Intelligent ventilation adjusting system and method for building

Through deep learning technology, temperature, humidity and carbon dioxide data are analyzed, combined with the spatial correlation information of the building, intelligent ventilation adjustment is achieved, which solves the problem that existing systems cannot adjust according to real-time conditions, and improves the comfort and energy efficiency of the indoor environment.

CN118998960AInactive Publication Date: 2024-11-22JIANGXI WANGKAI CONSTR CO LTD
View PDF 0 Cites 2 Cited by

Patent Information

Application Number
CN202411182614.6
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-27
Publication Date
2024-11-22
Estimated Expiration
Not applicable · inactive patent

AI Technical Summary

Technical Problem

The existing building ventilation system cannot be accurately adjusted according to real-time indoor environmental conditions, and mechanical ventilation is difficult to meet the preferences and needs of individual users, resulting in insufficient indoor environment comfort and energy efficiency.

Method used

Using deep learning technology, the data collected by temperature, humidity and carbon dioxide sensors, combined with the building's adjacency matrix, extract environmental parameters and room correlation feature matrix to determine whether ventilation and adjustment are needed to achieve automated control.

Benefits of technology

It improves the comfort and energy efficiency of the indoor environment, realizes the intelligent level of building management, meets the needs of individual users, and improves air quality and energy-saving effects.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118998960A_ABST
    Figure CN118998960A_ABST
Patent Text Reader

Abstract

The invention relates to the field of intelligent ventilation adjustment, and particularly discloses an intelligent ventilation adjustment system and method for a building. The method comprises the following steps: firstly, acquiring indoor environment temperature data of each room in a building acquired by a temperature sensor, and indoor environment humidity data of each room in the building acquired by a humidity sensor; according to the method, indoor environment carbon dioxide data of each room in a building and an adjacent matrix of all rooms in the building are collected by a carbon dioxide sensor, then feature extraction and correlation analysis are performed on the four parts by using a deep learning technology, and finally a classification result is obtained through a classifier so as to judge whether the building needs to be subjected to ventilation adjustment. And therefore, the comfort of the indoor environment is improved, energy conservation, emission reduction and automatic control are realized, and the intelligent level of building management is improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the field of intelligent ventilation and regulation, and more specifically, to an intelligent ventilation and regulation system and method for a building. Background Art

[0002] Building ventilation refers to the exchange of indoor air with outdoor air through reasonable design and control means to maintain the freshness, temperature and humidity of indoor air, remove indoor pollutants and odors, and provide a comfortable and healthy indoor environment. Good building ventilation can effectively reduce harmful substances, moisture and heat in indoor air, improve indoor air quality, prevent the spread of diseases, and improve people's quality of life and work efficiency. At the same time, reasonable ventilation design can also save energy and reduce the operating costs of buildings.

[0003] Building ventilation can be achieved in a variety of ways, including artificial ventilation and mechanical ventilation. Artificial ventilation relies on natural airflow and pressure differences to achieve air exchange, and the ventilation effect can be controlled by manually opening windows and doors. Mechanical ventilation actively regulates indoor air flow through the use of ventilation equipment such as fans and air conditioning systems. However, artificial ventilation cannot provide precise indoor environmental control. Mechanical ventilation operates according to fixed settings and schedules, and cannot be adjusted according to real-time indoor environmental conditions. In addition, mechanical ventilation is usually centrally controlled and difficult to meet the preferences and needs of individual users.

[0004] Therefore, an intelligent ventilation and adjustment system and method for a building is desired. Summary of the invention

[0005] In order to solve the above technical problems, the present application is proposed. The embodiment of the present application provides a building intelligent ventilation and adjustment system and method, which first obtains the indoor environment temperature data of each room in the building collected by the temperature sensor, the indoor environment humidity data of each room in the building collected by the humidity sensor, the indoor environment carbon dioxide data of each room in the building collected by the carbon dioxide sensor, and the adjacency matrix of all rooms in the building, and then uses deep learning technology to perform feature extraction and association analysis on the four, and finally obtains the classification result through the classifier to determine whether the building needs ventilation and adjustment, thereby improving the comfort of the indoor environment, achieving energy saving and emission reduction and automatic control, and thus improving the intelligent level of building management.

[0006] According to one aspect of the present application, there is provided an intelligent ventilation and adjustment system for a building, comprising:

[0007] A building data acquisition module, used to acquire indoor environmental temperature data of each room in the building collected by a temperature sensor, indoor environmental humidity data of each room in the building collected by a humidity sensor, indoor environmental carbon dioxide data of each room in the building collected by a carbon dioxide sensor, and an adjacency matrix of all rooms in the building;

[0008] A building data extraction module, for extracting an environmental parameter global correlation feature matrix and a room adjacency correlation feature matrix from the indoor environmental temperature data of each room in the building collected by the temperature sensor, the indoor environmental humidity data of each room in the building collected by the humidity sensor, the indoor environmental carbon dioxide data of each room in the building collected by the carbon dioxide sensor, and the adjacency matrix of all rooms in the building;

[0009] The ventilation adjustment judgment generation module is used to judge whether the building needs ventilation adjustment based on the environmental parameter global correlation feature matrix and the room adjacent correlation feature matrix.

