Building maximum energy consumption demand evaluation method, device, equipment and storage medium

Abstract

The application relates to a building maximum energy consumption demand evaluation method and device, computer equipment, a computer readable storage medium and a computer program product. The method comprises the following steps: obtaining building data of a building, wherein the building data comprises map data, topographic data of a location, an internal structure image and an appearance image; determining building body data and building structure data according to the internal structure image and the appearance image; generating building energy consumption data under different scenes through a building energy consumption simulation mode according to the map data, the topographic data, the building body data and the building structure data; training a building maximum energy consumption demand evaluation model according to the building energy consumption data under different scenes; obtaining target building data of a building to be evaluated; and outputting a maximum energy consumption demand value of the building to be evaluated by using the maximum energy consumption demand evaluation model according to the target building data. The method can realize accurate evaluation of the maximum energy consumption demand of the building.

Description

Technical Field

[0001] This application relates to the field of building simulation technology, and in particular to a method, apparatus, computer equipment, computer-readable storage medium, and computer program product for assessing the maximum energy consumption demand of a building. Background Technology

[0002] The building sector plays a vital role in global energy consumption, and its energy flexibility is crucial for balancing power supply and demand. With increasing climate change and electrification, energy demand fluctuations under extreme weather conditions are becoming increasingly prominent. For example, during extreme heat waves, building air conditioning loads can account for more than 50% of total urban energy consumption. In the summers of 2022 and 2024, excessive air conditioning loads led to varying degrees of power shortages, causing significant economic losses. To address this challenge, a comprehensive survey of industrial and commercial users is needed, along with the establishment of a database of grid load interaction potential and an expansion of the scope of demand response programs.

[0003] Currently, there are three main technical approaches to assessing the maximum energy demand of buildings. The first is the traditional physical modeling method, which assesses maximum energy demand by modeling the building's thermodynamic mechanisms and HVAC systems in detail. However, high data collection costs, privacy protection, and data ownership issues hinder its widespread application. The second is an aggregate modeling approach, which builds a temperature-controlled load (TCL) model and optimizes its operational reserves, but it cannot achieve accurate assessments at the individual building level. The third is a data-driven approach, which uses linear regression and unsupervised classification techniques, combined with household electricity consumption and outdoor temperature data, to assess the maximum energy demand of air conditioning operation. However, obtaining fine-grained historical data is difficult for external organizations.

[0004] Therefore, while traditional methods each have their advantages, they are insufficient in terms of data acquisition and assessment accuracy. There is an urgent need for a method, device, computer equipment, computer-readable storage medium, and computer program product for assessing the maximum energy consumption demand of buildings, which can achieve accurate assessment of the maximum energy consumption demand of buildings. Summary of the Invention

[0005] Therefore, it is necessary to provide a method, apparatus, computer equipment, computer-readable storage medium, and computer program product for assessing the maximum energy consumption demand of a building, which can accurately assess the maximum energy consumption demand of a building, in response to the above-mentioned technical problems.

[0006] Firstly, this application provides a method for assessing the maximum energy consumption demand of a building, including:

[0007] Acquire building data, including map data, terrain data of the location, internal structure images, and exterior images;

[0008] Based on the internal structure image and the external image, determine the building body data and building structure data;

[0009] Based on the map data, terrain data, building body data, and building structure data, energy consumption data of buildings under different scenarios are generated through building energy consumption simulation.

[0010] Train a model to assess the maximum energy consumption demand of buildings based on building energy consumption data in different scenarios.

[0011] Obtain the target building data for the building to be evaluated;

[0012] Based on the target building data, the maximum energy consumption demand value of the building to be evaluated is output using the maximum energy consumption demand assessment model.

[0013] In one embodiment, the building structure data includes wall material type and window-to-wall coverage ratio; determining the building body data and building structure data based on the internal structure image and the external image includes:

[0014] Pixel-level semantic segmentation is performed on the building facade in the appearance image to determine the wall material type and the segmentation results between windows, walls and other areas;

[0015] Based on the segmentation results, calculate the window-to-wall coverage ratio of the building;

[0016] Historical description information of the building is obtained, and the internal structure images and historical description information are analyzed to obtain building body data. The historical description information includes the building's purpose, construction year, internal equipment information, and maintenance records.

[0017] In one embodiment, the step of generating building energy consumption data under different scenarios based on the map data, terrain data, building body data, and building structure data through building energy consumption simulation includes:

[0018] Assess the importance of all features contained in the building body data and the building structure data;

[0019] The target building feature vector is determined based on the ranking of the importance of all features.

[0020] Based on the map data, the terrain data, and the target building feature vector, building energy consumption data under different scenarios is generated through building energy consumption simulation.

[0021] In one embodiment, the step of generating building energy consumption data under different scenarios based on the map data, terrain data, building body data, and building structure data through building energy consumption simulation includes:

[0022] Based on the map data and the terrain data, obtain the historical and future meteorological data of the area where the building is located;

[0023] Based on the historical meteorological data and the future meteorological data, the key climate scenarios for the area where the building is located are determined. The key climate scenarios include the climate average year, the year of extreme high temperature events, the year of extreme low temperature events, and the climate scenarios of seasonal representative days.

[0024] Based on the building body data and the building structure data, the internal dynamic load data of the building is obtained. The internal dynamic load data includes personnel density, personnel activity time periods, power of various electrical appliances, and usage time periods of various electrical appliances.

[0025] Based on the internal dynamic load data, determine the building's energy consumption usage scenario;

[0026] Simulate the key climate scenarios and energy consumption scenarios, and set various combinations of meteorological parameters and energy consumption parameters in the simulation software to generate building energy consumption data under different scenarios.

[0027] In one embodiment, training a maximum energy demand assessment model for a building based on building energy consumption data under different scenarios includes:

[0028] The corresponding parameter values ​​of building energy consumption data under different scenarios are processed to have a unified dimension.

[0029] The processed building energy consumption data is used as the true label for model training;

[0030] The building entity data, the building structure data, and the real labels are matched to establish a mapping relationship table;

[0031] Based on the mapping table, train a model to assess the maximum energy consumption demand of buildings.

[0032] In one embodiment, training a maximum energy demand assessment model for a building based on building energy consumption data under different scenarios includes:

[0033] Calculate the regression prediction loading and classification prediction potential level of at least one machine learning algorithm;

[0034] The target machine learning algorithm is determined based on the regression prediction loading value and the classification prediction potential level;

[0035] A k-fold cross-validation strategy was adopted to optimize and validate model parameters for building energy consumption data under different scenarios.

[0036] Using validated building energy consumption data, a target machine learning algorithm is sampled to train a model for assessing the building's maximum energy consumption demand.

[0037] Secondly, this application also provides a device for assessing the maximum energy consumption demand of a building, comprising:

[0038] The data acquisition module is used to acquire building data, which includes map data, terrain data of the location, internal structure images, and exterior images.

