipation power of S1 is P1, the heat radiation superposition impact factor of S1 is (A12 × P1) + (A13 × P1). This calculation is repeated for all devices to obtain the heat radiation superposition impact factor for each device. Then, a thermal environment energy efficiency assessment matrix is ​​constructed based on these heat radiation superposition impact factors and the temperature distribution prediction results. The rows and columns of the matrix correspond to different channel areas, and each element in the matrix represents the heat dissipation efficiency per unit area to the energy consumption ratio of the corresponding channel area. For example, for a channel area, the heat dissipation efficiency per unit area can be calculated by dividing the total heat dissipated by the equipment in that area by the area of ​​the area. Energy consumption is determined by the energy required by air conditioners and other equipment to cool the area. The ratio of the two is the value of the element. Next, the thermal environment energy efficiency assessment matrix is ​​divided into regions based on a preset energy efficiency threshold. For example, if the preset energy efficiency threshold is E0, each element in the matrix is ​​traversed, and channel regions corresponding to elements with an E0 value less than E0 are selected. These regions are designated as inefficient channel regions. For these selected inefficient channel regions, the equipment involved is identified based on their spatial coordinates. If an inefficient channel region is caused by improper equipment installation orientation, resulting in interference in heat radiation and affecting heat dissipation efficiency and energy consumption ratio, the installation orientation of the relevant equipment is adjusted to optimize air flow and heat dissipation. If the low energy efficiency of the region is due to insufficient heat dissipation capacity, auxiliary heat dissipation devices are added, such as small fans installed near the equipment. Through these operations, a second optimized layout solution is generated, further improving the thermal environment energy efficiency of the hot and cold aisles in the data center computer room.

[0082] As an independently implementable technical solution, after adjusting the channel layout information in the initial layout plan according to the layout optimization strategy to obtain the target layout plan, it also includes: obtaining real-time equipment load data of the target building area, and generating a dynamic heat source intensity correction coefficient according to the load change cycle; fusing the dynamic heat source intensity correction coefficient with the heat dissipation information set in the BIM building model, and updating the equipment power data and heat radiation range data; dynamically dividing the channel partitions of the target layout plan based on the updated heat dissipation information set, adjusting the channel width priority and the isolation distance between hot and cold channels according to the peak period of heat source intensity; performing periodic fluid dynamics verification based on the dynamically divided channel partitions, and generating a flexible channel layout plan that adapts to load fluctuations.

[0083] In a data center computer room, a dedicated data acquisition system acquires real-time equipment load data for the target building area. This data acquisition system, connected to the monitoring ports of each device, monitors the load conditions of equipment such as server cabinets and air conditioning units in real time. For example, real-time load data for server cabinet S1 can be obtained through its internal monitoring module. This load data may be presented in the form of operating power values, task processing capacity, and other information. Based on this real-time equipment load data, its load variation cycle is analyzed. For example, after a period of monitoring, it was found that server cabinet S1 has a high load between 10:00 AM and 4:00 PM daily, exhibiting a certain periodic pattern. Using a corresponding algorithm, a dynamic heat source intensity correction factor is generated based on the load variation cycle. For example, by analyzing historical load data and current real-time load data, the algorithm considers factors such as peak load, average load, and load variation trends to determine the dynamic heat source intensity correction factor k1 for server cabinet S1 during peak load periods.

[0084] The generated dynamic heat source intensity correction factor is merged with the cooling information set in the BIM building model. For server cabinet S1, its original device power data is P1, and the heat radiation range data is a spherical area with a radius of R1 centered on the cabinet. During fusion, the device power data is updated to P1 × k1, and the heat radiation range data is also adjusted based on the correction factor and the relevant physical model. For example, the heat radiation range radius becomes R1 × k1' (k1' is an adjustment factor determined based on the thermal radiation principle and the correction factor). This fusion operation is repeated for all devices, thus updating the cooling information set in the entire BIM building model.

[0085] Dynamically partition the target layout plan's aisles based on the updated cooling information. Analyze the cooling requirements of different devices during peak heat intensity periods. For example, during peak server cabinet load periods, the heat intensity of multiple cabinets increases significantly, requiring more aisle space to ensure air flow and heat dissipation. Therefore, aisle width priority is adjusted, prioritizing increased aisle width in areas with high cooling requirements. Furthermore, the spacing between hot and cold aisles is adjusted to prevent excessive mixing of hot and cold air, which could affect cooling efficiency.