[0010] According to another aspect of the present application, a method for intelligent ventilation adjustment of a building is provided, comprising:

[0011] Acquire indoor environmental temperature data of each room in the building collected by a temperature sensor, indoor environmental humidity data of each room in the building collected by a humidity sensor, indoor environmental carbon dioxide data of each room in the building collected by a carbon dioxide sensor, and an adjacency matrix of all rooms in the building;

[0012] Extracting an environmental parameter global correlation feature matrix and a room adjacency correlation feature matrix from the indoor environmental temperature data of each room in the building collected by the temperature sensor, the indoor environmental humidity data of each room in the building collected by the humidity sensor, the indoor environmental carbon dioxide data of each room in the building collected by the carbon dioxide sensor, and the adjacency matrix of all rooms in the building;

[0013] Based on the environmental parameter global correlation feature matrix and the room adjacency correlation feature matrix, it is determined whether the building needs ventilation adjustment.

[0014] Compared with the prior art, the present application provides an intelligent ventilation and adjustment system and method for a building, which first obtains the indoor environmental temperature data of each room in the building collected by a temperature sensor, the indoor environmental humidity data of each room in the building collected by a humidity sensor, the indoor environmental carbon dioxide data of each room in the building collected by a carbon dioxide sensor, and the adjacency matrix of all rooms in the building, and then uses deep learning technology to perform feature extraction and association analysis on the four, and finally obtains the classification result through a classifier to determine whether the building needs ventilation adjustment, thereby improving the comfort of the indoor environment, achieving energy conservation and emission reduction and automatic control, and thus improving the level of intelligence in building management. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] By describing the embodiments of the present application in more detail in conjunction with the accompanying drawings, the above and other purposes, features and advantages of the present application will become more apparent. The accompanying drawings are used to provide a further understanding of the embodiments of the present application and constitute a part of the specification. Together with the embodiments of the present application, they are used to explain the present application and do not constitute a limitation of the present application. In the accompanying drawings, the same reference numerals generally represent the same components or steps.

[0016] Figure 1 4 is a block diagram of an intelligent ventilation and adjustment system for a building according to an embodiment of the present application.

[0017] Figure 2 It is a block diagram of a building data extraction module in a building intelligent ventilation and conditioning system according to an embodiment of the present application.

[0018] Figure 3 It is a block diagram of a ventilation adjustment judgment generation module in a building intelligent ventilation adjustment system according to an embodiment of the present application.

[0019] Figure 4 Flow chart of intelligent ventilation regulation method for buildings.

[0020] Figure 5 is a block diagram of an electronic device according to an embodiment of the present application. DETAILED DESCRIPTION

[0021] Below, the exemplary embodiments according to the present application will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments of the present application, and it should be understood that the present application is not limited to the exemplary embodiments described here.

[0022] Figure 1 FIG. 1 is a block diagram of a building intelligent ventilation and adjustment system according to an embodiment of the present application. Figure 1As shown, the intelligent ventilation and adjustment system 100 for buildings according to the embodiment of the present application includes: a building data acquisition module 110, which is used to acquire the indoor environment temperature data of each room in the building collected by the temperature sensor, the indoor environment humidity data of each room in the building collected by the humidity sensor, the indoor environment carbon dioxide data of each room in the building collected by the carbon dioxide sensor, and the adjacency matrix of all rooms in the building; a building data extraction module 120, which is used to extract the environmental parameter global correlation feature matrix and the room adjacency association feature matrix from the indoor environment temperature data of each room in the building collected by the temperature sensor, the indoor environment humidity data of each room in the building collected by the humidity sensor, the indoor environment carbon dioxide data of each room in the building collected by the carbon dioxide sensor, and the adjacency matrix of all rooms in the building; a ventilation adjustment judgment generation module 130, which is used to judge whether the building needs ventilation adjustment based on the environmental parameter global correlation feature matrix and the room adjacency association feature matrix.

[0023] In the above-mentioned intelligent ventilation and adjustment system 100 for buildings, the building data acquisition module 110 is used to obtain the indoor environmental temperature data of each room in the building collected by the temperature sensor, the indoor environmental humidity data of each room in the building collected by the humidity sensor, the indoor environmental carbon dioxide data of each room in the building collected by the carbon dioxide sensor, and the adjacency matrix of all rooms in the building. As mentioned above, artificial ventilation cannot provide accurate indoor environmental control. Mechanical ventilation is operated according to fixed settings and schedules, and cannot be adjusted according to real-time indoor environmental conditions. Moreover, mechanical ventilation is usually centrally controlled and it is difficult to meet the preferences and needs of individual users. In the technical solution of the present application, by obtaining the key factors affecting indoor comfort and air quality such as temperature, humidity and carbon dioxide, and providing the adjacency matrix of spatial association information between rooms in the building, the indoor environment of the building is comprehensively monitored and analyzed, so as to perform intelligent ventilation adjustment and improve the quality and comfort of the indoor environment.

[0024] In the above-mentioned intelligent ventilation and adjustment system 100 for buildings, the building data extraction module 120 is used to extract the global correlation feature matrix of environmental parameters and the room adjacency correlation feature matrix from the indoor environmental temperature data of each room in the building collected by the temperature sensor, the indoor environmental humidity data of each room in the building collected by the humidity sensor, the indoor environmental carbon dioxide data of each room in the building collected by the carbon dioxide sensor, and the adjacency matrix of all rooms in the building. It should be understood that by extracting these feature matrices, the environmental correlation and air flow conditions between different rooms can be revealed. The global correlation feature matrix reflects the overall change trend of the internal environmental parameters of the entire building, while the room adjacency correlation feature matrix reveals the local environmental relationship between each room. The extraction and analysis of these feature matrices help to determine whether the building needs ventilation adjustment to provide better indoor environmental quality and comfort.

