Building performance multi-objective optimization method and equipment based on intelligent algorithm

Abstract

The invention belongs to the technical field of building construction, and particularly discloses a building performance multi-objective optimization method and equipment based on an intelligent algorithm. Comprising the steps of performing parametric simulation on building performance to obtain a sample data set; constructing a BO-CatBoost prediction model, predicting the building lighting performance, the building energy consumption, the building thermal comfort and the building carbon emission by adopting the BO-CatBoost prediction model, and establishing a nonlinear mapping relationship between each influence factor and the building energy consumption, the carbon emission, the lighting and the thermal comfort; and constructing a multi-target prediction function by taking different climate zone variables as constraint conditions. According to the method, efficient collaborative optimization of performance in multiple aspects of energy consumption, carbon emission, lighting, thermal comfort and the like of the building is realized, so that the building can effectively adapt to different climate zone environments while meeting multi-target performance requirements, and an efficient, universal and widely applicable optimization strategy is provided for sustainable development of the building industry.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of building construction, and more specifically relates to a building performance multi-objective optimization method and device based on an intelligent algorithm. BACKGROUND

[0002] With the rapid development of artificial intelligence, machine learning models have gradually become popular in the field of building carbon emissions due to their excellent learning ability. In the paper Considering critical building materials for embodied carbon emissions in buildings: A machine learning-based prediction model and tool, three algorithms, artificial neural networks, support vector regression, and extreme gradient boosting, were used to develop machine learning models to predict the consumption of key building materials and embodied carbon emissions. In the paper Assessment of operational carbon emissions for residential buildings comparing different machine learning approaches, five machine learning models were trained to predict the operational carbon emissions of residential buildings in 34 representative cities in China, and it was found that the building climate zone, population, urbanization rate, and building electrification rate were the most important features affecting emissions. In the paper Transition towards carbon neutrality: Forecasting Hong Kong's buildings carbon footprint by 2050 using a machine learning approach. Sustainable Production and Consumption, the STIRPAT model and machine learning were combined to predict the carbon footprint of the Hong Kong construction industry in 2050, and the results showed that the Hong Kong construction industry could achieve near-zero emissions.

[0003] In addition, the energy-saving design of single buildings is the key to achieving emission reduction targets in the future building field. However, energy-saving design is not just about reducing building energy consumption and carbon emissions, but also needs to meet the design requirements of various parameters in various specifications and standards. It is necessary to improve the performance of buildings while ensuring low-carbon and energy-saving buildings. According to the research of the University of Kansas on the weight of the building indoor environment, the thermal environment accounts for the largest proportion (45.9%), followed by the light environment (34.1%), and finally the wind environment, air quality and sound environment. The thermal and light environment of the building is closely related to the building energy consumption and carbon emissions, so the building performance target is mainly carbon emissions, energy consumption, thermal comfort and light performance. The building performance target is mainly controlled by the design parameters of the building envelope and the setting parameters of the heating, ventilation and air conditioning system, and there are many influencing factors and complex correlations between them. However, there are conflicts between building performance targets, so further exploration is needed to obtain the optimal building design scheme efficiently and reliably. SUMMARY

[0004] In view of the above defects or improvement needs of the prior art, the present application provides a building performance multi-objective optimization method and device based on intelligent algorithm, which realizes efficient and collaborative optimization of building energy consumption, carbon emissions, lighting, thermal comfort and other performances by combining a precise prediction model with a multi-objective optimization algorithm. It is based on data-driven and fine-tuned according to the characteristics of different climate zones, which not only significantly improves the scientificity and rationality of building design, but also greatly improves the optimization efficiency, so that the building can effectively adapt to different climate zones while meeting the requirements of multi-objective performance, providing an efficient, universal and widely applicable optimization strategy for the sustainable development of the building industry, and helping to promote the overall improvement of building performance and the green transformation of the industry.

[0005] To achieve the above-mentioned purpose, according to one aspect of the present application, a building performance multi-objective optimization method based on intelligent algorithm is provided, comprising the following steps:

[0006] Step one, parameterize simulation of building performance by Grasshopper to obtain sample data set;

[0007] Step two, build a BO-CatBoost prediction model, use the BO-CatBoost prediction model to predict the building lighting performance, building energy consumption, building thermal comfort and building carbon emissions, and establish the nonlinear mapping relationship between each influencing factor and building energy consumption, carbon emissions, lighting, thermal comfort;

[0008] Step three, taking the nonlinear mapping relationship as the fitness function of NSGA-III, taking the variables of different climate zones as constraint conditions, constructing a multi-objective prediction function, and using the prediction function to balance and optimize the four targets of building energy consumption, carbon emission, lighting, and thermal comfort to obtain an optimal solution set.

[0009] As a further preferred, in the sample data set, the input variables include:

[0010] window-wall ratio, external wall heat transfer coefficient, roof heat transfer coefficient, external window heat transfer coefficient, external wall solar radiation absorption coefficient, roof solar radiation absorption coefficient, external window solar radiation absorption coefficient, air change frequency, air conditioning and heating temperature, air conditioning refrigeration temperature, lighting power density, shading angle;

[0011] The input variables include: building energy consumption, carbon emission, lighting, and thermal comfort.

[0012] As a further preferred, in step one, the building carbon emission calculation model includes:

[0013]

[0014] wherein, E build (t) is the carbon emission index of the building at time t, a is the first adjustment factor of the building carbon emission calculation model, η(t) is the energy efficiency coefficient of the building at time t, ρ(t) is the personnel density of the building at time t, β is the second adjustment factor of the building carbon emission calculation model, γ is the third adjustment factor of the building carbon emission calculation model, ξ(t) is the energy consumption of the building at time t, which is used to describe the energy consumption of the building affected by the season, δ is the fourth adjustment factor of the building carbon emission calculation model, κ' is the fifth adjustment factor of the building carbon emission calculation model, E e (t) is the influence index of the environment on carbon emission at time t, μ is the sixth adjustment factor of the building carbon emission calculation model, U(t) is the actual energy demand of the building at time t, v is the seventh adjustment factor of the building carbon emission calculation model.