[0086] Periodic fluid dynamics verification is performed based on the dynamically partitioned channel zones. A verification cycle is set, such as every hour. During each verification, the dynamically partitioned channel zones are converted into boundary condition inputs required for fluid dynamics analysis, including updated channel dimensions, equipment locations, and new heat dissipation information. Meshing parameters for the fluid dynamics simulation are set based on the channel geometry complexity, and a fluid dynamics simulation is performed. After the simulation, leakage status verification data and temperature distribution verification data are output. This verification data is analyzed to check whether it meets pre-defined performance indicators, such as leakage rate and temperature uniformity. If the verification results meet the requirements, the current channel layout can be used. If not, the channel layout is further adjusted based on the verification results, and fluid dynamics verification is re-performed until a flexible channel layout that adapts to load fluctuations is generated. This flexible channel layout can adjust the channel layout in real time based on dynamic changes in equipment load, ensuring optimal heat dissipation and air flow in the data center room under varying load conditions.

[0087] It should be noted that, when implementing the embodiments of the present invention, those skilled in the art may rely on existing mature technology systems to perform algorithm adaptation and engineering implementation of each link.

[0088] First, during the BIM modeling phase, the building information modeling (BIM) software's built-in geometry engine was used to analyze building structural data. The IFC standard protocol was used to achieve semantic alignment and topological reconstruction of multi-source data, and point cloud data processing technology was combined to extract equipment spatial coordinates. Fluid dynamics analysis employed computational fluid dynamics (CFD) methods based on the Reynolds-averaged Navier-Stokes equations. The standard k-ε turbulence model was used to simulate air flow. The governing flow equations were discretized using the finite volume method, and the inlet velocity and outlet pressure were defined using the built-in boundary condition setting module in commercial CFD software (such as ANSYS Fluent).

[0089] Secondly, thermodynamic analysis utilizes a coupled calculation method combining the energy conservation equation with Fourier's law of heat conduction, combined with equipment heat source intensity data to simulate the heat transfer process. A multivariate linear regression algorithm is used to construct the air leakage prediction model. The least squares method is used to fit the variable relationships in the historical data, and standardized preprocessing is used to eliminate dimensional differences. A multi-objective optimization process employs a gradient descent algorithm or a genetic algorithm, with parameter sensitivity analysis determining the iteration direction. Pareto frontier theory is then incorporated to balance conflicts among multiple optimization objectives.

[0090] Sensor data processing then incorporates a Kalman filter algorithm to suppress noise, and dynamic time warping (DTW) technology is employed to analyze the match between measured data and predicted streamlines. The Monte Carlo radiation heat transfer method is used to calculate the overlap range of equipment thermal radiation, and ray tracing is used to simulate the heat radiation path. Dynamic load response utilizes time series prediction algorithms (such as LSTM neural networks) to analyze equipment load cycle characteristics, and combines real-time data stream processing technology to dynamically modify heat source parameters.

[0091] In addition, all physical quantity calculations follow the International System of Units, and key parameter assignments refer to industry standard manuals (such as the ASHRAE Thermal Environment Design Guide).

[0092] It is worth mentioning that the embodiment of the present invention constructs a BIM model including equipment distribution and spatial topological relationship through building information modeling technology, combines computational fluid dynamics method to simulate air flow in hot and cold channels, uses finite volume method to discretely solve Navier-Stokes equation to obtain fluid distribution data, establishes quantitative relationship between equipment heat source intensity and air flow through thermodynamic coupling analysis, innovatively proposes a channel layout optimization method based on thermal radiation superposition range calculation and multi-objective optimization function (including air leakage rate, temperature uniformity and thermal interference index), adopts gradient descent algorithm to iteratively optimize channel geometric parameters, and constructs a channel layout optimization method based on dynamic sensor data feedback. The closed-loop optimization system forms a complete technical system covering data collection, model construction, simulation analysis, parameter optimization and verification implementation. The technical means involved, such as the spatial grid generation algorithm, k-ε turbulence model, and Monte Carlo thermal radiation calculation method, all have clear physical meanings and engineering implementation paths. By setting grid division standards, heat source intensity calculation formulas and verification thresholds, the repeatable implementation of the technical solution is achieved. It creatively solves technical problems such as air leakage control, temperature balance adjustment and dynamic load adaptation in the layout of hot and cold channels in data center computer rooms, significantly improves the heat dissipation efficiency of equipment and reduces energy consumption, which is a substantial technical improvement to the building environment control method.

[0093] In summary, the embodiment of the present invention first generates a BIM building model containing equipment distribution information and spatial topological relationships by acquiring a set of building structure data, and divides the BIM building model into channels and combines it with fluid dynamics analysis to simulate the air flow state and obtain fluid distribution data; secondly, the BIM building model is coupled with the fluid distribution data for analysis to predict the air leakage state and temperature distribution, and through cross-domain data fusion analysis, it provides strong support for accurately grasping the conditions of hot and cold channels; then, based on the prediction results, a layout optimization strategy is generated and the channel layout is adjusted to obtain the target layout solution, which effectively improves the rationality of the hot and cold channel layout, optimizes the air flow and temperature distribution inside the building, improves the overall functionality and comfort of the building, and realizes a more efficient and scientific building layout planning method.