[0025] Figure 2 FIG. 1 is a block diagram of a building data extraction module in a building intelligent ventilation and control system according to an embodiment of the present application. Figure 2 As shown, in a specific embodiment of the present application, the building data extraction module 120 includes: an indoor environment data feature extraction unit 121, which is used to perform feature extraction on the indoor environment temperature data of each room in the building collected by the temperature sensor and the indoor environment humidity data of each room in the building collected by the humidity sensor to obtain the building indoor environment temperature-humidity association matrix; an indoor environment carbon dioxide data feature extraction unit 122, which is used to perform feature extraction on the indoor environment carbon dioxide data of each room in the building collected by the carbon dioxide sensor to obtain the building carbon dioxide concentration association feature matrix; a building indoor environment data feature association unit 123, which is used to perform feature fusion on the building indoor environment temperature-humidity association matrix and the building carbon dioxide concentration association feature matrix to obtain the environmental parameter global association feature matrix; a building adjacency matrix feature extraction unit 124, which is used to perform feature extraction on the adjacency matrix of all rooms in the building to obtain the room adjacency association feature matrix. It should be understood that the indoor environment temperature-humidity correlation matrix of the building obtained by extracting features from the indoor environment temperature data of each room in the building collected by the temperature sensor and the indoor environment humidity data of each room in the building collected by the humidity sensor can reflect the temperature and humidity relationship between different rooms and help understand the heat and humidity exchange inside the building. By extracting features and establishing a correlation matrix, the correlation between temperature and humidity can be found, and the overall change trend and local differences of the indoor environment can be understood. Such a correlation matrix helps to evaluate the temperature and humidity distribution inside the building, and provides a basis for the intelligent ventilation and adjustment system to improve indoor comfort and energy saving.

[0026] Furthermore, feature extraction is performed on the indoor environmental carbon dioxide data of each room in the building collected by the carbon dioxide sensor, aiming to obtain the building carbon dioxide concentration correlation feature matrix. This feature matrix can reveal the relationship between carbon dioxide concentrations in different rooms and help understand the air quality distribution inside the building. By extracting features and establishing a correlation matrix, the overall trend and local differences in carbon dioxide concentration can be found, thereby evaluating the air circulation and ventilation effect inside the building. Furthermore, by fusing these two correlation matrices, the relationship between temperature, humidity and carbon dioxide concentration can be comprehensively considered, thereby gaining a more comprehensive understanding of the state of the building's indoor environment. The fused global correlation feature matrix can reveal the comprehensive influence and interaction between different environmental parameters, helping to more accurately evaluate indoor air quality and comfort.

[0027] In particular, by extracting the features of the adjacency matrix, we can understand the spatial layout and connection relationship between different rooms. This adjacency association feature matrix can reveal the relative position, mutual dependence and degree of communication between rooms, and help understand the spatial structure and layout inside the building. By analyzing the adjacency association feature matrix, the ventilation transmission path, air flow direction and possible pollution transmission path between rooms can be determined. Such an association feature matrix is ​​very important for intelligent ventilation and control systems. It can provide the system with a reference for air circulation patterns and ventilation requirements between rooms, and achieve more effective indoor air quality control and comfort optimization.

[0028] In a specific embodiment of the present application, the indoor environment data feature extraction unit 121 includes: the indoor environment temperature data of each room in the building collected by the temperature sensor and the indoor environment humidity data of each room in the building collected by the humidity sensor are respectively processed through the environmental parameter semantic encoder based on the context encoder model to obtain multiple building indoor environment temperature feature vectors and multiple building indoor environment humidity feature vectors; calculate the association matrix of the multiple building indoor environment temperature feature vectors and the multiple building indoor environment humidity feature vectors to obtain the building indoor environment temperature-humidity association matrix. It should be understood that the indoor environment temperature data and the indoor environment humidity data of each room in the building are respectively processed through the environmental parameter semantic encoder based on the context encoder model, aiming to obtain multiple building indoor environment temperature feature vectors and multiple building indoor environment humidity feature vectors. Such a processing method can convert the original temperature and humidity data into a feature vector with semantic information, and better express the characteristics and change trends of the indoor environment. By using the context encoder model, the environmental information and mutual relationship of the surrounding rooms can be taken into account, thereby enhancing the expression ability of the feature vector. Specifically, the indoor environmental temperature data of each room in the building collected by the temperature sensor is converted into a temperature input vector using the embedding layer of the environmental parameter semantic encoder based on the context encoder model to obtain a sequence of temperature input vectors; the sequence of temperature input vectors is globally semantically encoded based on context to generate the multiple building indoor environmental temperature feature vectors; the indoor environmental humidity data of each room in the building collected by the humidity sensor is converted into a humidity input vector using the embedding layer of the environmental parameter semantic encoder based on the context encoder model to obtain a sequence of humidity input vectors; and the sequence of humidity input vectors is globally semantically encoded based on context to generate the multiple building indoor environmental humidity feature vectors.

[0029] Furthermore, calculating the correlation matrix can reveal the correlation and mutual influence between indoor temperature and humidity. By calculating the correlation between eigenvectors, we can understand the degree of correlation between temperature and humidity, such as whether they are positively correlated, negatively correlated, or unrelated. Such a correlation matrix can provide comprehensive information about the indoor environment and help understand the dynamic changes and interactions between temperature and humidity.