[0015] As a further preferred, in step one, the energy efficiency coefficient η(t) of the building energy consumption includes:

[0016]

[0017] wherein, η0 is the initial energy efficiency coefficient, α T is the weight of the environment temperature, T(t) is the environment temperature at time t, T base is the reference temperature, λ T is the adjustment factor of the environment temperature, β H is the weight of the environment humidity, H(t) is the environment humidity at time t, H baseγ' is the adjustment factor of the ambient humidity, α W is the weight of the ambient wind speed, W(t) is the ambient wind speed at time t, α L is the weight of the light intensity, L(t) is the light intensity at time t, L max is the maximum value of the light intensity.

[0018] As a further preferred, in step three, in the process of building the BO-CatBoost prediction model, BO is used to obtain the optimal CatBoost hyperparameter combination:

[0019] (1) Randomly initialize n0 points

[0020] (2) Obtain the corresponding function value f(X init ), then the initial point set D0 = {X init , f(X init )}, let t = n0, D t-1 = D0, loop this process until the maximum number of iterations;

[0021] (3) According to the obtained point set D t-1 , build a surrogate model S(x), based on the surrogate model S(x), maximize the acquisition function θ(x|D t-1 ), obtain the next evaluation point x t = argmax θ(x|D t-1 );

[0022] (4) Add the function value f(X t ) of the evaluation point to the current evaluation point set, D t = D t-1 ∪{x t , f(x t )};

[0023] (5) Output the optimal CatBoost hyperparameter combination.

[0024] As a further preferred, in step three, the objective function of the multi-objective prediction function includes:

[0025] f1 = min f EUI (CatBoost regression(x1,x2,x3,x4,x5,x6,x7,x8,x9,x 10 ,x 11 ,x 12 ,x 13 ,x 14 ,x 15 ))

[0026] f2 = min f CEICatBoost regression(x1, x2, x3, x4, x5, x6, x7, x8, x9, x 10 , x 11 , x 12 , x 13 , x 14 , x 15 )

[0027] f3 = min f PMV |CatBoost regression(x1, x2, x3, x4, x5, x6, x7, x8, x9, x 10 , x 11 , x 12 , x 13 , x 14 , x 15 )|

[0028] f4 = min f UDI (CatBoost regression(x1, x2, x3, x4, x5, x6, x7, x8, x9, x 10 , x 11 , x 12 , x 13 , x 14 , x 15 )

[0029] wherein x1, x2, x3, x4, x5, x6, x7, x8, x9, x 10 , x 11 , x 12 , x 13 , x 14 , x 15 are window-wall ratio-east, window-wall ratio-west, window-wall ratio-south, window-wall ratio-north, external wall heat transfer coefficient, roof heat transfer coefficient, external window heat transfer coefficient, external wall solar radiation absorption coefficient, roof solar radiation absorption coefficient, external window solar radiation absorption coefficient, air change frequency, air conditioning and heating temperature, air conditioning cooling temperature, lighting power density, shading angle, f EUI is building energy consumption, f CEI is carbon emission, f PMW is thermal comfort, f UDI is daylighting.

[0030] As a further preferred, in step three, the NSGA-III algorithm is used to obtain the optimal solution set, including:

[0031] The ideal point method is used to calculate the distance of each solution in the Pareto optimal solution set from the ideal point, and the point closest to the ideal point is found, which is the unique optimal solution;

[0032] The calculation model of the distance of each solution ideal point comprises:

[0033]

[0034] In the formula, (x pareto , y pareto , z pareto , w pareto ) is the coordinate corresponding to the optimal Pareto front, and (x Epoint , y Epoint , z Epoint , w Epoint ) is the coordinate corresponding to the ideal point.

[0035] According to another aspect of the present application, there is also provided a building performance multi-objective optimization system based on an intelligent algorithm, comprising:

[0036] A first master control module is configured to perform parametric simulation on building performance through Grasshopper to obtain a sample data set;

[0037] A second master control module is configured to construct a BO-CatBoost prediction model, to predict building lighting performance, building energy consumption, building thermal comfort and building carbon emission by using the BO-CatBoost prediction model, and to establish a nonlinear mapping relationship between each influencing factor and building energy consumption, carbon emission, lighting, and thermal comfort;

[0038] A third master control module is configured to use the nonlinear mapping relationship as a fitness function of NSGA-III, to use different climate zone variables as constraint conditions, to construct a multi-objective prediction function, and to use the prediction function to balance and optimize the four targets of building energy consumption, carbon emission, lighting, and thermal comfort, so as to obtain an optimal solution set.

[0039] According to another aspect of the present application, there is also provided an electronic device comprising at least one processor, at least one memory and a communication interface; wherein the processor, memory and communication interface communicate with each other; the memory stores program instructions executable by the processor, and the processor invokes the program instructions to perform a building performance multi-objective optimization method based on an intelligent algorithm according to any one of the above embodiments or a combination of multiple embodiments.

[0040] According to another aspect of the present application, there is also provided a non-transitory computer readable storage medium storing computer instructions, which cause the computer to perform a building performance multi-objective optimization method based on an intelligent algorithm according to any one of the above embodiments or a combination of multiple embodiments.