[0039] The data processing module is used to determine the building body data and building structure data based on the internal structure image and the external image;

[0040] The data processing module is also used to generate building energy consumption data under different scenarios based on the map data, the terrain data, the building body data and the building structure data through building energy consumption simulation.

[0041] The model training module is used to train a model for assessing the maximum energy consumption demand of a building based on building energy consumption data in different scenarios.

[0042] The data acquisition module is also used to acquire target building data of the building to be evaluated;

[0043] The maximum energy consumption demand assessment module is used to output the maximum energy consumption demand value of the building to be assessed based on the target building data and the maximum energy consumption demand assessment model.

[0044] Thirdly, this application also provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to perform the following steps:

[0045] Acquire building data, including map data, terrain data of the location, internal structure images, and exterior images;

[0046] Based on the internal structure image and the external image, determine the building body data and building structure data;

[0047] Based on the map data, terrain data, building body data, and building structure data, energy consumption data of buildings under different scenarios are generated through building energy consumption simulation.

[0048] Train a model to assess the maximum energy consumption demand of buildings based on building energy consumption data in different scenarios.

[0049] Obtain the target building data for the building to be evaluated;

[0050] Based on the target building data, the maximum energy consumption demand value of the building to be evaluated is output using the maximum energy consumption demand assessment model.

[0051] Fourthly, this application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, performs the following steps:

[0052] Acquire building data, including map data, terrain data of the location, internal structure images, and exterior images;

[0053] Based on the internal structure image and the external image, determine the building body data and building structure data;

[0054] Based on the map data, terrain data, building body data, and building structure data, energy consumption data of buildings under different scenarios are generated through building energy consumption simulation.

[0055] Train a model to assess the maximum energy consumption demand of buildings based on building energy consumption data in different scenarios.

[0056] Obtain the target building data for the building to be evaluated;

[0057] Based on the target building data, the maximum energy consumption demand value of the building to be evaluated is output using the maximum energy consumption demand assessment model.

[0058] Fifthly, this application also provides a computer program product, including a computer program that, when executed by a processor, performs the following steps:

[0059] Acquire building data, including map data, terrain data of the location, internal structure images, and exterior images;

[0060] Based on the internal structure image and the external image, determine the building body data and building structure data;

[0061] Based on the map data, terrain data, building body data, and building structure data, energy consumption data of buildings under different scenarios are generated through building energy consumption simulation.

[0062] Train a model to assess the maximum energy consumption demand of buildings based on building energy consumption data in different scenarios.

[0063] Obtain the target building data for the building to be evaluated;

[0064] Based on the target building data, the maximum energy consumption demand value of the building to be evaluated is output using the maximum energy consumption demand assessment model.

[0065] The aforementioned method, apparatus, computer equipment, computer-readable storage medium, and computer program product for assessing the maximum energy consumption demand of buildings, based on building entity data and structural data, combined with map and terrain data, generate building energy consumption data under different scenarios through building energy consumption simulation. This refined simulation method can fully consider the energy consumption performance of buildings under different environmental conditions (such as different climates and seasons), thus more realistically reflecting the actual energy consumption characteristics of buildings. Training the maximum energy consumption demand assessment model of buildings using energy consumption data from different scenarios can effectively capture the inherent laws and changing trends of building energy consumption demand. This model can quickly and accurately output the maximum energy consumption demand value of the building to be assessed. Accurate assessment of the maximum energy consumption demand of buildings can help building managers and energy planners better formulate energy allocation strategies, optimize energy use efficiency, reduce energy waste, and lower operating costs. This application is applicable to buildings of different types and sizes, and the scope of data collection and model training can be flexibly adjusted according to actual needs. With data accumulation and model optimization, the assessment accuracy will be further improved, giving it good scalability and adaptability. Attached Figure Description

[0066] To more clearly illustrate the technical solutions in the embodiments of this application or related technologies, the drawings used in the description of the embodiments of this application or related technologies will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0067] Figure 1 This is a diagram illustrating the application environment of a method for assessing the maximum energy consumption demand of a building in one embodiment.

[0068] Figure 2 This is a flowchart illustrating a method for assessing the maximum energy consumption demand of a building in one embodiment;

[0069] Figure 3 This is a flowchart illustrating a method for assessing the maximum energy consumption demand of a building in another embodiment;

[0070] Figure 4 A structural block diagram of a building's maximum energy consumption demand assessment device in one embodiment;

[0071] Figure 5 This is an internal structural diagram of a computer device in one embodiment. Detailed Implementation

[0072] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0073] It should be noted that the terms "first," "second," etc., used in this application can be used to describe various elements, but these elements are not limited by these terms. These terms are only used to distinguish the first element from the second element. The terms "comprising" and "having," and any variations thereof, used in this application, are intended to cover non-exclusive inclusion. The term "multiple" used in this application refers to two or more. The term "and / or" used in this application refers to one of the embodiments, or any combination of multiple embodiments.

[0074] The method for assessing the maximum energy consumption demand of buildings provided in this application can be applied to, for example... Figure 1 In the application environment shown, terminal 102 communicates with server 104 via a network. A data storage system can store the data that server 104 needs to process. The data storage system can be integrated onto server 104, or it can be located in the cloud or on another network server.

[0075] Server 104 acquires building data of the building through terminal 102. The building data includes map data, terrain data of the location, internal structure images, and exterior images. Server 104 determines the building body data and building structure data based on the internal structure images and exterior images. Based on the map data, terrain data, building body data, and building structure data, server 104 generates building energy consumption data under different scenarios through building energy consumption simulation. Based on the building energy consumption data under different scenarios, server 104 trains a maximum energy consumption demand assessment model for the building. Server 104 acquires target building data of the building to be assessed through terminal 102. Based on the target building data, server 104 uses the maximum energy consumption demand assessment model to output the maximum energy consumption demand value of the building to be assessed.

[0076] The terminal 102 can be, but is not limited to, various personal computers, laptops, smartphones, tablets, IoT devices, and portable wearable devices. The server 104 can be a standalone physical server, a server cluster or distributed system consisting of multiple physical servers, or a cloud server providing cloud computing services.

[0077] In one exemplary embodiment, such as Figure 2 As shown, a method for assessing the maximum energy consumption demand of a building is provided, and this method is applied to... Figure 1Taking the server in the example, the explanation includes the following steps S202 to S212. Wherein:

[0078] Step S202: Obtain the building data, which includes map data, terrain data of the location, internal structure images, and exterior images.

[0079] Specifically, map data refers to geographic information related to the geographical location of a building, usually presented in the form of electronic maps, including location coordinates, surrounding environment, and transportation networks. Topographic data refers to the topographic features of the building's location, including ground elevation, slope, and landform information. Internal structure images refer to images of the building's internal layout and structure, typically including floor plans, spatial height, materials, and construction. Exterior images refer to images of the building's exterior, including elevations, roof designs, and external facilities.

