A step-by-step prediction method, system, terminal and medium for indoor space thermal environment

Abstract

The present invention provides a method, system, terminal and medium for predicting the indoor space thermal environment step by step, which specifically relates to the field of building energy-saving technology. The scheme includes: obtaining a room topology relationship diagram based on the target apartment diagram, and determining a space layout information matrix based on the room topology relationship diagram; obtaining wind pressure values ​​at each inner door plane based on the room topology relationship diagram and preset outdoor door and window wind pressure values; using a neural network to fuse the space layout information matrix and the wind pressure values ​​at all inner door planes to obtain indoor temperature and wind speed distribution data; based on the indoor temperature and wind speed distribution data and the target apartment diagram, generating indoor space thermal environment prediction results. The scheme simplifies the complex overall indoor thermal comfort prediction problem into the indoor thermal comfort prediction problem of a single room by means of step-by-step prediction, reduces the computational complexity, significantly improves the prediction efficiency while ensuring the prediction accuracy, and improves the generalization ability of the model.

Description

Technical Field

[0001] The present invention relates to the technical field of building energy conservation, and in particular to a method, system, terminal and medium for predicting indoor space thermal environment step by step. Background Art

[0002] Thermal comfort prediction plays a vital role in creating a high-quality indoor environment and achieving efficient energy management. It not only ensures the comfort of occupants, but also achieves the dual benefits of energy saving and consumption reduction.

[0003] Traditional thermal comfort methods use statistical models based on experimental data to predict thermal comfort, but sometimes there are significant prediction errors. Some scholars use CFD computational fluid dynamics to simulate indoor thermal environments to achieve thermal comfort prediction, but this method often has the disadvantage of slow simulation speed. Unlike traditional thermal comfort models, data-driven models based on machine learning are more accurate and faster.

[0004] At present, thermal comfort prediction methods mainly use computational fluid dynamics (CFD) or machine learning (such as multi-layer perceptron, support vector machine, random forest and Gaussian process regression, etc.) to simulate and predict the indoor thermal environment. Although the method of predicting thermal comfort using machine learning has high prediction accuracy, they all use an overall prediction method, which has the problems of high computational complexity and insufficient generalization ability of the model. Summary of the invention

[0005] In view of the above-mentioned deficiencies in the prior art, the object of the present invention is to provide a method, system, terminal and medium for step-by-step prediction of the thermal environment of an indoor space, aiming to solve the problems of high computational complexity and insufficient generalization ability of the model in the prior art.

[0006] In order to achieve the above object, the first aspect of the present invention provides a step-by-step prediction method for indoor space thermal environment, comprising:

[0007] Get the target floor plan;

[0008] Based on the target floor plan, a room topology relationship diagram is obtained, and based on the room topology relationship diagram, a space layout information matrix is ​​determined;

[0009] Based on the room topology diagram and the preset outdoor door and window wind pressure values, obtain the wind pressure values ​​at each inner door plane;

[0010] The spatial layout information matrix and the wind pressure values ​​at all the inner door planes are integrated using a preset neural network to obtain indoor temperature and wind speed distribution data;

[0011] Based on the indoor temperature and wind speed distribution data and the target floor plan, an indoor space thermal environment prediction result is generated.

[0012] Optionally, obtaining a room topology diagram based on the target floor plan includes:

[0013] Based on the door and window relationship diagram in the target floor plan, obtaining a spatial connection relationship;

[0014] The spatial connection relationship is converted into a spatial relationship list, and based on the spatial relationship list, a room topology relationship diagram is obtained.

[0015] Optionally, determining a space layout information matrix based on the room topology relationship diagram includes:

[0016] Constructing a room model by using the spatial connection relationship and the indoor component relationship in the room topological relationship diagram;

[0017] Based on the room model, a space layout information matrix is ​​determined.

[0018] Optionally, obtaining the wind pressure value at each inner door plane based on the room topology relationship diagram and preset outdoor door and window wind pressure values ​​includes:

[0019] Based on the room topology diagram and the preset outdoor door and window wind pressure values, an indoor airflow network model is constructed;

[0020] Based on the indoor airflow network model, calculating the pressure drop between each room;

[0021] The ventilation volume of the corresponding room is calculated based on the pressure drop, and the wind pressure value at each inner door plane is obtained based on the ventilation volume.

[0022] Optionally, the using of a preset neural network to fuse the spatial layout information matrix and the wind pressure values ​​at all the inner door planes to obtain indoor temperature and wind speed distribution data includes:

[0023] Using a first neural network, the spatial layout information matrix and the wind pressure values ​​at all the inner door planes are merged to obtain a spatial pressure feature matrix;

[0024] The spatial pressure characteristic matrix is ​​processed by a second neural network to obtain indoor temperature and wind speed distribution data.