[0041] Overall, compared with the prior art, the above technical solutions conceived by the present application mainly have the following technical advantages:

[0042] 1. The present application can accurately predict building daylighting performance, building energy consumption, building thermal comfort and building carbon emissions by constructing a BO-CatBoost prediction model, and establish a nonlinear mapping relationship between various influencing factors and building performance indicators. Compared with the traditional parameter-by-parameter adjustment and simulation method, this model can quickly and accurately obtain building performance prediction results under different parameter combinations, greatly reducing the simulation calculation amount and time cost, and improving the optimization efficiency. The nonlinear mapping relationship is used as the fitness function of NSGA-III, and different climate zone variables are used as constraint conditions to construct a multi-objective prediction function, achieving the balance and optimization of building energy consumption, carbon emissions, daylighting and thermal comfort. This multi-objective collaborative optimization method avoids the problem of deterioration of other performance indicators that may be caused by single objective optimization, and can quickly find the optimal solution set that meets the multi-objective requirements in the limited design space, so that the building can achieve a better level in multiple key performance aspects.

[0043] 2. The present application obtains building information based on parameterized modeling, obtains sample data set through parameterized simulation by Grasshopper, and then constructs a prediction model and an optimization algorithm. The whole process is based on a large amount of data for analysis and decision-making. This makes the building design no longer rely on the experience and subjective judgment of the designer, but has more scientific and objective data support, which can more accurately evaluate the advantages and disadvantages of different design schemes, provide more reliable decision basis for building design, and improve the scientificity and rationality of building design. Specifically, after modeling according to the initial drawing of the building, the building carbon emission evaluation model is set and the carbon emission index is calculated, and the initial drawing is adjusted to make the carbon emission index lower than the preset threshold. This fine design adjustment based on the carbon emission evaluation model can ensure that the building design meets the performance requirements while effectively controlling carbon emissions, in line with the concept of sustainable development.

[0044] 3. The different climate zone variables are included in the optimization process as constraint conditions, fully considering the different requirements and influences of different climate zones on building performance. This means that when optimizing building performance, building design parameters can be adjusted according to the characteristics of different climate zones, such as temperature, humidity, sunshine duration, etc., so that the building can achieve the best performance in the respective climate environment. BRIEF DESCRIPTION OF DRAWINGS

[0045] Figure 1 is a flowchart of a building performance multi-objective optimization method based on intelligent algorithm according to an embodiment of the present application;

[0046] Figure 2is a schematic diagram of model prediction result visualization related to an embodiment of the present application;

[0047] Figure 3 is a schematic diagram of different model accuracy results related to an embodiment of the present application;

[0048] Figure 4 is a five-region Pareto frontier distribution diagram related to an embodiment of the present application. DETAILED DESCRIPTION

[0049] In order to make the objectives, technical solutions and advantages of the present application clearer, the present application will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and do not limit the present application. In addition, the technical features involved in each embodiment of the present application described below can be combined with each other as long as they do not conflict with each other.

[0050] As shown in Figure 1 , the building performance multi-objective optimization method based on intelligent algorithm provided by the embodiment of the present application comprises the following steps:

[0051] Step one, obtain building information, and perform parameterized simulation on building performance through Grasshopper to obtain a sample data set.

[0052] The building information includes: ambient temperature, ambient humidity, ambient wind speed, light intensity, thermal conductivity of the building, air permeability of the building, air density inside the building, and volume of the building.

[0053] The building carbon emission calculation model comprises:

[0054] E build (t)=exp(α·ln(η(t)·ρ(t)))+β·(1+γ·ξ(t)) δ ·(1+κ′·E e (t) μ ·U(t) v )

[0055] Wherein, E build(t) is the carbon emission index of the building at time t, a is the first adjustment factor of the building carbon emission calculation model, which mainly adjusts the comprehensive influence degree of energy efficiency coefficient and personnel density on building carbon emission, embodies the comprehensive adjustment effect of building on carbon emission in energy efficiency utilization and personnel density distribution, the higher the energy efficiency and the more reasonable the personnel density distribution, the lower the carbon emission is generally, a plays a quantitative adjustment function on this relationship, η(t) is the energy efficiency coefficient of the building at time t, p(t) is the personnel density of the building at time t, β is the second adjustment factor of the building carbon emission calculation model, which is used to adjust the influence degree of building carbon emission under the joint action of many factors such as seasonal influenced energy consumption, environmental influence index on carbon emission and actual energy demand of building. It comprehensively considers the complex influence of environmental factors (such as energy consumption change caused by seasonal change of temperature, humidity, etc.) and energy demand on building carbon emission, quantifies and adjusts the comprehensive influence of these factors on carbon emission by combining with other parameters, so that the building carbon emission calculation model is closer to the actual carbon emission situation caused by the change of many factors. γ is the third adjustment factor of the building carbon emission calculation model, which mainly adjusts the influence intensity of seasonal influenced energy consumption of building on carbon emission, which can reflect the contribution degree of seasonal energy consumption change to carbon emission, so as to make the model more accurately consider the influence of seasonal factors on building carbon emission and make the calculation result more accurate. ξ(t) is the energy consumption of the building at time t, which is used to describe the seasonal influenced energy consumption of the building, δ is the fourth adjustment factor of the building carbon emission calculation model, which further adjusts and corrects the carbon emission under the cross influence of many factors such as seasonal influenced energy consumption, environmental influence index on carbon emission and actual energy demand of building, κ' is the fifth adjustment factor of the building carbon emission calculation model, which is used to adjust the comprehensive influence intensity of environmental influence index on carbon emission and actual energy demand of building in carbon emission calculation, which reflects the comprehensive influence degree of the interaction between environmental factors (such as wind direction, sunshine, etc.) and building energy demand on carbon emission, E e (t) is the environmental influence index on carbon emission at time t, μ is the sixth adjustment factor of the building carbon emission calculation model, which mainly adjusts the weight of environmental influence index on carbon emission in carbon emission calculation. Environmental factors have direct or indirect influence on building carbon emission, such as air quality, climate conditions, etc., which will affect the energy utilization efficiency and carbon emission, U(t) is the actual energy demand of the building at time t, v is the seventh adjustment factor of the building carbon emission calculation model, which is used to adjust the influence of actual energy demand of building in carbon emission calculation.

