Urban grid building energy consumption correction method and system based on space-time feature coupling

Abstract

The application provides a city grid building energy consumption correction method and system based on space-time feature coupling, which comprises the following steps: obtaining the texture feature of a first target building, the ground heat intensity of a neighborhood space, a dominant wind vector, the spatial relationship between the first target building and the buildings in its neighborhood space, the design energy consumption of the first target building, and the time-series thermal inertia feature of the first target region; determining the neighborhood interaction feature of the first target building based on the dominant wind vector, the spatial relationship between the first target building and the buildings in its neighborhood space, and the ground heat intensity of the neighborhood space; inputting the texture feature of the first target building, the neighborhood interaction feature, and the time-series thermal inertia feature of the first target region into a pre-trained target neural network model to obtain an environmental thermal penalty index; and determining the corrected energy consumption of the first target building based on the design energy consumption of the first target building and the environmental thermal penalty index. Through the implementation of the application, the calculation accuracy of city building energy consumption can be improved.

Description

Technical Field

[0001] This invention belongs to the field of smart energy management technology, specifically relating to a method and system for correcting urban grid-based building energy consumption based on spatiotemporal feature coupling. Background Technology

[0002] As the core carriers of energy consumption, the accurate calculation and control of urban buildings' energy consumption has become a key aspect of smart energy management, urban planning optimization, and carbon emission control. Current building energy consumption assessments typically rely on data from suburban standard weather stations (such as TMY files). Studies show that ignoring urban microclimates (such as urban heat islands, UHI) during the high-temperature summer months can lead to a 11.5% to 16.5% underestimation of building cooling load forecasts in high-density urban areas, with peak load deviations even higher.

[0003] Among related technologies, WRF (Weather Research and Forecasting) combined with BEP / BEM schemes can accurately simulate urban weather, but its computational load is huge and cannot be used as a real-time engineering tool. Furthermore, existing fast estimation models usually consider a certain region independently, ignoring interaction effects, resulting in large errors in building energy consumption calculations. Summary of the Invention

[0004] In view of this, the purpose of the present invention is to provide a method and system for correcting urban gridded building energy consumption based on spatiotemporal feature coupling, so as to meet the need to improve the accuracy of urban building energy consumption calculation.

[0005] To achieve the above objectives, the present invention provides the following technical solution: According to a first aspect, the present invention provides a method for correcting the energy consumption of urban gridded buildings based on spatiotemporal feature coupling, comprising: acquiring the textural features of a first target building, the surface heat intensity of its neighborhood space, the prevailing wind vector, the spatial relationship between the first target building and buildings in its neighborhood space, the design energy consumption of the first target building, and the temporal thermal inertia features of a first target area, wherein the first target area includes the first target building and its neighborhood space, and the design energy consumption of the first target building is determined based on the suburban climate environment; determining the neighborhood interaction features of the first target building based on the prevailing wind vector, the spatial relationship between the first target building and buildings in its neighborhood space, and the surface heat intensity of the neighborhood space; inputting the textural features of the first target building, the neighborhood interaction features, and the temporal thermal inertia features of the first target area into a pre-trained target neural network model to obtain an environmental heat penalty index; and determining the corrected energy consumption of the first target building based on its design energy consumption and the environmental heat penalty index.

[0006] According to a second aspect, the present invention provides an urban gridded building energy consumption correction system based on spatiotemporal feature coupling, comprising: a data acquisition module for acquiring the textural features of a first target building, the surface heat intensity of its neighborhood space, the prevailing wind vector, the spatial relationship between the first target building and buildings in its neighborhood space, the design energy consumption of the first target building, and the temporal thermal inertia features of a first target area, wherein the first target area includes the first target building and its neighborhood space, and the design energy consumption of the first target building is determined based on the suburban climate environment; a neighborhood interaction feature determination module for determining the neighborhood interaction features of the first target building based on the prevailing wind vector, the spatial relationship between the first target building and buildings in its neighborhood space, and the surface heat intensity of the neighborhood space; an environmental heat penalty index determination module for inputting the textural features of the first target building, the neighborhood interaction features, and the temporal thermal inertia features of the first target area into a pre-trained target neural network model to obtain an environmental heat penalty index; and an energy consumption correction module for determining the corrected energy consumption of the first target building based on its design energy consumption and the environmental heat penalty index.

