CN117606106A - Method for realizing building heating and ventilation automatic control based on deep learning graph neural network - Google Patents
Method for realizing building heating and ventilation automatic control based on deep learning graph neural network Download PDFInfo
- Publication number
- CN117606106A CN117606106A CN202311807670.XA CN202311807670A CN117606106A CN 117606106 A CN117606106 A CN 117606106A CN 202311807670 A CN202311807670 A CN 202311807670A CN 117606106 A CN117606106 A CN 117606106A
- Authority
- CN
- China
- Prior art keywords
- neural network
- room
- data
- building
- ventilation
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000009423 ventilation Methods 0.000 title claims abstract description 56
- 238000000034 method Methods 0.000 title claims abstract description 50
- 238000010438 heat treatment Methods 0.000 title claims abstract description 26
- 238000013528 artificial neural network Methods 0.000 title claims abstract description 21
- 238000013135 deep learning Methods 0.000 title claims abstract description 19
- 238000003062 neural network model Methods 0.000 claims abstract description 31
- 238000012544 monitoring process Methods 0.000 claims abstract description 24
- 239000002356 single layer Substances 0.000 claims abstract description 19
- CURLTUGMZLYLDI-UHFFFAOYSA-N Carbon dioxide Chemical compound O=C=O CURLTUGMZLYLDI-UHFFFAOYSA-N 0.000 claims description 58
- 229910002092 carbon dioxide Inorganic materials 0.000 claims description 29
- 239000001569 carbon dioxide Substances 0.000 claims description 29
- 239000000809 air pollutant Substances 0.000 claims description 24
- 231100001243 air pollutant Toxicity 0.000 claims description 24
- 238000005516 engineering process Methods 0.000 claims description 15
- 238000012545 processing Methods 0.000 claims description 15
- 238000012549 training Methods 0.000 claims description 11
- 238000004458 analytical method Methods 0.000 claims description 10
- 230000008859 change Effects 0.000 claims description 10
- 230000001105 regulatory effect Effects 0.000 claims description 10
- 238000013527 convolutional neural network Methods 0.000 claims description 9
- 230000008569 process Effects 0.000 claims description 9
- 230000007613 environmental effect Effects 0.000 claims description 8
- 238000007781 pre-processing Methods 0.000 claims description 7
- 239000000356 contaminant Substances 0.000 claims description 6
- 230000009471 action Effects 0.000 claims description 5
- 230000006399 behavior Effects 0.000 claims description 5
- 238000012300 Sequence Analysis Methods 0.000 claims description 3
- 238000004140 cleaning Methods 0.000 claims description 3
- 238000012937 correction Methods 0.000 claims description 3
- 238000007405 data analysis Methods 0.000 claims description 3
- 238000013499 data model Methods 0.000 claims description 3
- 238000003709 image segmentation Methods 0.000 claims description 3
- 238000002372 labelling Methods 0.000 claims description 3
- 238000005457 optimization Methods 0.000 claims description 3
- 238000003909 pattern recognition Methods 0.000 claims description 3
- 238000004088 simulation Methods 0.000 claims description 3
- 238000005065 mining Methods 0.000 claims description 2
- 230000009286 beneficial effect Effects 0.000 abstract description 2
- 238000002474 experimental method Methods 0.000 description 3
- MWUXSHHQAYIFBG-UHFFFAOYSA-N nitrogen oxide Inorganic materials O=[N] MWUXSHHQAYIFBG-UHFFFAOYSA-N 0.000 description 3
- 230000001276 controlling effect Effects 0.000 description 2
- 238000013500 data storage Methods 0.000 description 2
- 238000001514 detection method Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 239000007789 gas Substances 0.000 description 2
- 238000010801 machine learning Methods 0.000 description 2
- 239000002912 waste gas Substances 0.000 description 2
- UGFAIRIUMAVXCW-UHFFFAOYSA-N Carbon monoxide Chemical compound [O+]#[C-] UGFAIRIUMAVXCW-UHFFFAOYSA-N 0.000 description 1
- 241001465754 Metazoa Species 0.000 description 1
- CBENFWSGALASAD-UHFFFAOYSA-N Ozone Chemical compound [O-][O+]=O CBENFWSGALASAD-UHFFFAOYSA-N 0.000 description 1
- 230000004075 alteration Effects 0.000 description 1
- 229910002091 carbon monoxide Inorganic materials 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 238000005265 energy consumption Methods 0.000 description 1