[0030] In a specific embodiment of the present application, the indoor environment carbon dioxide data feature extraction unit 122 includes: passing the indoor environment carbon dioxide data of each room in the building collected by the carbon dioxide sensor through the environmental parameter semantic encoder based on the context encoder model to obtain multiple building indoor environment carbon dioxide feature vectors; two-dimensionally splicing the multiple building indoor environment carbon dioxide feature vectors to obtain the building indoor environment carbon dioxide feature matrix; passing the building indoor environment carbon dioxide feature matrix through the carbon dioxide concentration feature extractor based on the convolutional neural network to obtain the building carbon dioxide concentration correlation feature matrix. It should be understood that by using the context encoder model, the environmental information and mutual relationship of the surrounding rooms can be taken into account, thereby enhancing the expression ability of the feature vector. In this way, the original carbon dioxide data can be converted into a feature vector with semantic information to better express the carbon dioxide level and change trend of the indoor environment. Specifically, the embedding layer of the environmental parameter semantic encoder based on the context encoder model is used to convert the indoor environmental carbon dioxide data of each room in the building collected by the carbon dioxide sensor into a carbon dioxide data input vector to obtain a sequence of carbon dioxide data input vectors; and the converter of the environmental parameter semantic encoder based on the context encoder model is used to perform context-based global semantic encoding on the sequence of carbon dioxide data input vectors to generate the indoor environmental carbon dioxide feature vectors of the multiple buildings.

[0031] Furthermore, by concatenating the indoor environmental CO2 feature vectors of multiple buildings in two dimensions, they can be combined into a larger matrix, where each row represents a building and each column represents the CO2 feature of the corresponding building. Such a feature matrix can provide a comprehensive perspective to show the differences and similarities of CO2 features between different buildings.

[0032] Furthermore, by applying convolutional neural networks, abstract features related to the building's carbon dioxide concentration can be extracted from the feature matrix. Convolutional neural networks are sensitive to spatial structures and can capture local patterns and spatial correlations in the feature matrix. Through the network's multi-layer convolution and pooling operations, more advanced feature representations can be gradually extracted. Such a feature extraction process can help discover the distribution characteristics of carbon dioxide concentration inside the building, such as high-concentration areas, concentration gradient changes, etc., and further understand the air quality conditions inside the building. By obtaining the building carbon dioxide concentration-related feature matrix, more accurate building carbon dioxide concentration information can be provided to the intelligent ventilation and control system, thereby achieving more refined ventilation control and air quality management.

[0033] In a specific embodiment of the present application, the building indoor environment data feature association unit 123 includes: fusing the building indoor environment temperature-humidity association matrix and the building carbon dioxide concentration association feature matrix to obtain an environmental parameter association feature matrix; wherein the fusion formula is: c =Concat[M a ,M b ], where M a represents the building indoor environment temperature-humidity association matrix, M b represents the building carbon dioxide concentration correlation feature matrix, M c Represents the environmental parameter association feature matrix, and Concat[·,·] represents a cascade function. It should be understood that by appropriately combining the two feature matrices, the information of environmental parameters such as the temperature, humidity and carbon dioxide concentration of the building can be integrated to obtain an environmental parameter association feature matrix. The fused feature matrix can provide a more comprehensive description of the indoor environment, including the correlation and comprehensive characteristics between different environmental parameters. By analyzing the environmental parameter association feature matrix, a deeper understanding can be obtained, such as the impact of temperature and humidity on carbon dioxide concentration, the interaction of environmental parameters between different areas, etc. Such a comprehensive feature matrix can provide a more comprehensive decision-making basis for the intelligent ventilation and adjustment system, enabling the system to more accurately determine whether the building needs ventilation adjustment and how to adjust it.

[0034] In a specific embodiment of the present application, the building adjacency matrix feature extraction unit 124 includes: passing the adjacency matrix of all rooms in the building through the adjacency matrix feature extractor to obtain the room adjacency feature matrix; passing the environmental parameter global association feature matrix and the room adjacency feature matrix through the graph neural network to obtain the room adjacency association feature matrix. It should be understood that by analyzing the adjacency relationship between rooms, the spatial layout and relative position information between rooms inside the building can be captured. The adjacency matrix feature extractor can process the adjacency matrix to extract the adjacency features between rooms. By performing graph theory analysis and feature extraction operations on the adjacency matrix, information such as the topological structure, connection relationship and spatial distance between rooms can be obtained. Such a feature extraction process can help understand the correlation and layout rules between rooms inside the building, so as to better describe the relationship between rooms inside the building. Such a feature matrix can provide more comprehensive spatial information for the intelligent ventilation and adjustment system, such as ventilation flow paths, air propagation paths, etc. between different rooms. Such information can help the system more accurately determine the ventilation needs inside the building and how to perform ventilation adjustment. Specifically, each layer except the last layer of the adjacency matrix feature extractor performs convolution processing, pooling processing and activation processing on the input data in the forward pass of the layer to output a topological feature map by the second last layer of the adjacency matrix feature extractor; and, the last layer of the adjacency matrix feature extractor performs convolution processing, global mean pooling processing along the channel dimension and activation processing on the topological feature map to generate the room adjacency association feature matrix. More specifically, the eigenvalue of each position on the non-diagonal position in the adjacency matrix indicates whether there is a connected relationship between the corresponding two rooms, and the eigenvalue of each position on the diagonal position in the adjacency matrix is ​​1, and if there is no connected relationship between the corresponding two rooms, the eigenvalue of the corresponding matrix position is 0.

