A method for generating architectural prototypes based on urban big data

Abstract

A method for generating building prototypes based on urban big data includes the following steps: step 1: calculating characteristic parameters of residential land and residential buildings; step 2: deleting redundant characteristic parameters based on the correlation between the characteristic parameters of residential land and residential buildings; determining cluster characteristic parameters of residential land and residential buildings for clustering based on the correlation between the retained characteristic parameters of residential land and residential buildings and the predicted energy consumption values ​​of residential buildings; step 3: performing two-stage clustering based on the determined residential land cluster centers and residential building cluster centers to obtain the optimal residential building clustering results focused on the optimal residential land cluster centers to generate a building prototype set; step 4: calculating building energy demand indicators based on the building prototype set; and when the comprehensive value of the building energy demand indicators is less than a set threshold, repeating step 3 to regenerate the building prototype set. The building prototypes meet the urban building energy demand forecast.

Description

Technical Field

[0001] The present invention belongs to the technical field of urban building model construction, relates to the operation and management stage of urban buildings, and is suitable for generating and constructing urban-scale building prototype models. Background Art

[0002] In recent years, with the increasing diversity of urban building forms, the differences between buildings have posed significant challenges to city-scale building modeling and energy demand analysis. This highlights the critical role of city-scale building prototypes, which can accurately capture the energy consumption characteristics of different building types and support the assessment and analysis of the overall energy performance of urban building complexes. Furthermore, standardized urban building prototypes provide a scientific basis for urban planning and energy-saving policies, facilitating the optimization of design and the implementation of energy-saving strategies. However, there is currently a lack of a comprehensive city-scale modeling system, as well as objective calculation methods and standards to effectively guide the analysis and prototyping of building energy demand.

[0003] Existing methods for automatically generating urban building cluster forms based on artificial intelligence (AI) and those based on building vector outlines generate urban residential building cluster models based on building vector outlines and height morphology. The selected building characteristic parameters focus on geometric form and building topology. The building prototypes generated by these methods fail to fully consider non-geometric topological factors in urban planning (such as green space ratio, housing density, and land form, which are important indicators in urban planning), which significantly impact building demand analysis. Consequently, ignoring these indicators can lead to inaccurate assessments of building cluster energy consumption, limiting the models' applicability in real-world urban-scale modeling. Furthermore, comprehensive regional building cluster load forecasting methods and dynamic evolutionary regional building cluster load forecasting methods both construct representative building clusters to achieve energy demand analysis. While these methods can optimize and reduce modeling workload to a certain extent, their classification criteria are often general and crude. These techniques fail to fully capture the diversity of building forms at the urban scale and lack specificity and adaptability, particularly in complex and diverse urban environments. This limitation makes existing models inadequate for large-scale urban building energy consumption analysis, making it difficult to provide accurate energy assessment results. Therefore, there is an urgent need to develop a model construction method that can take into account the architectural diversity at the scale of urban architectural big data and the impact of urban planning indicators on architectural form, so as to improve the construction accuracy and energy consumption representativeness of architectural prototypes at the urban scale. Summary of the Invention

[0004] To address the shortcomings of existing technologies, this paper provides a building prototype generation method based on urban big data. This method combines urban planning indicators with building form parameters to construct a representative building prototype model of energy consumption. This model can be used to accurately and effectively predict the energy demand of urban buildings.

[0005] The present invention adopts the following technical solutions.

[0006] The present invention proposes a method for generating building prototypes based on urban big data, comprising:

[0007] Step 1: Obtain the number and area of ​​each residential land, and obtain the building base area, number of floors, and height of each residential building belonging to each residential land to calculate various characteristic parameters of the residential land and residential buildings; the characteristic parameters of the residential land include: volume ratio, building density, building height limit, land compactness index, and high-rise ratio; the characteristic parameters of the residential buildings include: building shape coefficient;

[0008] Step 2: based on the correlation between each characteristic parameter of the residential land and each characteristic parameter of the residential building, redundant characteristic parameters are deleted; after the deletion process, based on the correlation between each characteristic parameter of the retained residential land and the predicted energy consumption value of the residential building, cluster characteristic parameters of the residential land for clustering are determined; at the same time, based on the correlation between each characteristic parameter of the retained residential building and the predicted energy consumption value of the residential building, cluster characteristic parameters of the residential building for clustering are determined;

[0009] Step 3: Determine the residential land cluster center and the residential building cluster center respectively based on the evaluation indicators of residential land and residential buildings; based on the determined residential land cluster center and residential building cluster center, use the clustering characteristic parameters of residential land and residential buildings to perform two-stage clustering to obtain the optimal residential building clustering result focused on the optimal residential land cluster center; generate a building prototype set based on the optimal residential building clustering result focused on the optimal residential land cluster center;

[0010] Step 4: Based on the building prototype set, calculate the building energy demand indicators separately, including: the peak load density of cooling and heating demand throughout the year, and the total intensity of cooling and heating load demand throughout the year; when the comprehensive value of the building energy demand indicators is less than the set threshold, repeat step 3 to regenerate the building prototype set; when the comprehensive value of the building energy demand indicators is not less than the set threshold, output the building prototype set.

