A distributed intelligent design system and method for building structures under a low-carbon background

Abstract

This invention relates to the field of computer-aided design technology, specifically to a distributed intelligent building structure design system and method under a low-carbon background. The system includes: a BIM parameter optimization module that constructs a building cluster adjacency matrix and solar shading coefficient based on building structure data, calculates ventilation scores, and optimizes the window-to-wall ratio using the SQP algorithm in conjunction with window heat transfer coefficients; a route generation module that constructs energy nodes and connection paths using the building cluster adjacency matrix; a carbon emission analysis module that includes construction carbon emission calculation, multi-factor coupled prediction, and net carbon emission aggregation, capable of predicting future carbon emissions based on the LSTM algorithm and assessing the building cluster's net carbon emissions in conjunction with ecological restoration; and a route optimization module that, based on the carbon emission analysis results, uses a heuristic algorithm to generate an initial feasible solution and optimizes the building cluster's energy node configuration and optimal connection paths through mixed-integer programming, thereby improving the efficiency of intelligent building structure design.

Description

Technical Field

[0001] This invention relates to the field of computer-aided design technology, specifically to an intelligent design system and method for distributed building structures under a low-carbon background. Background Technology

[0002] Existing building-aided design systems, based on tools such as Revit and AutoCAD, provide basic functional support for architectural design through parametric modeling engines and static energy consumption analysis modules. These systems can generate building geometric models, automatically compile bills of materials, and estimate carbon emissions for fixed scenarios, meeting the needs of traditional design scenarios. However, for the dynamic simulation and collaborative optimization requirements of distributed building structures under low-carbon goals, existing systems have certain technical shortcomings in data processing architecture, algorithm engine integration, and multi-module collaborative design.

[0003] First, most building design systems, based on BIM platforms, employ individual building energy analysis engines without constructing a graph-based energy network topology model for building clusters. This prevents them from using node optimization algorithms to reduce cross-building energy transmission losses, thus hindering the efficient collaborative utilization of regional renewable energy. Second, these systems operate ecological restoration modules and carbon accounting engines independently, failing to integrate soil carbon sequestration rate calculation models and vegetation carbon absorption dynamic databases. This prevents the design scheme from simultaneously assessing carbon emissions during the construction phase and the site's carbon sink offsetting benefits. Furthermore, these systems lack multi-objective optimization engines, potentially impacting the iteration speed and global optimality of design schemes.

[0004] To address this, a distributed building structure intelligent design system and method under a low-carbon background are proposed. Summary of the Invention

[0005] The purpose of this invention is to provide a distributed intelligent building structure design system and method under a low-carbon background, so as to improve the efficiency of intelligent building structure design. The system utilizes a BIM parameter optimization module to construct a building cluster adjacency matrix and solar shading coefficient based on building structure data, calculates ventilation scores, and optimizes the window-to-wall ratio using the SQP algorithm in conjunction with window heat transfer coefficients. A route generation module uses the building cluster adjacency matrix to construct energy nodes and connection paths. A carbon emission analysis module includes construction carbon emission calculation, multi-factor coupled prediction, and net carbon emission aggregation. It can predict future carbon emissions based on the LSTM algorithm and assess the net carbon emissions of the building cluster in conjunction with ecological restoration. Based on the carbon emission analysis results, the route optimization module uses a heuristic algorithm to generate an initial feasible solution and optimizes the configuration of energy nodes and the optimal connection paths of the building cluster through mixed-integer programming, thereby improving the efficiency of intelligent building structure design.

[0006] To achieve the above objectives, the present invention provides the following technical solution:

[0007] A distributed building structure intelligent design system and method under a low-carbon background, comprising:

[0008] The BIM parameter optimization module includes: a spatial topology modeling unit, used to generate a building cluster adjacency matrix and solar shading coefficient based on building structure data; and a building parameter optimization unit, used to calculate ventilation scores and, based on the solar shading coefficient and window heat transfer coefficient, to solve for the optimal window-to-wall ratio using the SQP algorithm.

[0009] The routing generation module is used to generate energy nodes and connection paths for the building group based on the building group adjacency matrix.

[0010] The carbon emission analysis module includes: a carbon emission calculation unit for quantifying the carbon footprint of building material production and transportation; a multi-factor coupled prediction unit for generating future carbon emissions using the LSTM algorithm based on user behavior logs; and a net carbon emission aggregation unit for assessing ecological restoration potential and outputting the net carbon emissions of the building complex.

[0011] The routing optimization module generates an initial feasible solution using a heuristic algorithm based on the net carbon emissions of the building complex, and solves for the optimal energy nodes and optimal connection paths of the building complex through mixed integer programming.

[0012] Furthermore, the building structure data includes building data, user behavior logs, ecological restoration data, and meteorological data.