[0094] See also Figure 2 As shown, this figure is a schematic diagram of the basic structure of a hot and cold channel simulation analysis system 200 provided by an embodiment of the present invention. The hot and cold channel simulation analysis system 200 includes:

[0095] Processor 201;

[0096] a storage device 202 having a computer program 2020 stored thereon;

[0097] When the computer program 2020 is executed by the processor 201, the processor 201 implements any of the hot and cold channel simulation analysis methods based on BIM and CFD.

[0098] Based on the above, a readable storage medium is provided, on which a program or instruction is stored. When the program or instruction is executed by a processor, the steps of the above method are implemented.

[0099] It should be noted that the various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. Reference can be made to the common and similar parts between the various embodiments. For the systems or devices disclosed in the embodiments, since they correspond to the methods disclosed in the embodiments, the description is relatively simple, and the relevant parts can be referred to the method description.

Claims

1. A cold and hot channel simulation analysis method based on BIM and CFD, characterized by: include: Acquire a building structure data set of a target building area, and generate a BIM building model based on the building structure data set, wherein the BIM building model includes equipment distribution information and spatial topology relationships of the target building area; Performing channel division processing on the BIM building model to generate an initial layout plan for hot and cold channels, performing fluid dynamics analysis based on the initial layout plan to simulate air flow conditions, and obtaining fluid distribution data for the hot and cold channels; Coupling the BIM building model with the fluid distribution data for analysis to generate air leakage state prediction results and temperature distribution prediction results for the hot and cold aisles; generating a layout optimization strategy based on the air leakage state prediction result and the temperature distribution prediction result, and adjusting the channel layout information in the initial layout plan according to the layout optimization strategy to obtain a target layout plan; The adjusting the channel layout information in the initial layout solution according to the layout optimization strategy to obtain a target layout solution includes: Modify the channel width and channel inclination angle of the initial layout scheme according to the parameter adjustment instruction in the layout optimization strategy to obtain a modified channel layout scheme; Re-determining the device spacing feature in the modified channel layout scheme to obtain a channel layout division scheme, wherein the device spacing feature is used to constrain the heat radiation ranges between adjacent devices to have no overlapping areas; Performing a fluid dynamics verification analysis on the channel layout division scheme to obtain a verification result, and if the verification result meets a preset air leakage tolerance threshold and a temperature uniformity threshold, generating the target layout scheme; Otherwise, the coupling analysis steps are executed again until the verification condition is met.

2. The hot and cold channel simulation analysis method based on BIM and CFD according to claim 1 is characterized in that: Generating a BIM building model based on the building structure data set includes: Performing geometric analysis on the building structure data set to extract structural contour features and equipment location coordinates of the target building area; generating a spatial grid model of the target building area based on the structural contour features, and marking a device identifier corresponding to the device position coordinates in the spatial grid model; Determining a heat dissipation information set of the target building area according to the device attribute data associated with the device identifier, the heat dissipation information set including device power data, heat dissipation efficiency data, and heat radiation range data; The spatial grid model is data-fused with the heat dissipation information set to generate a BIM building model including the device distribution information and the spatial topological relationship, where the spatial topological relationship is used to describe the relative positions and channel connectivity between devices.

3. The hot and cold channel simulation analysis method based on BIM and CFD according to claim 2 is characterized in that: The performing of fluid dynamics analysis based on the initial layout scheme to simulate air flow conditions and obtain fluid distribution data of the hot and cold aisles includes: Determining fluid inlet boundary conditions and fluid outlet boundary conditions according to the channel division information in the initial layout scheme, wherein the boundary conditions include initial values ​​of inlet flow rate, outlet pressure, and temperature; Setting obstacle constraint features based on equipment distribution information in the BIM building model, wherein the obstacle constraint features include equipment geometric shape features and surface roughness features; Determining the air velocity distribution and pressure gradient distribution in the hot and cold aisles by combining the obstacle constraint characteristics and the target fluid dynamics control model; The fluid distribution data is generated according to the air velocity distribution and the pressure gradient distribution. The fluid distribution data includes a streamline trajectory set, a vortex region position, and a turbulence intensity index.