[0080] This process involves: collecting basic information about the building, such as location, floor area, and building type, from open data platforms using web crawling technology; interacting with open mapping software (such as OpenStreetMap) to obtain detailed information such as the building's outline, number of floors, and height; utilizing remote sensing technology to acquire information on the building's exterior features and surrounding environment to supplement the building's spatial data; and collecting exterior and interior photographs of the building from street view and social media platforms to obtain detailed information on building materials and type. Feature engineering is then employed to preprocess the collected multimodal data, including data cleaning, deduplication, and format conversion, to ensure data consistency and accuracy. By acquiring these four types of data, a comprehensive understanding of the building's internal and external characteristics and its surrounding environment can be achieved.

[0081] Step S204: Determine the building body data and building structure data based on the internal structure image and the external image.

[0082] Specifically, a fully convolutional network (FCN) model can be used to process internal structure images and external images to obtain building ontology data and building structure data.

[0083] The building data includes basic building information, functional zoning, and exterior features.

[0084] Basic building information: such as building name, type (residential, office building, shopping mall, etc.), purpose (residential, office, commercial, etc.), building area, number of floors, height, etc.

[0085] Building functional zoning: such as room layout, location and area of ​​public areas (corridors, stairs, elevators, etc.).

[0086] Architectural features: such as facade design, roof form, and external facilities (balconies, canopies, etc.).

[0087] Building structural data includes structural type, structural materials, structural details, and structural performance.

[0088] Structural types: such as frame structure, shear wall structure, hybrid structure, etc.

[0089] Structural materials: such as wall materials (brick, concrete, glass curtain walls, etc.), floor materials (concrete, steel structures, etc.), roof materials, etc.

[0090] Structural details: such as wall thickness, floor slab thickness, beam and column dimensions, and node connection methods.

[0091] Structural performance: such as seismic performance, fire resistance, thermal insulation performance, etc.

[0092] Step S206: Based on map data, terrain data, building body data, and building structure data, generate building energy consumption data for different scenarios through building energy consumption simulation.

[0093] Specifically, building energy consumption simulation methods, such as using building energy consumption simulation software (e.g., EnergyPlus, DeST), can simulate the energy consumption performance of a building based on input building data and environmental conditions.

[0094] The simulation process includes:

[0095] Climate conditions setting: Based on historical meteorological data and future climate scenarios of the target area, different combinations of meteorological parameters (such as temperature, humidity, solar radiation, wind speed, etc.) are set to simulate various climate scenarios such as typical years, extreme hot / cold years, and different seasons.

[0096] Usage mode settings: Define the dynamic load inside the building, including personnel density, lighting power density, equipment power and operating schedule, air conditioning system operation strategy, etc.

[0097] Energy consumption calculation: Run the simulation model to calculate the building’s total energy consumption, sub-item energy consumption (such as heating, cooling, lighting, equipment, ventilation) and key system parameters (such as peak, valley and variation curves of heating and cooling loads) under different climate conditions and usage patterns.

[0098] Through multi-dimensional data and multi-scenario simulation, the energy consumption performance of buildings under different conditions can be comprehensively evaluated.

[0099] Step S208: Train a model to assess the maximum energy consumption demand of a building based on building energy consumption data under different scenarios.

[0100] Specifically, in order to construct a model that can predict the maximum energy consumption demand of a building under any given conditions, the input features of the model include:

[0101] Architectural features: such as building area, window-to-wall ratio, wall materials, number of floors, building type, etc.

[0102] Scene characteristics: such as climate conditions (temperature, humidity, solar radiation, etc.) and usage patterns (personnel density, equipment usage time, etc.).

[0103] The model's output features include the building's maximum energy demand under specific conditions.

[0104] Prior to this, building energy consumption data under different scenarios needs to be normalized or standardized to eliminate the influence of unit of measurement and improve the efficiency and stability of model training. Appropriate machine learning algorithms should be selected, such as Multilayer Perceptron (MLP) or Random Forest. The preprocessed building energy consumption data is then used as training data to train the model to learn the mapping relationship between building characteristics and energy consumption requirements.

[0105] Step S210: Obtain the target building data of the building to be evaluated.

[0106] Specifically, target building data refers to various characteristic and attribute data directly related to the building to be evaluated. This data comprehensively describes the building's basic information, structural features, operational status, and surrounding environment. It is a necessary input for assessing the building's energy consumption requirements and directly affects the accuracy and reliability of the assessment results.

[0107] The target building data includes map data, terrain data, building body data, building structure data, operational status data, and other relevant information, including the year of construction and energy management status.

[0108] Step S212: Based on the target building data, use the maximum energy consumption demand assessment model to output the maximum energy consumption demand value of the building to be assessed.

[0109] Specifically, the target building data is formatted to match the training data and standardized as necessary before being input into the evaluation model. Based on the input building data and the learned mapping relationships, the evaluation model calculates the maximum energy demand of the building under specific conditions. The predicted value output by the model is the maximum energy demand of the building being evaluated.

[0110] The aforementioned method for assessing the maximum energy consumption demand of buildings, based on building body data and structural data, combined with map and terrain data, generates building energy consumption data under different scenarios through building energy consumption simulation. This refined simulation method can fully consider the energy consumption performance of buildings under different environmental conditions (such as different climates and seasons), thus more realistically reflecting the actual energy consumption characteristics of buildings. Training the maximum energy consumption demand assessment model of buildings using energy consumption data from different scenarios can effectively capture the inherent laws and changing trends of building energy consumption demand. This model can quickly and accurately output the maximum energy consumption demand value of the building to be assessed. Accurate assessment of the maximum energy consumption demand of buildings can help building managers and energy planners better formulate energy allocation strategies, optimize energy use efficiency, reduce energy waste, and lower operating costs. This application is applicable to buildings of different types and sizes, and the scope of data collection and model training can be flexibly adjusted according to actual needs. With data accumulation and model optimization, the assessment accuracy will be further improved, giving it good scalability and adaptability.

[0111] In one exemplary embodiment, such as Figure 3 As shown, the building structure data includes wall material type and window-to-wall coverage ratio; based on internal and external structural images, the building body data and building structure data are determined, including:

[0112] Step S302: Perform pixel-level semantic segmentation on the building facade in the exterior image to determine the wall material type and the segmentation results between windows, walls and other areas;

[0113] Step S304: Calculate the window-to-wall coverage ratio of the building based on the segmentation results;

[0114] Step S306: Obtain historical description information of the building, analyze the internal structure images and historical description information to obtain building body data, wherein the historical description information includes building use, building year, internal equipment information and maintenance records.

[0115] Specifically, image processing techniques (such as fully convolutional networks, FCN) are used to perform pixel-level semantic segmentation on the building facade in the exterior image, classifying each pixel in the image into different categories, such as windows, walls, balconies, etc. Based on the semantic segmentation results, the wall portion is identified, and its texture, color, and other features are further analyzed to determine the wall material type. According to the semantic segmentation results, the areas of windows and walls are calculated separately, and then their ratio, i.e., the window-to-wall coverage area ratio, is calculated.