[0007] According to a third aspect, an embodiment of the present invention provides an electronic device, the device comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the steps of the urban gridded building energy consumption correction method based on multi-scale spatiotemporal feature coupling described in the first aspect or any embodiment of the first aspect.

[0008] According to a fourth aspect, embodiments of the present invention provide a computer storage medium storing computer instructions that, when executed by a processor, implement the steps of the urban gridded building energy consumption correction method based on multi-scale spatiotemporal feature coupling as described in the first aspect or any embodiment of the first aspect.

[0009] This invention provides a method for correcting energy consumption of urban gridded buildings based on spatiotemporal feature coupling. It uses urban-specific environmental parameters such as the texture features of the first target building, the surface heat intensity of the surrounding space, and the prevailing wind vector as influencing elements to capture the impact of heat island effect and wind corridor shielding on building energy consumption. Furthermore, by quantifying the coupling effect between the prevailing wind vector and the spatial relationship of neighboring buildings, it uses neighborhood interaction features as key inputs for energy consumption correction, effectively characterizing spatial interaction effects such as upwind building thermal advection and neighborhood shielding. Based on the nonlinear mapping capability of the target neural network model for texture features, neighborhood interaction features, and temporal thermal inertia features, the generated environmental heat penalty index can dynamically adapt to the spatiotemporal differences of different urban grids. Combined with design energy consumption based on suburban climate, it improves the accuracy of urban gridded building energy consumption calculation.

[0010] Other advantages, objectives, and features of the invention will be set forth in the following description and will be apparent to those skilled in the art in some respects, or may be learned by practice of the invention. The objectives and other advantages of the invention can be realized and obtained through the following description. Attached Figure Description

[0011] To make the objectives, technical solutions, and beneficial effects of this invention clearer, the following figures are provided for illustration: Figure 1 This is a flowchart illustrating a specific example of the urban gridded building energy consumption correction method based on spatiotemporal feature coupling in this invention. Figure 2 This is a schematic diagram of the neighborhood space in this invention; Figure 3 This is a flowchart illustrating the overall scheme of the urban gridded building energy consumption correction method based on spatiotemporal feature coupling in this invention. Figure 4 This is a schematic block diagram of a specific example of an electronic device in an embodiment of the present invention. Detailed Implementation

[0012] The technical solution of the present invention will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0013] In the description of this invention, it should be noted that, unless otherwise explicitly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; they can also refer to the internal connection of two components; and they can refer to a wireless connection or a wired connection. Those skilled in the art can understand the specific meaning of the above terms in this invention based on the specific circumstances.

[0014] Furthermore, the technical features involved in the different embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.

[0015] This invention provides a method for correcting urban gridded building energy consumption based on spatiotemporal feature coupling, such as... Figure 1 As shown, it includes: S101, acquire the textural features of the first target building, the surface heat intensity of the surrounding space, the prevailing wind vector, the spatial relationship between the first target building and the buildings in its surrounding space, the design energy consumption of the first target building, and the temporal thermal inertia characteristics of the first target area. The first target area includes the first target building and the surrounding space. The design energy consumption of the first target building is determined based on the suburban climate environment. S102, Based on the dominant wind vector, the spatial relationship between the first target building and the buildings in its neighborhood, and the surface heat intensity of the neighborhood, determine the neighborhood interaction characteristics of the first target building. S103, input the texture features, neighborhood interaction features, and temporal thermal inertia features of the first target building into the pre-trained target neural network model to obtain the environmental thermal penalty index; S104, Based on the design energy consumption and environmental heat penalty index of the first target building, determine the corrected energy consumption of the first target building.