[0025] Optionally, the spatial layout information matrix and the wind pressure values ​​at all the inner door planes are fused using a preset neural network prediction model to obtain a spatial pressure feature matrix, including:

[0026] Using the wind pressure values ​​at all the inner door planes, assigning values ​​to corresponding positions of the room topology map corresponding to the space layout information matrix to obtain an initial space pressure information matrix;

[0027] The spatial layout information matrix and the initial spatial pressure information matrix are fused and feature extracted using a preset neural network prediction model to obtain a spatial pressure feature matrix.

[0028] Optionally, generating an indoor space thermal environment prediction result based on the indoor temperature and wind speed distribution data and the target floor plan includes:

[0029] Based on the indoor temperature and wind speed distribution data, obtaining the human body thermal response index value at each position in each room;

[0030] Based on the human body thermal reaction index values ​​at each of the positions and the target floor plan, a prediction result of the indoor space thermal environment is generated.

[0031] A second aspect of the present invention provides a step-by-step prediction system for indoor space thermal environment, the system comprising:

[0032] An information acquisition module is used to obtain a target floor plan;

[0033] A room topology relationship module, used to obtain a room topology relationship diagram based on the target floor plan, and determine a space layout information matrix based on the room topology relationship diagram;

[0034] An airflow network module, used to obtain the wind pressure value at each inner door plane based on the room topology diagram and the preset outdoor door and window wind pressure value;

[0035] A thermal environment parameter prediction module, used to fuse the spatial layout information matrix and the wind pressure values ​​at all the inner door planes using a preset neural network to obtain indoor temperature and wind speed distribution data;

[0036] The indoor thermal comfort prediction module is used to generate indoor space thermal environment prediction results based on the indoor temperature and wind speed distribution data and the target floor plan.

[0037] A third aspect of the present invention provides an intelligent terminal, which includes a memory, a processor, and a step-by-step prediction program for the indoor space thermal environment stored in the memory and executable on the processor, wherein the step-by-step prediction program for the indoor space thermal environment, when executed by the processor, implements any one of the steps of the above-mentioned step-by-step prediction method for the indoor space thermal environment.

[0038] A fourth aspect of the present invention provides a computer-readable storage medium, on which is stored a step-by-step prediction program for an indoor space thermal environment. When the step-by-step prediction program for an indoor space thermal environment is executed by a processor, the step-by-step prediction program for an indoor space thermal environment implements any one of the steps of the above-mentioned step-by-step prediction method for an indoor space thermal environment.

[0039] Compared with the prior art, the beneficial effects of this solution are as follows:

[0040] The present invention first divides the overall target floor plan into individual rooms to obtain a room topology relationship diagram, establishes the topological relationship between key components such as doors, windows, and walls, and then uses the room topology relationship diagram and the outdoor door and window wind pressure values ​​to calculate the wind pressure values ​​at each inner door plane of each room, and then uses a neural network to fuse the spatial layout information matrix corresponding to the room topology relationship diagram and the wind pressure values ​​at all inner door planes, and calculates the indoor temperature and wind speed distribution point by point, and finally combines the indoor temperature and wind speed distribution data with the spatial layout of the target floor plan to obtain the indoor space thermal environment prediction result. It can be seen that the method of the present invention simplifies the complex overall indoor thermal comfort prediction problem into the indoor thermal comfort prediction problem of a single room by a step-by-step prediction method, effectively reduces the computational complexity, significantly improves the prediction efficiency while ensuring the prediction accuracy, and greatly improves the generalization ability of the model by evaluating a single room. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0042] Figure 1 It is a flow chart of the step-by-step prediction method of indoor space thermal environment of the present invention;

[0043] Figure 2 It is a schematic diagram of a step-by-step prediction framework of the indoor space thermal environment of the present invention;

[0044] Figure 3 A schematic diagram of a spatial topological relationship generation process of the present invention;

[0045] Figure 4 The flow chart of calculating the wind pressure value at the inner door plane of the present invention;

[0046] Figure 5 It is a thermal environment parameter prediction flow chart of the present invention;

[0047] Figure 6 It is the indoor thermal comfort prediction flow chart of the present invention;

[0048] Figure 7 It is a schematic diagram of the module of the indoor space thermal environment step-by-step prediction system of the present invention;

[0049] Figure 8 It is a schematic diagram of the structure of the intelligent terminal of the present invention. DETAILED DESCRIPTION

[0050] In the following description, specific details such as specific system structures, technologies, etc. are provided for the purpose of illustration rather than limitation, so as to provide a thorough understanding of the embodiments of the present invention. However, it should be clear to those skilled in the art that the present invention may be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to prevent unnecessary details from obstructing the description of the present invention.