[0116] By leveraging the Large Language Model API to perform deep text analysis, historical descriptive information about buildings can be obtained. This information may come from architectural design documents, maintenance records, and publicly available data platforms, capturing details such as building purpose, construction year, internal equipment information, and maintenance records. Specific prompts are designed to guide the Large Language Model in extracting key structured numerical features (such as construction year, air conditioning system type code, window type code, and whether energy storage equipment is present).

[0117] By analyzing images of a building's internal structure, such as floor plans, cross-sections, or interior photographs, information about the building's layout, room functions, and equipment arrangement can be extracted. Furthermore, by analyzing these internal structural images and historical descriptive information, natural language processing techniques (such as large language model APIs) can be used to extract key information from the text, generating building ontology data.

[0118] The building structure data and building ontology data extracted from images and text will be integrated to form a complete building feature dataset.

[0119] In this embodiment, key building structure data and building ontology data are extracted from the building's exterior and interior images and historical description information through image processing technology (such as pixel-level semantic segmentation) and text analysis technology (such as natural language processing). This data can be used as input to train machine learning models, predict the building's energy consumption demand, and support the formulation of energy-saving strategies.

[0120] In an exemplary embodiment, based on map data, terrain data, building body data, and building structure data, building energy consumption data under different scenarios is generated through building energy consumption simulation, including:

[0121] Assess the importance of all features contained in the building body data and building structure data;

[0122] The target building feature vector is determined based on the ranking of the importance of all features.

[0123] Based on map data, terrain data, and target building feature vectors, energy consumption data of buildings under different scenarios is generated through building energy consumption simulation.

[0124] Specifically, machine learning algorithms (such as random forest, LASSO regression, etc.) are used to evaluate the importance of all features contained in the building ontology data and building structure data. The purpose is to determine which features have the greatest impact on building energy consumption and to quantify the contribution of each feature to energy consumption.

[0125] Based on the evaluation results of feature importance, all features are sorted, and the top N features with the highest importance are selected to form the target building feature vector. The target building feature vector is the most critical input parameter in subsequent energy consumption simulation.

[0126] Input data: Map data and terrain data: providing environmental information about the building's location. Target building feature vector: containing key features that have the greatest impact on energy consumption.

[0127] Simulation process:

[0128] Building energy consumption simulation software: Use professional building energy consumption simulation software (such as EnergyPlus, DeST, etc.) to simulate energy consumption.

[0129] Different scenario settings: Simulate the energy consumption performance of buildings under various conditions, including:

[0130] Climate scenarios: typical year, extreme hot year, extreme cold year, different seasons, etc.

[0131] Usage scenarios: weekdays, weekends, holidays, etc.

[0132] Operating strategy scenarios: different air conditioning temperature settings, lighting control strategies, etc.

[0133] Output data: Generate building energy consumption data for different scenarios, including total energy consumption, sub-item energy consumption (such as heating, cooling, lighting, equipment, ventilation) and key system parameters (such as peak, valley and change curves of heating and cooling loads).

[0134] In this embodiment, by evaluating feature importance and selecting key features, the data input into the energy consumption simulation model is ensured to be the most representative and influential, thereby improving the accuracy and reliability of the simulation results. Unnecessary feature inputs are reduced, computational complexity is lowered, and simulation efficiency is improved. By simulating various scenarios, the energy consumption performance of buildings under different conditions is comprehensively evaluated.

[0135] In an exemplary embodiment, based on map data, terrain data, building body data, and building structure data, building energy consumption data under different scenarios is generated through building energy consumption simulation, including:

[0136] Based on map and terrain data, obtain historical and future weather data for the area where the building is located;

[0137] Based on historical and future meteorological data, the key climate scenarios for the area where the building is located are determined. The key climate scenarios include the average climate year, the year of extreme high temperature events, the year of extreme low temperature events, and the climate scenarios of seasonal representative days.

[0138] Based on the building body data and building structure data, obtain the internal dynamic load data of the building, which includes personnel density, personnel activity time periods, power of various electrical appliances, and usage time periods of various electrical appliances;

[0139] Based on the internal dynamic load data, determine the building's energy consumption usage scenarios;

[0140] Simulate key climate and energy consumption scenarios, and set various combinations of meteorological and energy consumption parameters in the simulation software to generate building energy consumption data under different scenarios.

[0141] Specifically, historical meteorological data for the building's location is obtained based on map and topographic data. This data typically includes temperature, humidity, solar radiation, wind speed, and wind direction. Future meteorological data for the building's location is also obtained; this data can be predicted using climate models and used to assess the impact of future climate change on energy consumption. Key climate scenarios include:

[0142] Average annual climate: Reflects the average climate conditions of the area where the building is located.

[0143] Extreme heat event year: Simulates the energy consumption performance of buildings under extreme heat conditions.

[0144] Extreme low temperature event year: Simulates the energy consumption performance of buildings under extreme low temperature conditions.

[0145] Seasonal representative day: Select typical weather conditions for each season to assess the impact of seasonal changes on energy consumption.

[0146] Personnel density refers to the number and distribution of people within a building. Personnel activity periods refer to the time people spend within the building, such as weekdays, weekends, and holidays. Power consumption of various electrical appliances refers to the power output of various electrical devices within the building (such as office equipment and servers). Usage periods for various electrical appliances refer to the usage schedules of various electrical devices (such as set temperature, start / stop times, and time-of-use control).

[0147] Based on the internal dynamic load data, determine the energy consumption scenarios of the building at different times (such as weekdays, weekends, and holidays) and under different operating modes (such as air conditioning operation strategies and lighting control strategies).

[0148] In building energy consumption simulation software, combinations of various meteorological parameters (such as dry-bulb temperature, wet-bulb temperature, solar radiation intensity, wind speed, and wind direction) and various energy consumption parameters (such as occupancy density, electrical appliance power, and equipment operating time) are set according to key climate scenarios and energy usage scenarios. The simulation model is run to generate energy consumption data for the building under different scenarios, including total energy consumption, component energy consumption (such as heating, cooling, lighting, equipment, and ventilation), and key system parameters (such as peak and valley values ​​of heating and cooling loads, and their variation curves). All simulation results constitute the basic dataset for subsequent model training.

[0149] In this embodiment, a comprehensive assessment of a building's energy consumption performance under different conditions is achieved by simulating various climate and energy usage scenarios. By combining historical and future meteorological data with detailed internal dynamic load data, the accuracy and reliability of energy consumption prediction are improved. Simulating energy consumption performance under extreme weather conditions allows for the development of proactive countermeasures, ensuring the building's energy supply and operational safety.

[0150] In one exemplary embodiment, a model for assessing the maximum energy consumption demand of a building is trained based on building energy consumption data under different scenarios, including:

[0151] The corresponding parameter values ​​of building energy consumption data under different scenarios are processed to have a unified dimension.