[0013] Furthermore, the spatial topology modeling unit specifically includes:

[0014] Obtain the geographical coordinates of the buildings and use the Euclidean distance formula to calculate the building spacing.

[0015] The solar shading coefficient is calculated based on the building spacing and expressed as follows:

[0016]

[0017] Among them, SC i，j H is the solar shading coefficient of building j to building i. j and H i Let θ be the height of building j and building i, respectively, and let cos() be the cosine function. sun SC is the solar altitude angle. i Let D be the total solar shading coefficient of building i, n be the total number of buildings, and D be the total solar shading coefficient of building i. ij The building spacing between building j and building i;

[0018] If the building spacing is less than the spacing threshold and the solar shading coefficient is less than the shading threshold, then it is determined to be in a connected state; the building group adjacency matrix is ​​generated based on the connected state.

[0019] Furthermore, the building parameter optimization unit specifically includes:

[0020] Wind field correction coefficients for building clusters were obtained based on computational fluid dynamics simulations.

[0021] Multiply the wind field correction coefficient of the building complex by the wind speed at the meteorological station to obtain the actual ventilation wind speed. Combine this with the window area and room volume to obtain a ventilation score, and then calculate the objective function, expressed as:

[0022] min WWR [α×(1-f vent )+β×(1-f light )+γ×U window [×WWR];

[0023] in, f is the function that minimizes the objective function. vent For the ventilation score, f light For lighting rating, U window is the heat transfer coefficient of the window, WWR is the window-to-wall ratio, and α, β, and γ are weighting coefficients;

[0024] Under the constraint of window-to-wall ratio, the optimal window-to-wall ratio is output using the SQP algorithm according to the objective function.

[0025] Furthermore, the assessment process for the ecological restoration potential includes:

[0026] The construction site is divided into grids, and the soil carbon sequestration, vegetation carbon sink, water carbon sink and microbial carbon sequestration of each grid are calculated to obtain the preliminary ecological restoration potential.

[0027] The affected area is divided with the building as the center, and the initial ecological restoration potential corresponding to the grid in the affected area is weakened to obtain the ecological restoration potential.

[0028] Furthermore, the building cluster energy nodes include: power generation nodes, energy storage nodes, and load nodes. The power generation nodes include photovoltaic capacity and wind power capacity; the energy storage nodes include energy storage system capacity and maximum energy storage charging and discharging power; and the load nodes include building baseline load and building peak load.

[0029] Furthermore, the solution process for the mixed integer programming problem includes:

[0030] Obtain the connection status of the adjacency matrix of the building group to obtain the connection path; calculate the cable transmission loss based on the connection path; calculate the energy cost of the energy nodes of the building group; and weight and sum the energy cost, the cable transmission loss, and the net carbon emissions of the building group to obtain the energy network optimization value.

[0031] The connection path is set as a discrete optimization variable, and the building cluster energy node is set as a continuous optimization variable. The heuristic algorithm is used to generate the initial feasible solution, and the initial feasible solution is input into the mixed integer programming to generate the optimal building cluster energy node and the optimal connection path.

[0032] A method for intelligent design of distributed building structures under a low-carbon background includes:

[0033] The building cluster adjacency matrix and solar shading coefficient are generated based on the building structure data; the ventilation score is calculated, and the optimal window-to-wall ratio is solved using the SQP algorithm based on the solar shading coefficient and the window heat transfer coefficient.

[0034] Establish an energy network topology to generate energy nodes and connection paths for the building group based on the building group adjacency matrix;

[0035] Quantify the carbon footprint of building material production and transportation; generate future carbon emissions using the LSTM algorithm based on the energy network topology; assess ecological restoration potential and output the net carbon emissions of the building complex;

[0036] Based on the net carbon emissions of the building complex, an initial feasible solution is generated using a heuristic algorithm, and the optimal energy nodes and optimal connection paths of the building complex are solved using mixed integer programming.

[0037] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0038] 1. This invention utilizes a BIM parameter optimization module to calculate the adjacency matrix and solar shading coefficient of a building complex through spatial topology modeling, and employs computational fluid dynamics simulation to evaluate building wind field characteristics, thereby optimizing daylighting performance, ventilation performance, and window-to-wall ratio. Combined with an LSTM prediction model, it generates future carbon emission trends based on user behavior logs, enabling dynamic prediction of building energy consumption optimization during the design phase, reducing later energy control costs, and improving the energy-saving effect of the building complex.

[0039] 2. This invention introduces an ecological restoration potential assessment mechanism. It calculates the natural carbon sequestration capacity of a building site through soil carbon sequestration, vegetation carbon sinks, water body carbon sinks, and microbial carbon sequestration. Combined with gridded analysis, it assesses the weakening effect of buildings on the ecosystem, thereby optimizing building density and green space layout to reduce the carbon emission impact of building sites. Furthermore, the carbon emission analysis module calculates the carbon footprint of building material production and transportation, and, combined with future carbon emission predictions, supports the design of minimizing carbon emissions throughout the entire life cycle of building complexes.