[0051] It should be understood that when used in this specification and the appended claims, the term "comprising" indicates the presence of described features, integers, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or groups thereof.

[0052] It should also be understood that the terms used in the present specification are only for the purpose of describing specific embodiments and are not intended to limit the present invention. As used in the present specification and the appended claims, unless the context clearly indicates otherwise, the singular forms "a", "an" and "the" are intended to include plural forms.

[0053] It should be further understood that the term "and / or" used in the present description and the appended claims refers to any and all possible combinations of one or more of the associated listed items, and includes these combinations.

[0054] The following is a clear and complete description of the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0055] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0056] In view of the problems of high computational complexity and insufficient generalization ability of the model in the prior art, the present invention proposes a step-by-step prediction method for the thermal environment of indoor space, which mainly simplifies the problem of directly calculating the indoor thermal comfort of the entire target apartment type into the problem of calculating the indoor thermal comfort of each room in the target apartment type, which can greatly reduce the computational complexity and difficulty and effectively improve the generalization ability of the model.

[0057] The embodiment of the present invention provides a step-by-step prediction method for indoor space thermal environment, which is deployed on electronic devices such as computers and servers and applied to the scenario of indoor thermal environment prediction of complete sets of housing. It is aimed at dividing the complete sets of housing into multiple rooms according to the room structure, thereby simplifying the calculation of complex indoor thermal comfort problems. The implementation of the method of the present invention includes four stages, and the tasks of these four stages are completed in sequence by the spatial topology relationship module, the airflow network module, the thermal environment parameter prediction module and the indoor thermal comfort prediction module. Specifically, Figure 1 and Figure 2 As shown, the steps of the method in this embodiment include:

[0058] Step S100: Obtaining a target floor plan;

[0059] Specifically, obtain a plan or three-dimensional diagram of a complete apartment consisting of an indoor space with multiple rooms, that is, a target apartment diagram. In order to simplify the calculation, if it is a three-dimensional diagram, it can be converted into a plan diagram and then the same research and analysis as the plan diagram can be performed. Therefore, the target apartment diagram mentioned below is the target apartment floor plan mentioned below. Among them, the target apartment diagram is applicable to an indoor layout with multiple rooms, where there is a connection between the rooms (used to represent internal doors and windows), and at least one room has a connection with the outdoors (used to represent external doors and windows). However, there is no restriction on the number of rooms included and the application places. It can be an apartment of public building types such as foyers, conference rooms, offices, etc., or it can be a residential apartment with one bedroom and one living room, two bedrooms and one living room, three bedrooms and two living rooms, or multiple bedrooms. Figure 3 The floor plan shown is a two-bedroom, one-living room residential apartment, including a living room, bathroom, kitchen, toilet, master bedroom and second bedroom.

[0060] Step S200: obtaining a room topology relationship diagram based on the target floor plan, and determining a space layout information matrix based on the room topology relationship diagram;

[0061] Specifically, the target floor plan is input into the spatial topology relationship module, which extracts the spatial connection relationship from the target floor plan, that is, the position relationship between rooms and rooms and between rooms and the outdoors, and generates a spatial position relationship list; a room topology relationship graph is generated based on the spatial position relationship list, and a room model is constructed based on the room topology relationship graph, and the spatial layout information matrix is ​​obtained from this model. The room topology relationship graph is then transmitted to the airflow network module, and the spatial layout information matrix is ​​transmitted to the thermal environment parameter prediction module.

[0062] Step S300: based on the room topology diagram and preset outdoor door and window wind pressure values, obtain the wind pressure value at each inner door plane;

[0063] Specifically, the airflow network module calculates the wind pressure value at the inner door plane of each room in the target apartment diagram according to the received room topology diagram and the preset outdoor door and window wind pressure value, using the energy equation, the law of conservation of mass of incompressible fluids, and the Bernoulli equation, and generates an indoor wind pressure list according to the room topology diagram using all wind pressure values, and transmits the indoor wind pressure list to the thermal environment parameter prediction module. Among them, the wind pressure value at the inner door plane refers to the pressure loss and the pressure difference between the rooms when the airflow passes through the doors, windows, walls and other components of each room in the closed space of the target apartment diagram. If both sides of a component are in the internal space of the target apartment, then the wind pressure value of the plane where it is located represents the difference in wind pressure values ​​on both sides of the plane; if one side of a component of a room is in the internal space of the target apartment and the other is exposed to the external space, then the wind pressure value of the plane where it is located represents the outdoor pressure difference value, and its calibration method is based on the lowest value of all preset outdoor door and window wind pressure values, and the pressure difference value of the corresponding plane is the difference between its preset outdoor door and window wind pressure value and the lowest value. The preset outdoor door and window wind pressure value refers to the outdoor door and window wind pressure value of the target apartment type obtained by actual measurement or simulation.