[0152] The processed building energy consumption data is used as the true label for model training;

[0153] The building data, building structure data and real labels are matched to establish a mapping table;

[0154] Based on the mapping table, train a model to assess the maximum energy consumption demand of buildings.

[0155] Specifically, since the parameters in building energy consumption data under different scenarios may have different units and dimensions (e.g., temperature in degrees Celsius, energy consumption in kilowatt-hours), it is necessary to unify the dimensions of these parameter values. This is usually achieved through normalization (such as Z-score normalization or Min-Max normalization) to eliminate the influence of dimensions and make the data more suitable for model training.

[0156] The true label refers to the processed building energy consumption data, that is, the actual energy consumption value of the building under different scenarios. This data will serve as the target for model training, guiding the model to learn the relationship between building features and energy consumption.

[0157] Building ontology data and building structure data: These data describe the features of the building, including building type, area, number of floors, wall materials, window-to-wall ratio, etc. The building ontology data and building structure data are matched with processed energy consumption data (real-world labels) to form a structured data table. This table describes the correspondence between building features and energy consumption, providing the input-output relationship for model training.

[0158] Using the data in the mapping table above, a machine learning model is trained to establish a mapping relationship between building features and maximum energy consumption demand. The goal of the model is to learn the complex relationship between building features and energy consumption, thereby enabling accurate prediction of the maximum energy consumption demand of new buildings.

[0159] A variety of machine learning algorithms can be chosen, such as linear regression, random forest, support vector machine (SVM), and neural networks. The specific choice depends on the characteristics of the data and the model's prediction objective. Model performance is evaluated using methods such as cross-validation, and hyperparameters are optimized to improve the model's generalization ability and prediction accuracy.

[0160] In this embodiment, standardized dimensional processing ensures data consistency and comparability, improving the efficiency and stability of model training. By matching building features with energy consumption data and establishing a mapping table, a clear input-output relationship is provided for model training, improving the model's prediction accuracy.

[0161] In one exemplary embodiment, a model for assessing the maximum energy consumption demand of a building is trained based on building energy consumption data under different scenarios, including:

[0162] Calculate the regression prediction loading and classification prediction potential level of at least one machine learning algorithm;

[0163] The target machine learning algorithm is determined based on the regression prediction loading value and the classification prediction potential level;

[0164] A k-fold cross-validation strategy was adopted to optimize and validate model parameters for building energy consumption data under different scenarios.

[0165] Using validated building energy consumption data, a target machine learning algorithm is sampled to train a model for assessing the building's maximum energy consumption demand.

[0166] Specifically, regression-predicted load values ​​refer to the use of machine learning algorithms to predict the energy load of a building under different scenarios. These values ​​are usually continuous, such as the building's total energy consumption (kilowatt-hours) or peak load (kilowatts) over a certain period of time.

[0167] Classification and prediction of energy potential levels refers to classifying a building's energy consumption potential into different levels (e.g., low, medium, high) and using classification algorithms to predict which energy consumption potential level the building belongs to. These levels are typically discrete categories.

[0168] Suppose we have the following algorithms:

[0169] Algorithm A: Random Forest Regression and Random Forest Classification; Algorithm B: Gradient Boosting Regression and Gradient Boosting Classification; Algorithm C: Multilayer Perceptron Regression and Multilayer Perceptron Classification.

[0170] For each algorithm, we train and validate it, and calculate its mean squared error, root mean squared error, mean absolute error, and absolute coefficient for regression tasks, and its accuracy, recall, precision, and F1 score (harmonic mean of recall and precision) for classification tasks.

[0171] Based on the evaluation metrics, compare the performance of different algorithms. Select the algorithm that performs well in both regression and classification tasks. Taking into account the model's prediction accuracy, robustness, computational complexity, and practical application scenarios, select the most suitable algorithm as the target machine learning algorithm.

[0172] K-fold cross-validation involves randomly dividing the building energy consumption dataset into k subsets of similar size. One subset is used as the validation set, and the remaining k-1 subsets are used as the training set, repeating the training and validation process k times. K-fold cross-validation evaluates the model's performance on different subsets and optimizes the model's hyperparameters (such as learning rate, number of trees, maximum depth, etc.) to improve the model's generalization ability and prediction accuracy.

[0173] A validated and optimized target machine learning algorithm is used to train a model for assessing the maximum energy demand of buildings by inputting a complete dataset of building energy consumption data. The model's performance is then evaluated using validated building energy consumption data to ensure the reliability of its predictions on new data.

[0174] In this embodiment, the most suitable algorithm is selected by calculating the regression-predicted load value and classification-predicted potential level of different machine learning algorithms to ensure the accuracy and reliability of the model. A k-fold cross-validation strategy is employed to optimize and validate the model's parameters, improving its generalization ability and avoiding overfitting. The optimized target machine learning algorithm is then used to train an assessment model capable of accurately predicting the building's maximum energy consumption demand.

[0175] The most detailed embodiment of this application is as follows:

[0176] Enter the building's address or name into the assessment system corresponding to the assessment model for the building's maximum energy consumption demand. For example, in this example, the assessment system input is "Section 2 of the Third Classroom Building of a certain university".

[0177] The geocode function is used to obtain the latitude and longitude information of the input location. Then, using the map API (getpic function), the latitude and longitude are used as input to obtain a planar map of the target location. OpenCV image processing tools are used to identify building outlines by specifying building colors and boundary colors in the planar map, resulting in a contour-processed planar map of the buildings. To ensure the stability and accuracy of the recognition, the following method is employed:

[0178] Unsupervised learning methods using sklearn are employed to cluster colors in the central region of a planar map, identifying the color with the highest central proportion as potential building colors. Different colored outlines are compared, and the outline closest to the center of the planar map is selected as the outline for the designated building. The rationality of the building outline is judged by its area, and small color blocks are removed as interference.

[0179] Based on the identified enclosed areas, OpenCV tools were used to calculate the length ratio and pixel area in the east-west and north-south directions. The pixel area was then converted to the actual floor area using the pixel scale of a planar map. Prior information about the building, such as the window-to-wall coverage ratio, was collected and organized online by calling a large model (the `ai()` function) and combining it with a search engine. Specific window-to-wall coverage ratio data requires further calculation and correction.

[0180] By calling the Street View map API, a 360° panoramic image of the target location is obtained for window-to-wall coverage ratio identification. Alternatively, photos of building exteriors can be collected on-site to improve recognition accuracy. A fully convolutional network (FCN) model is used to identify the window-to-wall ratio from the building exterior photos, and the results meet the system's input data requirements. For images from non-frontal viewpoints, OpenCV's four-point projection technique is used to project them to a frontal viewpoint, reducing image acquisition requirements and improving the accuracy of window-to-wall ratio calculation.

[0181] The acquired building data (such as building area, number of floors, opening hours, etc.) and building structural data (such as window-to-wall ratio) are integrated and input into the building's maximum energy demand assessment model. The model is then run to obtain the predicted results of the building's electricity demand response potential.