[0035] Furthermore, the global correlation feature matrix of environmental parameters and the room adjacency feature matrix are processed by a graph neural network, where nodes represent rooms in a building and edges represent adjacency relationships between them, which can capture the correlation and interaction between rooms. The graph neural network can transfer and aggregate information in a graph structure, and can integrate local room adjacency features into a global scope through connections and message passing between nodes. Through multi-layer graph convolution and pooling operations, the graph neural network can gradually extract more advanced room adjacency correlation feature representations. Such a processing process can help discover correlation patterns between different rooms in a building, such as ventilation paths, air flow, and temperature transfer. Specifically, the graph neural network processes the global correlation feature matrix of environmental parameters and the room adjacency feature matrix through learnable neural network parameters to obtain the room adjacency correlation feature matrix containing Euclidean distance topological features and semantic understanding feature information described by each thread.

[0036] In the above-mentioned intelligent ventilation and adjustment system 100 for buildings, the ventilation adjustment judgment generation module 130 is used to judge whether the building needs ventilation adjustment based on the global correlation feature matrix of the environmental parameters and the room adjacency correlation feature matrix. It should be understood that the global correlation feature matrix of the environmental parameters contains the environmental parameters of each room in the building, such as temperature, humidity and air quality. By analyzing the global correlation of these parameters, the overall environmental state inside the building can be understood. For example, if the temperature of the entire building is too high or the air quality decreases, it may mean that ventilation adjustment is needed to improve the indoor environment. The room adjacency correlation feature matrix reflects the adjacency relationship and interaction between different rooms in the building. By analyzing these correlation features, the ventilation paths, air flow and temperature transfer between different rooms can be understood. If the adjacency relationship between some rooms leads to poor air flow or large temperature differences, ventilation adjustment may be needed to balance the indoor environment. Such judgments can help the intelligent ventilation and adjustment system make accurate decisions and take corresponding control measures to adjust the ventilation state of the building.

[0037] Figure 3 FIG. 1 is a block diagram of a ventilation adjustment judgment generation module in a building intelligent ventilation adjustment system according to an embodiment of the present application. Figure 3As shown, in a specific embodiment of the present application, the ventilation adjustment judgment generation module 130 includes: a building indoor data fusion unit 131, which is used to fuse the environmental parameter global correlation feature matrix and the room adjacent correlation feature matrix to obtain a building ventilation classification feature matrix; a building indoor data matrix expansion unit 132, which is used to expand the building ventilation classification feature matrix to obtain a building ventilation classification feature vector; an optimization unit 133, which is used to perform backward learning preference optimization based on inverse error analysis on the building ventilation classification feature vector to obtain an optimized building ventilation classification feature vector; a building ventilation judgment unit 134, which is used to pass the optimized building ventilation classification feature vector through a classifier to obtain a classification result, and the classification result is used to judge whether the building needs ventilation adjustment. It should be understood that by fusing these two feature matrices, the correlation information between the internal environmental parameters of the building and the rooms can be comprehensively considered to obtain a building ventilation classification feature matrix. Such a feature matrix can provide more comprehensive and accurate information for judging whether a building needs ventilation adjustment. Based on such a feature matrix, a more intelligent and effective ventilation strategy can be formulated to achieve a healthier, more comfortable and energy-saving indoor environment.

[0038] Furthermore, considering that the building ventilation classification feature matrix is ​​usually a multi-dimensional array, by expanding it into a building ventilation classification feature vector, the multi-dimensional data can be simplified to one dimension, thereby reducing the computational complexity. For large-scale data, this simplification can significantly improve the processing speed.

[0039] In the technical solution of the present application, it is considered that the sensor data of temperature, humidity and carbon dioxide concentration need to be accurate. If the sensor accuracy is insufficient or is disturbed by the environment, the quality of the original data may be reduced. The environmental parameter semantic encoder may not be able to fully capture all subtle changes related to the indoor environment, especially for complex environmental dynamics. The method for calculating the temperature-humidity association matrix needs to be able to accurately reflect the relationship between the two. If the calculation method is inappropriate, key information may be lost. Convolutional neural networks and graph neural networks may not be able to capture all features related to ventilation needs when extracting features, especially complex patterns in space and time. When fusing the environmental parameter association feature matrix and the room adjacency association feature matrix, the importance and complementarity of different features need to be considered. If the fusion strategy is inappropriate, information loss or feature conflict may occur. Graph neural networks may have limitations when processing adjacency matrices and environmental parameter global association feature matrices, such as insufficient sensitivity to topological structures. This leads to a class harmony offset in the building ventilation classification feature vector relative to the predetermined class label of the classifier. In order to reduce the class harmony offset, in the technical solution of the present application, the building ventilation classification feature vector is subjected to backward learning preference optimization based on inverse error analysis.

[0040] The method comprises: performing a backward learning preference optimization based on an inverse error analysis on the building ventilation classification feature vector to obtain an optimized building ventilation classification feature vector, comprising: performing a probabilistic activation based on a Sigmoid function on the building ventilation classification feature vector to obtain a probabilistic building ventilation classification feature vector; inputting the building ventilation classification feature vector into a classifier to obtain an initial class label probability value; calculating the absolute value of the difference between the feature value at each position in the probabilistic building ventilation classification feature vector and the initial class label probability value to obtain a building ventilation classification fine-grained error metric vector; and using the building ventilation classification fine-grained error metric vector as the basis for the optimization of the building ventilation classification feature vector. The metric vector is used as a key vector, the building ventilation classification feature vector is used as a query vector and a value vector, and the key vector, the query vector and the value vector are input into a converter structure to obtain a backward learning type attention optimized building ventilation classification feature vector; the ratio between the sum of the building ventilation classification fine-grained error metric vector and the length of the building ventilation classification fine-grained error metric vector is calculated, and the value of a natural exponential function with the ratio as a power is calculated as an optimization scaling factor; the product between the optimization scaling factor and the backward learning type attention optimized building ventilation classification feature vector is calculated to obtain the optimized building ventilation classification feature vector.

[0041] The optimization steps are specifically expressed as follows:

[0042]

[0043] Among them, V c represents the probabilistic building ventilation classification feature vector, P represents the initial class label probability value, Indicates subtraction by position, V e represents the fine-grained error metric vector for building ventilation classification, v i represents the eigenvalue of the i-th position of the building ventilation classification fine-grained error measurement vector, L represents the length of the building ventilation classification fine-grained error measurement vector, T represents the transpose of the vector, represents vector multiplication, Softmax represents the normalized exponential function, ⊙ represents point multiplication by position, V' c represents the optimized building ventilation classification feature vector.

[0044] Based on this, in the technical solution of the present application, the building ventilation classification feature vector is subjected to backward learning preference optimization based on reverse error analysis, which first implements probabilistic activation based on the Sigmoid function on the building ventilation classification feature vector to obtain a probabilistic building ventilation classification feature vector, and inputs the building ventilation classification feature vector into the classifier to obtain an initial class label probability value. Then, the error between the probabilistic building ventilation classification feature vector and the initial class label probability value is calculated through reverse error analysis to generate a fine-grained error metric vector, which quantifies the viewpoint metric contribution of each feature position to the classification error.

[0045] Then, the fine-grained error metric vector of the building ventilation classification is used as a key vector, the building ventilation classification feature vector is used as a query vector and a value vector, and the key vector, the query vector and the value vector are input into a transformer structure to perform backward machine learning based on an error metric-based class attention mechanism on the building ventilation classification feature vector through a class Transformer architecture. In this way, a backward interpretability analysis is performed on the building ventilation classification feature vector so that the obtained backward learning class attention optimized building ventilation classification feature vector has a backward learning sensitive preference relative to the predetermined class label of the classifier. And further, the ratio of the sum of the error metric vector and the vector length is calculated, and the optimization scaling factor obtained by calculating the natural exponential function value with this ratio as an exponent is used to perform backward scaling optimization on the backward learning class attention optimized building ventilation classification feature vector. In this way, the sensitive preference and robustness of the building ventilation classification feature vector relative to the predetermined class label of the classifier are improved, thereby improving the accuracy of the classification result.

[0046] Furthermore, by using a classifier, the ventilation needs of a building can be classified and categorized according to various features and patterns in the optimized building ventilation classification feature vector. The classification result can tell whether the building needs ventilation adjustment. For example, if the classification result shows that the building is in a state where ventilation is required, then corresponding measures can be taken to turn on the ventilation system to improve the indoor air quality and temperature. On the contrary, if the classification result shows that the building is in a state where ventilation is not required, energy can be saved and the indoor environment can be kept stable. Therefore, by inputting the optimized building ventilation classification feature vector into the classifier and obtaining the classification results, it can be judged based on these results whether the building needs ventilation adjustment, thereby achieving a healthier, more comfortable and energy-saving indoor environment.

[0047] In summary, the embodiment of the present application first obtains the indoor environmental temperature data of each room in the building collected by the temperature sensor, the indoor environmental humidity data of each room in the building collected by the humidity sensor, the indoor environmental carbon dioxide data of each room in the building collected by the carbon dioxide sensor, and the adjacency matrix of all rooms in the building, and then uses deep learning technology to perform feature extraction and association analysis on the four, and finally obtains the classification result through the classifier to determine whether the building needs ventilation adjustment, thereby improving the comfort of the indoor environment, achieving energy conservation and emission reduction and automatic control, and thus improving the level of intelligence of building management.

[0048] As described above, the building intelligent ventilation and adjustment system 100 according to the embodiment of the present application can be implemented in various terminal devices, such as a server deployed with a building intelligent ventilation and adjustment control algorithm. In one example, the building intelligent ventilation and adjustment system 100 can be integrated into the terminal device as a software module and / or a hardware module. For example, the building intelligent ventilation and adjustment system 100 can be a software module in the operating system of the terminal device, or can be an application developed for the terminal device; of course, the building intelligent ventilation and adjustment system 100 can also be one of the many hardware modules of the terminal device.

[0049] Alternatively, in another example, the building intelligent ventilation and conditioning system 100 and the terminal device may also be separate devices, and the building intelligent ventilation and conditioning system 100 may be connected to the terminal device via a wired and / or wireless network and transmit interactive information in an agreed data format.

[0050] Figure 4 The following is a flow chart of the intelligent ventilation adjustment method for buildings. Figure 4 As shown, according to the intelligent ventilation and adjustment method for buildings in an embodiment of the present application, it includes: S110, acquiring indoor environmental temperature data of each room in the building collected by a temperature sensor, indoor environmental humidity data of each room in the building collected by a humidity sensor, indoor environmental carbon dioxide data of each room in the building collected by a carbon dioxide sensor, and an adjacency matrix of all rooms in the building; S120, extracting an environmental parameter global correlation feature matrix and a room adjacency correlation feature matrix from the indoor environmental temperature data of each room in the building collected by the temperature sensor, the indoor environmental humidity data of each room in the building collected by the humidity sensor, the indoor environmental carbon dioxide data of each room in the building collected by the carbon dioxide sensor, and the adjacency matrix of all rooms in the building; S130, judging whether the building needs ventilation and adjustment based on the environmental parameter global correlation feature matrix and the room adjacency correlation feature matrix.

[0051] Here, those skilled in the art can understand that the specific operations of each step in the above-mentioned building intelligent ventilation adjustment method have been referred to above. Figures 1 to 3 The invention has been introduced in detail in the description of the intelligent ventilation and conditioning system for buildings, and therefore, its repeated description will be omitted.