[0056] The energy efficiency coefficient η(t) of the building at time t includes:

[0057]

[0058] wherein η0is an initial energy efficiency coefficient, a T is a weight of the ambient temperature, T(t) is the ambient temperature at time t, T base is a reference temperature, λ T is an adjustment factor of the ambient temperature, β H is a weight of the ambient humidity, H(t) is the ambient humidity at time t, H base is a reference humidity, γ' is an adjustment factor of the ambient humidity, a W is a weight of the ambient wind speed, W(t) is the ambient wind speed at time t, a L is a weight of the light intensity, L(t) is the light intensity at time t, L max is a maximum value of the light intensity.

[0059] The energy consumption ξ(t) of the building at time t comprises:

[0060]

[0061] wherein ζ0is a reference energy consumption, λ κ is a weight of the thermal conductivity, κ(t) is the thermal conductivity of the building at time t, a κ is an adjustment factor of the thermal conductivity, δ C is a weight of the air permeability, C(t) is the air permeability of the building at time t, β C is an adjustment factor of the air permeability, p a (t) is the air density inside the building at time t, γ ρ is an adjustment factor of the air density, V is the volume of the building, δ V is an adjustment factor of the volume of the building, μ F is a weight of the wind speed passing through the interior of the building, F (t) is the wind speed passing through the interior of the building at time t, a F is an adjustment factor of the wind speed passing through the interior of the building.

[0062] The environmental impact index E e (t) on carbon emissions at time t comprises:

[0063]

[0064] wherein a1is a first adjustment factor of the environmental impact index on carbon emissions, b1is a second adjustment factor of the environmental impact index on carbon emissions, g1is a third adjustment factor of the environmental impact index on carbon emissions, A veg (t) is the vegetation coverage of the region where the building is located at time t, d1is a fourth adjustment factor of the environmental impact index on carbon emissions, z1is a fifth adjustment factor of the environmental impact index on carbon emissions, C abs(t) is the CO2 absorption efficiency of the region where the building is located at time t.

[0065] The actual energy demand U(t) of the building at time t includes:

[0066]

[0067] wherein a2 is the first adjustment factor of the actual energy demand, b2 is the second adjustment factor of the actual energy demand, T in (t) is the temperature inside the building at time t, g2 is the third adjustment factor of the actual energy demand, d2 is the fourth adjustment factor of the actual energy demand, H in (t) is the humidity inside the building at time t, k" is the fifth adjustment factor of the actual energy demand.

[0068] In addition, in the sample data set, the input variables include:

[0069] window-wall ratio, external wall heat transfer coefficient, roof heat transfer coefficient, external window heat transfer coefficient, external wall solar radiation absorption coefficient, roof solar radiation absorption coefficient, external window solar radiation absorption coefficient, air change frequency, air conditioning heating temperature, air conditioning cooling temperature, lighting power density, shading angle;

[0070] The input variables include: building energy consumption, carbon emission, daylighting, thermal comfort.

[0071] Preferably, in step one, based on the building design parameter range of different climate zones, Latin hypercube sampling method is used to sample the input parameters, and for each climate zone, a set of test schemes are obtained, then Grasshopper is used to simulate the building performance, and the data set obtained by simulation is used as the sample of the prediction model;

[0072] Then Ladybug and Honeybee are used to simulate the building energy consumption, carbon emission, daylighting, and thermal comfort.

[0073] Step two, build a BO-CatBoost prediction model, use the BO-CatBoost prediction model to predict the building daylighting performance, building energy consumption, building thermal comfort, and building carbon emission, and establish the nonlinear mapping relationship between each influencing factor and building energy consumption, carbon emission, daylighting, and thermal comfort;

[0074] In this step, in order to improve the prediction accuracy of CatBoost, the three hyperparameters of depth, l2_leaf_reg and learning_rate are optimized by Bayesian. Then the BO-CatBoost prediction model is trained with sample data set to learn the relationship between building design parameters and building performance, that is, to establish the nonlinear mapping relationship between each influencing factor and building energy consumption, carbon emission, lighting, thermal comfort.

[0075] Step three, the nonlinear mapping relationship is taken as the fitness function of NSGA-III, and different climate zone variables are taken as constraint conditions to construct a multi-objective prediction function. The prediction function is used to balance and optimize the four targets of building energy consumption, carbon emission, lighting and thermal comfort to obtain the optimal solution set.

[0076] In the construction of BO-CatBoost prediction model, the optimal hyperparameter combination of CatBoost is obtained by BO:

[0077] (1) Randomly initialize n0 points

[0078] (2) Obtain the corresponding function value f(X init ), then the initial point set D0={X init , f(X init )}, let t=n0, D t-1 =D0, loop this process until the maximum iteration number;

[0079] (3) According to the obtained point set D t-1 , construct the surrogate model S(x), based on the surrogate model S(x), maximize the acquisition function θ(x|D t-1 ), obtain the next evaluation point x t =argmaxθ(x|D t-1 );

[0080] (4) Add the function value f(X t ) of the evaluation point to the current evaluation point set, D t =D t-1 ∪{x t , f(x t )};

[0081] (5) Output the optimal hyperparameter combination of CatBoost.