[0040] 3. This invention employs a routing optimization module to generate initial feasible solutions through heuristic algorithms and combines this with mixed-integer programming to optimize the energy node layout and connection paths of building clusters. Simultaneously, it utilizes a cable transmission loss model to calculate the optimal energy sharing scheme, improving energy transmission efficiency between buildings. Finally, by combining power generation, energy storage, and load optimization strategies, it optimizes the investment return cycle of building clusters while ensuring low-carbon goals, thereby improving the efficiency of intelligent building structural design. Attached Figure Description

[0041] Figure 1 This is a schematic diagram of the structure of a distributed building structure intelligent design system under a low-carbon background according to the present invention;

[0042] Figure 2 This is a schematic diagram of the structure for calculating net carbon emissions in this invention;

[0043] Figure 3 This is a flowchart illustrating an intelligent design method for distributed building structures under a low-carbon background, as proposed in this invention. Detailed Implementation

[0044] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0045] Please see Figures 1 to 3 This invention provides an intelligent design system and method for distributed building structures under a low-carbon background, the technical solution of which is as follows:

[0046] Example 1:

[0047] Driven by the global goal of carbon neutrality, carbon emissions in the construction industry have received widespread attention. Traditional building design methods often fail to fully consider the synergistic optimization of building layout, energy utilization, and green carbon emissions, resulting in low overall energy efficiency and high carbon emissions for building complexes, making it difficult to meet the needs of green and low-carbon development. To address these issues, a distributed intelligent building structure design system under a low-carbon context is proposed, aiming to improve the efficiency of intelligent building structure design, such as... Figure 1 Shown, including:

[0048] refer to Figure 1 The BIM parameter optimization module includes: a spatial topology modeling unit, used to generate a building cluster adjacency matrix and solar shading coefficient based on building structure data; and a building parameter optimization unit, used to calculate ventilation scores and, based on the solar shading coefficient and window heat transfer coefficient, to solve for the optimal window-to-wall ratio using the SQP algorithm.

[0049] Furthermore, the building structure data includes building data, user behavior logs, ecological restoration data, and meteorological data.

[0050] Specifically, building data includes: BIM model data: building geometry information, such as building height, length, width, and spacing; material information, such as heat transfer coefficients and carbon footprint data of building components such as walls, windows, roofs, and foundations; functional information, such as building use (office, residential, and commercial), room distribution, and space utilization. Energy equipment parameters: parameters of photovoltaic systems, energy storage systems, and electrical loads. GIS geographic data: topographic information, geographic coordinates, and soil depth.

[0051] User behavior logs include: energy consumption history, space utilization, and user preference settings.

[0052] Ecological restoration data includes: vegetation coverage, aquatic microbial data, vegetation types, and water quality monitoring data.

[0053] Meteorological data includes temperature, humidity, and wind speed.

[0054] Building data forms the foundation for building energy consumption analysis and energy system optimization. Accurate collection and in-depth analysis of building structural design data provide strong support for reducing energy consumption and improving energy efficiency. User behavior logs help optimize intelligent building management and carbon emission prediction. Mining and analyzing this data allows for a better understanding of building users' needs and habits, leading to more efficient building management and carbon emission prediction. Ecological restoration data helps comprehensively assess the impact of buildings on the ecological environment and provides a scientific basis for optimizing carbon neutrality strategies. Furthermore, meteorological data plays a crucial role in building energy consumption calculations and ventilation optimization. Analyzing and applying meteorological information can further improve building energy efficiency and environmental adaptability.

[0055] Furthermore, the spatial topology modeling unit specifically includes:

[0056] Obtain the geographical coordinates of the buildings and use the Euclidean distance formula to calculate the building spacing.

[0057] The solar shading coefficient is calculated based on the building spacing and expressed as follows:

[0058]

[0059] Among them, SC i，j H is the solar shading coefficient of building j to building i. j and H i Let θ be the height of building j and building i, respectively, and let cos() be the cosine function. sum SC is the solar altitude angle.i Let D be the total solar shading coefficient of building i, n be the total number of buildings, and D be the total solar shading coefficient of building i. ij The building spacing between building j and building i;

[0060] If the building spacing is less than the spacing threshold and the solar shading coefficient is less than the shading threshold, then it is determined to be in a connected state; the building group adjacency matrix is ​​generated based on the connected state.

[0061] Specifically, as shown in Table 1, assuming there are three buildings with geographical coordinates (X, Y, Z), the distance between the buildings is calculated using the Euclidean distance formula. The distance between A1 and A2 is 47.17m, and the distance between A2 and A3 is 53.85m.