[0052] Figure 5 A structural diagram of an electronic device provided in an embodiment of the present application, the electronic device 100 includes: a processor 11, a memory 12, a bus 13 and a communication interface 14, wherein the processor 11, the communication interface 14 and the memory 12 are connected via the bus 13; the processor 11 is used to execute an executable module stored in the memory 12, such as a computer program.

[0053] The memory 12 may include a high-speed random access memory (RAM), and may also include a non-volatile memory (non-volatile memory), such as at least one disk memory. The communication connection between the system network element and at least one other network element is realized through at least one communication interface 14 (which may be wired or wireless), and the Internet, wide area network, local area network, metropolitan area network, etc. may be used.

[0054] The bus 13 may be an ISA bus, a PCI bus, or an EISA bus, etc. The bus may be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 5 Only one bidirectional arrow is used in the diagram, but this does not mean that there is only one bus or only one type of bus.

[0055] Among them, the memory 12 is used to store programs, and the processor 11 executes the program after receiving the execution instruction. The method executed by the device for flow process definition disclosed in any embodiment of the above-mentioned embodiment of the present invention can be applied to the processor 11, or implemented by the processor 11.

[0056] The processor 11 may be an integrated circuit chip with signal processing capabilities. In the implementation process, each step of the above method can be completed by the hardware integrated logic circuit or software instructions in the processor 11. The above processor 11 can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components. The methods, steps and logic block diagrams disclosed in the embodiments of the present invention can be implemented or executed. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor. The steps of the method disclosed in the embodiment of the present invention can be directly embodied as a hardware decoding processor for execution, or a combination of hardware and software modules in the decoding processor for execution. The software module may be located in a storage medium mature in the art such as a random access memory, a flash memory, a read-only memory, a programmable read-only memory, or an electrically erasable programmable memory, a register, etc. The storage medium is located in the memory 12, and the processor 11 reads the information in the memory 12 and completes the steps of the above method in combination with its hardware.

[0057] Corresponding to the above method, an embodiment of the present application also provides a computer-readable storage medium, which stores machine-executable instructions. When the computer-executable instructions are called and executed by a processor, the computer-executable instructions prompt the processor to execute the steps of the above method.

[0058] In the embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. The device embodiments described above are merely schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some communication interfaces, and the indirect coupling or communication connection of the devices or units can be electrical, mechanical or other forms.

[0059] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0060] In addition, each functional unit in the embodiments provided in the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0061] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, electronic device, or network device, etc.) to perform all or part of the steps of the methods described in each embodiment of the present invention. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0062] It should be noted that similar numbers and letters represent similar items in the accompanying drawings. Therefore, once an item is defined in one drawing, it does not need to be further defined and explained in subsequent drawings. In addition, the terms "first", "second", "third", etc. are only used to distinguish the description and cannot be understood as indicating or implying relative importance.

[0063] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. An intelligent ventilation and adjustment system for a building, It is characterized in that include: A building data acquisition module, used to acquire indoor environmental temperature data of each room in the building collected by a temperature sensor, indoor environmental humidity data of each room in the building collected by a humidity sensor, indoor environmental carbon dioxide data of each room in the building collected by a carbon dioxide sensor, and an adjacency matrix of all rooms in the building; A building data extraction module, for extracting an environmental parameter global correlation feature matrix and a room adjacency correlation feature matrix from the indoor environmental temperature data of each room in the building collected by the temperature sensor, the indoor environmental humidity data of each room in the building collected by the humidity sensor, the indoor environmental carbon dioxide data of each room in the building collected by the carbon dioxide sensor, and the adjacency matrix of all rooms in the building; The ventilation adjustment judgment generation module is used to judge whether the building needs ventilation adjustment based on the environmental parameter global correlation feature matrix and the room adjacent correlation feature matrix.

2. The intelligent ventilation and conditioning system for buildings according to claim 1, It is characterized in that The building data extraction module comprises: An indoor environment data feature extraction unit is used to extract features from the indoor environment temperature data of each room in the building collected by the temperature sensor and the indoor environment humidity data of each room in the building collected by the humidity sensor to obtain a building indoor environment temperature-humidity association matrix; An indoor environment carbon dioxide data feature extraction unit, used to extract features from the indoor environment carbon dioxide data of each room in the building collected by the carbon dioxide sensor to obtain a building carbon dioxide concentration correlation feature matrix; A building indoor environment data feature association unit, used for performing feature fusion on the building indoor environment temperature-humidity association matrix and the building carbon dioxide concentration association feature matrix to obtain the environmental parameter global association feature matrix; The building adjacency matrix feature extraction unit is used to extract features from the adjacency matrices of all rooms in the building to obtain the room adjacency association feature matrix.

3. The intelligent ventilation and conditioning system for buildings according to claim 2, It is characterized in that The indoor environment data feature extraction unit comprises: The indoor environment temperature data of each room in the building collected by the temperature sensor and the indoor environment humidity data of each room in the building collected by the humidity sensor are respectively passed through an environmental parameter semantic encoder based on a context encoder model to obtain multiple building indoor environment temperature feature vectors and multiple building indoor environment humidity feature vectors; The correlation matrix of the plurality of building indoor environment temperature feature vectors and the plurality of building indoor environment humidity feature vectors is calculated to obtain the building indoor environment temperature-humidity correlation matrix.