[0011] Preferably, the floor area ratio satisfies the following relationship:

[0012]

[0013] Where FAR i is the volume ratio of the i-th residential land, BBA i,j is the building base area of ​​the jth residential building belonging to the i-th residential land, Storeys i,j is the number of floors of the jth residential building belonging to the i-th residential land, LAAi is the area of ​​the i-th residential land, m i is the set of residential buildings belonging to the i-th residential land;

[0014] The building density satisfies the following relationship:

[0015]

[0016] Where, BCR i is the building density of the i-th residential land;

[0017] The building height limit satisfies the following relationship:

[0018] MAH i =max(BUH i,j ) (3)

[0019] Where, MAH i is the building height limit for the i-th residential land, BUH i,j is the height of the jth residential building belonging to the i-th residential land;

[0020] The land compactness index satisfies the following relationship:

[0021]

[0022] Where, NCI i is the compactness index of the i-th residential land, LAP i is the perimeter of the i-th residential land, LAA i is the area of ​​the i-th residential land;

[0023] The proportion of high-rise buildings satisfies the following relationship:

[0024]

[0025] Where, HRP i is the proportion of high-rise buildings in the i-th residential land;

[0026] The building shape coefficient satisfies the following relationship:

[0027]

[0028] Where, Building surface area i,j It is the surface area of ​​the jth residential building belonging to the i-th residential land that is in contact with the outdoor atmosphere.

[0029] Preferably, step 2 includes:

[0030] When the Spearman correlation coefficient between the characteristic parameters of the residential land and the characteristic parameters of the residential buildings is less than the set first limit value, delete the corresponding characteristic parameters of the residential land and the characteristic parameters of the residential buildings; otherwise, retain the corresponding characteristic parameters of the residential land and the characteristic parameters of the residential buildings;

[0031] When the Spearman correlation coefficient between the retained characteristic parameters of the residential land and the energy consumption prediction values of each residential building belonging to the residential land is not less than the set second limit value, determine the retained characteristic parameters of the residential land as the clustering characteristic parameters; otherwise, the retained characteristic parameters of the residential land are the standby characteristic parameters;

[0032] When the Spearman correlation coefficient between the retained characteristic parameters of the residential buildings and the energy consumption prediction values of the residential buildings is not less than the set second limit value, determine the retained characteristic parameters of the residential buildings as the clustering characteristic parameters; otherwise, the retained characteristic parameters of the building land are the standby characteristic parameters.

[0033] Preferably, the first limit value is 0.5.

[0034] Preferably, the second limit value is 0.7.

[0035] Preferably, step 3 includes:

[0036] Step 3.1, determine the residential land clustering centers according to the various evaluation indexes of the residential land, and determine the residential building clustering centers according to the various evaluation indexes of the residential buildings;

[0037] Step 3.2, extract n1 residential land clustering centers as the first cluster centers of the residential land, where 3 ≤ n1 < N, and N is the number of residential land clustering centers; use the K-medoids clustering method to cluster the clustering characteristic parameters of the residential buildings with respect to the first cluster centers of the residential land, and use the Bayesian information criterion scoring method to score the clustering results to obtain the first score corresponding to the first cluster centers of the residential land; randomly extract residential land clustering centers different from each first cluster center of the residential land, use the K-medoids clustering method to cluster the clustering characteristic parameters of the residential buildings that have been clustered in the first cluster centers of the residential land with respect to the extracted residential land clustering centers, and use the Bayesian information criterion scoring method to score the clustering results to obtain the second score corresponding to each first cluster center of the residential land; when the second score is greater than or equal to the first score, retain the extracted residential land clustering centers; otherwise, when the second score is less than the first score, delete the extracted residential land clustering centers;

[0038] After repeating the above operations, obtain the optimal residential land clustering centers and the residential buildings focused by the optimal residential land clustering centers;

[0039] Step 3.3: Use the K-medoids clustering method to cluster the residential buildings focused on the optimal residential land cluster center into the residential building cluster center. Use the Bayesian Information Criterion scoring method to score the clustering results. When the score is greater than the set threshold, the optimal residential building cluster result focused on the optimal residential land cluster center is obtained.

[0040] Step 3.4: Generate a set of building prototypes based on the optimal residential building clustering results focused on the optimal residential land cluster center.