[0064] Step S400: using a preset neural network to fuse the spatial layout information matrix and the wind pressure values ​​at all the inner door planes to obtain indoor temperature and wind speed distribution data;

[0065] Specifically, the thermal environment parameter prediction module obtains the position connection relationship between the doors, windows, walls and other components of each room in the target apartment through the spatial layout information matrix, obtains the wind pressure values ​​of the doors, windows, walls and other components in the target apartment at the inner door plane of the corresponding room through the indoor wind pressure list, and uses the wind pressure value to assign pressure to the grid consistent with the spatial layout information matrix. The specific method is to take the lowest pressure value at the inner door and window plane contained in the room as 0, and the pressure of the remaining planes is the difference between the actual pressure value of the plane and the lowest pressure value. The assignment is performed according to the grid, and the grid connected to the door and window plane is assigned the plane pressure value, and the non-contact grid is assigned -1, so as to obtain the pressure information matrix. Then, the spatial layout information matrix and the pressure information matrix are input into a preset neural network (i.e., the indoor thermal environment prediction neural network), and the temperature field and wind speed field at the location of each component in the target floor plan corresponding to the spatial layout information matrix are predicted point by point. The present invention does not limit the type of indoor thermal environment prediction neural network, and any type that can predict the internal temperature field and wind speed field of each room in the target floor plan based on the spatial layout information matrix and the pressure information matrix is ​​acceptable. Finally, the temperature and wind speed distribution data of the entire indoor space of the target floor plan are obtained and transmitted to the indoor thermal comfort prediction module.

[0066] Step S500: Generate indoor space thermal environment prediction results based on the indoor temperature and wind speed distribution data and the target floor plan.

[0067] Specifically, the indoor thermal comfort prediction module inputs the temperature field and wind speed field corresponding to each position in the indoor space temperature and wind speed distribution data into the heat balance equation in sequence, calculates the human thermal response index value at the key positions of each room, and then corresponds the human thermal response index value at each position to the position in the target floor plan one by one, to generate the indoor space thermal environment prediction result, that is, the indoor space thermal comfort distribution result.

[0068] In this embodiment, the present invention can quickly calculate the wind pressure value at each inner door plane by generating a spatial topological relationship diagram and a spatial layout information matrix, and combining the outdoor door and window wind pressure value; then use a neural network to fuse the spatial layout information matrix and the wind pressure value at the inner door plane of the corresponding room to obtain the indoor temperature and wind speed distribution data of each room; finally, based on the indoor temperature and wind speed distribution data and the target apartment diagram, generate the indoor space thermal environment prediction result. Compared with directly predicting the thermal comfort distribution of the internal space of the target apartment as a whole, the present invention simplifies the complex indoor thermal comfort problem, that is, by splitting the entire target apartment into several rooms in units of rooms, and calculating the thermal comfort distribution results in each room, it can improve the prediction efficiency while ensuring the calculation accuracy, can greatly reduce the calculation complexity and difficulty, and effectively improve the generalization ability of the model, thereby helping to improve the efficiency of indoor thermal environment assessment in the early stage of building design.

[0069] In a preferred embodiment, the step S200 of obtaining a room topology diagram based on the target floor plan includes:

[0070] Step S210: obtaining a spatial connection relationship based on the door and window relationship diagram in the target floor plan;

[0071] Step S220: converting the spatial connection relationship into a spatial relationship list, and obtaining a room topology relationship diagram based on the spatial relationship list.

[0072] Specifically, the spatial relationship of the target floor plan is identified, that is, the positional relationship between each adjacent room in the target floor plan and the positional relationship between each room and the outdoors are identified, and space identification, wall identification, and door and window identification are performed based on the target floor plan. The spatial relationship represented by the identification result and the positional relationship between components such as walls, doors and windows are converted into a spatial position relationship list, as shown in Table 1, where 0 indicates no connection relationship and 1 indicates a connection relationship. The positional relationship between components such as walls, doors and windows is represented in detail by the spatial position relationship list.

[0073] Table 1:

[0074]

[0075] Then, a room topology diagram is generated based on the spatial position relationship list, including a spatial relationship diagram and a component relationship diagram, such as Figure 3 As shown in the figure. The component relationship diagram represents the positional relationship between rooms and doors, windows, and walls. It is a graph structure, with nodes representing rooms, doors, windows, or walls, and edges representing walls. The spatial relationship diagram represents the positional relationship between rooms and rooms and between rooms and the outdoors. It is also a graph structure, with nodes representing indoor or outdoor spaces, and edges representing door and window connections. The room topology diagram can intuitively show the spatial relationship between each room and the outdoors in the target apartment, as well as the connection relationship between components in each room.