[0182] The confidence level of the multilayer perceptron (MLP) machine learning model was evaluated using a simulation dataset. The median error of the MLP model was less than 10%, and the root mean square error was 12.9%, indicating that the model's predictions were highly reliable in most cases. The model was relatively accurate in predicting buildings with low demand response potential. Although the accuracy in predicting buildings with high demand response potential was slightly lower, the model still has high prediction accuracy in large-scale practical applications due to the small number of high-potential buildings.

[0183] Assess the accuracy of the input data, including building area, number of floors, opening hours, operating hours, and window-to-wall ratio. Confirm the reasonableness of the data through online searches and on-site verification.

[0184] Building area: The building area calculated from the floor plan is close to the data queried online, which verifies its reasonableness.

[0185] Number of floors: The actual number of floors is 4. Considering the internal structure, it is reasonable to estimate that there are 4 or 5 floors.

[0186] Opening hours and operating hours: The data is reasonable based on actual opening hours and duration.

[0187] Window-to-wall coverage ratio: Window-to-wall ratio data identified through the FCN model has high reliability.

[0188] According to peak load reduction forecasting methods, the demand response potential assessment of Section 2 of the third classroom building at a university shows that if the air conditioning temperature is raised by 2°C during peak electricity consumption periods, the building's electricity load can be reduced by 20.49 kW, equivalent to the electricity demand of 14 households during the same period, or by 16.1 kg / h of carbon dioxide emissions. Combined with the peak-valley electricity pricing mechanism, its annual peak-shaving potential can reduce electricity expenses by approximately 143,000 yuan, significantly improving the economic efficiency of power grid operation.

[0189] This application provides a method for assessing building demand response potential based on artificial intelligence and simulation. It enables low-cost, large-scale assessment of building demand response potential without relying on detailed building parameters and operational data, providing strong support for the building sector's demand response planning.

[0190] It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.

[0191] Based on the same inventive concept, this application also provides a maximum energy demand assessment device for buildings to implement the above-described method for assessing the maximum energy demand of buildings. The solution provided by this device is similar to the solution described in the above-described method. Therefore, the specific limitations of one or more embodiments of the maximum energy demand assessment device for buildings provided below can be found in the limitations of the maximum energy demand assessment method for buildings described above, and will not be repeated here.

[0192] In one exemplary embodiment, such as Figure 4 As shown, a device for assessing the maximum energy consumption demand of a building is provided, comprising:

[0193] The data acquisition module 402 is used to acquire building data, which includes map data, terrain data of the location, internal structure images and exterior images;

[0194] Data processing module 404 is used to determine building body data and building structure data based on internal structure images and external images;

[0195] The data processing module 404 is also used to generate building energy consumption data under different scenarios based on map data, terrain data, building body data and building structure data through building energy consumption simulation.

[0196] Model training module 406 is used to train a building maximum energy consumption demand assessment model based on building energy consumption data in different scenarios.

[0197] The data acquisition module 402 is also used to acquire target building data of the building to be evaluated;

[0198] The maximum energy consumption demand assessment module 408 is used to output the maximum energy consumption demand value of the building to be assessed based on the target building data and the maximum energy consumption demand assessment model.

[0199] In one exemplary embodiment, the building structure data includes wall material type and window-to-wall coverage ratio;

[0200] The data processing module 404 is also used to perform pixel-level semantic segmentation on the building facade in the exterior image, determine the wall material type, and the segmentation results between windows, walls and other areas; and calculate the window-to-wall coverage ratio of the building based on the segmentation results.

[0201] The data acquisition module 402 is also used to acquire historical description information of the building, and the data processing module 404 is also used to analyze the internal structure images and historical description information to obtain building body data. The historical description information includes the building's use, construction year, internal equipment information, and maintenance records.

[0202] In an exemplary embodiment, the data processing module 404 is further configured to evaluate the importance of all features contained in the building ontology data and building structure data; determine the target building feature vector based on the ranking of the importance of all features; and generate building energy consumption data under different scenarios based on map data, terrain data, and the target building feature vector through building energy consumption simulation.

[0203] In an exemplary embodiment, the data processing module 404 is further configured to: acquire historical and future meteorological data of the area to which the building is located based on map data and terrain data; determine key climate scenarios for the area to which the building is located based on the historical and future meteorological data, including climate average year, extreme high temperature event year, extreme low temperature event year, and seasonal representative day climate scenarios; acquire internal dynamic load data of the building based on building body data and building structure data, including personnel density, personnel activity periods, power of various electrical appliances, and usage periods of various electrical appliances; determine energy consumption scenarios of the building based on internal dynamic load data; simulate key climate scenarios and energy consumption scenarios, and set combinations of various meteorological parameters and various energy consumption parameters in simulation software to generate building energy consumption data under different scenarios.

[0204] In an exemplary embodiment, the model training module 406 is specifically used to perform unified dimension processing on the corresponding parameter values ​​of building energy consumption data under different scenarios; use the processed building energy consumption data as the real label for model training; match the building ontology data, building structure data and real label to establish a mapping relationship table; and train the maximum energy consumption demand assessment model of the building according to the mapping relationship table.

[0205] In an exemplary embodiment, the model training module 406 is specifically used to calculate the regression predicted load value and classification predicted potential level of at least one machine learning algorithm; determine the target machine learning algorithm based on the regression predicted load value and classification predicted potential level; optimize and validate the model parameters of building energy consumption data under different scenarios using a k-fold cross-validation strategy; and use the validated building energy consumption data to sample the target machine learning algorithm and train the maximum energy consumption demand assessment model of the building.

[0206] Each module in the aforementioned maximum energy demand assessment device for buildings can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device, or stored in the memory of a computer device as software, so that the processor can call and execute the corresponding operations of each module.

[0207] In one exemplary embodiment, a computer device is provided, which may be a server, and its internal structure diagram may be as follows: Figure 5 As shown, this computer device includes a processor, memory, input / output (I / O) interfaces, and a communication interface. The processor, memory, and I / O interfaces are connected via a system bus, and the communication interface is also connected to the system bus via the I / O interfaces. The processor provides computational and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system, computer programs, and a database. The internal memory provides the environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The database stores building data. The I / O interfaces are used for exchanging information between the processor and external devices. The communication interface is used for communication with external terminals via a network connection. When executed by the processor, the computer program implements a method for assessing the maximum energy consumption demand of a building.

[0208] Those skilled in the art will understand that Figure 5 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0209] In one exemplary embodiment, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to perform the following steps:

[0210] Acquire building data, which includes map data, terrain data of the location, internal structure images, and exterior images;

[0211] Based on the internal structure images and the external images, determine the building body data and building structure data;

[0212] Based on map data, terrain data, building body data, and building structure data, energy consumption data of buildings under different scenarios are generated through building energy consumption simulation.

[0213] Train a model to assess the maximum energy consumption demand of buildings based on building energy consumption data in different scenarios.

