Method, device, equipment and medium for generating optimal building energy retrofit strategy

Abstract

The present invention relates to the field of artificial intelligence technology, and provides a method, device, equipment and medium for generating an optimal renovation strategy for building energy. On the one hand, based on multi-dimensional data, the energy simulation results of urban buildings are obtained, and training samples are constructed according to the energy simulation results of urban buildings to perform integrated training on multiple machine learning models to obtain a surrogate model. In this way, during subsequent building performance simulations, the surrogate model can be directly used for prediction without the need for long-term energy consumption simulations each time, improving the processing efficiency. On the other hand, multi-objective optimization processing is carried out based on the surrogate model and the parallel genetic algorithm plug-in to quickly generate the optimal renovation strategy for the target building, which not only realizes the quantification of the renovation results, but also does not rely on empiricism, making the generated renovation strategy more reasonable.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence technology, and particularly to a method, device, equipment and medium for generating an optimal building energy renovation strategy. Background Art

[0002] With the continuous advancement of the urbanization process, the urban heat island effect has become increasingly serious, and the urban heat island effect has a significant impact on the energy consumption of urban buildings.

[0003] In response to the above problems, implementing effective building energy renovation is crucial for improving urban energy efficiency and reducing carbon emissions.

[0004] However, the existing building energy renovation solutions only rely on empiricism, cannot quantify the renovation effect, and cannot give specific renovation strategies according to specific cases. At the same time, the existing building energy renovation solutions also need to use energy consumption simulation as the evaluation basis, so there is also a problem of long time consumption. Summary of the Invention

[0005] In view of the above, it is necessary to provide a method, device, equipment and medium for generating an optimal building energy renovation strategy, aiming to solve the problem of being unable to generate an optimal building energy renovation strategy efficiently and specifically.

[0006] A method for generating an optimal building energy renovation strategy, the method for generating an optimal building energy renovation strategy includes:

[0007] Obtain building data of a target city to construct a city model;

[0008] Obtain data of a typical meteorological year of the city and greening data of the target city, and input the city model, the data of the typical meteorological year of the city and the greening data into a city weather generator to obtain meteorological data of the target city under the influence of the urban heat island effect;

[0009] Call a building energy consumption simulation engine to simulate the building energy consumption of the target city based on the city model, the data of the typical meteorological year of the city or the meteorological data of the target city under the influence of the urban heat island effect, and obtain a simulation result of the urban building energy of the target city;

[0010] Construct a training sample according to the simulation result of the urban building energy to perform integrated training on multiple machine learning models to obtain a proxy model;

[0011] In response to a renovation strategy generation instruction for a target building in the target city, obtain building data of the target building based on the data dimension of the training sample, and call a parallel genetic algorithm plug-in to generate multiple initial renovation strategies for the target building;

[0012] Input the building data and the multiple initial retrofit strategies into the surrogate model, and perform multi-objective optimization on the multiple initial retrofit strategies based on the output of the parallel genetic algorithm plug-in and the surrogate model to obtain the optimal retrofit strategy for the target building.

[0013] According to a preferred embodiment of the present invention, the obtaining of the building data of the target city to construct the city model includes:

[0014] Obtain the building characteristics, building surrounding characteristics, and building thermal characteristics of the specified type of buildings in the target city to construct the city model;

[0015] Among them, the building characteristics are used to characterize the geometric attributes of the building;

[0016] Among them, the building surrounding characteristics are used to characterize the geometric attributes of the surrounding buildings;

[0017] Among them, the building thermal characteristics are used to characterize the thermal attributes of the building envelope materials;

[0018] Among them, the specified type of buildings includes office buildings and commercial buildings.

[0019] According to a preferred embodiment of the present invention, the constructing of the training samples based on the urban building energy simulation results to perform integrated training on multiple machine learning models to obtain the surrogate model includes:

[0020] Generate the building characteristics, building surrounding characteristics, and building thermal characteristics of each building according to the urban building energy simulation results;

[0021] Obtain the retrofit strategy and the total building load of each building;

[0022] Retrieve a preset number of the machine learning models;

[0023] Use the building characteristics, building surrounding characteristics, and building thermal characteristics of each building, as well as the retrofit strategy of each building as inputs, and the total building load of each building as the training target to perform integrated training on the machine learning models to obtain the surrogate model;

[0024] Among them, during the training process, the Bayesian optimization algorithm is used to optimize the model parameters.

[0025] According to a preferred embodiment of the present invention, after obtaining the surrogate model, the method further includes:

[0026] Perform sensitivity detection, interpretability detection, and uncertainty detection on the surrogate model;

[0027] When the proxy model fails the sensitivity detection, and / or the interpretability detection, and / or the uncertainty detection, optimize and train the proxy model;

[0028] Among them, when the proxy model fails the sensitivity test, determine that the data volume of the training samples is insufficient, supplement the training samples based on the urban building energy simulation results, and optimize and train the proxy model based on the supplemented training samples.

[0029] According to a preferred embodiment of the present invention, the calling the parallel genetic algorithm plug-in to generate multiple initial renovation strategies for the target building includes:

[0030] Obtain multiple renovation parameters under the renovation strategy and the sampling range of the value of each renovation parameter;

[0031] Based on the parallel genetic algorithm plug-in, sample the values of each renovation parameter within the corresponding sampling range to obtain multiple sets of renovation parameter sets;

[0032] Take each set of renovation parameters as one of the initial renovation strategies to obtain the multiple initial renovation strategies.

[0033] According to a preferred embodiment of the present invention, the inputting the building data and the multiple initial renovation strategies into the proxy model, and performing multi-objective optimization processing on the multiple initial renovation strategies based on the outputs of the parallel genetic algorithm plug-in and the proxy model to obtain the optimal renovation strategy for the target building includes:

[0034] Construct constraint conditions based on the ideal renovation cost;

[0035] Determine the predicted total building load of each initial renovation strategy according to the output of the proxy model;

[0036] Under the constraint conditions, use the parallel genetic algorithm plug-in to obtain the initial renovation strategy with the lowest predicted total building load as the optimal renovation strategy.