[0062] Table 1. Geographical Coordinates of Buildings

[0063] Building Number X-coordinate (m) Y coordinate (m) Z-height (m) A1 10 15 30 A2 50 40 25 A3 70 90 35

[0064] The solar altitude angle is calculated based on the local geographical coordinates, date, and time, and is expressed as:

[0065] θ sun =arcsin[sin()×sin(δ)+cos()×cos(δ)×cos(h)];

[0066] Where, θ sun denoted as solar altitude angle, denoted as local latitude, δ as solar declination angle, h as hour angle, arcsin() as inverse trigonometric function, sin() as sine function, and cos() as cosine function.

[0067] Assume θ sun At 45°, the solar shading coefficients SC of A1 and A2 are... 1，2 It is -0.075, due to SC 1，2 A negative value indicates that A2 will not block A1. The solar shading coefficients SC of A1 and A3 are... 1，3 It is 0.065, due to SC 1，3 If the result is positive, A3 blocks A1. The total shading coefficient of A1 is calculated and obtained as 0.065. In this embodiment, monitoring is performed continuously for 7 days, and the maximum value among the 7 days is taken as the solar shading coefficient.

[0068] Based on the building spacing threshold of 50m and the occlusion threshold of 0.15, it is determined whether buildings can be connected. Since A1 and A2 are unobstructed and the distance is less than 50m, A1 and A2 are connected. Based on the connection relationship, an adjacency matrix is ​​generated, represented as follows:

[0069]

[0070] In this context, 1 indicates that the buildings can be connected, and 0 indicates that they cannot be connected.

[0071] Precise calculations of building spacing and shading coefficients lay the foundation for avoiding excessive shading and improving natural lighting. Simultaneously, the generated adjacency matrix can be used to optimize the microgrid connecting adjacent buildings, enabling energy sharing between buildings and supporting the rational layout of photovoltaic and energy storage systems, thereby improving the efficiency of intelligent building structural design.

[0072] Furthermore, the building parameter optimization unit specifically includes:

[0073] The wind field correction coefficient k of the building complex was obtained based on computational fluid dynamics (CFD) simulation. CFD ;

[0074] Multiplying the wind field correction factor of the building complex by the wind speed at the meteorological station yields the actual ventilation wind speed, expressed as:

[0075] v act =v met ×k CFD ;

[0076] Among them, v act v represents the actual ventilation velocity. met The wind speed at a height of 10m at the weather station;

[0077] The ventilation score f is obtained by combining the window area and the room volume. vent , represented as:

[0078]

[0079] Where ACH is the ventilation rate, A window v is the area of ​​the window opening. room For room volume, ACH target Let exp() be the natural exponential function, where ACH is the target ventilation rate. When ACH is lower than ACH... target When ventilation is excessive, the score decreases linearly, indicating that ventilation needs to be increased; conversely, when ventilation is excessive, the score index decays, indicating that ventilation needs to be reduced.

[0080] The light intensity is calculated and expressed as:

[0081] E room =T vis ×WWR×A warr ×I sorar ×(1-SC i ):

[0082] Among them, E room For light intensity, T vis The visible light transmittance of the window glass is given by A, WWR by the window-to-wall ratio. wall For the wall area, I solarSC represents outdoor solar radiation intensity. i Let be the total solar shading coefficient of building i;

[0083] The daylighting score is calculated and expressed as follows:

[0084]

[0085] Among them, f light For daylighting rating, E target The target illumination value;

[0086] And calculate the objective function, expressed as:

[0087] min WWR [α×(1-f vent )+β×(1-f light )+γ×U window [×WWR];

[0088] in, f is the function that minimizes the objective function. vent For the ventilation score, f light For lighting rating, U window The window heat transfer coefficient is denoted as WWR, the window-to-wall ratio is denoted as WWR, and α, β, and γ are weighting coefficients.

[0089] Under the constraint of window-to-wall ratio, the optimal window-to-wall ratio is output using the SQP algorithm according to the objective function.

[0090] Specifically, in this embodiment, target values ​​are set according to the season, such as ACH. target The score is 5 in summer and 2 in winter, with segmented scoring. The window-to-wall ratio constraint ranges from 0.2 to WWR to 0.6. The SQP algorithm approximates the objective function quadratically, causing the solution to gradually converge to the optimal window-to-wall ratio.

[0091] CFD wind field correction technology enables more accurate calculation of natural ventilation capacity, providing strong support for optimizing the window-to-wall ratio and improving air circulation efficiency. Simultaneously, by calculating the total shading coefficient, the window-to-wall ratio can be dynamically adjusted to ensure optimal lighting effects, thereby reducing lighting energy consumption. Furthermore, optimizing the window heat transfer coefficient effectively reduces unnecessary heat loss and improves the energy-saving performance of the building envelope. Building upon this, the SQP algorithm is used to automatically optimize the window-to-wall ratio, achieving an optimal balance between ventilation, lighting, and energy conservation, further enhancing the efficiency and intelligence of intelligent building structural design.