- 230000005183 environmental health Effects 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 238000003905 indoor air pollution Methods 0.000 description 1
- 238000012417 linear regression Methods 0.000 description 1
- 238000007726 management method Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 210000002569 neuron Anatomy 0.000 description 1
- 235000019645 odor Nutrition 0.000 description 1
- 230000008520 organization Effects 0.000 description 1
- 230000036284 oxygen consumption Effects 0.000 description 1
- 229910052704 radon Inorganic materials 0.000 description 1
- SYUHGPGVQRZVTB-UHFFFAOYSA-N radon atom Chemical compound [Rn] SYUHGPGVQRZVTB-UHFFFAOYSA-N 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 230000011218 segmentation Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- XTQHKBHJIVJGKJ-UHFFFAOYSA-N sulfur monoxide Chemical class S=O XTQHKBHJIVJGKJ-UHFFFAOYSA-N 0.000 description 1
- 229910052815 sulfur oxide Inorganic materials 0.000 description 1
- 238000012706 support-vector machine Methods 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
- 239000002699 waste material Substances 0.000 description 1
Classifications
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F24—HEATING; RANGES; VENTILATING
- F24F—AIR-CONDITIONING; AIR-HUMIDIFICATION; VENTILATION; USE OF AIR CURRENTS FOR SCREENING
- F24F11/00—Control or safety arrangements
- F24F11/30—Control or safety arrangements for purposes related to the operation of the system, e.g. for safety or monitoring
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F24—HEATING; RANGES; VENTILATING
- F24F—AIR-CONDITIONING; AIR-HUMIDIFICATION; VENTILATION; USE OF AIR CURRENTS FOR SCREENING
- F24F11/00—Control or safety arrangements
- F24F11/50—Control or safety arrangements characterised by user interfaces or communication
- F24F11/56—Remote control
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F24—HEATING; RANGES; VENTILATING
- F24F—AIR-CONDITIONING; AIR-HUMIDIFICATION; VENTILATION; USE OF AIR CURRENTS FOR SCREENING
- F24F11/00—Control or safety arrangements
- F24F11/62—Control or safety arrangements characterised by the type of control or by internal processing, e.g. using fuzzy logic, adaptive control or estimation of values
- F24F11/63—Electronic processing
- F24F11/64—Electronic processing using pre-stored data
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F24—HEATING; RANGES; VENTILATING
- F24F—AIR-CONDITIONING; AIR-HUMIDIFICATION; VENTILATION; USE OF AIR CURRENTS FOR SCREENING
- F24F11/00—Control or safety arrangements
- F24F11/70—Control systems characterised by their outputs; Constructional details thereof
- F24F11/72—Control systems characterised by their outputs; Constructional details thereof for controlling the supply of treated air, e.g. its pressure
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F24—HEATING; RANGES; VENTILATING
- F24F—AIR-CONDITIONING; AIR-HUMIDIFICATION; VENTILATION; USE OF AIR CURRENTS FOR SCREENING
- F24F11/00—Control or safety arrangements
- F24F11/88—Electrical aspects, e.g. circuits
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F24—HEATING; RANGES; VENTILATING
- F24F—AIR-CONDITIONING; AIR-HUMIDIFICATION; VENTILATION; USE OF AIR CURRENTS FOR SCREENING
- F24F2110/00—Control inputs relating to air properties
- F24F2110/10—Temperature
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F24—HEATING; RANGES; VENTILATING
- F24F—AIR-CONDITIONING; AIR-HUMIDIFICATION; VENTILATION; USE OF AIR CURRENTS FOR SCREENING
- F24F2110/00—Control inputs relating to air properties
- F24F2110/50—Air quality properties
- F24F2110/65—Concentration of specific substances or contaminants
- F24F2110/70—Carbon dioxide
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F24—HEATING; RANGES; VENTILATING
- F24F—AIR-CONDITIONING; AIR-HUMIDIFICATION; VENTILATION; USE OF AIR CURRENTS FOR SCREENING
- F24F2120/00—Control inputs relating to users or occupants
- F24F2120/10—Occupancy
- F24F2120/14—Activity of occupants
Landscapes
- Engineering & Computer Science (AREA)
- Chemical & Material Sciences (AREA)
- Combustion & Propulsion (AREA)
- Mechanical Engineering (AREA)
- General Engineering & Computer Science (AREA)
- Signal Processing (AREA)
- Human Computer Interaction (AREA)
- Physics & Mathematics (AREA)
- Fuzzy Systems (AREA)
- Mathematical Physics (AREA)
- Air Conditioning Control Device (AREA)
Abstract