[0041] Preferably, the evaluation indicators for residential land include but are not limited to: land area, volume ratio, building density, building height limit, green space ratio, land compactness index, unit density, and high-rise ratio; the evaluation indicators for residential buildings include but are not limited to: projected area, building height, and building shape coefficient;

[0042] The number N of residential land cluster centers is consistent with the number of evaluation indicators of residential land; the number M of residential building cluster centers is consistent with the number of evaluation indicators of residential buildings.

[0043] Preferably, in step 3.2, the residential land evaluation index of the optimal residential land cluster center is used as a key index for evaluating residential land and energy consumption prediction.

[0044] Preferably, the scores of the clustering results scored by the Bayesian Information Criterion scoring method satisfy the following relationship:

[0045]

[0046] Where, is the maximum log-likelihood function of the clustering feature parameters of the i-th group, is the clustering characteristic parameter x of the i-th group i Belongs to cluster i C i The maximum likelihood estimate of the clustering characteristic parameter x of the i-th group i Including clustering characteristic parameters of residential land and residential buildings; n i is the i-th cluster C i The number of samples in , p is the number of features of the cluster.

[0047] Preferably, in step 3.3, the scores of the clustering results using the Bayesian Information Criterion scoring method are processed into a range of [0, 1], and the threshold value is set to be no less than 0.62.

[0048] Preferably, the building prototype set includes K building prototypes. For the k-th building prototype, k=1, 2, ..., K, the building energy demand index satisfies the following relationship:

[0049]

[0050]

[0051] Where, is the total intensity of cooling and heating load demand of the k-th building prototype throughout the year, is the peak load density of cooling and heating demand of the k-th building prototype throughout the year, h is the hour, is the hourly energy consumption intensity of the building.

[0052] Preferably, the normalized value of the weighted sum of the total intensity of the cooling and heating load demand of the k-th building prototype throughout the year and the peak load density of the cooling and heating demand throughout the year is used as the comprehensive value of the building energy demand index; the threshold is set to 0.88.

[0053] The present invention also provides a terminal, comprising a processor and a storage medium; the storage medium is used to store instructions; and the processor is used to operate according to the instructions to execute steps of the method.

[0054] The present invention also relates to a computer-readable storage medium having a computer program stored thereon, which implements the steps of the method when the program is executed by a processor.

[0055] The present invention offers significant advantages over existing technologies, including at least one: a method combining a two-stage clustering approach based on urban big data with an automated building performance simulation program to generate building prototypes at the city scale. Compared to existing technologies, the method proposed in this paper can be applied to building cluster-scale model construction and energy demand analysis at the city scale. It leverages the advantages of clustering algorithms in data mining and feature extraction, significantly reducing the computational cost of building energy demand simulation at the city scale, improving the scientific nature and accuracy of system evaluation, and more objectively, comprehensively, and efficiently assessing building energy demand characteristics at the city scale. This method utilizes a two-stage plot-building clustering approach to obtain a set of residential building prototypes. Compared to existing technologies, the data-driven clustering algorithm introduced in this paper enables fully objective feature analysis and exhibits good scalability. It is applicable to building clusters at different regional scales, ensuring that the calculation results truly reflect the differences in energy usage properties of different residential buildings, and achieving city-scale building energy demand feature extraction at an acceptable computational cost. This method can more efficiently assess urban energy demand and quantify and index urban building energy consumption indicators.

[0056] The method proposed in this invention can more accurately construct building prototypes, thereby providing a scientific basis for city-scale energy demand forecasting and helping to achieve the optimization and sustainable development of urban energy systems. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] Figure 1The present invention provides a flow chart of a method for generating a building prototype based on urban big data. DETAILED DESCRIPTION

[0058] To make the objectives, technical solutions, and advantages of the present invention more clear, the technical solutions of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention. The embodiments described in this application are only part of the embodiments of the present invention, not all of them. Based on the spirit 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.

[0059] This paper provides a big data-based building prototype generation method applicable to urban scales, specifically for analyzing building energy demand. Compared to existing technologies, this method considers both site characteristics and the impact of non-geometric topological factors on building energy consumption, making it applicable to energy demand forecasting at all stages, from planning and design to building operation.

[0060] like Figure 1 As shown, including:

[0061] Step 1: Obtain the number and area of ​​each residential land, and obtain the building base area, number of floors and height of each residential building belonging to each residential land to calculate the characteristic parameters of the residential land and the characteristic parameters of the residential buildings; among which, the characteristic parameters of the residential land include: volume ratio, building density, building height limit, land compactness index, and high-rise proportion, and the characteristic parameters of the residential buildings include: building shape coefficient.

[0062] Considering that residential areas are the most common land units in built residential communities, the land and building clustering method proposed in this invention refers to a two-stage clustering method from residential land (in plots) to residential buildings. This method can be applied to the prototype generation of urban building complexes of different scales, and comprehensively considers the impact of building geometric parameters and non-geometric parameters on energy consumption. Therefore, the clustering parameter group of land and buildings is first determined.