[0214] Obtain the target building data for the building to be evaluated;

[0215] Based on the target building data, the maximum energy consumption demand assessment model is used to output the maximum energy consumption demand value of the building to be assessed.

[0216] In one embodiment, the processor, when executing a computer program, also performs the following steps:

[0217] Building structure data includes wall material type and window-to-wall coverage ratio;

[0218] Pixel-level semantic segmentation is performed on the building facade in the exterior image to determine the wall material type and the segmentation results between windows, walls and other areas;

[0219] Based on the segmentation results, calculate the window-to-wall coverage ratio of the building;

[0220] The historical description information of the building is obtained, and the internal structural images and historical description information are analyzed to obtain the building body data. The historical description information includes the building's use, construction year, internal equipment information, and maintenance records.

[0221] In one embodiment, the processor, when executing a computer program, also performs the following steps:

[0222] Assess the importance of all features contained in the building body data and building structure data;

[0223] The target building feature vector is determined based on the ranking of the importance of all features.

[0224] Based on map data, terrain data, and target building feature vectors, energy consumption data of buildings under different scenarios is generated through building energy consumption simulation.

[0225] In one embodiment, the processor, when executing a computer program, also performs the following steps:

[0226] Based on map and terrain data, obtain historical and future weather data for the area where the building is located;

[0227] Based on historical and future meteorological data, the key climate scenarios for the area where the building is located are determined. The key climate scenarios include the average climate year, the year of extreme high temperature events, the year of extreme low temperature events, and the climate scenarios of seasonal representative days.

[0228] Based on the building body data and building structure data, obtain the internal dynamic load data of the building, which includes personnel density, personnel activity time periods, power of various electrical appliances, and usage time periods of various electrical appliances;

[0229] Based on the internal dynamic load data, determine the building's energy consumption usage scenarios;

[0230] Simulate key climate and energy consumption scenarios, and set various combinations of meteorological and energy consumption parameters in the simulation software to generate building energy consumption data under different scenarios.

[0231] In one embodiment, the processor, when executing a computer program, also performs the following steps:

[0232] The corresponding parameter values ​​of building energy consumption data under different scenarios are processed to have a unified dimension.

[0233] The processed building energy consumption data is used as the true label for model training;

[0234] The building data, building structure data and real labels are matched to establish a mapping table;

[0235] Based on the mapping table, train a model to assess the maximum energy consumption demand of buildings.

[0236] In one embodiment, the processor, when executing a computer program, also performs the following steps:

[0237] Calculate the regression prediction loading and classification prediction potential level of at least one machine learning algorithm;

[0238] The target machine learning algorithm is determined based on the regression prediction loading value and the classification prediction potential level;

[0239] A k-fold cross-validation strategy was adopted to optimize and validate model parameters for building energy consumption data under different scenarios.

[0240] Using validated building energy consumption data, a target machine learning algorithm is sampled to train a model for assessing the building's maximum energy consumption demand.

[0241] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon, the computer program performing the following steps when executed by a processor:

[0242] Acquire building data, which includes map data, terrain data of the location, internal structure images, and exterior images;

[0243] Based on the internal structure images and the external images, determine the building body data and building structure data;

[0244] Based on map data, terrain data, building body data, and building structure data, energy consumption data of buildings under different scenarios are generated through building energy consumption simulation.

[0245] Train a model to assess the maximum energy consumption demand of buildings based on building energy consumption data in different scenarios.

[0246] Obtain the target building data for the building to be evaluated;

[0247] Based on the target building data, the maximum energy consumption demand assessment model is used to output the maximum energy consumption demand value of the building to be assessed.

[0248] In one embodiment, when the computer program is executed by a processor, it also performs the following steps:

[0249] Building structure data includes wall material type and window-to-wall coverage ratio;

[0250] Pixel-level semantic segmentation is performed on the building facade in the exterior image to determine the wall material type and the segmentation results between windows, walls and other areas;

[0251] Based on the segmentation results, calculate the window-to-wall coverage ratio of the building;

[0252] The historical description information of the building is obtained, and the internal structural images and historical description information are analyzed to obtain the building body data. The historical description information includes the building's use, construction year, internal equipment information, and maintenance records.

[0253] In one embodiment, when the computer program is executed by a processor, it also performs the following steps:

[0254] Assess the importance of all features contained in the building body data and building structure data;

[0255] The target building feature vector is determined based on the ranking of the importance of all features.

[0256] Based on map data, terrain data, and target building feature vectors, energy consumption data of buildings under different scenarios is generated through building energy consumption simulation.

[0257] In one embodiment, when the computer program is executed by a processor, it also performs the following steps:

[0258] Based on map and terrain data, obtain historical and future weather data for the area where the building is located;

[0259] Based on historical and future meteorological data, the key climate scenarios for the area where the building is located are determined. The key climate scenarios include the average climate year, the year of extreme high temperature events, the year of extreme low temperature events, and the climate scenarios of seasonal representative days.

[0260] Based on the building body data and building structure data, obtain the internal dynamic load data of the building, which includes personnel density, personnel activity time periods, power of various electrical appliances, and usage time periods of various electrical appliances;

[0261] Based on the internal dynamic load data, determine the building's energy consumption usage scenarios;

[0262] Simulate key climate and energy consumption scenarios, and set various combinations of meteorological and energy consumption parameters in the simulation software to generate building energy consumption data under different scenarios.

[0263] In one embodiment, when the computer program is executed by a processor, it also performs the following steps:

[0264] The corresponding parameter values ​​of building energy consumption data under different scenarios are processed to have a unified dimension.

[0265] The processed building energy consumption data is used as the true label for model training;

[0266] The building data, building structure data and real labels are matched to establish a mapping table;

[0267] Based on the mapping table, train a model to assess the maximum energy consumption demand of buildings.

[0268] In one embodiment, when the computer program is executed by a processor, it also performs the following steps:

[0269] Calculate the regression prediction loading and classification prediction potential level of at least one machine learning algorithm;

[0270] The target machine learning algorithm is determined based on the regression prediction loading value and the classification prediction potential level;

[0271] A k-fold cross-validation strategy was adopted to optimize and validate model parameters for building energy consumption data under different scenarios.

[0272] Using validated building energy consumption data, a target machine learning algorithm is sampled to train a model for assessing the building's maximum energy consumption demand.

[0273] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, performs the following steps:

[0274] Acquire building data, which includes map data, terrain data of the location, internal structure images, and exterior images;

[0275] Based on the internal structure images and the external images, determine the building body data and building structure data;

[0276] Based on map data, terrain data, building body data, and building structure data, energy consumption data of buildings under different scenarios are generated through building energy consumption simulation.

[0277] Train a model to assess the maximum energy consumption demand of buildings based on building energy consumption data in different scenarios.

[0278] Obtain the target building data for the building to be evaluated;

[0279] Based on the target building data, the maximum energy consumption demand assessment model is used to output the maximum energy consumption demand value of the building to be assessed.