The invention discloses a method for realizing building heating and ventilation automatic control based on a deep learning graph neural network, which comprises the following steps: the input end obtains data: acquiring the number of people and the flowing condition of each room by using a monitoring camera, and simultaneously acquiring the area of each room, the number of air vents and the ventilation speed data from a whole building model or a single-layer plan, wherein the data are used as input characteristics to provide basic data for a subsequent graph neural network model; the beneficial effects of the invention are as follows: dynamic prediction and adjustment of internal environment parameters of a building are realized through the deep learning graph neural network, and the intelligent degree of a heating ventilation control system is improved; the influence of the number of people and the flow condition on the heating and ventilation requirements is considered, so that the control system meets the actual requirements better; the unified management of the whole building or a single-layer plan is realized, and the overall efficiency of the heating ventilation control system is improved.
Description
Technical Field
The invention belongs to the technical field of green buildings, and particularly relates to a method for realizing building heating and ventilation automatic control based on a deep learning map neural network.
Background
Indoor air quality (indoor air quality, IAQ) refers to the content of various components in the air inside and around a building and its structure and is an important indicator for indicating environmental health and proper occupancy. The building needs to reduce the indoor air pollution level to the allowable limit through ventilation to improve the indoor air quality, and the necessary external air quantity required to be introduced is the necessary ventilation quantity. Wherein the air pollutants comprise carbon dioxide, carbon monoxide, nitrogen oxides, sulfur oxides, ozone, radon, organic gases, odors, airborne particulates, etc., and the criteria thereof are as the air Pollutant Standard (PSI)。
In places where people use, the carbon dioxide concentration is one of the most commonly used evaluation indexes for measuring the indoor air quality, and along with the popularization of the mode of totally-enclosed centralized ventilation depending on a central air conditioner, the energy-saving and accurate regulation of ventilation to the necessary ventilation amount is an essential majority of modern buildings; however, the traditional building heating ventilation is completely dependent on manual regulation, and the temperature and ventilation of the central air conditioner of many large buildings are constant throughout the year, which cannot meet the basic standard of indoor air quality, and causes waste and maldistribution of energy sources; on the other hand, generating an countermeasure network (Generative adversarial networks, abbreviated GAN) is a new field in machine learning research, data is learned by simulating animal neuron transfer modes, including images, sounds, texts, etc., however, generating an countermeasure network in the building field is limited to learning and filling of images, generating an countermeasure network based on graph-based generation (graph-based generation) is rarely applied in the building field at present, and further, in a graph-based generation (graph-based generation) method, a graph neural network (Graph neural network, abbreviated gn) can be applied to non-euclidean space as compared with a more general convolutional neural network (convolutional neural network, abbreviated CNN), so the application range is wider and more flexible, and thus can be applied to more complex and various space types in the building field, and at the same time, a technique for regulating building ventilation in the intelligent green building field is not yet applied.
Disclosure of Invention
The invention aims to provide a method for realizing building heating and ventilation automatic control based on a deep learning map neural network, which predicts and adjusts the concentration and temperature change of carbon dioxide or other air pollutants in each room in real time through the map neural network according to the quantity and flow condition of personnel monitoring each room in a building, and realizes intelligent green building in the heating and ventilation direction.