[0063] Compared with the existing technology, this invention combines the calculation of existing residential land characteristic parameters in a more systematic way, accurately describing the coupling relationship between land characteristics and building forms. The calculation methods of some characteristic parameters are shown in the following formulas:

[0064] The floor area ratio satisfies the following relationship:

[0065]

[0066] Where FAR i is the volume ratio of the i-th residential land, BBA i,jis the building base area of ​​the jth residential building belonging to the i-th residential land, Storeys i,j is the number of floors of the jth residential building belonging to the i-th residential land, LAA i is the area of ​​the i-th residential land, m i is the set of residential buildings belonging to the i-th residential land.

[0067] The building density satisfies the following relationship:

[0068]

[0069] Where, BCR i is the building density of the i-th residential land.

[0070] The building height limit satisfies the following relationship:

[0071] MAH i =max(BUH i,j ) (3)

[0072] Where, MAH i is the building height limit for the i-th residential land, BUH i,j is the height of the jth residential building belonging to the i-th residential land.

[0073] The land compactness index satisfies the following relationship:

[0074]

[0075] Where, NCI i is the compactness index of the i-th residential land, LAP i is the perimeter of the i-th residential land, LAA i is the area of ​​the i-th residential land.

[0076] The proportion of high-rise buildings satisfies the following relationship:

[0077]

[0078] Where, HRP i is the proportion of high-rise buildings in the i-th residential land.

[0079] The building shape coefficient satisfies the following relationship:

[0080]

[0081] Where, Building surface area i,j It is the surface area of ​​the jth residential building belonging to the i-th residential land that is in contact with the outdoor atmosphere.

[0082] Step 2: Delete redundant characteristic parameters based on the correlation between each characteristic parameter of residential land and each characteristic parameter of residential buildings; after the deletion process, determine the clustering characteristic parameters of residential land used for clustering based on the correlation between each characteristic parameter of retained residential land and the predicted energy consumption value of residential buildings, and at the same time determine the clustering characteristic parameters of residential buildings used for clustering based on the correlation between each characteristic parameter of retained residential buildings and the predicted energy consumption value of residential buildings.

[0083] The Spearman correlation coefficient is a typical indicator for evaluating the correlation between continuous or ordered variables. It does not make special requirements on the distribution characteristics of the original variables and has a wide range of applications. It can be applied even to discrete variables. Therefore, considering that the Spearman correlation coefficient has stronger data adaptability, the present invention introduces the Spearman correlation coefficient to analyze the strength of the correlation between each characteristic parameter and between the characteristic parameter and the energy consumption prediction value. This method avoids overfitting by eliminating redundant variables. Among them, the numerical range of the Spearman correlation coefficient of any two parameters is [-1,1]. Its absolute value is less than 0.3, indicating that the correlation is negligible; the absolute value is not less than 0.3 and less than 0.5, indicating low correlation; the absolute value is not less than 0.5 and less than 0.7, indicating moderate correlation; the absolute value is not less than 0.7 and less than 0.9, indicating high correlation; and the absolute value is not less than 0.9 and less than 1, indicating extremely high correlation. It is worth noting that the present invention uses the Spearman correlation coefficient to calculate the correlation between two parameters, which is a non-restrictive and preferred choice. Those skilled in the art can choose different correlation coefficients and calculation methods according to the characteristics of the parameters to be analyzed for correlation.

[0084] Before clustering urban residential building data, it is necessary to conduct correlation analysis and screening of characteristic parameters, including:

[0085] When the Spearman correlation coefficient between the characteristic parameters of residential land and the characteristic parameters of residential buildings is less than the set first limit, the corresponding characteristic parameters of residential land and residential buildings are deleted; otherwise, the corresponding characteristic parameters of residential land and residential buildings are retained; wherein the first limit is 0.5;

[0086] When the Spearman correlation coefficient between the characteristic parameter of the retained residential land and the predicted energy consumption value of each residential building belonging to the residential land is not less than the set second limit value, the characteristic parameter of the retained residential land is determined to be the cluster characteristic parameter; otherwise, the characteristic parameter of the retained residential land is the backup characteristic parameter; wherein the second limit value is 0.7;

[0087] When the Spearman correlation coefficient between the characteristic parameters of the retained residential building and the predicted energy consumption value of the residential building is not less than the set second limit, the characteristic parameters of the retained residential building are determined to be cluster characteristic parameters; otherwise, the characteristic parameters of the retained building land are determined to be spare characteristic parameters; wherein the second limit is 0.7;

[0088] Through the above deletion process, the influence of redundant parameters on the credibility of cluster analysis results is eliminated, and only parameters closely related to building energy consumption prediction are extracted and used as cluster feature parameters for subsequent cluster analysis and modeling. At the same time, parameters that are generally correlated with building energy consumption prediction but are still related are retained as spare feature parameters. When there are problems such as a small number of cluster feature parameters, the spare feature parameters can be used as approximate cluster feature parameters through data processing methods, thereby improving the reliability of cluster analysis results.