[0037] According to a preferred embodiment of the present invention, the constructing constraint conditions based on the ideal renovation cost includes:

[0038] Construct the constraint conditions using the following formula:

[0039] ;

[0040] Among them, Cost represents the ideal renovation cost, represents the cost weight of the i th renovation parameter under the adopted renovation strategy, Indicates the linear cost index corresponding to the value of the i th retrofit parameter when it is assumed that the retrofit cost of each retrofit parameter within the corresponding sampling range varies linearly under the adopted retrofit strategy, where n is a positive integer;

[0041] Among them, ;

[0042] Among them, Indicates the value of the i th retrofit parameter; Indicates the upper limit of the sampling range of the i th retrofit parameter; Indicates the lower limit of the sampling range of the i th retrofit parameter; Indicates the correlation between the value of the i th retrofit parameter and the retrofit cost. Among them, when , it indicates that the i th retrofit parameter is positively correlated with the retrofit cost, and the higher the value of the i th retrofit parameter, the higher the corresponding retrofit cost. When , it indicates that the i th retrofit parameter is negatively correlated with the retrofit cost, and the lower the value of the i th retrofit parameter, the higher the corresponding retrofit cost.

[0043] A device for generating an optimal building energy retrofit strategy, the device for generating an optimal building energy retrofit strategy includes:

[0044] An acquisition unit for acquiring building data of a target city to construct a city model;

[0045] An input unit for acquiring data of a typical meteorological year of the city and greening data of the target city, and inputting the city model, the data of the typical meteorological year of the city, and the greening data into a city weather generator to obtain meteorological data of the target city under the influence of the urban heat island effect;

[0046] A simulation unit for calling a building energy consumption simulation engine to simulate the building energy consumption of the target city based on the city model, the data of the typical meteorological year of the city, or the meteorological data of the target city under the influence of the urban heat island effect to obtain a simulation result of the urban building energy of the target city;

[0047] A training unit for constructing a training sample according to the simulation result of the urban building energy to perform integrated training on multiple machine learning models to obtain a surrogate model;

[0048] A generation unit, configured to, in response to an instruction for generating a renovation strategy for a target building in the target city, obtain building data of the target building based on the data dimension of the training sample, and call a parallel genetic algorithm plug-in to generate a plurality of initial renovation strategies for the target building;

[0049] An optimization unit, configured to input the building data and the plurality of initial renovation strategies into the proxy model, and perform multi-objective optimization processing on the plurality of initial renovation strategies based on the outputs of the parallel genetic algorithm plug-in and the proxy model to obtain an optimal renovation strategy for the target building.

[0050] A computer device, comprising:

[0051] A memory, storing at least one instruction; and

[0052] A processor, configured to execute the instruction stored in the memory to implement the method for generating an optimal renovation strategy for building energy.

[0053] A computer-readable storage medium, storing at least one instruction therein, where the at least one instruction is executed by a processor in a computer device to implement the method for generating an optimal renovation strategy for building energy.

[0054] As can be seen from the above technical solutions, on the one hand, based on multi-dimensional data, a simulation result of urban building energy is obtained, and a training sample is constructed according to the simulation result of urban building energy to perform integrated training on multiple machine learning models to obtain a proxy model. In this way, during subsequent building performance simulations, the proxy model can be directly used for prediction, without the need for long-term energy consumption simulations each time, improving the processing efficiency. On the other hand, multi-objective optimization processing is performed based on the proxy model and the parallel genetic algorithm plug-in to quickly generate an optimal renovation strategy for the target building, which not only realizes the quantification of the renovation results but also does not rely on empiricism, making the generated renovation strategy more reasonable. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] Figure 1 is a flowchart of a preferred embodiment of the method for generating an optimal renovation strategy for building energy according to the present invention;

[0056] Figure 2 is a functional module diagram of a preferred embodiment of the device for generating an optimal renovation strategy for building energy according to the present invention;

[0057] Figure 3 is a schematic structural diagram of a computer device of a preferred embodiment for implementing the method for generating an optimal renovation strategy for building energy according to the present invention. DETAILED DESCRIPTION

[0058] To make the objectives, technical solutions, and advantages of the present invention clearer, the present invention will be described in detail below with reference to the accompanying drawings and specific embodiments.

[0059] As Figure 1 shown, it is a flowchart of a preferred embodiment of the method for generating the optimal building energy retrofit strategy of the present invention. According to different requirements, the order of steps in this flowchart can be changed, and some steps can be omitted.

[0060] The method for generating the optimal building energy retrofit strategy is applied to one or more computer devices. The computer device is a device that can automatically perform numerical calculations and / or information processing according to pre-set or stored instructions. Its hardware includes, but is not limited to, microprocessors, application specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), digital signal processors (DSPs), embedded devices, etc.

[0061] The computer device can be a computer or a cloud platform, and the computer device can call devices such as rhino and grasshopper.

[0062] Among them, artificial intelligence (AI) is a theory, method, technology, and application system that uses digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceive the environment, acquire knowledge, and use knowledge to obtain the best results.

[0063] Artificial intelligence basic technologies generally include technologies such as sensors, dedicated artificial intelligence chips, cloud computing, distributed storage, big data processing technologies, operation / interaction systems, and mechatronics. Artificial intelligence software technologies mainly include several major directions such as computer vision technology, robotics, biometric technology, speech processing technology, natural language processing technology, and machine learning / deep learning.

[0064] The network where the computer device is located includes, but is not limited to, the Internet, wide area network, metropolitan area network, local area network, virtual private network (VPN), etc.

[0065] S10, obtain building data of the target city to construct a city model.

[0066] In this embodiment, the obtaining building data of the target city to construct a city model includes:

[0067] Obtain the building characteristics, the characteristics of the surrounding buildings, and the building thermal characteristics of the specified type of buildings in the target city to construct the city model;

[0068] Among them, the building characteristics are used to characterize the geometric attributes of the building;

[0069] Among them, the characteristics of the surrounding buildings are used to characterize the geometric attributes of the surrounding buildings;

[0070] Among them, the building thermal characteristics are used to characterize the thermal attributes of the building envelope materials (including wall materials, glass, etc.);

[0071] Among them, the specified type of buildings includes office buildings and commercial buildings.

[0072] For example: The building characteristics may include, but are not limited to, one or a combination of the following features: total floor area, building volume, building height, window area, shape factor, etc.

[0073] The total floor area refers to the total area of all floors of the building;

[0074] The building volume refers to the total volume of the space occupied by the building, including the total volume of all floors;

[0075] The building height refers to the distance from the highest point of the building to the ground or the foundation;

[0076] The window area refers to the total area of all windows of the building;

[0077] The shape factor refers to the ratio of the surface area of the building in contact with the air to the volume.

[0078] For example: The characteristics of the surrounding buildings may include, but are not limited to, one or a combination of the following features: the number of surrounding buildings, building density, building spacing, sky view factor, perimeter-area ratio.

[0079] The number of surrounding buildings refers to the total number of buildings in the buffer zone;

[0080] The building density refers to the ratio of the sum of the bottom areas of the buildings in the buffer zone to the area of the buffer zone;

[0081] The building spacing refers to the average value of the distances between the buildings in the buffer zone;

[0082] The sky view factor refers to the ratio of the sky area observed on the ground to the total visible area;

[0083] The perimeter-area ratio refers to the ratio of the perimeter of the building outline to the floor area of the building.

[0084] For example, the building thermal characteristics may include, but are not limited to, one or a combination of the following characteristics: the thermal resistance of the external wall, the thermal resistance of the internal wall, etc.

[0085] Among them, the building characteristics and the building surrounding characteristics can be calculated through a GIS (Geographic Information System) platform.

[0086] In the above embodiment, the urban model is constructed based on multi-dimensional variables of building characteristics, building surrounding characteristics, and building thermal characteristics, so that the established model can more accurately express the attributes of urban buildings.

[0087] S11. Obtain the data of the typical meteorological year of the city and the greening data of the target city, and input the urban model, the data of the typical meteorological year of the city, and the greening data into an Urban Weather Generator (UWG) to obtain the meteorological data of the target city under the influence of the urban heat island effect.

[0088] In this embodiment, the data of the typical meteorological year of the city can be sourced from a highly reliable meteorological database such as the EnergyPlus typical meteorological year database.

[0089] In this embodiment, the greening data may include greening indicators such as tree coverage rate and grass coverage rate.

[0090] In this embodiment, inputting the urban model, the data of the typical meteorological year of the city, and the greening data into the Urban Weather Generator can quickly generate the meteorological data of the target city under the influence of the urban heat island effect.

[0091] S12. Call a building energy consumption simulation engine to simulate the building energy consumption of the target city based on the urban model, the data of the typical meteorological year of the city, or the meteorological data under the influence of the urban heat island effect, and obtain the urban building energy simulation result of the target city.

[0092] For example, the building energy consumption simulation engine can be an EnergyPlus engine. By performing UBEM (Urban Building Energy Modeling), the energy consumption in the following four scenarios can be simulated: the energy consumption of an office building under the data of the typical meteorological year of the city, the energy consumption of a commercial building under the data of the typical meteorological year of the city, the energy consumption of an office building under the meteorological data under the influence of the urban heat island effect, and the energy consumption of a commercial building under the meteorological data under the influence of the urban heat island effect.

[0093] After obtaining the urban building energy simulation results of the target city, in order to further ensure the accuracy of the model, it is also possible to obtain the building energy consumption report of the target city, and determine the accuracy of the urban building energy simulation results by comparing the urban building energy simulation results with the building energy consumption report, further ensuring the availability of the urban building energy simulation results.

[0094] S13. Construct training samples based on the urban building energy simulation results to perform integrated training on multiple machine learning models, and obtain a surrogate model.

[0095] In this embodiment, the step of constructing training samples based on the urban building energy simulation results to perform integrated training on multiple machine learning models and obtain a surrogate model includes:

[0096] Generate building features, building surrounding features, and building thermal characteristics of each building according to the urban building energy simulation results;

[0097] Obtain the renovation strategy and total building load of each building;

[0098] Retrieve a preset number of the machine learning models;

[0099] Use the building features, building surrounding features, and building thermal characteristics of each building, as well as the renovation strategy of each building as inputs, and use the total building load of each building as the training target to perform integrated training on the machine learning models to obtain the surrogate model;

[0100] Among them, during the training process, the Bayesian optimization algorithm is used to optimize the model parameters.

[0101] Among them, the preset number can be configured according to actual accuracy requirements.

[0102] For example: when the preset number is 6, the machine learning models may include six integrated learning algorithm models: BaggingClassifier (BR), Extremely randomized trees (ETR), RandomForest Regression (RFR), Gradient Boosting Regression (GBR), Adaptive Boosting (ADB), and Extreme Gradient Boosting (XGB). By performing integrated training on the above six models, it is possible to comprehensively obtain more accurate prediction results based on the outputs of the six models.

[0103] In this embodiment, a machine learning model is used to construct a data-driven energy model. Building features, building surrounding features, and building thermal characteristics are used as model features. The energy consumption under various scenarios obtained by batch simulation of the urban building energy simulation results (such as the energy consumption of office buildings under the data of the typical meteorological year in the city, the energy consumption of commercial buildings under the data of the typical meteorological year in the city, the energy consumption of office buildings under the meteorological data affected by the urban heat island effect, and the energy consumption of commercial buildings under the meteorological data affected by the urban heat island effect) are used as targets to train the machine learning model respectively. And based on the ensemble algorithm, the outputs of each model are integrated to make the prediction results of the obtained surrogate model more accurate.