[0082] In step three, the objective function of the multi-objective prediction function includes:

[0083] f1=min f EUI (CatBoost regression(x1,x2,x3,x4,x5,x6,x7,x8,x9,x10 x11, x 12 x 13 x 14 x 15 ))

[0084] f2 = min f CEI (CatBoost regression(x1, x2, x3, x4, x5, x6, x7, x8, x9, x 10 x 11 x 12 x 13 x 14 x 15 ))

[0085] f3 = min f PMV |CatBoost regression(x1, x2, x3, x4, x5, x6, x7, x8, x9, x 10 x 11 x 12 x 13 x 14 x 15 )|

[0086] f4 = min f UDI (CatBoost regression(x1, x2, x3, x4, x5, x6, x7, x8, x9, x 10 x 11 x 12 x 13 x 14 x 15 ))

[0087] wherein x1, x2, x3, x4, x5, x6, x7, x8, x9, x 10 x 11 x 12 x 13 x 14 x 15 are window-wall ratio-east, window-wall ratio-west, window-wall ratio-south, window-wall ratio-north, external wall heat transfer coefficient, roof heat transfer coefficient, external window heat transfer coefficient, external wall solar radiation absorption coefficient, roof solar radiation absorption coefficient, external window solar radiation absorption coefficient, air change frequency, air conditioning and heating temperature, air conditioning cooling temperature, lighting power density, shading angle, f EUI is building energy consumption, f CEI is carbon emission, f PMV is thermal comfort, f UDI is daylighting.

[0088] In step three, the NSGA-III algorithm is used to obtain the optimal solution set, including:

[0089] The ideal point method is used to calculate the distance of each solution in the Pareto optimal solution set from the ideal point, and the point closest to the ideal point is found, which is the unique optimal solution;

[0090] The calculation model of the distance of each solution from the ideal point includes:

[0091]

[0092] In the formula, (x pareto , y pareto , z pareto , w pareto ) is the coordinate corresponding to the optimal Pareto front, and (x Epoint , y Epoint , z Epoint , w Epoint ) is the coordinate corresponding to the ideal point.

[0093] Example 2

[0094] This embodiment is a newly built two-class multi-storey office building, which consists of office buildings, public buildings, service buildings and equipment buildings, etc. The building form is "L" type, with 4 layers, the building height is 15.3m, the building area is 1223.62m 2 , the total building area is 4701.39m 2 , the volume rate is 0.58, and the electricity is the main energy of the whole building. Rhino and Grasshopper are used for parametric modeling of the building. The "Code for Design of Thermal Engineering of Civil Buildings" divides the country into five climate zones, namely, severe cold region, cold region, hot summer and cold winter region, hot summer and warm winter region, and moderate region. In this paper, Shenyang, Beijing, Wuhan, Guangzhou and Kunming are selected as the representative cities of the five climate zones, and the corresponding climate data is imported into Grasshopper to analyze the influence of building design parameters on building performance under different climate conditions.

[0095] Building performance simulation and data set acquisition:

[0096] 1. Parametric model construction

[0097] In this embodiment, the geometric model of the building is constructed using the parametric modeling battery of Grasshopper; the daylighting simulation is calculated using the Radiance kernel in the Honeybee plug-in; the energy consumption simulation is constructed using the Openstudio battery with Energyplus as the kernel in the Honeybee plug-in; and the reading and writing of a large amount of data is completed using the Lunchbox plug-in.

[0098] The geometric model is established by using the Grasshopper platform. First, the model is captured by using Geometry and then split by layers. HB Room from solid is used to create rooms and set the building envelope parameters. Then, HB Apertures by Ratio is used to set the external windows according to the window-wall ratio, and HB Louver Shades is used to customize the shading components. Finally, HB Model is used to create the performance model.

[0099] The light performance evaluation model is established by using the Grasshopper platform. Based on the geometric model, first, HB Program Type is used to set the optical properties of the building, such as lighting power. Then, LB Download Weather is used to import the climate file. Next, HB Sensor Grid from Rooms is used to create the lighting calculation grid, which is set to 1m considering the model size and calculation time. Finally, all the cells are connected to the HB Annual Daylight simulation operator to perform light performance simulation.

[0100] The energy consumption evaluation model is established by using the Grasshopper platform. Based on the geometric model, first, LB Download Weather is used to import the climate file. Then, HB Program Type is used to set the device operation parameters and operation schedule, such as heating and cooling set temperature, device power density, and personnel density. Finally, all the cells are connected to the HB Model to OSM simulation operator to perform energy consumption simulation.

[0101] The thermal comfort evaluation model is established by using the Grasshopper platform. Based on the energy consumption evaluation model, first, HB Read Room Comfort Result is used to set the environmental parameters, such as air temperature, humidity, and wind speed. Then, the human metabolic rate and clothing insulation value are set. Finally, LB PMV Comfort simulation operator is used to simulate the human thermal comfort and obtain the PMV value.

[0102] The carbon emission evaluation model is established by using the Grasshopper platform. First, based on the energy consumption evaluation model, HB Carbon Emission Intensity is used to set the carbon emission factor of the power grid and output the operation carbon emission intensity. Then, Layer Import in the Cardinal LCA plugin is used to extract material information, and Report plugin is used to output the carbon emission in the production stage of building materials.

[0103] 2、Model parameter setting

[0104] By consulting the Public Building Energy Efficiency Design Standard (GB 50189-2015), the General Specification for Building Energy Conservation and Renewable Energy Utilization (GB 55015-2021), the General Specification for Building Environment (GB 55016-2021), the Standard for Building Lighting Design (GB 50034-2004), and other building envelope and energy system design requirements in China, the constraint conditions for building design parameters in different climate zones are obtained, as shown below.