[0089] Step 3: Determine the residential land cluster center and the residential building cluster center respectively based on the evaluation indicators of residential land and residential buildings; based on the determined residential land cluster center and residential building cluster center, use the clustering characteristic parameters of residential land and residential buildings to perform two-stage clustering to obtain the optimal residential building clustering result focused on the optimal residential land cluster center; generate a building prototype set based on the optimal residential building clustering result focused on the optimal residential land cluster center.

[0090] Compared with the existing technology, the present invention proposes a two-stage X-medoids clustering method, which combines the Bayesian Information Criterion scoring method with the K-medoids clustering method. The Bayesian Information Criterion scoring method evaluates and scores each cluster obtained by the K-medoids clustering method in each stage, and determines whether the clustering result is optimal based on the scoring result, thereby achieving the best clustering result in a completely objective manner.

[0091] Among them, the Bayesian information criterion BIC satisfies the following relationship:

[0092]

[0093] Where, is the maximum log-likelihood function of the clustering feature parameters of the i-th group, is the clustering characteristic parameter x of the i-th group i Belongs to cluster i C i The maximum likelihood estimate of the clustering characteristic parameter x of the i-th group i Including clustering characteristic parameters of residential land and residential buildings; n i is the i-th cluster C i The number of samples in , p is the number of features of the cluster.

[0094] Specifically, step 3 includes:

[0095] Step 3.1: Determine the clustering centers of residential land according to various evaluation indexes of residential land, and determine the clustering centers of residential buildings according to various evaluation indexes of residential buildings.

[0096] Specifically, the various evaluation indexes of residential land include but are not limited to: land area, plot ratio, building density, building height limit, green space rate, land compactness index, unit density, and high-rise proportion; the various evaluation indexes of residential buildings include but are not limited to: projected area, building height, and building shape coefficient.

[0097] The number N of the clustering centers of residential land is consistent with the number of the evaluation indexes of residential land; the number M of the clustering centers of residential buildings is consistent with the number of the evaluation indexes of residential buildings.

[0098] Step 3.2: Extract n1 clustering centers of residential land as the first cluster centers of residential land, where 3 ≤ n1 < N. Use the K-medoids clustering method to cluster the clustering characteristic parameters of residential buildings with respect to the first cluster centers of residential land, and use the Bayesian information criterion scoring method to score the clustering results to obtain the first score S1 corresponding to the first cluster centers of residential land. Arbitrarily extract the clustering centers of residential land different from the first cluster centers of residential land, use the K-medoids clustering method to cluster the clustering characteristic parameters of the residential buildings that have been clustered in the first cluster centers of residential land with respect to the extracted clustering centers of residential land, and use the Bayesian information criterion scoring method to score the clustering results to obtain the second score S2 corresponding to the first cluster centers of residential land. When the second score is greater than or equal to the first score, retain the extracted clustering centers of residential land; conversely, when the second score is less than the first score, delete the extracted clustering centers of residential land.

[0099] After repeating the above operations, obtain the optimal clustering centers of residential land and the residential buildings focused by the optimal clustering centers of residential land.

[0100] Through the above operations, the division and determination of the optimal clusters of residential land are achieved, and the optimization of the evaluation indexes of residential land is also achieved. The evaluation indexes of residential land corresponding to the optimal clustering centers of residential land can be used as key indexes for evaluating residential land and energy consumption prediction.

[0101] Step 3.3: Use the K-medoids clustering method to cluster the residential buildings focused by the optimal clustering centers of residential land with respect to the clustering centers of residential buildings, and use the Bayesian information criterion scoring method to score the clustering results. When the score is greater than the set threshold, obtain the optimal clustering result of the residential buildings focused by the optimal clustering centers of residential land.

[0102] In the embodiment, the scores of the clustering results using the Bayesian Information Criterion scoring method are processed into a range of [0, 1], and the threshold value is set to be no less than 0.62.

[0103] The method proposed in the present invention uses the X-medoids clustering method to objectively divide the sample set into several typical clusters during the first stage of plot feature clustering. Residential buildings are considered to be the accompanying attributes of each individual plot. Although they are not selected as clustering features in this stage, they still need to be classified into each plot cluster. Afterwards, the second stage clustering process focuses on the residential building level. At the same time, Manhattan distance is selected as the measurement indicator for both stage clustering. Compared with the current clustering methods and energy consumption prediction models based on the geometric topological characteristics of building complexes, the two-stage clustering method adopted by the present invention not only takes into account the land use characteristics, but also takes into account the building form, and introduces non-geometric parameters as clustering features, giving a systematic, complete, and quantifiable parameter description, which provides a more scientific and accurate reference solution for energy consumption prediction at the scale of building complexes.