In order to achieve the above purpose, the present invention provides the following technical solutions: a method for realizing building heating and ventilation automatic control based on a deep learning graph neural network comprises the following steps:
the input end obtains data: acquiring the number of people and the flowing condition of each room by using a monitoring camera, and simultaneously acquiring the area of each room, the number of air vents and the ventilation speed data from a whole building model or a single-layer plan, wherein the data are used as input characteristics to provide basic data for a subsequent graph neural network model;
constructing a graph neural network model: using the above-described input features, in combination with building internal structure and environmental parameters, a graphic neural network model is constructed that can learn and simulate the heat conduction and air flow processes inside the building, thereby predicting carbon dioxide or other air contaminant concentration and temperature changes in each room;
training a graph neural network model: training the graph neural network model by utilizing historical data, and accurately predicting the concentration and temperature change of carbon dioxide or other air pollutants in each room by adjusting model parameters;
output end regulation control: and predicting and regulating the concentration and temperature change of carbon dioxide or other air pollutants in each room in real time through a graph neural network model, and regulating the ventilation quantity according to the prediction result so as to maintain the comfort level of the indoor environment and meet the energy-saving requirement.
As a preferable technical scheme of the invention, the specific method for acquiring the number of people and the flow condition of each room by using the monitoring camera comprises the following steps:
a monitoring camera is arranged in the building to cover the entrance and the exit of each room;
analyzing and processing the video by utilizing image processing and computer vision technology through the video stream captured by the monitoring camera;
the video is monitored and analyzed in real time by using a personnel flow statistics system, so that the personnel number and the flow condition of each room are obtained;
automatically analyzing the monitoring video by using an intelligent analysis technology;
and deep mining is carried out on the monitoring video through a time sequence analysis and pattern recognition technology.
As a preferred technical solution of the present invention, a specific method for analyzing and processing video using image processing and computer vision techniques includes the steps of: preprocessing the acquired video, including denoising, image enhancement and color correction, so as to improve the quality and definition of the video; detecting and tracking a target in the video by using the R-CNN, and obtaining a motion trail and behavior analysis of the target; separating a foreground from a background in the video through an image segmentation technology; identifying and classifying actions in the video by using the 3D CNN; and encoding and decoding the video.
As a preferable technical scheme of the invention, the method for acquiring the data of the area of each room, the number of the air vents and the ventilation speed from the whole building model or the single-layer plan view comprises the following steps:
obtaining a building model or a single-layer plan of a building from the hands of an architect or designer;
measuring and calculating the area of each room, the number of air vents and the ventilation speed parameters on the building model or the single-layer plan by using a measuring tool;
recording the measured and calculated data in a table or a database, and performing data processing and analysis;
and (3) establishing a data model by using the measured and calculated data, and realizing the prediction and simulation of the internal environment parameters of the building.
As a preferable technical scheme of the invention, the specific method for constructing the graph neural network model is as follows: the method comprises the steps of obtaining the number of personnel and the flowing condition of each room through a monitoring camera, and obtaining the area of each room, the number of air vents and the ventilation speed data from a whole building model or a single-layer plan view to obtain the internal structure and the environmental parameters of a building; cleaning, preprocessing and labeling the collected data; and inputting the preprocessed data into the graph neural network model.
As a preferable technical scheme of the invention, a gradient descent and random gradient descent optimization algorithm is used for model training.
As a preferable technical scheme of the invention, the system also comprises a recording module, and the concentration and the temperature of carbon dioxide or other air pollutants in each room are recorded through the whole process of the recording module.
As a preferable technical scheme of the invention, the specific method for recording the concentration and the temperature of carbon dioxide or other air pollutants in each room in the whole process through the recording module is as follows:
a carbon dioxide or other air contaminant concentration and temperature sensor network covering each room;
recording the data by a recording module;
data analysis tools are used to analyze the data.