[0280] In one embodiment, when the computer program is executed by a processor, it also performs the following steps:

[0281] Building structure data includes wall material type and window-to-wall coverage ratio;

[0282] Pixel-level semantic segmentation is performed on the building facade in the exterior image to determine the wall material type and the segmentation results between windows, walls and other areas;

[0283] Based on the segmentation results, calculate the window-to-wall coverage ratio of the building;

[0284] The historical description information of the building is obtained, and the internal structural images and historical description information are analyzed to obtain the building body data. The historical description information includes the building's use, construction year, internal equipment information, and maintenance records.

[0285] In one embodiment, when the computer program is executed by a processor, it also performs the following steps:

[0286] Assess the importance of all features contained in the building body data and building structure data;

[0287] The target building feature vector is determined based on the ranking of the importance of all features.

[0288] Based on map data, terrain data, and target building feature vectors, energy consumption data of buildings under different scenarios is generated through building energy consumption simulation.

[0289] In one embodiment, when the computer program is executed by a processor, it also performs the following steps:

[0290] Based on map and terrain data, obtain historical and future weather data for the area where the building is located;

[0291] Based on historical and future meteorological data, the key climate scenarios for the area where the building is located are determined. The key climate scenarios include the average climate year, the year of extreme high temperature events, the year of extreme low temperature events, and the climate scenarios of seasonal representative days.

[0292] Based on the building body data and building structure data, obtain the internal dynamic load data of the building, which includes personnel density, personnel activity time periods, power of various electrical appliances, and usage time periods of various electrical appliances;

[0293] Based on the internal dynamic load data, determine the building's energy consumption usage scenarios;

[0294] Simulate key climate and energy consumption scenarios, and set various combinations of meteorological and energy consumption parameters in the simulation software to generate building energy consumption data under different scenarios.

[0295] In one embodiment, when the computer program is executed by a processor, it also performs the following steps:

[0296] The corresponding parameter values ​​of building energy consumption data under different scenarios are processed to have a unified dimension.

[0297] The processed building energy consumption data is used as the true label for model training;

[0298] The building data, building structure data and real labels are matched to establish a mapping table;

[0299] Based on the mapping table, train a model to assess the maximum energy consumption demand of buildings.

[0300] In one embodiment, when the computer program is executed by a processor, it also performs the following steps:

[0301] Calculate the regression prediction loading and classification prediction potential level of at least one machine learning algorithm;

[0302] The target machine learning algorithm is determined based on the regression prediction loading value and the classification prediction potential level;

[0303] A k-fold cross-validation strategy was adopted to optimize and validate model parameters for building energy consumption data under different scenarios.

[0304] Using validated building energy consumption data, a target machine learning algorithm is sampled to train a model for assessing the building's maximum energy consumption demand.

[0305] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of the relevant data must comply with relevant regulations.

[0306] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, artificial intelligence (AI) processors, etc., and are not limited to these.

[0307] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this application.

[0308] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A method for assessing the maximum energy consumption demand of a building, characterized in that, The method includes: Acquire building data, including map data, terrain data of the location, internal structure images, and exterior images; Based on the internal structure image and the external image, determine the building body data and building structure data; Based on the map data, terrain data, building body data, and building structure data, building energy consumption data under different scenarios is generated through building energy consumption simulation, including: Based on the map data and terrain data, historical and future meteorological data for the area where the building is located are obtained. Based on the historical and future meteorological data, key climate scenarios for the area where the building is located are determined, including climate average year, extreme high-temperature event year, extreme low-temperature event year, and seasonal representative day climate scenarios. Based on the building body data and building structure data, internal dynamic load data for the building is obtained, including personnel density, personnel activity periods, power consumption of various electrical appliances, and usage periods of various electrical appliances. Based on the internal dynamic load data, the building's energy consumption scenarios are determined. The key climate scenarios and energy consumption scenarios are simulated, and combinations of various meteorological parameters and energy consumption parameters are set in simulation software to generate building energy consumption data under different scenarios. Based on building energy consumption data under different scenarios, a model for assessing the maximum energy consumption demand of buildings is trained, including: The corresponding parameter values ​​of building energy consumption data under different scenarios are processed to a unified dimension; the processed building energy consumption data is used as the real label for model training; the building body data, the building structure data and the real label are matched to establish a mapping relationship table; and the maximum energy consumption demand assessment model of the building is trained according to the mapping relationship table. Obtain the target building data for the building to be evaluated; Based on the target building data, the maximum energy consumption demand value of the building to be evaluated is output using the maximum energy consumption demand assessment model.

2. The method according to claim 1, characterized in that, The building structure data includes wall material type and window-to-wall coverage ratio; the process of determining the building body data and building structure data based on the internal structure image and the external image includes: Pixel-level semantic segmentation is performed on the building facade in the appearance image to determine the wall material type and the segmentation results between windows, walls and other areas; Based on the segmentation results, calculate the window-to-wall coverage ratio of the building; Historical description information of the building is obtained, and the internal structure images and historical description information are analyzed to obtain building body data. The historical description information includes the building's purpose, construction year, internal equipment information, and maintenance records.

3. The method according to claim 2, characterized in that, The step of generating building energy consumption data under different scenarios based on the map data, terrain data, building body data, and building structure data through building energy consumption simulation may include: Assess the importance of all features contained in the building body data and the building structure data; The target building feature vector is determined based on the ranking of the importance of all features. Based on the map data, the terrain data, and the target building feature vector, building energy consumption data under different scenarios is generated through building energy consumption simulation.

4. The method according to claim 1, characterized in that, The step of training a maximum energy demand assessment model for a building based on building energy consumption data under different scenarios may include: Calculate the regression prediction loading and classification prediction potential level of at least one machine learning algorithm; The target machine learning algorithm is determined based on the regression prediction loading value and the classification prediction potential level; A k-fold cross-validation strategy was adopted to optimize and validate model parameters for building energy consumption data under different scenarios. Using validated building energy consumption data, a target machine learning algorithm is sampled to train a model for assessing the building's maximum energy consumption demand.

5. A device for assessing the maximum energy consumption demand of a building, characterized in that, The apparatus comprising, when using the method according to any one of claims 1-4, includes: The data acquisition module is used to acquire building data, which includes map data, terrain data of the location, internal structure images, and exterior images. The data processing module is used to determine the building body data and building structure data based on the internal structure image and the external image; The data processing module is also used to generate building energy consumption data under different scenarios based on the map data, the terrain data, the building body data and the building structure data through building energy consumption simulation. The model training module is used to train a model for assessing the maximum energy consumption demand of a building based on building energy consumption data in different scenarios. The data acquisition module is also used to acquire target building data of the building to be evaluated; The maximum energy consumption demand assessment module is used to output the maximum energy consumption demand value of the building to be assessed based on the target building data and the maximum energy consumption demand assessment model.

6. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 4.

7. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 4.

8. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 4.

Images