[0104] Step 3.4: Generate a set of building prototypes based on the optimal residential building clustering results focused on the optimal residential land cluster center.

[0105] Step 4: Based on the building prototype set, calculate the building energy demand indicators separately, including: the peak load density of cooling and heating demand throughout the year, and the total intensity of cooling and heating load demand throughout the year; when the comprehensive value of the building energy demand indicators is less than the set threshold, repeat step 3 to regenerate the building prototype set; when the comprehensive value of the building energy demand indicators is not less than the set threshold, output the building prototype set.

[0106] The building prototype set consists of K building prototypes. For the kth building prototype, where k = 1, 2, …, K, its energy demand is calculated using energy demand simulation, i.e., 8760 hourly cooling and heating loads. Generally, building energy system designers focus only on annual peak load and annual total demand. Compared to time-series hourly cooling and heating load data, cooling and heating demand per unit building area can eliminate the interference of floor area ratio on plot energy demand indicators. Therefore, this paper selects annual peak cooling and heating load density (PLI) and annual total cooling and heating load intensity (EUI) as performance evaluation indicators for residential building cooling and heating energy demand. The annual peak cooling and heating load density (PLI) describes the peak load characteristics of the corresponding building prototype, while the annual total cooling and heating load intensity (EUI) describes the total load characteristics of the corresponding building prototype. The introduction of these indicators eliminates the influence of time-series data and solves the difficulty of comparing energy demand data between plots. They also consider both the real-time load characteristics of buildings and their cumulative load demand throughout the year, providing a quantifiable assessment scheme for building complex-scale energy demand forecasting.

[0107] The building energy demand index satisfies the following relationship:

[0108]

[0109] Where, is the total intensity of cooling and heating load demand of the k-th building prototype throughout the year, is the peak load density of cooling and heating demand of the k-th building prototype throughout the year, h is the hour, is the hourly energy consumption intensity of the building.

[0110] The normalized value of the weighted sum of the total annual cooling and heating load demand intensity and the annual cooling and heating demand peak load density of the k-th building prototype is used as the comprehensive value of the building energy demand index; the threshold is set at 0.88.

[0111] This invention provides important theoretical and technical support for energy demand analysis during the operation and maintenance phase of urban buildings. The method of the present invention can provide an assessment basis for energy demand in urban-scale buildings, facilitating scenario comparison and optimal decision-making. A building prototype generation method based on urban big data is proposed. A two-stage clustering analysis method based on plots and buildings is used to determine the optimal number of building prototypes. Furthermore, based on the classification of building clusters and a building energy demand simulation method, building performance simulation is performed on the set of building prototypes to obtain their energy demand characteristics. This method significantly reduces the computational cost of predicting and assessing urban building energy demand.

[0112] By evaluating the energy demand characteristics of urban residential buildings, this method can provide a solid theoretical basis and technical support for the design and optimization of building energy systems. This helps to make more scientific optimization decisions for system design and operation, thereby achieving accurate configuration of energy system capacity. This not only helps to improve the reliability and economic efficiency of system operation, but also has important significance for achieving energy conservation and emission reduction goals.

[0113] The present disclosure may be a system, method and / or computer program product. The computer program product may include a computer-readable storage medium carrying computer-readable program instructions for causing a processor to implement various aspects of the present disclosure.

[0114] A computer-readable storage medium can be a tangible device that can hold and store instructions for use by an instruction execution device. A computer-readable storage medium can be, for example, but not limited to, an electrical storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device, or any suitable combination thereof. More specific examples (a non-exhaustive list) of computer-readable storage media include: a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), a static random access memory (SRAM), a portable compact disc read-only memory (CD-ROM), a digital versatile disk (DVD), a memory stick, a floppy disk, a mechanical encoding device, such as a punch card or a raised structure in a groove on which instructions are stored, and any suitable combination thereof. As used herein, a computer-readable storage medium is not to be construed as a transient signal per se, such as a radio wave or other freely propagating electromagnetic wave, an electromagnetic wave propagating through a waveguide or other transmission medium (e.g., a light pulse through a fiber optic cable), or an electrical signal transmitted through an electrical wire.

[0115] The computer-readable program instructions described herein can be downloaded from a computer-readable storage medium to each computing / processing device, or downloaded to an external computer or external storage device via a network, such as the Internet, a local area network, a wide area network, and / or a wireless network. The network can include copper transmission cables, fiber optic transmission, wireless transmission, routers, firewalls, switches, gateway computers, and / or edge servers. The network adapter card or network interface in each computing / processing device receives the computer-readable program instructions from the network and forwards the computer-readable program instructions to be stored in the computer-readable storage medium in each computing / processing device.