Compared with the prior art, the invention has the beneficial effects that:
dynamic prediction and adjustment of internal environment parameters of a building are realized through the deep learning graph neural network, and the intelligent degree of a heating ventilation control system is improved;
the influence of the number of people and the flowing condition on the heating and ventilation requirements is considered, so that the control system is more in line with the actual requirements and is suitable for indoor environment standards of various places;
the unified management of the whole building or a single-layer plan is realized, and the overall efficiency of the heating ventilation control system is improved;
the system provides more refined and personalized heating and ventilation service, and improves the comfort level of users and the energy utilization efficiency;
meanwhile, the method can be widely used for precisely regulating and controlling any one or a plurality of indoor air pollutants simultaneously by changing the types and the numerical standards of the sensors, and can be suitable for any environmental air quality standard; therefore, the system can be widely used for controlling the exhaust gas amount and air pollutants in factories and other places besides places for human use.
Drawings
FIG. 1 is a flow chart of a method for realizing building heating and ventilation automatic control according to the invention.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
Example 1
Referring to fig. 1, a first embodiment of the present invention provides a method for realizing automatic heating and ventilation control in places with frequent indoor activity changes, such as kindergarten, school, office building, etc., based on a neural network of a deep learning map, comprising the following steps:
the input end obtains data: acquiring the personnel number, the flowing condition and the activity state of each room by using a monitoring camera, and simultaneously acquiring the area, the number of air vents and the ventilation speed data of each room from a whole kindergarten, school or office building model or a single-layer plan, wherein the data are used as input characteristics to provide basic data for a subsequent graphic neural network model;
constructing a graph neural network model: by using the input characteristics and combining internal structures and environmental parameters of kindergarten, school or office building buildings, a graphic neural network model is constructed, and the model can learn and simulate the heat conduction and air flow processes in the buildings so as to predict the carbon dioxide concentration and temperature change of each room;
training a graph neural network model: training the graph neural network model by utilizing historical data, and accurately predicting the carbon dioxide concentration and the temperature change of each room by adjusting model parameters;
output end regulation control: the carbon dioxide concentration and the temperature change of each room are predicted and regulated in real time through a graphic neural network model, and the ventilation quantity is regulated according to the prediction result so as to keep the comfort level of the indoor environment (the carbon dioxide concentration is controlled within the range of 700 or 1000ppm, and the indoor temperature is controlled within the indoor thermal comfort index range of PMV-0.5 to 0.5 specified by the International organization for standardization ISO) and meet the energy saving requirement。
Example 2
Referring to fig. 1, a method for implementing floor heating ventilation self-control of a factory and the like based on a deep learning map neural network is provided in a second embodiment of the present invention, and includes the following steps:
the input end obtains data: acquiring the number of people and the flowing condition of each room of a factory by using a monitoring camera, acquiring the content of waste gas of each workshop of the factory by using a sensor, and acquiring the area of each room, the number of air vents and the ventilation speed data from a whole building model or a single-layer plan, wherein the data are used as input characteristics to provide basic data for a subsequent graphic neural network model;
constructing a graph neural network model: using the above-mentioned input features, in combination with the internal structure of the building and environmental parameters, constructing a graphic neural network model which can learn and simulate the heat conduction and air flow processes in the interior of the building so as to predict the air pollutants and temperature changes of each room;
training a graph neural network model: training the graphic neural network model by utilizing historical data, and accurately predicting air pollutants and temperature changes of each room by adjusting model parameters;
output end regulation control: through the graphic neural network model, air pollutants and temperature changes of each room and workshop of the factory are predicted and regulated in real time, and the ventilation quantity is regulated according to the prediction result, so that the comfort level of the indoor environment is kept, the waste gas concentration is prevented from being too high, and the energy-saving requirement is met.