[0076] In a preferred embodiment, the determining of the space layout information matrix based on the room topology relationship diagram in step S200 includes:

[0077] Step S230: constructing a room model by using the spatial connection relationship and the indoor component relationship in the room topology relationship diagram;

[0078] Step S240: Determine a space layout information matrix based on the room model.

[0079] Specifically, the room model is constructed by using the spatial connection relationship and the indoor component relationship in the room topology diagram, and the spatial layout information matrix is ​​obtained from the room model. For example, according to the outline and area of ​​the target floor plan, a grid network with a resolution that meets the application requirements is established, such as a resolution of 0.2×0.2 meters, to ensure that the outline of the target floor plan falls within the grid network. The center of each grid in the grid network is used as a grid point, and each grid is assigned a value according to the indoor space and wall position corresponding to the spatial layout information matrix. If the grid point is located indoors, it is assigned a value of 1, and if the grid point is located outdoors or on the wall, it is assigned a value of 0. In this way, a spatial layout information matrix is ​​constructed to represent the flow field range.

[0080] like Figure 4 As shown, in a preferred embodiment, the step S300 of obtaining the wind pressure value at each inner door plane based on the room topology diagram and the preset outdoor door and window wind pressure value includes:

[0081] Step S310: constructing an indoor airflow network model based on the room topology diagram and preset outdoor door and window wind pressure values;

[0082] Step S320: Calculating the pressure drop between rooms based on the indoor airflow network model;

[0083] Step S330: Calculate the ventilation volume of the corresponding room based on the pressure drop, and obtain the wind pressure value at each inner door plane based on the ventilation volume.

[0084] Specifically, based on the room topology diagram and the preset outdoor door and window wind pressure values, an indoor airflow network model is constructed. This model is used to represent how airflow flows between rooms through doors and windows. Since the airflow changes in buildings are mainly based on the conservation of mass, the pressure drop between regions is calculated according to the energy equation formula. , pressure drop is used to represent the pressure loss in each room when the airflow passes through obstacles (such as doors, windows, etc.). The magnitude of the pressure drop is related to the size, shape, degree of opening of the obstacle and the airflow speed. In fluid mechanics, the formula Used to describe the flow of traffic through an obstacle (such as a door, window, etc.) The pressure difference on both sides of the obstacle The relationship between is the flow coefficient, which represents a dimensionless constant whose magnitude depends on the geometry of the obstacle and the properties of the fluid; the exponent This embodiment sets a reference value for the ventilation volume in each room. The reference value refers to the indoor ventilation volume corresponding to a good indoor thermal comfort. Substitute into the formula Calculate the flow rate (i.e. ventilation volume) of each room , and compare the obtained flow value with the reference value (or the last calculated air flow value). If the deviation is greater than the deviation threshold (such as ), the initial value is replaced by the latest ventilation volume, and the ventilation volume is iteratively calculated until the deviation is less than or equal to the deviation threshold, and the pressure drop and ventilation volume corresponding to each room are obtained. Then, according to the law of conservation of mass of incompressible fluid, the ventilation volume Substitute into the formula ,in Indicates wind speed, Representing the area of ​​inner doors and windows, the wind speed at each inner door plane is obtained, and thus the corresponding wind pressure value is obtained.

[0085] Since the air flow changes in the building satisfy the principle of the law of conservation of mass of incompressible fluids, this embodiment uses the incompressible fluid mass conservation equation to utilize the room topology diagram of the target apartment diagram and the outdoor door and window wind pressure values ​​to accurately calculate the wind pressure values ​​at each inner door plane in the target apartment.

[0086] like Figure 5 As shown, in a preferred embodiment, the method of using a preset neural network in step S400 to fuse the spatial layout information matrix and the wind pressure values ​​at all the inner door planes to obtain indoor temperature and wind speed distribution data includes:

[0087] Step S410: using a first neural network to fuse the space layout information matrix and the indoor pressure value to obtain a space pressure feature matrix;