4. The intelligent ventilation and conditioning system for buildings according to claim 3, It is characterized in that The indoor environment carbon dioxide data feature extraction unit comprises: The indoor environmental carbon dioxide data of each room in the building collected by the carbon dioxide sensor is passed through the environmental parameter semantic encoder based on the context encoder model to obtain multiple building indoor environmental carbon dioxide feature vectors; Performing two-dimensional splicing on the plurality of building indoor environment carbon dioxide feature vectors to obtain a building indoor environment carbon dioxide feature matrix; The carbon dioxide feature matrix of the building indoor environment is passed through a carbon dioxide concentration feature extractor based on a convolutional neural network to obtain the building carbon dioxide concentration correlation feature matrix.

5. The intelligent ventilation and conditioning system for buildings according to claim 4, It is characterized in that The building indoor environment data feature association unit comprises: The building indoor environment temperature-humidity correlation matrix and the building carbon dioxide concentration correlation feature matrix are merged to obtain an environmental parameter correlation feature matrix; Wherein, the fusion formula is: M c =Concat[M a ,M b ] Among them, M a represents the building indoor environment temperature-humidity association matrix, M b represents the building carbon dioxide concentration correlation feature matrix, M c represents the environmental parameter association feature matrix, and Concat[·,·] represents the cascade function.

6. The intelligent ventilation and conditioning system for buildings according to claim 5, It is characterized in that The building adjacency matrix feature extraction unit comprises: Passing the adjacency matrix of all rooms in the building through an adjacency matrix feature extractor to obtain a room adjacency feature matrix; The environmental parameter global correlation feature matrix and the room adjacency feature matrix are passed through the graph neural network to obtain the room adjacency correlation feature matrix.

7. The intelligent ventilation and conditioning system for buildings according to claim 6, It is characterized in that The ventilation adjustment judgment generation module includes: A building indoor data fusion unit, used for fusing the environmental parameter global correlation feature matrix and the room adjacency correlation feature matrix to obtain a building ventilation classification feature matrix; A building indoor data matrix expansion unit, used for expanding the building ventilation classification feature matrix to obtain a building ventilation classification feature vector; an optimization unit, configured to perform backward learning preference optimization based on inverse error analysis on the building ventilation classification feature vector to obtain an optimized building ventilation classification feature vector; The building ventilation judgment unit is used to pass the optimized building ventilation classification feature vector through a classifier to obtain a classification result, and the classification result is used to judge whether the building needs ventilation adjustment.

8. The intelligent ventilation and conditioning system for buildings according to claim 7, It is characterized in that The optimization unit comprises: Performing probabilistic activation based on a Sigmoid function on the building ventilation classification feature vector to obtain a probabilistic building ventilation classification feature vector; Inputting the building ventilation classification feature vector into a classifier to obtain an initial class label probability value; Calculating the absolute value of the difference between the feature value of each position in the probabilistic building ventilation classification feature vector and the initial class label probability value to obtain a building ventilation classification fine-grained error metric vector; Using the building ventilation classification fine-grained error metric vector as a key vector, the building ventilation classification feature vector as a query vector and a value vector, inputting the key vector, the query vector and the value vector into a transformer structure to obtain a backward learning class attention optimized building ventilation classification feature vector; Calculating a ratio between a sum of the building ventilation classification fine-grained error metric vectors and a length of the building ventilation classification fine-grained error metric vector, and calculating a natural exponential function value raised to a power of the ratio as an optimization scaling factor; The product between the optimized scaling factor and the backward learning class attention optimized building ventilation classification feature vector is calculated to obtain the optimized building ventilation classification feature vector.

9. A method for intelligent ventilation regulation of a building, It is characterized in that include: Acquire indoor environmental temperature data of each room in the building collected by a temperature sensor, indoor environmental humidity data of each room in the building collected by a humidity sensor, indoor environmental carbon dioxide data of each room in the building collected by a carbon dioxide sensor, and an adjacency matrix of all rooms in the building; Extracting an environmental parameter global correlation feature matrix and a room adjacency correlation feature matrix from the indoor environmental temperature data of each room in the building collected by the temperature sensor, the indoor environmental humidity data of each room in the building collected by the humidity sensor, the indoor environmental carbon dioxide data of each room in the building collected by the carbon dioxide sensor, and the adjacency matrix of all rooms in the building; Based on the environmental parameter global correlation feature matrix and the room adjacency correlation feature matrix, it is determined whether the building needs ventilation adjustment.

10. The intelligent ventilation adjustment method for a building according to claim 9, It is characterized in that Extracting an environmental parameter global correlation feature matrix and a room adjacency correlation feature matrix from the indoor environmental temperature data of each room in the building collected by the temperature sensor, the indoor environmental humidity data of each room in the building collected by the humidity sensor, the indoor environmental carbon dioxide data of each room in the building collected by the carbon dioxide sensor, and the adjacency matrix of all rooms in the building, including: Performing feature extraction on the indoor environment temperature data of each room in the building collected by the temperature sensor and the indoor environment humidity data of each room in the building collected by the humidity sensor to obtain a building indoor environment temperature-humidity correlation matrix; Extracting features from the indoor environmental carbon dioxide data of each room in the building collected by the carbon dioxide sensor to obtain a building carbon dioxide concentration correlation feature matrix; Performing feature fusion on the building indoor environment temperature-humidity correlation matrix and the building carbon dioxide concentration correlation feature matrix to obtain the environmental parameter global correlation feature matrix; Feature extraction is performed on the adjacency matrix of all the rooms in the building to obtain the room adjacency association feature matrix.

Citation Information

Cited By

  • Building carbon emission real-time online monitoring system

    CN120102805A

  • Temperature and humidity adjusting method and system based on intelligent ventilation control

    CN120560411A