In this embodiment, preferably, the specific method for acquiring the number of people and the flow condition in each room by using the monitoring camera includes the following steps: a monitoring camera is arranged in the building to cover the entrance and the exit of each room so as to capture the flowing condition of personnel; analyzing and processing the video by utilizing image processing and computer vision technology through the video stream captured by the monitoring camera; the video is monitored and analyzed in real time by using special equipment such as a personnel flow statistics system and the like, so that the personnel number and the flow condition of each room are obtained; automatically analyzing the monitoring video by using an intelligent analysis technology, detecting and identifying personnel by using human body detection and face recognition technologies, and calculating the number of the personnel; in addition, the monitoring video can be deeply mined through time sequence analysis and pattern recognition technology, for example, the rule and trend of personnel flow are analyzed; the specific method for analyzing and processing the video by utilizing the image processing and computer vision technology comprises the following steps: video preprocessing: preprocessing the acquired video, including denoising, image enhancement and color correction operation, so as to improve the quality and definition of the video; target detection and tracking: detecting and tracking targets in the video by using a computer vision algorithm, such as YOLO and fast R-CNN, and obtaining motion trail and behavior analysis of the targets; object segmentation: separating a foreground from a background in the video through an image segmentation technology so as to facilitate the subsequent extraction and behavior analysis of specific targets; and (3) action recognition: recognizing and classifying actions in the video, such as human body posture estimation, by utilizing computer vision and deep learning algorithms, such as Two-Stream CNN and 3D CNN, so as to accurately calculate oxygen consumption by distinguishing different actions and behaviors; video encoding and decoding: the video is encoded using common codec standards such as h.264, h.265, and more efficient video codec algorithms such as AV1.
In this embodiment, preferably, the data of the area of each room, the number of ventilation openings and the ventilation speed are obtained from the whole building model or a single-layer plan view, and the specific method includes the following steps:
obtaining a building model or a single-layer plan of a building from the hands of an architect or designer;
measuring and calculating the area of each room, the number of air vents and the ventilation speed parameters on the building model or the single-layer plan by using a measuring tool; the measured and calculated data are recorded in a table or a database, and necessary data processing and analysis are performed, for example, the average area, the number of air vents and the ventilation speed of each room can be counted, and a data model, such as a linear regression model and a support vector machine model, is established by using the measured and calculated data, so as to realize the prediction and simulation of the internal environment parameters of the building.
In this embodiment, preferably, the specific method for constructing the graph neural network model is as follows: the method comprises the steps of obtaining the number of personnel and the flowing condition of each room through a monitoring camera, and obtaining the area of each room, the number of air vents and the ventilation speed data from a whole building model or a single-layer plan view to obtain the internal structure and the environmental parameters of a building; cleaning, preprocessing and labeling the collected data; and inputting the preprocessed data into the graph neural network model.
In this embodiment, it is preferable to perform model training using a gradient descent, random gradient descent optimization algorithm.
In this embodiment, it is preferable to further include a recording module, through which the concentration and temperature of carbon dioxide or other air pollutants in each room are recorded in the whole course.
In this embodiment, preferably, the specific method for recording the concentration and temperature of carbon dioxide or other air pollutants in each room through the recording module is as follows:
first, a network of sensors for carbon dioxide or other air contaminant concentration and temperature is required to cover each room. The network can be composed of a plurality of carbon dioxide or other air pollutant concentration sensors and temperature sensors, and each sensor can monitor and record the carbon dioxide or other air pollutant concentration and temperature of a room in real time;
these data may then be recorded by a recording module, which may be a data storage device, such as a database or a data storage server, which receives and stores data for each sensor, including carbon dioxide or other air contaminant concentrations and temperatures for each room;
finally, a data analysis tool, which may be a specialized data analysis software or a machine learning algorithm, may be used to analyze the data, through which the trend of changes in carbon dioxide concentration or other air pollutants and temperature, and the correlation between them, of each room may be known.
Control experiment 1 (empty control group): no ventilation measures are taken during the same time, scene and personnel flow, and the concentration and temperature change of carbon dioxide or other air pollutants in each room are recorded.
Control experiment 2 (natural ventilation control group): the ventilation quantity is adjusted in the same time, scene and personnel flowing process depending on subjective feelings of a user, and the concentration and temperature change of carbon dioxide or other air pollutants in each room and the energy consumption condition are recorded. The control experiment is used for verifying that the invention can achieve the minimum energy loss and simultaneously ensure that the concentration and the temperature of the carbon dioxide or other air pollutants in the space are kept within the standard interval.