[0092] The routing generation module is used to generate energy nodes and connection paths for the building group based on the building group adjacency matrix.

[0093] Among them, connection path C i，jThis indicates the feasibility of energy sharing. If, in the adjacency matrix, the connection path M between building i and building j... i，j If the value is 1, then energy sharing is allowed.

[0094] Furthermore, in order to achieve precise energy management, the energy system in the building complex is divided into three types of core energy nodes, namely: power generation nodes, energy storage nodes, and load nodes.

[0095] Power generation nodes primarily provide renewable energy, including photovoltaic (PV) capacity and wind power capacity. PV capacity refers to the maximum power that can be installed on a rooftop or facade; wind power capacity is suitable for building sites with abundant wind resources, enabling effective utilization of wind power generation. By rationally setting up power generation nodes, the dependence of building complexes on the traditional power grid can be effectively reduced, improving energy self-sufficiency.

[0096] The role of energy storage nodes is to balance the time difference between power generation and consumption, ensuring a stable energy supply. An energy storage node includes the energy storage system capacity and the maximum charge / discharge power. The energy storage system capacity determines how much electrical energy the energy storage device can store; while the maximum charge / discharge power limits the maximum amount of electrical energy the energy storage device can charge or discharge per unit time. Energy storage nodes can store energy during the day when there is excess photovoltaic power generation and release it at night, effectively smoothing out fluctuations in renewable energy, enhancing the stability of building microgrids, and avoiding the risk of power outages.

[0097] Load nodes are the main source of energy consumption and also possess intelligent scheduling capabilities to reduce peak loads. These load nodes include the building's baseline load and peak load. The baseline load, such as lighting and security systems, represents the basic electricity requirements necessary for the building's normal operation; peak loads, such as air conditioning and elevators operating during peak hours, typically have higher electricity demands. By precisely controlling the load nodes, power overload can be avoided, ensuring efficient energy use across the building complex.

[0098] This classification method is the foundation for achieving precise energy management, helps to establish an efficient energy sharing network, optimizes energy flow, and thus promotes the improvement of the efficiency of intelligent building structure design.

[0099] The carbon emission analysis module includes: a carbon emission calculation unit for quantifying the carbon footprint of building material production and transportation; a multi-factor coupled prediction unit for generating future carbon emissions using the LSTM algorithm based on user behavior logs; and a net carbon emission aggregation unit for assessing ecological restoration potential and outputting the net carbon emissions of the building complex.

[0100] The construction of carbon emission calculation units includes:

[0101] From the BIM model data, a list of building materials is extracted, and then the carbon emissions from material production are calculated, as shown below:

[0102] C prod =∑ m (Q m ×EF m );

[0103] Among them, C prod For carbon emissions during the building materials production stage, Q m EF represents the amount of material m used. m Carbon emission factor of materials;

[0104] Carbon emissions from material transportation are calculated and expressed as follows:

[0105] C trans =∑ m (Q m ×D m ×EF trans );

[0106] Among them, C trans For carbon emissions from the transportation of building materials, Q m D represents the amount of material m used. m For material transport distance, EF trans Carbon emission factor for transportation mode;

[0107] Adding the carbon emissions from building material transportation to those from the building material production stage yields the total carbon emissions during the construction phase, thereby supporting the selection of low-carbon materials. By combining GIS data, the building material supply chain can be optimized, reducing unnecessary long-distance transportation.

[0108] The multi-factor coupled prediction unit assumes the building complex is an office building complex and obtains energy usage patterns, equipment usage habits, personnel activity time, and carbon emissions from similar buildings. This data, combined with meteorological data, forms the input data. The input data is then divided into time windows and fed into an LSTM algorithm to predict future carbon emissions. This helps in the rational planning of the energy system and reduces the risk of exceeding carbon emission limits during later operational phases.

[0109] Furthermore, the assessment process for the ecological restoration potential includes:

[0110] The construction site is divided into grids, and the soil carbon sequestration, vegetation carbon sink, water carbon sink and microbial carbon sequestration of each grid are calculated to obtain the preliminary ecological restoration potential.

[0111] The affected area is divided with the building as the center, and the initial ecological restoration potential corresponding to the grid in the affected area is weakened to obtain the ecological restoration potential.

[0112] Specifically, traditional buildings often neglect the original ecological functions of a site, such as soil and vegetation, leading to a decline in the ecosystem's carbon sequestration capacity. Low-carbon buildings require an assessment of the site's ecological restoration potential. Regular grids (e.g., 10m × 10m) or adaptive grids are used, with each grid containing soil data, vegetation data, water body data, and microbial data.

[0113] The formula for calculating the initial ecological restoration potential after gridding is expressed as follows:

[0114] P eco (x, y) = P soil (x, y) + P veg (x, y) + P water (x, y) + P micro (x, y);

[0115] Among them, P eco (x, y) represents the initial ecological restoration potential of the grid (x, y), P soil (x, y), P veg (x, y), P water (x, y) and P micro (x, y) represent soil carbon sequestration, vegetation carbon sink, water carbon sink, and microbial carbon sequestration, respectively. x is the grid index in the east-west direction, and y is the grid index in the north-south direction.

[0116] Since the presence of buildings weakens ecological restoration capabilities, the impact on the surrounding area needs to be considered (due to human activities and microclimate changes, carbon sequestration capacity is weakened). The impact area is divided with building i as the center, and its radius of influence is calculated. In this embodiment, the radius of influence is twice the height of building i.

[0117] Within the building impact zone, the reduction of ecological carbon sinks is represented as:

[0118]

[0119] in, N has the potential for ecological restoration. i N represents the number of grid cells occupied by building i. grid The total number of grids in the affected area is represented by k, which is an empirical coefficient representing the reduction ratio of ecological carbon sink in the building-affected area. It can be set to 0.3, and gird is the grid sequence number.

[0120] like Figure 2 As shown, carbon emissions from construction and transportation are calculated based on building data, operational carbon emissions are calculated based on user behavior logs, and the net carbon emissions of the building complex are obtained based on ecological restoration data and the carbon offsetting potential assessment of energy nodes within the building complex. This is expressed as:

[0121]

[0122] Among them, C net For the net carbon emissions of the building complex, C prod，i Carbon emissions from the production of building materials, C trans，i Carbon emissions from the transportation of building materials, P PV C represents the potential to reduce carbon emissions at building cluster energy nodes. pre，i For the future carbon emissions of buildings, Q renewable，i E represents the amount of electricity that building i receives from renewable energy nodes. ele E represents the carbon emission factor of grid electricity. renewable Carbon emission factor of renewable energy, Q store，i E is the electricity provided by energy storage to building i. peak E represents the carbon emission factor of the power grid during peak hours. off-peak This represents the carbon emission factor of the power grid during off-peak hours.

[0123] Grid-based computing allows for the quantification of the impact on soil, vegetation, water bodies, and microbial carbon sinks. Optimizing building layouts based on this data can effectively reduce ecological damage. For example, in areas with high carbon sink capacity, building density should be reduced to strictly protect existing ecosystems; while in low-carbon sink areas, building distribution can be appropriately increased to improve land utilization. Furthermore, assessing the carbon offsetting capacity of buildings provides a basis for subsequent energy conservation and emission reduction measures, contributing not only to ecological protection but also improving the efficiency of intelligent building structural design.

[0124] The routing optimization module generates an initial feasible solution using a heuristic algorithm based on the net carbon emissions of the building complex, and solves for the optimal energy nodes and optimal connection paths of the building complex through mixed integer programming.

[0125] Furthermore, the solution process for the mixed integer programming problem includes:

[0126] Obtain the connection status of the adjacency matrix of the building group to obtain the connection path; calculate the cable transmission loss based on the connection path; calculate the energy cost of the energy nodes of the building group; and weight and sum the energy cost, the cable transmission loss, and the net carbon emissions of the building group to obtain the energy network optimization value.

[0127] The connection path is set as a discrete optimization variable, and the building cluster energy node is set as a continuous optimization variable. The heuristic algorithm is used to generate the initial feasible solution, and the initial feasible solution is input into the mixed integer programming to generate the optimal building cluster energy node and the optimal connection path.

[0128] By optimizing energy routing, energy loss is reduced, and overall energy efficiency is improved. Simultaneously, optimizing equipment selection and investment strategies reduces unnecessary cable and energy storage investments, lowering operating costs and thus enhancing the efficiency of intelligent building structural design.

[0129] First, based on BIM modeling, building spacing and shading coefficients can be accurately calculated to optimize ventilation and lighting in building complexes. Then, the SQP algorithm is used to optimize the window-to-wall ratio, maximizing natural lighting and ventilation to meet residential needs. Next, a carbon emission prediction model based on LSTM, combined with user behavior data, dynamically predicts carbon emission trends. Simultaneously, it calculates carbon emissions from building material production and transportation, supporting the selection of low-carbon materials and reducing the building's carbon footprint. Integrated ecological restoration potential assessment calculates biological carbon sink capacity, providing a basis for carbon neutrality strategies. Finally, by combining mixed-integer programming and heuristic algorithms, energy-sharing paths between buildings are automatically optimized, improving self-sufficiency and reducing reliance on the traditional power grid, thereby enhancing the efficiency of intelligent building structural design.