[0016] For example, this embodiment mainly describes the online usage process: The primary target building represents any real-world building, whose textural features may include building height, density, and impermeable surface ratio. Neighborhood space can be defined as the division of a city into... The spatial grid is formed with the first target building as the center of the grid. The area, excluding the first target building, in The Moore neighborhood is the neighborhood space for interactive computation. It can be 3 or 5, such as Figure 2 As shown, with Taking 3 as an example, the central grid represents the first target building, and the remaining spatial range represents the domain space. The surface heat intensity of the neighboring space can be obtained using deployed surface heat flux sensors; the prevailing wind vector can be used to determine the real-time wind direction and speed through sensors; the spatial relationship between the first target building and buildings in its neighboring space includes the neighbor-to-center vector and the distance, where the neighbor-to-center vector is a spatial position vector pointing from the neighboring grid to the central grid, used to determine the orientation of the neighboring grid relative to the central grid; the design energy consumption of the first target building is based on suburban standard meteorological data, and after locking the building parameters according to national standards, the baseline energy consumption value is simulated using standardized software; the temporal thermal inertia characteristics of the first target area can be obtained by acquiring meteorological parameters (such as air temperature) or surface temperature within a preset time period (such as 24 hours) in the first target area, and calculating their statistical values ​​(such as average or maximum values) to characterize the thermal hysteresis effect.

[0017] It should be noted that the texture features of the first target building, the surface heat intensity of the surrounding space, the prevailing wind vector, the spatial relationship between the first target building and the buildings in its surrounding space, the design energy consumption of the first target building, and the temporal thermal inertia characteristics of the first target area are all pre-collected and stored in the database. When using them, the database data can be directly called.

[0018] After obtaining the above data, it is necessary to determine the neighborhood interaction characteristics of the first target building. The process can be described as follows: Based on the spatial relationship between the first target building and the buildings in its neighborhood space, determine the neighbor pointing center vector and the distance between the first target building and the buildings in its neighborhood space; based on the prevailing wind vector and the neighbor pointing center vector, determine the angle between the prevailing wind vector and the neighbor pointing center vector; based on the distance between the first target building and the buildings in its neighborhood space, and the angle between the prevailing wind vector and the neighbor pointing center vector, determine the influence weight of the buildings in the neighborhood space on the first target building; based on the influence weight of the buildings in the neighborhood space on the first target building and the surface heat intensity of the neighborhood space, determine the neighborhood interaction characteristics of the first target building.

[0019] Specifically, such as Figure 2 As shown, for any grid n in the neighborhood, the dominant wind vector ,in, as well as Representing the east-west and north-south wind speed components respectively, calculate the wind direction angle. Then, based on this, the position vector P (neighbor-to-center vector) of the neighboring grid n pointing to the center grid is calculated. Next, the prevailing wind vector is calculated. The angle between the vector pointing to the center of the neighboring vector cosine value .

[0020] Next, the weight of its influence on the central grid (i.e., the first target building) is calculated. The calculation formula is as follows: (1) in, The angle between the dominant wind vector W and the neighboring vector pointing to the center. Based on the distance between the first target building and buildings in its neighborhood, Characterize the activation function to ensure that only In other words, the upwind grid generates thermal effects, while the downwind grid weight is automatically set to 0.

[0021] Assuming the prevailing wind vector is northerly, the above formula will automatically assign higher weights to neighboring grids north of the central grid, while the weights of neighboring grids south of the central grid will be zero.

[0022] After obtaining the influence weights of each grid cell on the central grid cell, the neighborhood interaction features are calculated using feature convolution. The specific formula is as follows: (2) in, Let be the surface heat intensity corresponding to the nth grid in the neighborhood space.

[0023] Next, the textural features, neighborhood interaction features, and temporal thermal inertia features of the first target building are used as a three-dimensional input feature set and input into a pre-trained target neural network model. The model uses a fusion structure of fully connected layers and convolutional layers to perform weighted fusion and nonlinear fitting of spatial interaction features and temporal thermal inertia features, and outputs a unique scalar, namely the environmental thermal penalty index, to achieve an end-to-end mapping from urban microclimate spatiotemporal features to energy consumption correction coefficients. The pre-trained target neural network model can be a regression model, such as XGBoost. It should be noted that in this embodiment, the regression model is not limited to single-output or multi-output. When the input is the textural features, neighborhood interaction features, and temporal thermal inertia features of all buildings, the regression model can output an environmental thermal penalty index vector, which contains environmental thermal penalty index vectors of multiple building types.

[0024] It should be further explained that the target neural network adopts a 4-layer fully connected regression network: the number of neurons in the input layer is consistent with the dimension of the input features; hidden layer 1 has 64 neurons with the ReLU activation function; hidden layer 2 has 32 neurons with the ReLU activation function; the output layer has 1 neuron (regressing the output environment hot penalty index); the loss function adopts mean squared error (MSE), and the optimizer is Adam, realizing a non-linear mapping from features to the hot penalty index.

[0025] When the input consists of the textural features, neighborhood interaction features, and temporal thermal inertia features of a target building, the regression model can output a single environmental thermal penalty index. The ratio of the energy consumption of a virtual standard building probe model constructed to characterize a certain building type in urban areas to the energy consumption of a suburban baseline, i.e.: (3) in, This represents the energy consumption of a virtual standard building probe model in a city. This represents the baseline energy consumption of a virtual standard building probe model in a suburban area.

[0026] The virtual standard building probe model is a building model constructed when acquiring samples for the target neural network model. The target neural network model determines the environmental heat penalty index based on the pre-constructed virtual standard building probe model, using this index as the training label. It also incorporates the collected textural features of real buildings, neighborhood interaction features calculated based on real building relationships, and temporal thermal inertia features as features. The training label and features form the training samples, enabling the target neural network to learn the nonlinear mapping relationship between the building's textural features, neighborhood interaction features, and temporal thermal inertia features and the environmental heat penalty index.

[0027] Based on the design energy consumption and environmental heat penalty index of the first target building, the corrected energy consumption of the first target building is determined according to the following formula: (4) in, Corrected energy consumption for the primary target building. Design energy consumption for the primary target building. The environmental heat penalty index for the primary target building. The shape factor is determined based on the similarity between the first target building and the corresponding type of virtual standard building probe model, and ranges from 0.95 to 1.05.

[0028] This invention provides a method for correcting energy consumption of urban gridded buildings based on spatiotemporal feature coupling. It uses urban-specific environmental parameters such as the texture features of the first target building, the surface heat intensity of the surrounding space, and the prevailing wind vector as influencing elements to capture the impact of heat island effect and wind corridor shielding on building energy consumption. Furthermore, by quantifying the coupling effect between the prevailing wind vector and the spatial relationship of neighboring buildings, it uses neighborhood interaction features as key inputs for energy consumption correction, effectively characterizing spatial interaction effects such as upwind building thermal advection and neighborhood shielding. Based on the nonlinear mapping capability of the target neural network model for texture features, neighborhood interaction features, and temporal thermal inertia features, the generated environmental heat penalty index can dynamically adapt to the spatiotemporal differences of different urban grids. Combined with design energy consumption based on suburban climate, it improves the accuracy of urban gridded building energy consumption calculation.

[0029] As an optional implementation, the process of constructing training samples for the target neural network model includes: Based on the building types at various locations throughout the city where the second target building is located, the corresponding set of virtual standard building probe models is called from the virtual probe library and deployed to the corresponding grid space center pre-divided according to the city area to obtain a virtual building model cluster; Urban climate environmental parameters and suburban climate environmental parameters are input into the virtual building model cluster respectively to obtain the energy consumption of the second target building under urban climate environmental parameters and the energy consumption of the second target building under suburban climate environmental parameters. The climate environmental parameters include the prevailing wind vector. Based on the energy consumption of the second target building under urban climate environmental parameters and the energy consumption of the second target building under suburban climate environmental parameters, the environmental heat penalty index of the second target building is determined. The second target area includes the second target building and its surrounding space. The study collected the textural features of the second target building, the surface heat intensity of the surrounding space, the spatial relationship between the second target building and the buildings in its surrounding space, and the temporal thermal inertia characteristics of the second target area. Based on the prevailing wind vector in urban climate environmental parameters, the spatial relationship between the second target building and buildings in its neighborhood, and the surface heat intensity of the neighborhood, the neighborhood interaction characteristics of the second target building are determined. Training samples for the target neural network model are constructed based on the textural features, neighborhood interaction features, temporal thermal inertia features of the second target building, and environmental thermal penalty index of the second target building.

[0030] For example, this embodiment mainly describes the offline training process. Offline training relies on a pre-built virtual probe library, which contains multiple types of virtual standard building probe models. The physical properties of each type of probe remain constant in the global grid. Specifically, it includes residential probe models and commercial probe models. Specifically, for the residential probe model, its geometry can be set as a standard cross-shaped or strip-shaped residential model with a floor height of 3m and a total height of 30m. The building envelope emphasizes nighttime heat dissipation lag, and the U-value of the exterior wall is locked at 1.0W / ( The wall-to-window ratio (WWR) can be locked at 0.3. The operating condition is set to intermittent air conditioning mode, for example, turning on between 18:00 and 8:00, with low internal load density, such as 5W / personnel. For commercial office probes, the geometry can be set as a standard large-depth rectangular model, which can be stacked using shoebox models. The building envelope emphasizes daytime radiative heat gain, with the U-value of the exterior walls locked at 0.5 W / ( The window-to-wall ratio (WWR) can be locked at 0.6, and the operating condition can be set to continuous daytime operation mode, for example, turning on from 8:00 to 19:00, with high internal load density, such as equipment and lighting density greater than 30W / m². .

[0031] For other building types, such as schools, hospitals, and industrial buildings, a similar paradigm expansion is adopted: corresponding standard probe models are constructed according to their function, operating period, load density, and building envelope characteristics, maintaining consistency between the model structure and the training process to achieve full type coverage. The virtual standard building probe model selects typical shape coefficients, orientations, and average window-to-wall ratios based on industry building energy-saving design standards. Through comparison and verification with a large number of actual building energy consumption samples, errors are controlled, and it can represent the average level of similar buildings.

[0032] In the embodiment, the entire city where the second target building is located is divided into... A spatial grid, where one grid can be 500m. The grid size is not limited to 500m in this embodiment and can be divided according to actual needs. For each building corresponding to a spatial grid, a pre-stored virtual standard building probe model is called from the virtual probe library, thus obtaining a cluster of virtual building models containing location and virtual standard building probe model type. It should be noted that the second target building is the building used when training the target neural network model, and is distinguished from the first target building only to differentiate different usage scenarios.

[0033] To obtain the environmental heat penalty index, urban and suburban climate environmental parameters are input into the virtual building model cluster to obtain the energy consumption of the virtual standard building probe model under urban climate conditions and the baseline energy consumption under suburban climate conditions. That is, the energy consumption of the second target building under urban climate environmental parameters and the energy consumption of the second target building under suburban climate environmental parameters. Among them, urban and suburban climate environmental parameters are generated by WRF-EnergyPlus coupled simulation. In order to construct a diverse training sample, it can cover typical weather types such as high temperature, sunny and hot weather, and the influence of the outer sinking airflow of typhoon. Specifically, the virtual building model cluster uses the EnergyPlus simulation engine to perform energy consumption calculation. The urban environmental parameters (temperature and humidity, solar radiation, wind speed and direction) and suburban environmental parameters are input into the virtual standard building probe model of the corresponding grid, and EnergyPlus performs hourly dynamic simulation based on the building envelope parameters, internal load, and air conditioning operation strategy, and outputs hourly cooling / heating load and cumulative energy consumption, thereby obtaining the energy consumption under urban climate and the suburban baseline energy consumption. Then, the environmental heat penalty index is calculated using formula (3).

[0034] The process of collecting the textural features of the second target building, the surface heat intensity of the surrounding space, the spatial relationship between the second target building and the buildings in its surrounding space, the temporal thermal inertia characteristics of the second target area, and the determination of the neighborhood interaction characteristics of the second target building is as described in the corresponding part of the above embodiments, and will not be repeated here.

[0035] Finally, based on the textural features, neighborhood interaction features, temporal thermal inertia features of the second target building, and the environmental thermal penalty index of the second target building, training samples for the target neural network model are constructed. The textural features, neighborhood interaction features, and temporal thermal inertia features of the second target building are used as sample features, and the environmental thermal penalty index of the second target building is used as sample labels. The above steps are repeated to obtain multiple training samples based on multiple types of second target buildings, so that the target neural network learns the nonlinear mapping relationship between the textural features, neighborhood interaction features, and temporal thermal inertia features of the building and the environmental thermal penalty index.

[0036] This invention provides a method for correcting urban gridded building energy consumption based on spatiotemporal feature coupling. It constructs a categorized probe library, which solves the drawbacks of the traditional single model and the one-size-fits-all approach. Furthermore, by constructing neighborhood interaction features, the target neural network is enabled to perceive upwind heat flow, effectively solving the problem of heat advection prediction under complex urban morphology and significantly improving the accuracy of urban building energy consumption calculation.

[0037] Combining the above online and offline processes, the overall scheme flow of the urban grid-based building energy consumption correction method based on spatiotemporal feature coupling is as follows: Figure 3 As shown, it includes: Step S1 involves constructing a city grid and virtual standard building probes. This step includes city area division and the construction of virtual standard building probes. For details, please refer to the corresponding sections of the above method embodiments, which will not be repeated here.

[0038] Step S2 involves extracting multi-scale meteorological-texture coupling features. This step includes extracting architectural texture features, neighborhood interaction features, and temporal thermal inertia features. For details, please refer to the corresponding sections of the above method embodiments; further elaboration is unnecessary.

[0039] Step S3 involves training the environmental thermal penalty residual model. This step includes coupled simulation using WRF-EnergyPlus, training a nonlinear regression model, and calculating the grid environment thermal penalty index. For details, please refer to the corresponding sections of the above method implementation examples, which will not be repeated here.

[0040] Step S4, Real Building Correction Based on Coefficient Mapping. This step includes the process of correcting the energy consumption of the actual building foundation design and the coefficients. For details, please refer to the corresponding sections of the above method embodiments, which will not be repeated here.

[0041] This invention provides an urban gridded building energy consumption correction system based on spatiotemporal feature coupling, comprising: The data acquisition module is used to acquire the textural features of the first target building, the surface heat intensity of the surrounding space, the prevailing wind vector, the spatial relationship between the first target building and the buildings in its surrounding space, the design energy consumption of the first target building, and the temporal thermal inertia characteristics of the first target area. The first target area includes the first target building and the surrounding space. The design energy consumption of the first target building is determined based on the suburban climate environment. The neighborhood interaction feature determination module is used to determine the neighborhood interaction features of the first target building based on the prevailing wind vector, the spatial relationship between the first target building and the buildings in its neighborhood space, and the surface heat intensity of the neighborhood space. The environmental heat penalty index determination module is used to input the texture features, neighborhood interaction features, and temporal thermal inertia features of the first target building into the pre-trained target neural network model to obtain the environmental heat penalty index. The energy consumption correction module is used to determine the corrected energy consumption of the first target building based on its design energy consumption and environmental heat penalty index.

[0042] This application also provides an electronic device, such as... Figure 4 As shown, processor 501 and memory 502 are connected via a bus or other means.