While embodiments of the present invention have been shown and described in detail with reference to the foregoing detailed description, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations may be made therein without departing from the principles and spirit of the invention, the scope of which is defined in the appended claims and their equivalents.
Claims (8)
1. A method for realizing building heating and ventilation automatic control based on a deep learning graph neural network is characterized by comprising the following steps of: the method comprises the following steps:
the input end obtains data: acquiring the number of people and the flowing condition of each room by using a monitoring camera, and simultaneously acquiring the area of each room, the number of air vents and the ventilation speed data from a whole building model or a single-layer plan, wherein the data are used as input characteristics to provide basic data for a subsequent graph neural network model;
constructing a graph neural network model: using the above-described input features, in combination with building internal structure and environmental parameters, a graphic neural network model is constructed that can learn and simulate the heat conduction and air flow processes inside the building, thereby predicting carbon dioxide or other air contaminant concentration and temperature changes in each room;
training a graph neural network model: training the graph neural network model by utilizing historical data, and accurately predicting the concentration and temperature change of carbon dioxide or other air pollutants in each room by adjusting model parameters;
output end regulation control: and predicting and regulating the concentration and temperature change of carbon dioxide or other air pollutants in each room in real time through a graph neural network model, and regulating the ventilation quantity according to the prediction result so as to maintain the comfort level of the indoor environment and meet the energy-saving requirement.
2. The method for realizing building heating and ventilation automatic control based on the deep learning map neural network according to claim 1, which is characterized in that: the specific method for acquiring the number of people and the flow condition of each room by using the monitoring camera comprises the following steps:
a monitoring camera is arranged in the building to cover the entrance and the exit of each room;
analyzing and processing the video by utilizing image processing and computer vision technology through the video stream captured by the monitoring camera;
the video is monitored and analyzed in real time by using a personnel flow statistics system, so that the personnel number and the flow condition of each room are obtained;
automatically analyzing the monitoring video by using an intelligent analysis technology;
and deep mining is carried out on the monitoring video through a time sequence analysis and pattern recognition technology.
3. The method for realizing building heating and ventilation automatic control based on the deep learning map neural network according to claim 2, which is characterized in that: the specific method for analyzing and processing the video by utilizing the image processing and computer vision technology comprises the following steps: preprocessing the acquired video, including denoising, image enhancement and color correction, so as to improve the quality and definition of the video; detecting and tracking a target in the video by using the R-CNN, and obtaining a motion trail and behavior analysis of the target; separating a foreground from a background in the video through an image segmentation technology; identifying and classifying actions in the video by using the 3D CNN; and encoding and decoding the video.
4. The method for realizing building heating and ventilation automatic control based on the deep learning map neural network according to claim 1, which is characterized in that: the specific method for acquiring the data of the area of each room, the number of the air vents and the ventilation speed from the whole building model or the single-layer plan view comprises the following steps of:
obtaining a building model or a single-layer plan of a building from the hands of an architect or designer;
measuring and calculating the area of each room, the number of air vents and the ventilation speed parameters on the building model or the single-layer plan by using a measuring tool;
recording the measured and calculated data in a table or a database, and performing data processing and analysis;
and (3) establishing a data model by using the measured and calculated data, and realizing the prediction and simulation of the internal environment parameters of the building.
5. The method for realizing building heating and ventilation automatic control based on the deep learning map neural network according to claim 1, which is characterized in that: the specific method for constructing the graph neural network model is as follows: the method comprises the steps of obtaining the number of personnel and the flowing condition of each room through a monitoring camera, and obtaining the area of each room, the number of air vents and the ventilation speed data from a whole building model or a single-layer plan view to obtain the internal structure and the environmental parameters of a building; cleaning, preprocessing and labeling the collected data; and inputting the preprocessed data into the graph neural network model.