[0116] The computer program instructions for performing the operations of the present disclosure may be assembly instructions, instruction set architecture (ISA) instructions, machine instructions, machine-dependent instructions, microcode, firmware instructions, state setting data, or source code or object code written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Smalltalk, C++, and conventional procedural programming languages ​​such as "C" language or similar programming languages. Computer-readable program instructions may be executed entirely on a user's computer, partially on a user's computer, as an independent software package, partially on a user's computer, partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer via any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., utilizing an Internet service provider to connect via the Internet). In some embodiments, an electronic circuit, such as a programmable logic circuit, a field programmable gate array (FPGA), or a programmable logic array (PLA), may be personalized by utilizing the state information of the computer-readable program instructions. The electronic circuit may execute the computer-readable program instructions, thereby realizing various aspects of the present disclosure.

[0117] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in the field should understand that the specific implementation methods of the present invention can still be modified or replaced by equivalents. Any modification or equivalent replacement that does not depart from the spirit and scope of the present invention should be covered by the scope of protection of the claims of the present invention.

Claims

1. A building prototype generation method based on urban big data, characterized in that: include: Step 1: Obtain the number and area of ​​each residential land, and obtain the building base area, number of floors, and height of each residential building belonging to each residential land to calculate various characteristic parameters of the residential land and residential buildings; the characteristic parameters of the residential land include: volume ratio, building density, building height limit, land compactness index, and high-rise ratio; the characteristic parameters of the residential buildings include: building shape coefficient; Step 2: based on the correlation between each characteristic parameter of the residential land and each characteristic parameter of the residential building, redundant characteristic parameters are deleted; after the deletion process, based on the correlation between each characteristic parameter of the retained residential land and the predicted energy consumption value of the residential building, cluster characteristic parameters of the residential land for clustering are determined; at the same time, based on the correlation between each characteristic parameter of the retained residential building and the predicted energy consumption value of the residential building, cluster characteristic parameters of the residential building for clustering are determined; Step 3: Determine the residential land cluster center and the residential building cluster center respectively based on the evaluation indicators of residential land and residential buildings; based on the determined residential land cluster center and residential building cluster center, use the clustering characteristic parameters of residential land and residential buildings to perform two-stage clustering to obtain the optimal residential building clustering result focused on the optimal residential land cluster center; generate a building prototype set based on the optimal residential building clustering result focused on the optimal residential land cluster center; Step 4: Based on the building prototype set, calculate the building energy demand indicators separately, including: the peak load density of cooling and heating demand throughout the year, and the total intensity of cooling and heating load demand throughout the year; when the comprehensive value of the building energy demand indicators is less than the set threshold, repeat step 3 to regenerate the building prototype set; when the comprehensive value of the building energy demand indicators is not less than the set threshold, output the building prototype set.

2. The method for generating building prototypes based on urban big data according to claim 1, characterized in that: The floor area ratio satisfies the following relationship: Where FAR i is the volume ratio of the i-th residential land, BBA i,j is the building base area of ​​the jth residential building belonging to the i-th residential land, Storeys i,j is the number of floors of the jth residential building belonging to the i-th residential land, LAA i is the area of ​​the i-th residential land, m i is the set of residential buildings belonging to the i-th residential land; The building density satisfies the following relationship: Where, BCR i is the building density of the i-th residential land; The building height limit satisfies the following relationship: MAH i =max(BUH i,j ) (3) Where, MAH i is the building height limit for the i-th residential land, BUH i,j is the height of the jth residential building belonging to the i-th residential land; The land compactness index satisfies the following relationship: Where, NCI i is the compactness index of the i-th residential land, ALP i is the perimeter of the i-th residential land; The proportion of high-rise buildings satisfies the following relationship: Where, HRP i is the proportion of high-rise buildings in the i-th residential land; The building shape coefficient satisfies the following relationship: Where, Building surface area i,j It is the surface area of ​​the jth residential building belonging to the i-th residential land that is in contact with the outdoor atmosphere.

3. The method for generating building prototypes based on urban big data according to claim 1, characterized in that: Step 2 includes: When the Spearman correlation coefficient between the characteristic parameters of residential land and the characteristic parameters of residential buildings is less than the set first limit, the corresponding characteristic parameters of residential land and residential buildings are deleted; otherwise, the corresponding characteristic parameters of residential land and residential buildings are retained; When the Spearman correlation coefficient between the characteristic parameter of the reserved residential land and the energy consumption prediction value of each residential building belonging to the residential land is not less than the set second limit, the characteristic parameter of the reserved residential land is determined to be the cluster characteristic parameter; otherwise, the characteristic parameter of the reserved residential land is the spare characteristic parameter; When the Spearman correlation coefficient between the characteristic parameters of the retained residential buildings and the predicted energy consumption value of the residential buildings is not less than the set second limit value, determine the characteristic parameters of the retained residential buildings as the clustering characteristic parameters; otherwise, the characteristic parameters of the retained building land are the alternative characteristic parameters.