[0043] Processor 501 can be a central processing unit (CPU). Processor 501 can also be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, or combinations of the above types of chips.

[0044] The memory 502, as a non-transitory computer-readable storage medium, can be used to store non-transitory software programs, non-transitory computer-executable programs, and modules, such as the program instructions / modules corresponding to the urban gridded building energy consumption correction method based on multi-scale spatiotemporal feature coupling in this embodiment of the invention. The processor executes various functional applications and data processing by running the non-transitory software programs, instructions, and modules stored in the memory.

[0045] Memory 502 may include a program storage area and a data storage area. The program storage area may store the operating system and applications required for at least one function; the data storage area may store data created by the processor, etc. Furthermore, the memory may include high-speed random access memory and non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. In some embodiments, memory 502 may optionally include memory remotely located relative to the processor, which can be connected to the processor via a network. Examples of such networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof.

[0046] The one or more modules are stored in the memory 502, and when executed by the processor 501, they perform actions such as... Figure 1 The illustrated embodiment presents a method for correcting urban gridded building energy consumption based on multi-scale spatiotemporal feature coupling.

[0047] For specific details regarding the aforementioned electronic devices, please refer to the relevant documentation. Figure 1 The relevant descriptions and effects in the illustrated embodiments are for understanding purposes only and will not be repeated here.

[0048] This embodiment also provides a computer storage medium storing computer-executable instructions that can execute the urban gridded building energy consumption correction method based on multi-scale spatiotemporal feature coupling in any of the above method embodiments. The storage medium can be a magnetic disk, optical disk, read-only memory (ROM), random access memory (RAM), flash memory, hard disk drive (HDD), or solid-state drive (SSD), etc.; the storage medium may also include combinations of the above types of memory.

[0049] Finally, it should be noted that the above preferred embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail through the above preferred embodiments, those skilled in the art should understand that various changes can be made to it in form and detail without departing from the scope defined by the claims of the present invention.

Claims

1. A method for correcting urban gridded building energy consumption based on spatiotemporal feature coupling, characterized in that, include: The textural features of the first target building, the surface heat intensity of the surrounding space, the prevailing wind vector, the spatial relationship between the first target building and the buildings in its surrounding space, the design energy consumption of the first target building, and the temporal thermal inertia characteristics of the first target area are obtained. The first target area includes the first target building and the surrounding space. The design energy consumption of the first target building is determined based on the suburban climate environment. Based on the prevailing wind vector, the spatial relationship between the first target building and buildings in its neighborhood, and the surface heat intensity of the neighborhood, the neighborhood interaction characteristics of the first target building are determined. The textural features, neighborhood interaction features, and temporal thermal inertia features of the first target building are input into a pre-trained target neural network model to obtain the environmental thermal penalty index. Based on the design energy consumption and environmental heat penalty index of the first target building, the corrected energy consumption of the first target building is determined.

2. The urban gridded building energy consumption correction method based on spatiotemporal feature coupling according to claim 1, characterized in that, The process of constructing training samples for the target neural network model includes: Based on the building types at various locations throughout the city where the second target building is located, the corresponding set of virtual standard building probe models is called from the virtual probe library and deployed to the corresponding grid space center pre-divided according to the city area to obtain a virtual building model cluster; Urban climate environmental parameters and suburban climate environmental parameters are input into the virtual building model cluster respectively to obtain the energy consumption of the second target building under urban climate environmental parameters and the energy consumption of the second target building under suburban climate environmental parameters. The climate environmental parameters include the prevailing wind vector. Based on the energy consumption of the second target building under urban climate environmental parameters and the energy consumption of the second target building under suburban climate environmental parameters, the environmental heat penalty index of the second target building is determined. The second target area includes the second target building and its surrounding space. The study collected the textural features of the second target building, the surface heat intensity of the surrounding space, the spatial relationship between the second target building and the buildings in its surrounding space, and the temporal thermal inertia characteristics of the second target area. Based on the prevailing wind vector in urban climate environmental parameters, the spatial relationship between the second target building and buildings in its neighborhood, and the surface heat intensity of the neighborhood, the neighborhood interaction characteristics of the second target building are determined. Training samples for the target neural network model are constructed based on the textural features, neighborhood interaction features, temporal thermal inertia features of the second target building, and environmental thermal penalty index of the second target building.

3. The urban gridded building energy consumption correction method based on spatiotemporal feature coupling according to claim 1, characterized in that, Based on the dominant wind vector, the spatial relationships between the first target building and its neighboring buildings, and the surface heat intensity of the neighboring space, the neighborhood interaction characteristics of the first target building are determined, including: Based on the spatial relationship between the first target building and the buildings in its neighborhood, determine the neighbor pointing center vector and the distance between the first target building and the buildings in its neighborhood; Based on the prevailing wind vector and the neighboring center vector, determine the angle between the prevailing wind vector and the neighboring center vector; The influence weights of buildings in the neighborhood space on the first target building are determined based on the distance between the first target building and buildings in its neighborhood space, the angle between the prevailing wind vector and the neighboring center vector. Based on the influence weights of buildings in the neighborhood space on the first target building and the surface heat intensity of the neighborhood space, the neighborhood interaction characteristics of the first target building are determined.

4. The urban gridded building energy consumption correction method based on spatiotemporal feature coupling according to claim 2, characterized in that, Based on the design energy consumption and environmental heat penalty index of the first target building, the corrected energy consumption of the first target building is determined, including: ； in, Corrected energy consumption for the primary target building. Design energy consumption for the primary target building. The environmental heat penalty index for the primary target building. The shape factor is determined based on the similarity between the first target building and the corresponding type of virtual standard building probe model.

5. The urban gridded building energy consumption correction method based on spatiotemporal feature coupling according to claim 3, characterized in that, Based on the influence weights of buildings in the neighborhood space on the first target building and the surface heat intensity of the neighborhood space, the neighborhood interaction characteristics of the first target building are determined, including: ； in, The influence weight of buildings in the neighborhood space on the first target building. , The angle between the prevailing wind vector and the neighboring vector pointing to the center. Based on the distance between the first target building and buildings in its neighborhood, The surface heat intensity of the surrounding space.

6. The urban gridded building energy consumption correction method based on spatiotemporal feature coupling according to claim 2, characterized in that, The virtual standard building probe model includes residential probe models and commercial / office probe models.

7. A city grid-based building energy consumption correction system based on spatiotemporal feature coupling, characterized in that, include: The data acquisition module is used to acquire the textural features of the first target building, the surface heat intensity of the surrounding space, the prevailing wind vector, the spatial relationship between the first target building and the buildings in its surrounding space, the design energy consumption of the first target building, and the temporal thermal inertia characteristics of the first target area. The first target area includes the first target building and the surrounding space. The design energy consumption of the first target building is determined based on the suburban climate environment. The neighborhood interaction feature determination module is used to determine the neighborhood interaction features of the first target building based on the prevailing wind vector, the spatial relationship between the first target building and the buildings in its neighborhood space, and the surface heat intensity of the neighborhood space. The environmental heat penalty index determination module is used to input the texture features, neighborhood interaction features, and temporal thermal inertia features of the first target building into the pre-trained target neural network model to obtain the environmental heat penalty index. The energy consumption correction module is used to determine the corrected energy consumption of the first target building based on its design energy consumption and environmental heat penalty index.

8. An electronic device, the device comprising: A memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that the processor performs the steps of the urban gridded building energy consumption correction method based on spatiotemporal feature coupling as described in any one of claims 1-6.

9. A computer storage medium storing computer instructions thereon, characterized in that, When executed by the processor, this instruction implements the steps of the urban gridded building energy consumption correction method based on spatiotemporal feature coupling as described in any one of claims 1-6.

Images