[0088] Specifically, since the spatial layout information matrix and the indoor door and window wind pressure list data have a large difference in dimension, this embodiment uses the indoor pressure value to assign the corresponding positions of the room topology map corresponding to the spatial layout information matrix to obtain the initial spatial pressure information matrix. Specifically, according to the grid network in the spatial layout information matrix, the grids at the door and window positions are assigned, and the assigned values ​​are the data in the indoor wind pressure list corresponding to the grids at the door and window positions. The grids at other non-door and window positions are assigned -1. The initial spatial pressure information matrix is ​​obtained through this assignment method, and the initial spatial pressure information matrix is ​​consistent in size with the spatial layout information matrix. The spatial layout information matrix and the initial spatial pressure information matrix are multimodally fused using a feature extraction model to obtain a spatial pressure feature matrix. Specifically, the initial spatial pressure information matrix and the spatial layout information matrix are input into a first convolutional neural network together, and the first convolutional neural network is used to extract features from the initial spatial pressure information matrix and the spatial layout information matrix using a convolutional layer, and then feature fusion is performed through a fully connected layer to obtain a spatial pressure feature matrix. The feature matrix can be used for subsequent prediction of the wind speed and temperature of the room, so that the feature extraction model can better identify the important features of the spatial pressure feature matrix. Among them, the feature fusion mechanism includes but is not limited to matrix splicing, dimensionality increase splicing, multi-channel feature extraction and post-fusion, orthogonal fusion, etc. All fusion mechanisms that can be implemented based on the feature extraction model of this embodiment are within the protection scope of the present invention.

[0089] Step S420: Using a second neural network to process the spatial pressure feature matrix to obtain indoor temperature and wind speed distribution data.

[0090] Specifically, the spatial pressure feature matrix is ​​input into the trained indoor temperature and wind speed prediction model to obtain the temperature value and wind speed field component of each point. The indoor temperature and wind speed prediction model is constructed based on the second convolutional neural network, and is trained by obtaining hundreds of CFD calculation cases generated based on actual room or ideal room models, as well as real or ideal wind speed components and temperatures to form a training data set. After traversing all points, the temperature and wind speed data of all points in the room are obtained, and all data are restored according to the coordinates of each position to obtain the distribution data of the temperature field and wind speed field of the entire target apartment.

[0091] It should be stated that the first convolutional neural network and the second convolutional neural network appearing in this embodiment represent neural network prediction models constructed by convolutional neural networks with the same or different network structures at different stages, namely, feature extraction models and indoor temperature and wind speed prediction models. The feature extraction model adopted in this embodiment is constructed based on a convolutional neural network, and its main purpose is to share the local receptive field and weights of the convolutional layer. As other preferred embodiments, other neural networks containing convolutional layers can be selected to construct feature extraction models according to application requirements such as calculation accuracy, calculation complexity, and calculation efficiency. For example, during the calculation process, the steady-state wind field calculation is converted into a non-steady-state wind field, and the LSTM-CNN model is used for calculation after adding the time dimension. The present invention does not impose specific restrictions on this.

[0092] Since the spatial layout information matrix and the indoor wind pressure list data have a large difference in dimension, this embodiment uses the indoor pressure value to assign the corresponding position of the room topology map corresponding to the spatial layout information matrix to obtain the initial spatial pressure information matrix. The assigned initial spatial pressure information matrix and the spatial layout information matrix have the same dimensional space. The neural network is then used to fuse the features of the two to obtain the spatial pressure feature matrix, which can effectively retain important feature information and improve calculation accuracy. Then, the indoor temperature and wind speed prediction model obtained based on neural network training can be used to quickly predict the indoor temperature and wind speed distribution data.

[0093] like Figure 6 As shown, in a preferred embodiment, the step S500 of generating indoor space thermal environment prediction results based on the indoor temperature and wind speed distribution data and the target floor plan includes:

[0094] Step S510: obtaining a human body thermal response index value at each position in each room based on the indoor temperature and wind speed distribution data;

[0095] Step S520: Generate indoor space thermal environment prediction results based on the human body thermal reaction index values ​​at each of the positions and the target floor plan.

[0096] Specifically, the temperature and wind speed data of the target apartment diagram are input into the heat balance equation point by point to obtain the PMV (Predicted Mean Vote) value of each point, which is used to characterize the thermal response of the human body. Each point represents the grid position corresponding to each set of temperature and wind speed data. The PMV value of each point is then matched one by one with the position of the corresponding grid of the target apartment diagram. Finally, the conditions of each room are integrated together as the output of the indoor thermal comfort distribution result of the target apartment.

[0097] The present invention uses a step-by-step prediction method to first divide the overall target floor plan into individual rooms to obtain a room topology relationship diagram, establish a topological relationship between key components such as doors, windows, and walls, and then use the room topology relationship diagram and the outdoor door and window wind pressure values ​​to construct an indoor airflow network model, establish a spatial connection relationship between each room, and calculate the wind speed of each room according to the law of conservation of mass of incompressible fluids, so as to obtain the wind pressure value at each inner door plane, so as to use the wind pressure value to indirectly reflect the gas flow (i.e., ventilation volume) in each room, and then use a neural network to fuse the spatial layout information matrix corresponding to the room topology relationship diagram and the wind pressure values ​​at all inner door planes, and calculate the indoor temperature and wind speed distribution point by point, and finally combine the indoor temperature and wind speed distribution data with the spatial layout of the target floor plan to obtain the indoor space thermal environment prediction result. It can be seen that the method of the present invention simplifies the complex overall indoor thermal comfort prediction problem into the indoor thermal comfort prediction problem of a single room, and with the help of the spatial topology relationship inside the room, the law of conservation of mass of incompressible fluids, and the neural network, the prediction efficiency can be significantly improved while ensuring the prediction accuracy, which is conducive to improving the efficiency of indoor thermal environment assessment in the early stage of architectural design.