[0105] Table 1 Range of building design parameters in different climate zones

[0106]

[0107] 3. Latin hypercube sampling to obtain test schemes

[0108] After determining the input parameters and their ranges, this paper uses ModeFrontier software to implement Latin hypercube sampling of input parameters. For each climate zone, 1000 test schemes are obtained, then Grasshopper is used for building performance simulation, and the simulation data set is used as the sample of the prediction model. As shown in Table 4 3, take the hot summer and cold winter area as an example to show the test schemes.

[0109] Table 2 Latin hypercube sampling test schemes in hot summer and cold winter area

[0110]

[0111] 4. Building performance simulation based on Ladybug+Honeybee

[0112] Based on the above test schemes, building energy consumption, carbon emissions, PMV, and lighting are simulated using Ladybug+Honeybee. According to the simulation results, it can be seen that building design parameters are very important for building energy saving. Under the same climate conditions, different design parameters will greatly affect the simulation results of building performance. For example: in Shenyang, a severe cold area, the minimum and maximum values of building carbon emission simulation results are 599.38 kg / m 2 and 884.11 kg / m 2 , with a difference of 294.73 kg / m 2At the same time, there are obvious differences in building performance under different climate conditions. For example, the carbon emissions of buildings in cold regions are significantly higher than those in other regions, but the building energy consumption is not the highest. This is mainly because the design parameters of the building envelope in cold regions are more stringent, which to some extent reduces the impact of climate conditions on building energy consumption. However, due to the lack of clean electricity in the power grid in cold regions, the carbon emission factor of the power grid is the highest among the five regions, which greatly increases the carbon emissions of buildings in cold regions. In temperate regions, the climate is pleasant, with small temperature differences between summer and winter, so the energy consumption and carbon emissions in this region are significantly lower than in other regions. In summary, for different climate conditions, reasonable optimization of building design parameters is very important to achieve carbon reduction in buildings.

[0113] Building performance prediction and accuracy analysis:

[0114] 1. Model hyperparameter setting

[0115] In order to improve the prediction accuracy of CatBoost, this embodiment uses Bayesian optimization to optimize the three hyperparameters of depth, l2_leaf_reg and learning_rate. Based on the Python platform, the number of Bayesian optimization iterations is set to 200. Taking the hot summer and cold winter region as an example, the Bayesian optimization hyperparameter table 4 gives the CatBoost model hyperparameter optimization results.

[0116] Table 3 CatBoost hyperparameter optimization range

[0117]

[0118] Table 4 Optimal hyperparameter combination in hot summer and cold winter region

[0119]

[0120] 2. Prediction accuracy analysis

[0121] Based on the 1000 sample data sets obtained by Latin hypercube and L+H simulation, 800 data sets are randomly selected as the training set and 200 data sets are randomly selected as the test set according to the ratio of 4:1, and the BO-CatBoost prediction model is trained. Figure 3 Taking the hot summer and cold winter region as an example, the model prediction results are visualized. The prediction evaluation index of building performance in five climate zones is calculated respectively, and the results are shown in Table 5. Figure 3 It can be seen intuitively that the BO-CatBoost model can well predict the building performance in the hot summer and cold winter region. The model has fully learned the relationship between building design parameters and building performance, and the fitting error of the training set and the test set is very low.

[0122] Table 5 BO-CatBoost prediction accuracy

[0123]

[0124] To verify the superiority of the BO-CatBoost model proposed in this paper, four commonly used machine learning algorithms, RF, GBDT, BPNN, CatBoost and BO-CatBoost model are selected for comparison. Taking the hot summer and cold winter region as an example, the prediction accuracy of each model is compared as shown in Figure 4 Figure 4 It can be seen that BO-CatBoost has the highest R2 and the lowest RMSE in predicting building energy consumption, lighting, thermal comfort and carbon emissions, which shows that the prediction performance of BO-CatBoost is better than the other four methods. In addition, compared with CatBoost model without Bayesian optimization, BO-CatBoost has better goodness of fit and smaller error, which shows that BO optimization of CatBoost's hyperparameters is an effective method that can significantly improve the prediction accuracy of the model.

[0125] Analysis of building performance influencing factors:

[0126] In order to further study the internal relationship between building design parameters and building carbon emissions, this paper uses the advantage of CatBoost model in explainability to quantitatively analyze the influence strength of 15 influencing factors, and obtains the importance ranking of 15 influencing factors.

[0127] Multi-objective optimization of building performance in different climate zones:

[0128] When using NSGA-III algorithm to predict the output results of optimization objectives, the mapping relationship between building design parameters and building energy consumption, carbon emissions, thermal comfort and lighting is established, and the objective functions of the four optimization objectives are obtained according to the following formula:

[0129] f1=min f EUI (CatBoost regression(x1,x2,x3,x4,x5,x6,x7,x8,x9,x 10 ,x 11 ,x 12 ,x 13 ,x 14 ,x 15 ))

[0130] f2=min f CEI (CatBoost regression(x1,x2,x3,x4,x5,x6,x7,x8,x9,x 10 ,x 11 ,x 12 ,x 13 ,x​14 ,x 15 ))

[0131] f3 = min f PMV |CatBoost regression(x1,x2,x3,x4,x5,x6,x7x8,x9,x 10 ,x 11 ,x 12 ,x 13 ,x 14 ,x 15 )|

[0132] f4 = min f UDI (CatBoost regression(x1,x2,x3,x4,x5,x6,x7,x8,x9,x 10 ,x 11 ,x 12 ,x 13 ,x 14 ,x 15 ))

[0133] Wherein, f1 refers to EUI (building energy density), f2 refers to CEI (carbon emission intensity), f3 refers to PMV (average thermal comfort index), and f4 refers to UDI (effective natural light intensity). x1, x2, x3, x4, x5, x6, x7, x8, x9, x 10 , x 11 , x 12 , x 13 , x 14 , x 15 respectively refer to WWR-E, WWR-W, WWR-S, WWR-N, EWHTC, RHTC, OWHTC, EWSAC, RSAC, OWSAC, AT, ACHT, ACCT, LPD, and SA.