[0104] Further, in order to verify the accuracy of the surrogate model, the support vector regression (SVR) of the basic model can also be used as a comparison to verify the advantages of the surrogate model obtained by ensemble learning.

[0105] In this embodiment, after obtaining the surrogate model, the method further includes:

[0106] Performing sensitivity detection, interpretability detection, and uncertainty detection on the surrogate model;

[0107] When the surrogate model fails the sensitivity detection, and / or the interpretability detection, and / or the uncertainty detection, perform optimized training on the surrogate model;

[0108] Among them, when the surrogate model fails the sensitivity test, it is determined that the data volume of the training samples is insufficient. Based on the urban building energy simulation results, the training samples are supplemented, and the surrogate model is optimized and trained based on the supplemented training samples.

[0109] Among them, the sensitivity detection refers to detecting whether the data volume is sufficient. For example: For each group of training, the data volume is increased by 2% each time and multiple models are trained. If the model performance tends to be stable, it means that the training data is sufficient.

[0110] Among them, the interpretability detection refers to detecting the contribution of each feature. For example: The SHapley Additive exPlanations (SHAP) method based on game theory can be used to explain the best model to judge the contribution of each feature to the building energy performance.

[0111] Among them, the uncertainty detection refers to detecting the average output and confidence of the model. For example: 95% bootstrap sampling can be performed on the training data set, and a certain number of models are trained. Based on the prediction results of the series of models, the average output and confidence of the model are determined.

[0112] Through the above embodiments, the accuracy of the model can be further verified.

[0113] S14. In response to an instruction for generating a renovation strategy for a target building in the target city, obtain the building data of the target building based on the data dimension of the training sample, and call a parallel genetic algorithm plug-in to generate multiple initial renovation strategies for the target building.

[0114] In this embodiment, the instruction for generating the renovation strategy can be triggered by relevant staff such as an architectural designer.

[0115] For example: the building features, building perimeter features, and building thermal performance features of the target building can be obtained according to the data dimension of the training sample.

[0116] In this embodiment, the calling of the parallel genetic algorithm plug-in to generate multiple initial renovation strategies for the target building includes:

[0117] Obtain multiple renovation parameters under the renovation strategy and the sampling range of the value of each renovation parameter;

[0118] Based on the parallel genetic algorithm plug-in, sample the values of each renovation parameter within the corresponding sampling range to obtain multiple sets of renovation parameter sets;

[0119] Use each set of renovation parameters as one of the initial renovation strategies to obtain the multiple initial renovation strategies.

[0120] For example: the multiple renovation parameters under the renovation strategy and the sampling range of the value of each renovation parameter can be seen in the following table:

[0121] Reform parameters Sampling range North window-wall ratio [0.2-0.8] East window-wall ratio [0.2-0.8] South window-wall ratio [0.2-0.8] West window-wall ratio [0.2-0.8] Exterior wall thermal resistance [0.667-2.80] Roof thermal resistance [0.604-2.350] Floor slab thermal resistance [0.667-1.42] Glass U-value [1.0-5.2] Glass heat gain coefficient [0.18- 0.52]

[0122] Furthermore, based on the parallel genetic algorithm plug-in, it is possible to extract a north window-wall ratio of 0.2, an east window-wall ratio of 0.2, a south window-wall ratio of 0.2, a west window-wall ratio of 0.2, an external wall thermal resistance of 0.667, a roof thermal resistance of 0.604, a floor thermal resistance of 0.667, a glass U-value of 1, and a glass heat gain coefficient of 0.18 from the sampling range of each renovation parameter to form an initial renovation strategy; extract a north window-wall ratio of 0.8, an east window-wall ratio of 0.8, a south window-wall ratio of 0.8, a west window-wall ratio of 0.8, an external wall thermal resistance of 2.80, a roof thermal resistance of 2.350, a floor thermal resistance of 1.42, a glass U-value of 5.2, and a glass heat gain coefficient of 0.52 to form an initial renovation strategy, and so on, finally forming multiple initial renovation strategies.

[0123] Among them, the parallel genetic algorithm plug-in can be the iGeneS parallel genetic algorithm plug-in developed by SD-II Lab of Tongji University, which is used to execute multi-objective optimization tasks. iGeneS is a new generation of genetic algorithm plug-in specially developed for deploying machine learning agent models. It realizes batch calling of agent models through matrix operations, and greatly shortens the optimization time compared with traditional optimization algorithm plug-ins.

[0124] Therefore, in this embodiment, the parallel genetic algorithm can automatically extract the values of each transformation parameter to form multiple initial transformation strategies without relying on artificial experience, so the efficiency is higher.

[0125] S15. Input the building data and the multiple initial transformation strategies into the agent model, and perform multi-objective optimization processing on the multiple initial transformation strategies based on the outputs of the parallel genetic algorithm plug-in and the agent model to obtain the optimal transformation strategy for the target building.

[0126] In this embodiment, the step of inputting the building data and the multiple initial transformation strategies into the agent model, and performing multi-objective optimization processing on the multiple initial transformation strategies based on the outputs of the parallel genetic algorithm plug-in and the agent model to obtain the optimal transformation strategy for the target building includes:

[0127] Construct constraint conditions based on the ideal transformation cost;

[0128] Determine the predicted total building load of each initial transformation strategy according to the output of the agent model;

[0129] Under the constraint conditions, use the parallel genetic algorithm plug-in to obtain the initial transformation strategy with the lowest predicted total building load as the optimal transformation strategy.