[0130] Example 2:

[0131] When developing distributed building complexes, an architectural design firm faced challenges such as complex energy system optimization, difficulty in quantifying the impact of building layout on carbon emissions, and the difficulty in predicting and optimizing carbon emissions. To address these issues, they employed an intelligent design method for distributed building structures under a low-carbon context, such as... Figure 3 Shown, including:

[0132] The building cluster adjacency matrix and solar shading coefficient are generated based on the building structure data; the ventilation score is calculated, and the optimal window-to-wall ratio is solved using the SQP algorithm based on the solar shading coefficient and the window heat transfer coefficient.

[0133] Establish an energy network topology to generate energy nodes and connection paths for the building group based on the building group adjacency matrix;

[0134] Quantify the carbon footprint of building material production and transportation; generate future carbon emissions using the LSTM algorithm based on the energy network topology; assess ecological restoration potential and output the net carbon emissions of the building complex;

[0135] Based on the net carbon emissions of the building complex, an initial feasible solution is generated using a heuristic algorithm, and the optimal energy nodes and optimal connection paths of the building complex are solved using mixed integer programming.

[0136] Furthermore, the solution process for the mixed integer programming problem includes:

[0137] Obtain the connection status of the adjacency matrix of the building group to obtain the connection path; calculate the cable transmission loss based on the connection path, expressed as:

[0138]

[0139] Where, n i，j For a cable transmission efficiency of 2% per 100m of line loss, D ij The building spacing between building j and building i;

[0140] The energy cost of the energy nodes in the building complex is calculated and expressed as follows:

[0141] C energy =∑ a (C cap，a +C op,a );

[0142] Among them, C energy For energy costs, C cap，a Let C be the investment cost of node a. op，a The operating and maintenance cost of node a.

[0143] The energy network optimization value is obtained by weighted summing of the energy cost, the cable transmission loss, and the net carbon emissions of the building complex.

[0144] The connection path is set as a discrete optimization variable, represented as:

[0145] M i，j ∈{0,1};

[0146] x a ∈{0,1};

[0147] Among them, M i，j To determine whether to establish a connection from building i to j, x a Whether to establish a building cluster energy node;

[0148] The energy nodes of the building complex are set as continuous optimization variables, namely the capacity of the power generation node and the capacity of the energy storage node.

[0149] The initial feasible solution is generated using the heuristic algorithm, and then input into the mixed integer programming algorithm to generate the optimal building cluster energy node and the optimal connection path.

[0150] The heuristic algorithm and mixed-integer programming satisfy the following constraints:

[0151] Carbon emission constraints mean that the net carbon emissions of the building complex do not exceed a preset carbon emission threshold of 1000 tCO2 / year;

[0152] The energy supply and demand balance constraint means that the sum of the capacity of the power generation node and the capacity of the energy storage node shall not exceed the capacity of the load node;

[0153] Investment recovery constraints are expressed as follows:

[0154]

[0155] Among them, C cap，a Let C be the investment cost of node a. op，a，t Let C be the operating and maintenance cost of node a in year t. saving，t Let r be the cost savings from replacing fossil fuels with renewable energy in year t, r be the discount rate (0.05), and R be the payback period (10 years).

[0156] A set of feasible energy node and connection path schemes is generated by using a genetic algorithm or tabu search. This scheme is used as the initial solution and input into a mixed integer programming algorithm to obtain the optimal energy node of the building group and the optimal connection path.

[0157] Specifically, an architectural design firm has buildings A, B, and C, whose locations form a triangle, and the initial adjacency matrix is ​​fully connected. As shown in Table 2, initial selectable energy node parameters are generated based on expert experience and actual conditions. Furthermore, the baseline load and peak load of each building load node are directly defined based on historical data.

[0158] Table 2 Energy node parameters of the building complex

[0159]

[0160] Genetic algorithms are suitable for exploring multi-objective trade-offs, while mixed-integer programming ensures global optimality. Genetic algorithms quickly filter out a small number of feasible solutions in a coarse search, while mixed-integer programming, based on the results of the genetic algorithm, provides a precise solution within a small range. Initial optional energy node parameters and fully connected paths are input into the genetic algorithm to obtain an initial solution. This initial solution is then input into the mixed-integer programming algorithm to obtain the final optimization results, as shown in Table 3. Furthermore, compared to using only the genetic algorithm, energy costs were reduced by 285,620 yuan, and carbon emissions were reduced by 81.6 tCO2.