4. A method for generating a building prototype based on urban big data according to claim 3, wherein The first limit value is 0.

5.

5. A method for generating a building prototype based on urban big data according to claim 3, wherein The second limit value is 0.

7.

6. A method for generating a building prototype based on urban big data according to claim 1, wherein Step 3 includes: Step 3.1, determine the residential land clustering center according to the various evaluation indexes of the residential land, and determine the residential building clustering center according to the various evaluation indexes of the residential buildings; Step 3.2, extract n1 residential land clustering centers as the first cluster centers of the residential land, where 3 ≤ n1 < N, and n is the number of residential land clustering centers; use the K-medoids clustering method to cluster the clustering characteristic parameters of the residential buildings on the first cluster centers of the residential land, and use the Bayesian information criterion scoring method to score the clustering results to obtain the first score corresponding to the first cluster centers of the residential land; randomly extract residential land clustering centers different from the first cluster centers of the residential land, and use the K-medoids clustering method to cluster the clustering characteristic parameters of the residential buildings that have been clustered on the first cluster centers of the residential land on the extracted residential land clustering centers, and use the Bayesian information criterion scoring method to score the clustering results to obtain the second score corresponding to the first cluster centers of the residential land; when the second score is greater than or equal to the first score, retain the extracted residential land clustering centers; otherwise, when the second score is less than the first score, delete the extracted residential land clustering centers; After repeating the above operations, obtain the optimal residential land clustering center and the residential buildings focused on by the optimal residential land clustering center; Step 3.3, use the K-medoids clustering method to cluster the residential buildings focused on by the optimal residential land clustering center on the residential building clustering center, and use the Bayesian information criterion scoring method to score the clustering results. When the score is greater than the set threshold, obtain the optimal residential building clustering result focused on by the optimal residential land clustering center; Step 3.4, generate a building prototype set based on the optimal residential building clustering result focused on by the optimal residential land clustering center.

7. A method for generating a building prototype based on urban big data according to claim 6, wherein The various evaluation indexes of the residential land include but are not limited to: land area, plot ratio, building density, building height limit, green space rate, land compactness index, unit density, and high-rise proportion; the various evaluation indexes of the residential buildings include but are not limited to: projected area, building height, and building shape coefficient; The number N of the residential land clustering centers is consistent with the number of the evaluation indexes of the residential land; The number M of residential building cluster centers is consistent with the number of evaluation indicators of residential buildings.

8. The method for generating building prototypes based on urban big data according to claim 7, characterized in that: In step 3.2, the residential land evaluation index of the optimal residential land cluster center is used as the key indicator for evaluating residential land and energy consumption prediction.

9. The method for generating building prototypes based on urban big data according to claim 6, characterized in that: The Bayesian Information Criterion scoring method scores the clustering results in the following relationship: Where, is the maximum log-likelihood function of the clustering feature parameters of the i-th group, is the clustering characteristic parameter x of the i-th group i Belongs to cluster i C i The maximum likelihood estimate of the clustering characteristic parameter x of the i-th group i Including clustering characteristic parameters of residential land and residential buildings; n i For cluster I C i The number of samples in , p is the number of features of the cluster.

10. The method for generating building prototypes based on urban big data according to claim 9, characterized in that: In step 3.3, the scores of the clustering results using the Bayesian Information Criterion scoring method are processed into the range of [0, 1], and the threshold value is set to be no less than 0.

62.

11. The method for generating building prototypes based on urban big data according to claim 1, characterized in that: The building prototype set includes K building prototypes. For the k-th building prototype, k = 1, 2, ..., K, the building energy demand index satisfies the following relationship: Where, is the total intensity of cooling and heating load demand of the k-th building prototype throughout the year, is the peak load density of cooling and heating demand of the k-th building prototype throughout the year, h is the hour, is the hourly energy consumption intensity of the building.

12. The method for generating building prototypes based on urban big data according to claim 11, characterized in that: The normalized value of the weighted sum of the total annual cooling and heating load demand intensity and the annual cooling and heating demand peak load density of the k-th building prototype is used as the comprehensive value of the building energy demand index; the threshold is set at 0.

88.

13. A terminal comprising a processor and a storage medium; characterized in that: The storage medium is used to store instructions; The processor is configured to operate according to the instructions to execute the steps of the method according to any one of claims 1 to 12.

14. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the steps of the method according to any one of claims 1 to 12 are implemented.

Images