[0130] Specifically, the step of constructing constraint conditions based on the ideal transformation cost includes:

[0131] Use the following formula to construct the constraint conditions:

[0132] ;

[0133] Among them, Cost represents the ideal transformation cost, represents the cost weight of the i th transformation parameter under the adopted transformation strategy, represents the linear cost index corresponding to the value of the i th transformation parameter when the transformation cost of each transformation parameter within the corresponding sampling range is assumed to vary linearly under the adopted transformation strategy, and n is a positive integer;

[0134] Among them, ;

[0135] Among them, represents the value of the i th retrofit parameter; represents the upper limit of the sampling range of the i th retrofit parameter; represents the lower limit of the sampling range of the i th retrofit parameter; represents the correlation between the value of the i th retrofit parameter and the retrofit cost. Among them, when , it means that the i th retrofit parameter is positively correlated with the retrofit cost, and the higher the value of the i th retrofit parameter, the higher the corresponding retrofit cost. When , it means that the i th retrofit parameter is negatively correlated with the retrofit cost, and the lower the value of the i th retrofit parameter, the higher the corresponding retrofit cost.

[0136] Among them, when , it means that the i-th retrofit parameter has no correlation with the retrofit cost.

[0137] In the above embodiment, the surrogate model is multi-objectively optimized with the total building load and the retrofit cost as the objectives. While ensuring a low total building load, it can also avoid the situation where all retrofit parameters converge to the thermally optimal values, resulting in cost overruns. Furthermore, the generated optimal retrofit strategy can achieve an ideal balance between the total building load and the retrofit cost.

[0138] Using this embodiment, the optimal building energy retrofit strategy for a typical city can be obtained, and the building energy retrofit plan for a similar city can be studied based on the optimal building energy retrofit strategy of the typical city. For example: By studying the optimal building energy retrofit strategy of Shenzhen City, it is expected to provide reference for other large cities suffering from the heat island effect.

[0139] It can be seen from the above technical solutions that, on the one hand, the urban building energy simulation results are obtained based on multi-dimensional data, and training samples are constructed according to the urban building energy simulation results to perform integrated training on multiple machine learning models to obtain a surrogate model. In this way, during subsequent building performance simulations, the surrogate model can be directly used for prediction, eliminating the need for long-term energy consumption simulations each time, thus improving the processing efficiency. On the other hand, multi-objective optimization processing is performed based on the surrogate model and the parallel genetic algorithm plug-in to quickly generate the optimal retrofit strategy for the target building, not only realizing the quantification of the retrofit results, but also not relying on empiricism, making the generated retrofit strategy more reasonable.

[0140] Such as Figure 2As shown in the figure, it is a functional block diagram of a preferred embodiment of the building energy optimal retrofit strategy generation device of the present invention. The building energy optimal retrofit strategy generation device 11 includes an acquisition unit 110, an input unit 111, a simulation unit 112, a training unit 113, a generation unit 114, and an optimization unit 115. The modules / units referred to in the present invention refer to a series of computer program segments that can be executed by a processor and can complete fixed functions, and are stored in a memory. In this embodiment, the functions of each module / unit will be described in detail in subsequent embodiments.

[0141] Among them, the acquisition unit 110 is used to acquire building data of a target city to construct a city model;

[0142] The input unit 111 is used to acquire typical meteorological year data of the city and the greening data of the target city, and input the city model, the typical meteorological year data of the city, and the greening data into a city weather generator to obtain the meteorological data of the target city under the influence of the urban heat island effect;

[0143] The simulation unit 112 is used to call a building energy consumption simulation engine to simulate the building energy consumption of the target city based on the city model, the typical meteorological year data of the city, or the meteorological data of the target city under the influence of the urban heat island effect, so as to obtain the urban building energy simulation result of the target city;

[0144] The training unit 113 is used to construct training samples according to the urban building energy simulation result to perform integrated training on multiple machine learning models to obtain a surrogate model;

[0145] The generation unit 114 is used to, in response to a retrofit strategy generation instruction for a target building in the target city, acquire the building data of the target building based on the data dimension of the training sample, and call a parallel genetic algorithm plug-in to generate multiple initial retrofit strategies for the target building;

[0146] The optimization unit 115 is used to input the building data and the multiple initial retrofit strategies into the surrogate model, and perform multi-objective optimization processing on the multiple initial retrofit strategies based on the output of the parallel genetic algorithm plug-in and the surrogate model to obtain the optimal retrofit strategy for the target building.

[0147] As can be seen from the above technical solutions, on the one hand, based on multi-dimensional data, the urban building energy simulation results are obtained, and training samples are constructed according to the urban building energy simulation results to integrally train multiple machine learning models to obtain a surrogate model. In this way, during subsequent building performance simulations, the surrogate model can be directly used for prediction, eliminating the need for long-term energy consumption simulations each time and improving the processing efficiency. On the other hand, multi-objective optimization processing is performed based on the surrogate model and the parallel genetic algorithm plug-in to quickly generate the optimal renovation strategy for the target building, not only realizing the quantification of the renovation results but also not relying on empiricism, making the generated renovation strategy more reasonable.

[0148] As Figure 3 shown, it is a schematic structural diagram of a computer device according to a preferred embodiment of the method for generating an optimal renovation strategy for building energy in the present invention.

[0149] The computer device 1 may include a memory 12, a processor 13, and a bus, and may also include a computer program stored in the memory 12 and executable on the processor 13, such as a program for generating an optimal renovation strategy for building energy.

[0150] Those skilled in the art can understand that the schematic diagram is only an example of the computer device 1 and does not constitute a limitation on the computer device 1. The computer device 1 may be a bus structure or a star structure. The computer device 1 may also include more or fewer other hardware or software than shown, or different component arrangements. For example, the computer device 1 may also include input / output devices, network access devices, etc.

[0151] It should be noted that the computer device 1 is only an example, and other existing or future possible electronic products that can be adapted to the present invention should also be included within the protection scope of the present invention and are hereby incorporated by reference.

[0152] Among them, the memory 12 includes at least one type of readable storage medium, and the readable storage medium includes flash memory, mobile hard disk, multimedia card, card-type memory (such as SD or DX memory, etc.), magnetic memory, magnetic disk, optical disc, etc. The memory 12 can be an internal storage unit of the computer device 1 in some embodiments, such as the mobile hard disk of the computer device 1. The memory 12 can also be an external storage device of the computer device 1 in other embodiments, such as a plug-in mobile hard disk, Smart Media Card (SMC), Secure Digital (SD) card, Flash Card, etc. equipped on the computer device 1. Further, the memory 12 can also include both the internal storage unit and the external storage device of the computer device 1. The memory 12 can not only be used to store application software installed in the computer device 1 and various types of data, such as the code of the building energy optimal retrofit strategy generation program, etc., but also be used to temporarily store the data that has been output or will be output.

[0153] The processor 13 can be composed of integrated circuits in some embodiments. For example, it can be composed of a single packaged integrated circuit, or can be composed of multiple integrated circuits with the same or different functions packaged, including the combination of one or more central processing units (CPU), microprocessors, digital processing chips, graphics processors, and various control chips, etc. The processor 13 is the control core (Control Unit) of the computer device 1, connects all components of the entire computer device 1 through various interfaces and lines, and by running or executing the programs or modules stored in the memory 12 (such as executing the building energy optimal retrofit strategy generation program, etc.), and calling the data stored in the memory 12, to execute various functions of the computer device 1 and process data.

[0154] The processor 13 executes the operating system of the computer device 1 and various installed application programs. The processor 13 executes the application program to implement the steps in the above-mentioned embodiments of each building energy optimal retrofit strategy generation method, such as Figure 1 the steps shown.

[0155] Exemplarily, the computer program may be divided into one or more modules / units, and the one or more modules / units are stored in the memory 12 and executed by the processor 13 to implement the present invention. The one or more modules / units may be a series of computer-readable instruction segments capable of performing specific functions, and these instruction segments are used to describe the execution process of the computer program in the computer device 1. For example, the computer program may be divided into an acquisition unit 110, an input unit 111, a simulation unit 112, a training unit 113, a generation unit 114, and an optimization unit 115.

[0156] The integrated units implemented in the form of software function modules may be stored in a computer-readable storage medium. The above software function modules stored in a storage medium include several instructions for causing a computer device (which may be a personal computer, a computer device, or a network device, etc.) or a processor to execute a part of the building energy optimal retrofit strategy generation method described in various embodiments of the present invention.

[0157] If the integrated module / unit of the computer device 1 is implemented in the form of a software function unit and sold or used as an independent product, it may be stored in a computer-readable storage medium. Based on such an understanding, to implement all or part of the processes in the above method embodiments of the present invention, it may also be completed by a computer program instructing relevant hardware devices. The computer program may be stored in a computer-readable storage medium, and when the computer program is executed by a processor, the steps of the above various method embodiments may be implemented.

[0158] Among them, the computer program includes computer program code, and the computer program code may be in the form of source code, object code, an executable file, or some intermediate form, etc. The computer-readable medium may include: any entity or device capable of carrying the computer program code, a recording medium, a USB flash drive, a mobile hard disk, a magnetic disk, an optical disc, a computer memory, a read-only memory (ROM, Read-Only Memory), a random access memory, etc.

[0159] Further, the computer-readable storage medium mainly includes a program storage area and a data storage area. Among them, the program storage area may store an operating system, application programs required for at least one function, etc.; the data storage area may store data created according to the use of the blockchain node, etc.

[0160] The blockchain referred to in the present invention is a new application mode of computer technologies such as distributed data storage, peer-to-peer transmission, consensus mechanism, and encryption algorithm. Blockchain, in essence, is a decentralized database, a series of data blocks generated by using cryptographic methods. Each data block contains information about a batch of network transactions, which is used to verify the validity of the information (anti-counterfeiting) and generate the next block. The blockchain can include the blockchain underlying platform, the platform product service layer, and the application service layer, etc.

[0161] The bus can be a peripheral component interconnect (PCI) bus or an extended industry standard architecture (EISA) bus, etc. This bus can be divided into an address bus, a data bus, a control bus, etc. For the sake of convenience in representation, in Figure 3 it is only represented by a straight line, but it does not mean that there is only one bus or one type of bus. The bus is arranged to realize the connection and communication between the memory 12 and at least one processor 13, etc.

[0162] Although not shown, the computer device 1 may further include a power source (such as a battery) for supplying power to each component. Preferably, the power source can be logically connected to the at least one processor 13 through a power management device, so as to realize functions such as charge management, discharge management, and power consumption management through the power management device. The power source may also include any components such as one or more DC or AC power sources, a recharge device, a power failure detection circuit, a power converter or inverter, a power status indicator, etc. The computer device 1 may also include a variety of sensors, a Bluetooth module, a Wi-Fi module, etc., which will not be elaborated here.

[0163] Furthermore, the computer device 1 may further include a network interface. Optionally, the network interface may include a wired interface and / or a wireless interface (such as a WI-FI interface, a Bluetooth interface, etc.), which is usually used to establish a communication connection between the computer device 1 and other computer devices.

[0164] Optionally, the computer device 1 may further include a user interface, which may be a display, an input unit (such as a keyboard), and optionally, the user interface may also be a standard wired interface or a wireless interface. Optionally, in some embodiments, the display may be an LED display, a liquid crystal display, a touch liquid crystal display, and an OLED (Organic Light-Emitting Diode) toucher, etc. Among them, the display may also be appropriately referred to as a display screen or a display unit, which is used to display the information processed in the computer device 1 and to display a visual user interface.

[0165] It should be understood that the above embodiments are only for illustration purposes and are not limited by this structure in the scope of the patent application.

[0166] Figure 3 Only the computer device 1 with components 12 - 13 is shown. Those skilled in the art can understand that Figure 3 the shown structure does not constitute a limitation on the computer device 1, and it may include fewer or more components than shown, or combine some components, or have different component arrangements.