[0134] Constraint: According to the relevant building energy-saving design specification, the value range of each design parameter is determined, and the constraint condition of the multi-objective optimization model in different climate zones is determined according to Table 6.

[0135] Table 6 Variable constraint conditions in different climate zones

[0136]

[0137] Multi-objective optimization result analysis

[0138] The EUI, UDI, PMV and CEI in five different climate zones are optimized by using NSGA-III algorithm, the maximum evolution number is set to 100, the population size is set to 100, and finally 84 optimal solutions are obtained, and the obtained Pareto front distribution is as followsFigure 4 As shown.

[0139] According to the ideal point method, the optimal solution in the Pareto solution set is calculated, and five optimal schemes of the regions are obtained. The optimized scheme and the initial design scheme are compared, and the results are shown in Table 7.

[0140] According to the table, for buildings in different climate zones, the BO-CatBoost-NSGA-III algorithm has an optimization effect on building performance. Among them, the EUI target is optimized by 18.79%-46.82%, the UDI target is optimized by 2.29%-6.43%, the PMV target is optimized by 11.74%-25.22%, and the CEI target is optimized by 19.97%-35.37%. Therefore, the BO-CatBoost-NSGA-III algorithm proposed in this paper can realize the optimization design of building energy saving in different climate zones by improving the building design parameters, which is helpful to promote the building energy saving and emission reduction work in different climate zones.

[0141] Table 7 Comparison of optimization scheme results

[0142]

[0143] Based on the content of the above device embodiment, as an optional embodiment, the BO-RF-NSGA-III-based shield underpass construction pier displacement optimization device provided in the embodiment of the present application is as follows: Figure 2 As shown, it also includes a fourth sub-module for implementing the TOPSIS method to determine the optimal shield construction parameters that meet the vertical displacement and horizontal displacement of the pier caused by the shield underpass construction according to the obtained Pareto optimal solution set, including: the actual working condition of the shield tunnel underpassing the high-speed railway bridge, obtaining the optimal shield construction parameters when the shield tunnel underpassing the high-speed railway bridge according to the parameter range obtained from the Pareto optimal solution set, and reducing the pier displacement caused by the shield underpass construction.

[0144] The method of the embodiment of the present application is realized by relying on an electronic device, so it is necessary to introduce the related electronic device. For this purpose, the embodiment of the present application provides an electronic device, which includes at least one processor, a communications interface, at least one memory, and a communications bus, wherein the at least one processor, the communications interface, and the at least one memory complete mutual communication through the communications bus. The at least one processor can call the logical instructions in the at least one memory to execute all or part of the steps of the method provided by the foregoing various method embodiments.

[0145] In addition, the logic instructions in the at least one memory described above can be implemented in the form of a software functional unit and sold or used as an independent product, and can be stored in a computer readable storage medium. Based on such understanding, the technical solutions of the present application essentially or the parts that contribute to the prior art or parts of the technical solutions can be embodied in the form of a software product. The computer software product is stored in a storage medium, and includes a plurality of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in the various method embodiments of the present application. The aforementioned storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), a magnetic disk or an optical disk, and various media that can store program codes.

[0146] The device embodiments described above are only schematic, wherein the units described as separate components can or can not be physically separate, and the components displayed as units can or can not be physical units, i.e., can be located in one place, or can be distributed on multiple network units. Part or all of the modules can be selected to achieve the purpose of the present embodiment according to actual needs. Those skilled in the art can understand and implement without creative labor.

[0147] From the description of the above embodiments, those skilled in the art can clearly understand that each embodiment can be realized by means of software and the necessary general hardware platform, and of course can also be realized by hardware. Based on such understanding, the above technical solutions essentially or the parts that contribute to the prior art can be embodied in the form of a software product. The computer software product can be stored in a computer readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes a plurality of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0148] Those skilled in the art can easily understand that the above description is only a preferred embodiment of the present application, and is not intended to limit the present application. Any modification, equivalent replacement and improvement made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A multi-objective optimization method for building performance based on intelligent algorithm, characterized in that: The following steps are involved: Step 1: parametric simulation of building performance is performed to obtain a sample data set; Step 2: Construct a BO-CatBoost prediction model. Use the BO-CatBoost prediction model to predict building lighting performance, building energy consumption, building thermal comfort, and building carbon emissions, and establish a nonlinear mapping relationship between each influencing factor and building energy consumption, carbon emissions, lighting, and thermal comfort. Step three: The nonlinear mapping relationship is used as the fitness function of NSGA-III, and the variables of different climate zones are used as constraints to construct a multi-objective prediction function. The prediction function is then used to balance and optimize the four objectives of building energy consumption, carbon emissions, daylighting, and thermal comfort to obtain the optimal solution set.

2. The multi-objective optimization method for building performance based on intelligent algorithm according to claim 1 is characterized in that: In step 1, the input variables in the sample data set include: Window-to-wall ratio, exterior wall heat transfer coefficient, roof heat transfer coefficient, exterior window heat transfer coefficient, exterior wall solar radiation absorption coefficient, roof solar radiation absorption coefficient, exterior window solar radiation absorption coefficient, ventilation rate, air conditioning heating temperature, air conditioning cooling temperature, lighting power density, shading angle; Input variables include: building energy consumption, carbon emissions, daylighting, and thermal comfort.