4. The hot and cold channel simulation analysis method based on BIM and CFD according to claim 3 is characterized in that: The coupling analysis of the BIM building model and the fluid distribution data to generate the air leakage state prediction results and temperature distribution prediction results of the hot and cold channels includes: Extracting heat source characteristics from the equipment distribution information in the BIM building model to obtain heat source intensity characteristics corresponding to each device; determining a spatial overlap region between an air flow path and the heat source intensity feature based on a set of streamline trajectories in the fluid distribution data, wherein the spatial overlap region represents a coverage range of the air flowing through the heat source; Performing thermodynamic equilibrium analysis in the spatial overlap region to obtain heat exchange state characteristics of the hot and cold channels, wherein the heat exchange state characteristics are used to quantify the contribution of air flow to heat dissipation of the equipment; generating an air leakage state prediction model based on the heat exchange state characteristics and the turbulence intensity index, and calculating the air leakage state prediction results of the cold and hot channels by using the air leakage state prediction model; The temperature distribution prediction result is generated according to the heat conduction relationship between the air flow path and the heat source intensity characteristic, and the temperature distribution prediction result reflects the temperature gradient change of different channel areas.

5. The hot and cold channel simulation analysis method based on BIM and CFD according to claim 4 is characterized in that: The generating of the air leakage state prediction model based on the heat exchange state characteristics and the turbulence intensity index includes: Obtaining quantitative information on air leakage status from historical engineering data, wherein the quantitative information on air leakage status includes air leakage rate values ​​corresponding to different heat exchange efficiencies and turbulence intensities; Performing a multivariate linear regression analysis on the air leakage state quantitative information to determine a first regression coefficient between the heat exchange state characteristic and the air leakage rate, and a second regression coefficient between the turbulence intensity index and the air leakage rate; Performing a product operation on the heat exchange state characteristic and the first regression coefficient to obtain a first prediction component, and performing a product operation on the turbulence intensity index and the second regression coefficient to obtain a second prediction component; The first prediction component and the second prediction component are superimposed to generate a prediction result of the air leakage state of the cold and hot channels.

6. The hot and cold channel simulation analysis method based on BIM and CFD according to claim 1 is characterized in that: Generating a layout optimization strategy based on the air leakage state prediction result and the temperature distribution prediction result includes: Identifying air leakage potential areas in the initial layout plan based on the air leakage state prediction result, and extracting channel boundary geometric vectors of the air leakage potential areas; Determine the temperature abnormality area in the hot and cold channels based on the temperature distribution prediction result, and calculate the thermal radiation superposition range of adjacent devices according to the spatial coordinates of the temperature abnormality area; Obtain a multi-objective optimization function including minimizing the air leakage rate index, maximizing the temperature uniformity index, and minimizing the equipment thermal interference index; The channel boundary geometric vector and the thermal radiation superposition range are adjusted through an iterative optimization algorithm so that the multi-objective optimization function reaches a preset convergence condition and the layout optimization strategy is generated.

7. The hot and cold channel simulation analysis method based on BIM and CFD according to claim 6 is characterized in that: The step of adjusting the channel boundary geometric vector and the thermal radiation superposition range by an iterative optimization algorithm so that the multi-objective optimization function reaches a preset convergence condition and generating the layout optimization strategy includes: Initializing the channel boundary geometric vector as a first initial characteristic parameter set, and the thermal radiation superposition range as a second initial characteristic parameter set; Calculating a gradient vector of the multi-objective optimization function at the first initial feature parameter set and the second initial feature parameter set, the gradient vector indicating a parameter adjustment direction; Stepwise updating the first initial feature parameter set and the second initial feature parameter set along the gradient vector direction until a change in the multi-objective optimization function is less than a set threshold; A layout optimization strategy including parameter adjustment instructions is generated according to the updated first initial feature parameter set and the second initial feature parameter set.

8. The hot and cold channel simulation analysis method based on BIM and CFD according to claim 1 is characterized in that: The verification results obtained by performing fluid dynamics verification analysis on the channel layout division scheme include: Converting the target layout plan into boundary condition input information required for fluid dynamics analysis, wherein the boundary condition input information includes updated channel dimensions and equipment position information; Setting meshing parameters for fluid dynamics simulations based on channel geometry complexity; Performing a fluid dynamics simulation based on the meshing parameters and the meshing parameters, and outputting air leakage status verification data and temperature distribution verification data; Performing a comparison analysis on the air leakage status verification data and the air leakage tolerance threshold, and performing a consistency analysis on the temperature distribution verification data and the temperature uniformity threshold; If the air leakage status verification data is less than the air leakage tolerance threshold and the temperature distribution verification data meets the temperature uniformity threshold, it is confirmed that the verification result meets the preset air leakage tolerance threshold and temperature uniformity threshold.

9. A cold and hot channel simulation analysis system, characterized in that: include: processor; A storage device having a computer program stored thereon, wherein when the computer program is executed by the processor, the processor implements the hot and cold channel simulation analysis method based on BIM and CFD as described in any one of claims 1 to 8.

Images