[0167] Combined with Figure 1 , the memory 12 in the computer device 1 stores a plurality of instructions to implement a method for generating an optimal renovation strategy for building energy. The processor 13 can execute the plurality of instructions to implement:

[0168] Obtain the building data of the target city to construct a city model;

[0169] Obtain the data of the typical meteorological year of the city and the greening data of the target city, and input the city model, the data of the typical meteorological year of the city, and the greening data into a city meteorological generator to obtain the meteorological data of the target city under the influence of the urban heat island effect;

[0170] Call a building energy consumption simulation engine to simulate the building energy consumption of the target city based on the city model, the data of the typical meteorological year of the city, or the meteorological data under the influence of the urban heat island effect to obtain the urban building energy simulation result of the target city;

[0171] Construct training samples according to the urban building energy simulation result to perform integrated training on multiple machine learning models to obtain a surrogate model;

[0172] In response to a renovation strategy generation instruction for a target building in the target city, obtain the building data of the target building based on the data dimension of the training sample, and call a parallel genetic algorithm plugin to generate a plurality of initial renovation strategies for the target building;

[0173] Input the building data and the multiple initial renovation strategies into the agent model, and perform multi-objective optimization processing on the multiple initial renovation strategies based on the output of the parallel genetic algorithm plug-in and the agent model to obtain the optimal renovation strategy for the target building.

[0174] Specifically, for the specific implementation method of the above instructions by the processor 13, reference can be made to Figure 1 the description of the relevant steps in the corresponding embodiments, which will not be elaborated here.

[0175] It should be noted that all the data involved in this case are legally obtained.

[0176] In several embodiments provided by the present invention, it should be understood that the disclosed systems, devices, and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of the modules is only a logical function division, and there can be other division methods in actual implementation.

[0177] The present invention can be used in many general or special computer system environments or configurations. For example: personal computers, server computers, multi-processor systems, microprocessor-based systems, set-top boxes, programmable consumer electronic devices, network PCs, minicomputers, mainframe computers, distributed computing environments including any of the above systems or devices, and so on. The present invention can be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform specific tasks or implement specific abstract data types. The present invention can also be practiced in distributed computing environments where tasks are performed by remote processing devices connected through a communication network. In a distributed computing environment, program modules can be located in local and remote computer storage media including storage devices.

[0178] The modules described as separate components may or may not be physically separated, and the components shown as modules may or may not be physical units, that is, they can be located in one place, or distributed to multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0179] In addition, in each embodiment of the present invention, the various functional modules can be integrated in a processing unit, or each unit can exist physically alone, or two or more units can be integrated in one unit. The above integrated unit can be implemented in the form of hardware, or in the form of a combination of hardware and software functional modules.

[0180] For those skilled in the art, it is obvious that the present invention is not limited to the details of the above-described exemplary embodiments, and the present invention can be implemented in other specific forms without departing from the spirit or basic characteristics of the present invention.

[0181] Therefore, in any aspect, the embodiments should be regarded as exemplary and non-limiting. The scope of the present invention is defined by the appended claims rather than the above description. Accordingly, all changes falling within the meaning and scope of the equivalent elements of the claims are intended to be embraced by the present invention. Any reference signs in the claims should not be construed as limiting the claims involved.

[0182] In addition, it is obvious that the term "comprising" does not exclude other units or steps, and the singular does not exclude the plural. The multiple units or devices described in the present invention can also be implemented by one unit or device through software or hardware. The terms such as "first" and "second" are used to denote names and do not denote any particular order.

[0183] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical solutions of the present invention can be modified or equivalently replaced without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A method for generating an optimal renovation strategy for building energy, characterized in that, The method for generating the optimal building energy retrofit strategy includes: Obtain the building data of the target city to construct a city model; Obtain the data of the typical meteorological year of the city and the greening data of the target city, and input the city model, the data of the typical meteorological year of the city, and the greening data into a city weather generator to obtain the meteorological data of the target city under the influence of the urban heat island effect; Call a building energy consumption simulation engine to simulate the building energy consumption of the target city based on the city model, the data of the typical meteorological year of the city, or the meteorological data of the target city under the influence of the urban heat island effect, and obtain the urban building energy simulation results of the target city; wherein, the urban building energy simulation results include: the energy consumption of office buildings under the data of the typical meteorological year of the city, the energy consumption of commercial buildings under the data of the typical meteorological year of the city, the energy consumption of office buildings under the meteorological data of the target city under the influence of the urban heat island effect, and the energy consumption of commercial buildings under the meteorological data of the target city under the influence of the urban heat island effect; Construct a training sample according to the urban building energy simulation results to perform integrated training on multiple machine learning models to obtain a surrogate model; In response to a retrofit strategy generation instruction for a target building in the target city, obtain the building data of the target building based on the data dimension of the training sample, and call a parallel genetic algorithm plug-in to generate multiple initial retrofit strategies for the target building; wherein, the step of calling the parallel genetic algorithm plug-in to generate multiple initial retrofit strategies for the target building includes: obtaining multiple retrofit parameters under the retrofit strategy and the sampling range of the value of each retrofit parameter; sampling the value of each retrofit parameter within the corresponding sampling range based on the parallel genetic algorithm plug-in to obtain multiple sets of retrofit parameter sets; using each set of retrofit parameters as an initial retrofit strategy to obtain the multiple initial retrofit strategies; Input the building data and the multiple initial retrofit strategies into the surrogate model, and perform multi-objective optimization processing on the multiple initial retrofit strategies based on the output of the parallel genetic algorithm plug-in and the surrogate model to obtain the optimal retrofit strategy of the target building.