[0161] Table 3 Energy node parameters of the building complex

[0162]

[0163] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. A distributed building structure intelligent design system under a low-carbon background, characterized in that, include: The BIM parameter optimization module includes: a spatial topology modeling unit, used to generate a building cluster adjacency matrix and solar shading coefficient based on building structure data; and a building parameter optimization unit, used to calculate ventilation scores and, based on the solar shading coefficient and window heat transfer coefficient, to solve for the optimal window-to-wall ratio using the SQP algorithm. The routing generation module is used to generate energy nodes and connection paths for the building group based on the building group adjacency matrix. The carbon emission analysis module includes: a carbon emission calculation unit for quantifying the carbon footprint of building material production and transportation; a multi-factor coupled prediction unit for generating future carbon emissions using the LSTM algorithm based on user behavior logs; and a net carbon emission aggregation unit for assessing ecological restoration potential and outputting the net carbon emissions of the building complex. The routing optimization module generates an initial feasible solution using a heuristic algorithm based on the net carbon emissions of the building complex, and solves for the optimal energy nodes and optimal connection paths of the building complex through mixed integer programming.

2. The intelligent design system for distributed building structures under a low-carbon background according to claim 1, characterized in that, The building structure data includes building data, user behavior logs, ecological restoration data, and meteorological data.

3. The intelligent design system for distributed building structures under a low-carbon background according to claim 1, characterized in that, The spatial topology modeling unit specifically includes: Obtain the geographical coordinates of the buildings and use the Euclidean distance formula to calculate the building spacing. The solar shading coefficient is calculated based on the building spacing and expressed as follows: Among them, SC i,j H is the solar shading coefficient of building j to building i. j and H i Let θ be the height of building j and building i, respectively, and let cos() be the cosine function. sun SC is the solar altitude angle. i Let D be the total solar shading coefficient of building i, n be the total number of buildings, and D be the total solar shading coefficient of building i. ij The building spacing between building j and building i; If the building spacing is less than the spacing threshold and the solar shading coefficient is less than the shading threshold, then it is determined to be in a connected state; the building group adjacency matrix is ​​generated based on the connected state.

4. The intelligent design system for distributed building structures under a low-carbon background according to claim 1, characterized in that, The building parameter optimization unit specifically includes: Wind field correction coefficients for building clusters were obtained based on computational fluid dynamics simulations. Multiply the wind field correction coefficient of the building complex by the wind speed at the meteorological station to obtain the actual ventilation wind speed. Combine this with the window area and room volume to obtain a ventilation score, and then calculate the objective function, expressed as: min WWR [α×(1-f vent )+β×(1-f light )+γ×U window ×WWR]; in, f is the function that minimizes the objective function. vent For the ventilation score, f light For lighting rating, U window is the heat transfer coefficient of the window, WWR is the window-to-wall ratio, and α, β, and γ are weighting coefficients; Under the constraint of window-to-wall ratio, the optimal window-to-wall ratio is output using the SQP algorithm according to the objective function.

5. The intelligent design system for distributed building structures under a low-carbon background according to claim 1, characterized in that, The assessment process for the ecological restoration potential includes: The construction site is divided into grids, and the soil carbon sequestration, vegetation carbon sink, water carbon sink and microbial carbon sequestration of each grid are calculated to obtain the preliminary ecological restoration potential. The affected area is divided with the building as the center, and the initial ecological restoration potential corresponding to the grid in the affected area is weakened to obtain the ecological restoration potential.

6. The intelligent design system for distributed building structures under a low-carbon background according to claim 1, characterized in that, The building cluster energy nodes include: power generation nodes, energy storage nodes, and load nodes. The power generation nodes include photovoltaic capacity and wind power capacity; the energy storage nodes include energy storage system capacity and maximum energy storage charging and discharging power; and the load nodes include building baseline load and building peak load.

7. The intelligent design system for distributed building structures under a low-carbon background according to claim 1, characterized in that, The solution process for the mixed integer programming problem includes: Obtain the connection status of the adjacency matrix of the building group to obtain the connection path; calculate the cable transmission loss based on the connection path; calculate the energy cost of the energy nodes of the building group; and weight and sum the energy cost, the cable transmission loss, and the net carbon emissions of the building group to obtain the energy network optimization value. The connection path is set as a discrete optimization variable, and the building cluster energy node is set as a continuous optimization variable. The heuristic algorithm is used to generate the initial feasible solution, and the initial feasible solution is input into the mixed integer programming to generate the optimal building cluster energy node and the optimal connection path.

8. A method for intelligent design of distributed building structures under a low-carbon background, characterized in that, include: Generate the building cluster adjacency matrix and solar shading coefficient based on building structure data; Calculate the ventilation score, and use the SQP algorithm to solve for the optimal window-to-wall ratio based on the solar shading coefficient and the window heat transfer coefficient. Establish an energy network topology to generate energy nodes and connection paths for the building group based on the building group adjacency matrix; Quantifying the carbon footprint of building material production and transportation; Based on the energy network topology, the LSTM algorithm is used to generate future carbon emissions; the potential for ecological restoration is assessed and the net carbon emissions of the building complex are output. Based on the net carbon emissions of the building complex, an initial feasible solution is generated using a heuristic algorithm, and the optimal energy nodes and optimal connection paths of the building complex are solved using mixed integer programming.

Images