[0098] like Figure 7 As shown, corresponding to the above-mentioned indoor space thermal environment step-by-step prediction method, an embodiment of the present invention further provides an indoor space thermal environment step-by-step prediction system, and the above-mentioned indoor space thermal environment step-by-step prediction system includes:

[0099] Information acquisition module 710, used to acquire a target floor plan;

[0100] The room topology relationship module 720 is used to obtain a room topology relationship diagram based on the target floor plan, and determine a space layout information matrix based on the room topology relationship diagram;

[0101] The airflow network module 730 is used to obtain the wind pressure value at each inner door plane based on the room topology relationship diagram and the preset outdoor door and window wind pressure value;

[0102] The thermal environment parameter prediction module 740 is used to fuse the spatial layout information matrix and the wind pressure values ​​at all the inner door planes using a preset neural network to obtain indoor temperature and wind speed distribution data;

[0103] The indoor thermal comfort prediction module 750 is used to generate indoor space thermal environment prediction results based on the indoor temperature and wind speed distribution data and the target floor plan.

[0104] Specifically, in this embodiment, the specific functions of the indoor space thermal environment step-by-step prediction system can also refer to the corresponding description in the indoor space thermal environment step-by-step prediction method, which will not be repeated here.

[0105] Based on the above embodiments, the present invention further provides an intelligent terminal, whose principle block diagram can be shown as follows: Figure 8 As shown. The above-mentioned intelligent terminal includes a processor, a memory, a network interface and a display screen connected through a system bus. Among them, the processor of the intelligent terminal is used to provide computing and control capabilities. The memory of the intelligent terminal includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a step-by-step prediction program for indoor space thermal environment. The internal memory provides an environment for the operation of the operating system in the non-volatile storage medium and the step-by-step prediction program based on the indoor space thermal environment. The network interface of the intelligent terminal is used to communicate with an external terminal through a network connection. When the step-by-step prediction program for indoor space thermal environment is executed by the processor, the steps of any one of the above-mentioned step-by-step prediction methods for indoor space thermal environment are implemented. The display screen of the intelligent terminal can be a liquid crystal display screen or an electronic ink display screen.

[0106] Those skilled in the art will understand that Figure 8 The principle block diagram shown in the figure is only a block diagram of a partial structure related to the scheme of the present invention, and does not constitute a limitation on the smart terminal to which the scheme of the present invention is applied. The specific smart terminal may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0107] In one embodiment, a smart terminal is provided, which includes a memory, a processor, and a step-by-step prediction program for indoor space thermal environment stored in the memory and executable on the processor. When the step-by-step prediction program for indoor space thermal environment is executed by the processor, the steps of any one of the step-by-step prediction methods for indoor space thermal environment provided in the embodiments of the present invention are implemented.

[0108] An embodiment of the present invention further provides a computer-readable storage medium, on which a step-by-step prediction program for an indoor space thermal environment is stored. When the step-by-step prediction program for an indoor space thermal environment is executed by a processor, the steps of any one of the step-by-step prediction methods for an indoor space thermal environment provided by an embodiment of the present invention are implemented.

[0109] It should be understood that the serial numbers of the steps in the above embodiments do not imply a sequence of execution, and the execution sequence of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0110] Those skilled in the art can clearly understand that, for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example for illustration. In practical applications, the above-mentioned function allocation can be completed by different functional units and modules as needed, that is, the internal structure of the above-mentioned device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiment can be integrated in a processing unit, or each unit can exist physically separately, or two or more units can be integrated in one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of software functional units. In addition, the specific names of the functional units and modules are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of the present invention. The specific working process of the units and modules in the above-mentioned system can refer to the corresponding process in the aforementioned method embodiment, which will not be repeated here.

[0111] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described or recorded in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0112] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0113] In the embodiments provided by the present invention, it should be understood that the disclosed apparatus / terminal equipment and method can be implemented in other ways. For example, the apparatus / terminal equipment embodiments described above are only illustrative, for example, the division of the above modules or units is only a logical function division, and in actual implementation, other division methods can be used, for example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed.

[0114] The embodiments described above are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, it should be understood by those skilled in the art that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features therein may be replaced by equivalents. However, these modifications or replacements do not deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should be included in the protection scope of the present invention.