2. The method for generating an optimal building energy retrofit strategy according to claim 1, wherein The step of obtaining the building data of the target city to construct a city model includes: Obtain the building characteristics, the building surrounding characteristics, and the building thermal characteristics of the specified type of buildings in the target city to construct the city model; Wherein, the building characteristics are used to characterize the geometric attributes of the building; Wherein, the building surrounding characteristics are used to characterize the geometric attributes of the surrounding buildings; Wherein, the building thermal characteristics are used to characterize the thermal attributes of the building envelope materials; Wherein, the specified type of buildings includes office buildings and commercial buildings.

3. The method for generating an optimal building energy retrofit strategy according to claim 2, characterized in that, The step of constructing a training sample according to the urban building energy simulation results to perform integrated training on multiple machine learning models to obtain a surrogate model includes: Generate the building characteristics, the building surrounding characteristics, and the building thermal characteristics of each building according to the urban building energy simulation results; Obtain the retrofit strategy and the total building load of each building; Retrieve a preset number of the machine learning models; Taking the building characteristics, the characteristics around the building, and the building thermal characteristics of each building as inputs, and taking the total building load of each building as the training target, the machine learning model is integratively trained to obtain the surrogate model; Among them, during the training process, the Bayesian optimization algorithm is used to optimize the model parameters.

4. The method for generating an optimal building energy retrofit strategy according to claim 3, characterized in that, After obtaining the surrogate model, the method further includes: Performing sensitivity detection, interpretability detection, and uncertainty detection on the surrogate model; When the surrogate model fails the sensitivity detection, and / or the interpretability detection, and / or the uncertainty detection, optimizing and training the surrogate model; Among them, when the surrogate model fails the sensitivity detection, it is determined that the data volume of the training samples is insufficient, the training samples are supplemented based on the urban building energy simulation results, and the surrogate model is optimized and trained based on the supplemented training samples.

5. The method for generating an optimal building energy retrofit strategy according to claim 1, characterized in that The step of inputting the building data and the multiple initial retrofit strategies into the surrogate model, and performing multi-objective optimization on the multiple initial retrofit strategies based on the outputs of the parallel genetic algorithm plug-in and the surrogate model to obtain the optimal retrofit strategy for the target building includes: Constructing a constraint condition based on the ideal retrofit cost; Determining the predicted total building load of each initial retrofit strategy according to the output of the surrogate model; Under the constraint condition, using the parallel genetic algorithm plug-in to obtain the initial retrofit strategy with the lowest predicted total building load as the optimal retrofit strategy.

6. The method for generating an optimal building energy retrofit strategy according to claim 5, characterized in that The constructing a constraint condition based on the ideal retrofit cost includes: Constructing the constraint condition using the following formula: ； Among them, Cost represents the ideal transformation cost, represents the cost weight of the i th transformation parameter under the adopted transformation strategy, represents the linear cost index corresponding to the value of the i th transformation parameter when it is assumed that the transformation cost of each transformation parameter within the corresponding sampling range under the adopted transformation strategy changes linearly, where n is a positive integer; Among them, ; Among them, represents the value of the i th transformation parameter; represents the upper limit of the sampling range of the i th transformation parameter; represents the lower limit of the sampling range of the i th transformation parameter; represents the correlation between the value of the i th transformation parameter and the transformation cost. Among them, when it indicates that the i th transformation parameter is positively correlated with the transformation cost. The greater the value of the i th transformation parameter, the higher the corresponding transformation cost. When it indicates that the i th transformation parameter is negatively correlated with the transformation cost. The smaller the value of the i th transformation parameter, the higher the corresponding transformation cost.

7. An apparatus for generating an optimal building energy retrofit strategy, characterized in that, The building energy optimal retrofit strategy generation device includes: An acquisition unit, configured to acquire building data of a target city to construct a city model; An input unit, configured to acquire urban typical meteorological year data and greening data of the target city, and input the city model, the urban typical meteorological year data, and the greening data into a city weather generator to obtain meteorological data of the target city under the influence of the urban heat island effect; A simulation unit, configured to call a building energy consumption simulation engine to simulate the building energy consumption of the target city based on the city model, the urban typical meteorological year data, or the meteorological data of the target city under the influence of the urban heat island effect, to obtain the urban building energy simulation results of the target city; among them, the urban building energy simulation results include: the energy consumption of office buildings under urban typical meteorological year data, the energy consumption of commercial buildings under urban typical meteorological year data, the energy consumption of office buildings under meteorological data of the target city under the influence of the urban heat island effect, and the energy consumption of commercial buildings under meteorological data of the target city under the influence of the urban heat island effect; A training unit, configured to construct training samples based on the urban building energy simulation results to integratively train multiple machine learning models to obtain a surrogate model; A generating unit, configured to, in response to an instruction for generating a renovation strategy for a target building in the target city, obtain building data of the target building based on the data dimension of the training sample, and call a parallel genetic algorithm plug-in to generate a plurality of initial renovation strategies for the target building; wherein, the calling of the parallel genetic algorithm plug-in to generate a plurality of initial renovation strategies for the target building includes: obtaining a plurality of renovation parameters under the renovation strategy and the sampling range of the value of each renovation parameter; sampling the value of each renovation parameter within the corresponding sampling range based on the parallel genetic algorithm plug-in to obtain multiple sets of renovation parameter sets; using each set of renovation parameters as one of the initial renovation strategies to obtain the plurality of initial renovation strategies. An optimizing unit, configured to input the building data and the plurality of initial renovation strategies into the proxy model, and perform multi-objective optimization processing on the plurality of initial renovation strategies based on the output of the parallel genetic algorithm plug-in and the proxy model to obtain the optimal renovation strategy for the target building.

8. A computer device, characterized in that, The computer device includes: a memory storing at least one instruction; and a processor, configured to execute the instruction stored in the memory to implement the method for generating an optimal renovation strategy for building energy according to any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that: At least one instruction is stored in the computer-readable storage medium, and the at least one instruction is executed by a processor in the computer device to implement the method for generating an optimal renovation strategy for building energy according to any one of claims 1 to 6.

Images