3. The multi-objective optimization method for building performance based on intelligent algorithm according to claim 1 is characterized in that: The building carbon emission calculation model includes: E build (t)=exp(α·ln(η(t)·ρ(t)))+β·(1+γ·ξ(t)) δ ·(1+κ′·E e (t) μ ·U(t) v ) Among them, E build (t) is the carbon emission index of the building at time t, α is the first adjustment factor of the building carbon emission calculation model, η(t) is the energy efficiency coefficient of the building at time t, ρ(t) is the occupant density of the building at time t, β is the second adjustment factor of the building carbon emission calculation model, γ is the third adjustment factor of the building carbon emission calculation model, ξ(t) is the energy consumption of the building at time t, which is used to describe the energy consumption of the building affected by the season, δ is the fourth adjustment factor of the building carbon emission calculation model, κ′ is the fifth adjustment factor of the building carbon emission calculation model, E e (t) is the environmental impact index on carbon emissions at time t, μ is the sixth adjustment factor of the building carbon emission calculation model, U(t) is the actual energy demand of the building at time t, and ν is the seventh adjustment factor of the building carbon emission calculation model.

4. The multi-objective optimization method for building performance based on intelligent algorithm according to claim 1 is characterized in that: In step 3, when constructing the BO-CatBoost prediction model, BO is used to obtain the optimal hyperparameter combination of CatBoost: (1) Randomly initialize n0 points (2) Obtain the corresponding function value f(X init ), then the initial point set D0={X init ,f(X init )}, let t = n0, D t-1 =D0, loop this process until the maximum number of iterations; (3) According to the obtained point set D t-1 , build a proxy model S(x), based on the proxy model S(x), maximize the acquisition function θ(x|D t-1 ), get the next evaluation point x t =argmaxθ(x|D t-1 ); (4) The function value f(X t ) is added to the current evaluation point set, D t =D t-1 ∪{x t ,f(x t )}; (5) Output the optimal hyperparameter combination of CatBoost.

5. The multi-objective optimization method for building performance based on intelligent algorithm according to claim 1 is characterized in that: In step 3, the objective function of the multi-objective prediction function includes: f1=min f EUI (CatBoost regression(x1,x2,x3,x4,x5,x6,x7,x8,x9,x 10 ,x 11 ,x 12 ,x 13 ,x 14 ,x 15 )) f2=min f CEI (CatBoost regression(x1,x2,x3,x4,x5,x6,x7,x8,x9,x 10 ,x 11 ,x 12 ,x 13 ,x 14 ,x 15 )) f3=min f PMV |CatBoost regression(x1,x2,x3,x4,x5,x6,x7,x8,x9,x 10 ,x 11 ,x 12 ,x 13 ,x 14 ,x 15 )| f4=min f UDI (CatBoost regression(x1,x2,x3,x4,x5,x6,x7,x8,x9,x 10 ,x 11 ,x 12 ,x 13 ,x 14 ,x 15 )) Where x1, x2, x3, x4, x5, x6, x7, x8, x9, x 10 , x 11 , x 12 , x 13 , x 14 , x 15 They are window-to-wall ratio-east, window-to-wall ratio-west, window-to-wall ratio-south, window-to-wall ratio-north, exterior wall heat transfer coefficient, roof heat transfer coefficient, exterior window heat transfer coefficient, exterior wall solar radiation absorption coefficient, roof solar radiation absorption coefficient, exterior window solar radiation absorption coefficient, ventilation rate, air conditioning heating temperature, air conditioning cooling temperature, lighting power density, shading angle, f EUI is the building energy consumption, f CEI is carbon emissions, f PMV is thermal comfort, f UDI For lighting.

6. The multi-objective optimization method for building performance based on intelligent algorithm according to claim 1 is characterized in that: In step 3, the NSGA-III algorithm is used to obtain the optimal solution set, including: The ideal point method is used to calculate the distance between each solution and the ideal point in the Pareto frontier optimal solution set, and the point closest to the ideal point is found, which is the only optimal solution.

7. The multi-objective optimization method for building performance based on intelligent algorithm according to claim 6 is characterized in that: The calculation model of the distance of each dissociation ideal point includes: In the formula, (x pareto ,y pareto , z pareto , w pareto ) is the coordinate corresponding to the optimal Pareto frontier, (x Epoint ,y Epoint , z Epoint , w Epoint ) are the coordinates corresponding to the ideal point.

8. A multi-objective building performance optimization system based on intelligent algorithms, characterized by: include: The first main control module is used to perform parametric simulation of building performance to obtain a sample data set; The second main control module is used to build a BO-CatBoost prediction model, which is used to predict building lighting performance, building energy consumption, building thermal comfort, and building carbon emissions, and to establish a nonlinear mapping relationship between each influencing factor and building energy consumption, carbon emissions, lighting, and thermal comfort; The third main control module is used to use the nonlinear mapping relationship as the fitness function of NSGA-III and the variables of different climate zones as constraints to construct a multi-objective prediction function, and use this prediction function to balance and optimize the four objectives of building energy consumption, carbon emissions, daylighting, and thermal comfort to obtain the optimal solution set.

9. An electronic device, characterized in that: include: At least one processor, at least one memory and a communication interface; wherein the processor, memory and communication interface communicate with each other; the memory stores program instructions that can be executed by the processor, and the processor calls the program instructions to execute a multi-objective optimization method for building performance based on an intelligent algorithm as described in any one of claims 1 to 7.

10. A non-transitory computer-readable storage medium, characterized in that The non-transitory computer-readable storage medium stores computer instructions, which enable the computer to execute the multi-objective optimization method for building performance based on an intelligent algorithm as described in any one of claims 1 to 7.

Images