Claims

1. A step-by-step prediction method for indoor space thermal environment, characterized in that: The following steps are involved: Get the target floor plan; Based on the target floor plan, a room topology relationship diagram is obtained, and based on the room topology relationship diagram, a space layout information matrix is ​​determined; Based on the room topology diagram and the preset outdoor door and window wind pressure values, obtain the wind pressure values ​​at each inner door plane; The spatial layout information matrix and the wind pressure values ​​at all the inner door planes are integrated using a preset neural network to obtain indoor temperature and wind speed distribution data; Based on the indoor temperature and wind speed distribution data and the target floor plan, an indoor space thermal environment prediction result is generated.

2. The indoor space thermal environment step-by-step prediction method according to claim 1 is characterized in that: The step of obtaining a room topology diagram based on the target floor plan includes: Based on the door and window relationship diagram in the target floor plan, obtaining a spatial connection relationship; The spatial connection relationship is converted into a spatial relationship list, and based on the spatial relationship list, a room topology relationship diagram is obtained.

3. The indoor space thermal environment step-by-step prediction method according to claim 1, characterized in that: The determining of the space layout information matrix based on the room topology relationship diagram includes: Constructing a room model by using the spatial connection relationship and the indoor component relationship in the room topological relationship diagram; Based on the room model, a space layout information matrix is ​​determined.

4. The indoor space thermal environment step-by-step prediction method according to claim 1, characterized in that: The step of obtaining the wind pressure value at each inner door plane based on the room topology diagram and the preset outdoor door and window wind pressure value comprises: Based on the room topology diagram and the preset outdoor door and window wind pressure values, an indoor airflow network model is constructed; Based on the indoor airflow network model, calculating the pressure drop between each room; The ventilation volume of the corresponding room is calculated based on the pressure drop, and the wind pressure value at each inner door plane is obtained based on the ventilation volume.

5. The indoor space thermal environment step-by-step prediction method according to claim 1, characterized in that: The method of using a preset neural network to fuse the spatial layout information matrix and the wind pressure values ​​at all the inner door planes to obtain indoor temperature and wind speed distribution data includes: Using a first neural network, the spatial layout information matrix and the wind pressure values ​​at all the inner door planes are merged to obtain a spatial pressure feature matrix; The spatial pressure characteristic matrix is ​​processed by a second neural network to obtain indoor temperature and wind speed distribution data.

6. The method for predicting indoor space thermal environment step by step according to claim 5, characterized in that: The method of using the first neural network to fuse the spatial layout information matrix and the wind pressure values ​​at all the inner door planes to obtain a spatial pressure feature matrix includes: Using the wind pressure values ​​at all the inner door planes, assigning values ​​to corresponding positions of the room topology map corresponding to the space layout information matrix to obtain an initial space pressure information matrix; The spatial layout information matrix and the initial spatial pressure information matrix are fused and feature extracted using a first neural network to obtain a spatial pressure feature matrix.

7. The method for predicting indoor space thermal environment step by step according to any one of claims 1 to 6, characterized in that: The generating of indoor space thermal environment prediction results based on the indoor temperature and wind speed distribution data and the target floor plan includes: Based on the indoor temperature and wind speed distribution data, obtaining the human body thermal response index value at each position in each room; Based on the human body thermal reaction index values ​​at each of the positions and the target floor plan, a prediction result of the indoor space thermal environment is generated.

8. A step-by-step prediction system for indoor space thermal environment, characterized in that: The system comprises: An information acquisition module is used to obtain a target floor plan; A room topology relationship module, used to obtain a room topology relationship diagram based on the target floor plan, and determine a space layout information matrix based on the room topology relationship diagram; An airflow network module, used to obtain the wind pressure value at each inner door plane based on the room topology diagram and the preset outdoor door and window wind pressure value; A thermal environment parameter prediction module, used to fuse the spatial layout information matrix and the wind pressure values ​​at all the inner door planes using a preset neural network to obtain indoor temperature and wind speed distribution data; The indoor thermal comfort prediction module is used to generate indoor space thermal environment prediction results based on the indoor temperature and wind speed distribution data and the target floor plan.

9. An intelligent terminal, characterized in that: The intelligent terminal includes a memory, a processor, and a step-by-step prediction program for the indoor space thermal environment stored in the memory and executable on the processor. When the step-by-step prediction program for the indoor space thermal environment is executed by the processor, the steps of the step-by-step prediction method for the indoor space thermal environment as described in any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a step-by-step prediction program for the indoor space thermal environment. When the step-by-step prediction program for the indoor space thermal environment is executed by the processor, the steps of the step-by-step prediction method for the indoor space thermal environment as claimed in any one of claims 1 to 7 are implemented.

Images