6. The method for realizing building heating and ventilation automatic control based on the deep learning map neural network according to claim 1, which is characterized in that: model training is performed by using a gradient descent and random gradient descent optimization algorithm.
7. The method for realizing building heating and ventilation automatic control based on the deep learning map neural network according to claim 1, which is characterized in that: the system also comprises a recording module, wherein the recording module is used for recording the concentration and the temperature of carbon dioxide or other air pollutants in each room in the whole process.
8. The method for realizing building heating and ventilation automatic control based on the deep learning map neural network according to claim 7, wherein the method comprises the following steps: the specific method for recording the concentration and the temperature of the carbon dioxide or other air pollutants in each room in the whole process through the recording module is as follows:
a carbon dioxide or other air contaminant concentration and temperature sensor network covering each room;
recording the data by a recording module;
data analysis tools are used to analyze the data.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311807670.XA CN117606106A (en) | 2023-12-26 | 2023-12-26 | Method for realizing building heating and ventilation automatic control based on deep learning graph neural network |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311807670.XA CN117606106A (en) | 2023-12-26 | 2023-12-26 | Method for realizing building heating and ventilation automatic control based on deep learning graph neural network |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117606106A true CN117606106A (en) | 2024-02-27 |
Family
ID=89958068
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311807670.XA Pending CN117606106A (en) | 2023-12-26 | 2023-12-26 | Method for realizing building heating and ventilation automatic control based on deep learning graph neural network |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117606106A (en) |
-
2023
- 2023-12-26 CN CN202311807670.XA patent/CN117606106A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Tushar et al. | Internet of things for green building management: disruptive innovations through low-cost sensor technology and artificial intelligence | |
| CN105190191B (en) | Energy-saving heating, ventilation, air-conditioner control system | |
| CN110377084B (en) | Building indoor environment regulation and control method based on intelligent control strategy | |
| KR20150112118A (en) | System for providing forcast of indor air quality in large building | |
| CN116681250A (en) | Building engineering progress supervisory systems based on artificial intelligence | |
| KR20170122043A (en) | Real-time indoor air quality outlier smoothing method and apparatus | |
| CN114858976A (en) | Intelligent analysis method and system for atmospheric quality of industrial park | |
| Ni et al. | Improving energy efficiency while preserving historic buildings with digital twins and artificial intelligence | |
| CN117354986B (en) | Intelligent control method and system for multifunctional LED lamp beads | |
| CN115272675A (en) | Energy management system and method based on multi-sensor information fusion | |
| CN117606106A (en) | Method for realizing building heating and ventilation automatic control based on deep learning graph neural network | |
| CN117029236A (en) | Clean room environment control method and system | |
| CN112381213A (en) | Industrial equipment residual life prediction method based on bidirectional long-term and short-term memory network | |
| KR102619200B1 (en) | Method and computer program for creating a neural network model that automatically controls environmental facilities based on artificial intelligence | |
| CN117128624A (en) | Control method and device for fresh air conditioner and air conditioner management system and storage medium | |
| CN109858576A (en) | The gradual self feed back concentration Entropy Changes prediction technique of gas, system and storage medium | |
| CN109657648B (en) | System and method for monitoring office building window opening and closing conditions in real time | |
| CN114912169B (en) | Industrial building heat supply autonomous optimization regulation and control method based on multisource information fusion | |
| Rastogi et al. | An IoT-based system to evaluate indoor air pollutants using grey relational analysis | |
| CN114298973A (en) | Intelligent heat supply monitoring method based on infrared image segmentation | |
| Chaudhary et al. | Predicting the performance of hybrid ventilation in buildings using a multivariate attention-based biLSTM Encoder–Decoder | |
| Lee et al. | Human-Centric Artificial Intelligence of Things–Based Indoor Environment Quality Modeling Framework for Supporting Student Well-Being in Educational Facilities | |
| Ranjitha et al. | Intensification in Internet of Robotic Things (IoRT) on Diverse | |
| CN118070664A (en) | Building energy consumption intelligent simulation method based on Internet of things and BIM | |
| CN114580798B (en) | Device point location prediction method and system based on transformer |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |