CN108413588B

CN108413588B - A Personalized Air Conditioning Control System and Method Based on Thermal Imaging and BP Neural Network

Info

Publication number: CN108413588B
Application number: CN201810144369.8A
Authority: CN
Inventors: 简毅文; 侯雨晨; 常小艳
Original assignee: Beijing University of Technology
Current assignee: Beijing University of Technology
Priority date: 2018-02-12
Filing date: 2018-02-12
Publication date: 2021-03-02
Anticipated expiration: 2038-02-12
Also published as: CN108413588A

Abstract

The present invention provides a personalized air conditioning control system and method based on infrared thermal imaging technology and BP neural network method. The infrared thermal imaging module is used to collect, process and transmit thermal imaging data; the information processing module is used to receive the information from the human-computer interaction module and the infrared thermal imaging module, and calculate and optimize through the BP neural network method The control parameters are transmitted to the air conditioner control module; the air conditioner control module is used to receive the transmission signal and realize the control of the air conditioner. By adopting the technical scheme of the present invention, the operation of the air-conditioning system can meet the dynamic thermal demands of the indoor environment of many people to the greatest extent.

Description

Personalized air conditioner control system and method based on thermal imaging and BP neural network

Technical Field

The invention belongs to the technical field of air conditioners, and particularly relates to a personalized air conditioner control system and method based on thermal imaging and a BP neural network.

Background

In order to control indoor hot and humid environment to meet the thermal comfort requirement of human body, air conditioning systems are widely applied. The design and operation of air conditioning systems is generally intended to create a uniform and stable indoor hot and humid environment. However, recent studies have found that even in a steady-state air-conditioning environment, human physiological parameters and human thermal evaluation change dynamically with time, and accordingly, the thermal demand of the human body for the indoor environment is not stabilized in a specific state. Thus, for air conditioning systems where the environment is not controllable, a series of problems arise due to the inability to provide users with the opportunity to feedback and interact with the environmental system. For example: in a teaching building provided with a central air conditioning system, the classroom temperature can not be adjusted by indoor personnel generally, and the situation that the indoor environment is overheated and the like when more indoor personnel exist is likely to occur, so that the learning efficiency of students is reduced, and energy waste is generated. For the air conditioning environment with changeable room temperature, the environmental control parameters of the system are often set by a certain user according to the comfort level of the user, the individual difference between people is ignored, and the air conditioning requirements of other people in the room can not be completely met, so that a series of negative effects are generated. For example: in office buildings, the primary air conditioner temperature setting value is too high or too low, which can cause secondary adjustment behaviors of staff to the air conditioner or clothes increase and decrease behaviors, not only affects working efficiency and generates unnecessary energy consumption, but also leads the staff to complain about psychology due to discomfort and reduces satisfaction rate.

The infrared thermal imaging technology is a comprehensive technology for receiving and processing infrared radiation emitted by an object, displaying the infrared radiation in a digital signal mode and utilizing the infrared radiation, and is widely applied to various fields of industry, agriculture, medical treatment and the like. When the infrared thermal imaging technology is applied to the technical field of building environment control, temperature values of different surfaces can be accurately obtained. With the development of the infrared thermal imaging technology, air conditioning systems based on infrared thermal imaging appear, and the air conditioning systems can accurately obtain the temperature value of the inner surface of an enclosure structure or the skin temperature value of a local part of a human body and adjust the refrigerating and heating quantity of an indoor tail end air conditioner, so that the control of the indoor environment can be linked with human physiological parameters, but the influence and the effect of human subjective thermal evaluation are not taken into consideration, the feedback and the interaction between a user and an environment system can not be realized, and the individual difference of the human subjective thermal environment evaluation is ignored.

The BP neural network is one of artificial neural networks, and is characterized in that a multi-layer neural network is trained by adopting an error back propagation algorithm, and the BP neural network is one of the most widely applied neural networks. By adopting the BP neural network algorithm and learning a certain rule through self-training, the result of obtaining the optimal output value when an input value is given can be realized, and the method has the advantages of large-scale parallel, distributed processing, self-organization, self-learning and the like. When the method is applied to the heating ventilation air-conditioning control field, the human body comfort degree can be accurately and efficiently predicted, and the automatic control level of the system is improved.

Disclosure of Invention

The invention aims to: the invention provides a personalized air conditioner control system and a method thereof based on an infrared thermal imaging technology and a BP neural network method, aiming at the problems caused by insufficient information feedback and inflexible control of the traditional air conditioner system and the existing air conditioner system applying thermal imaging, so that the operation of the air conditioner system can meet the dynamic heat requirement of multiple people on the indoor environment to the maximum extent.

In order to achieve the purpose of the invention, the technical scheme adopted by the invention is as follows:

a personalized air conditioner control system based on thermal imaging and BP neural network comprises: the human-computer interaction module is used for acquiring initial and thermal comfort feedback information of each user; the infrared thermal imaging module is used for acquiring and processing thermal imaging information; the information processing module is used for receiving information from the man-machine interaction module and the infrared thermal imaging module and calculating to obtain control parameters of the optimized air conditioning system through a BP neural network method; the air conditioner control module is used for receiving the transmission signal and realizing the control of the air conditioning system and the terminal equipment thereof.

Preferably, the gender, age, height, weight, habit taking, constitution, office position and TSV voting parameters of each user are collected by means of a mobile phone APP or computer software.

Preferably, the information processing module determines skin temperature data points for different users based on a collected analysis of the respective user data.

Preferably, the infrared thermal imaging module is used for generating human body thermal imaging information and converting the human body thermal imaging information into temperature data.

A control method of a personalized air conditioning system based on thermal imaging and BP neural network comprises the following steps:

s1, regularly inputting and updating personal basic information by a user, and performing real-time feedback of thermal comfort, wherein the input information is used for calculating corresponding evaluation indexes such as BMI and guiding infrared thermal imaging data acquisition;

step S2, the infrared thermal imaging module collects thermal imaging data;

step S3, converting the infrared thermography and the temperature field data;

step S4, extracting temperature data points aiming at different users from the temperature field data according to the user input information; step S5, carrying out BP neural network training on the input layer data by taking a user as a unit;

step S6, obtaining a control strategy of the air conditioning system through data analysis;

and step S7, the air conditioning system automatic controller receives the control signal from the information processing module and automatically controls the air conditioning system terminal equipment.

Preferably, in step S4, different human skin temperature monitoring points are selected for different users according to the gender and the habit of the user.

Preferably, in the BP neural network training in step S5, the skin test point temperature, sex, BMI, habit and physical parameters are used as input layer parameters, the number of neurons in the hidden layer that converges in the training is designed, the TSV index is used as an output layer parameter, and error correction is performed by using TSV data provided by a human-computer interaction system.

Preferably, the specific process of step S6 is:

step S61, the air conditioning system is started by taking the indoor design temperature in the specification as the initial set temperature;

step S62, the terminal air conditioning equipment acts and adjusts the air conditioning system according to the initial temperature set value;

step S63, obtaining a TSV feedback data group through a BP neural network method;

step S64, judging whether the TSV data group meets the thermal comfort requirement of the preset crowd probability;

step S65, if the set thermal comfort requirement is met, the debugging process is ended;

step S66, the debugging result is fed back to a human-computer interaction interface in the form of electronic signals;

and step S67, adjusting the environmental control parameters of the air conditioning system terminal equipment through the judgment of the TSV feedback data set, and returning to the step S62.

The personalized air conditioner control system and method based on thermal imaging and BP neural network have the advantages that:

the intelligent control of the air conditioning system realized by the air conditioning control system and the method improves the problem that the setting of the air conditioning system is not matched with the requirement of a user due to neglecting individual difference in a multi-user environment, and the problem that personalized parameter selection based on human-computer interaction is lacked in the existing air conditioning system applying thermal imaging. Therefore, the individual requirements of the human body in a multi-person space on the hot and humid environment can be met to the maximum extent, and unnecessary energy consumption is reduced.

Drawings

FIG. 1 is a functional block diagram of a thermal imaging and BP neural network based personalized air conditioner control system and method according to the present invention

FIG. 2 is a method schematic diagram of a thermal imaging and BP neural network based personalized air conditioning control system and method according to the present invention

FIG. 3 is a BP neural network method schematic diagram of a thermal imaging and BP neural network based personalized air conditioner control system and method according to the present invention

FIG. 4 is a control flow diagram of an air conditioning system of the personalized air conditioning control system and method based on thermal imaging and BP neural network according to the present invention

Detailed Description

The following describes a personalized air conditioning control system and method based on thermal imaging and BP neural network, with reference to the accompanying drawings and examples. As shown in fig. 1, an individualized air conditioner control system based on thermal imaging and BP neural network mainly includes the following four functional modules: the system comprises a human-computer interaction module, an infrared thermal imaging module, an information processing module and an air conditioner control module. The control logic of each functional module during the operation of the system is shown in fig. 1, but the interrelation between different modules is not limited to fig. 1. The man-machine interaction module realizes data acquisition of each user by relying on mobile phone APP or computer software of indoor personnel, and the acquired data comprise: sex, age, height, weight, habitual clothes, physique, office position, human body thermal comfort feedback information and the like are used as conditions for selecting an air conditioner terminal operation control mode and a basis for extracting infrared thermal imaging data; the infrared thermal imaging module generates an infrared thermal image by monitoring indoor personnel; after receiving and preprocessing the data of the human-computer interaction module and the infrared thermal imaging module, the information processing module analyzes and learns the processed data by adopting a BP neural network training method to obtain an optimized control parameter; the air conditioner control module receives the control instruction from the information processing module, takes control action on terminal equipment of the air conditioner system, and achieves automatic control on the air conditioning environment.

The invention also provides a control method of the personalized air conditioning system based on the thermal imaging and the BP neural network, which comprises the following steps as shown in figure 2. Step S1, in the operation process of a personalized air conditioning system based on thermal imaging and BP neural network, firstly, a user needs to regularly input and update personal basic information (including sex, age, height, weight, uniform, physical parameters, office position and the like) and perform real-time feedback of thermal comfort, and the input information is used for calculating corresponding evaluation indexes such as BMI and the like and guiding infrared thermal imaging data acquisition; step S2, the infrared thermal imaging module collects thermal imaging data; step S3, converting the infrared thermography and the temperature field data; step S4, extracting temperature data points aiming at different users from the temperature field data according to the user input information; step S5, carrying out BP neural network training on the input layer data by taking a user as a unit; step S6, obtaining a control strategy of the air conditioning system through data analysis; and step S7, the air conditioning system automatic controller receives the control signal from the information processing module and automatically controls the air conditioning system terminal equipment.

The steps S4, S5, and S6 are integrated in the information processing module by computer programming.

The selection manner of the temperature data points for different users in step S4 is as follows, after the information processing module receives the information from the human-computer interaction module, the information processing module determines the required skin temperature data points according to factors such as sex and habit of the user, and the computer program converts the infrared thermography into temperature field data to extract the skin temperature data points for different users.

The principle of the BP neural network method described in step S5 is as shown in fig. 3. The specific implementation mode is that the BP neural network method obtains the TSV prediction data of the output layer meeting all users through error feedback training of data of the input layer. For a personalized air conditioning system control method based on thermal imaging and BP neural network, skin measuring point temperature, gender, BMI, habitual clothes and physical parameters are selected as weighted evaluation parameters of an input layer, the number of neurons of a hidden layer for convergence of training is set, a prediction result is compared with thermal comfort feedback data extracted by a human-computer interaction module, and reverse transmission of errors is carried out, so that the weight of the input layer is corrected. After a period of training, the user thermal comfort feedback step in step S1 can be omitted, thereby achieving highly automated control of the air conditioning system.

The air conditioning system control strategy in step S6 is shown in fig. 4, and the specific process includes: step S61, the air conditioning system is started by taking the indoor design temperature in the specification as the initial set temperature; s62, the air conditioning equipment at the tail end operates, and the air conditioning system is adjusted according to the initial temperature set value; s63, obtaining a TSV feedback data group through a BP neural network method; s64, judging whether the TSV data group meets the thermal comfort requirement of the preset crowd probability; s65, if the set thermal comfort requirement is met, ending the debugging process; s66, feeding back the debugging result to the human-computer interaction interface in the form of electronic signals; and S67, adjusting the environmental control parameters of the air conditioning system terminal equipment through the judgment of the TSV feedback data set, and returning to the step S62.

The invention discloses a personalized air conditioner control system and method based on thermal imaging and a BP neural network. The invention mainly solves the problem that the setting of the air conditioning system is not matched with the requirement of a user due to neglecting individual difference in a multi-person environment, and the problem that personalized parameter selection based on human-computer interaction is lacked in the existing air conditioning system applying thermal imaging. The invention has the advantages that the individual requirements of a human body in a multi-person space on a hot and humid environment are met to the maximum extent by realizing the intelligent control of the air conditioning system, and unnecessary energy consumption is reduced.

The above embodiments are merely illustrative of the present invention and are not intended to limit the present invention. Although the embodiments of the present invention have been described, those skilled in the art may make changes, modifications, substitutions and alterations to the embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A control method of a personalized air conditioning system based on thermal imaging and BP neural network is disclosed, the system for realizing the method comprises: the human-computer interaction module is used for acquiring initial information and thermal comfort feedback information of users of a multi-user group; the infrared thermal imaging module is used for acquiring and processing thermal imaging information; the information processing module is used for receiving information from the man-machine interaction module and the infrared thermal imaging module and calculating to obtain control parameters of the optimized air conditioning system through a BP neural network method; the air conditioner control module is used for receiving the transmission signal and realizing the control of the air conditioner system and the terminal equipment thereof;

collecting sex, age, height, weight, clothes, physique, office position and TSV voting parameters of users of a multi-user group by means of a mobile phone APP or computer software;

the information processing module determines skin temperature data points for different users in the multi-crowd users based on the acquisition and analysis of the multi-crowd user data;

the method is characterized by comprising the following steps:

s1, regularly inputting and updating personal basic information by a user, and performing real-time feedback of thermal comfort, wherein the input information is used for calculating corresponding evaluation indexes of BMI and guiding infrared thermal imaging data acquisition;

step S2, the infrared thermal imaging module collects thermal imaging data;

step S3, converting the infrared thermography and the temperature field data;

step S4, extracting temperature data points aiming at different users from the temperature field data according to the user input information; step S5, carrying out BP neural network training on the input layer data by taking a user as a unit; in the BP neural network training in the step S5, the skin measuring point temperature, the sex, the BMI, the habit and the physique parameters are used as input layer parameters, the number of neurons in a hidden layer which enables training to be converged is designed, the TSV index is used as an output layer parameter, and error correction is performed by adopting TSV data provided by a human-computer interaction system;

step S6, obtaining a control strategy of the air conditioning system through data analysis; wherein, the specific flow of step S6 is:

step S67, adjusting the environmental control parameters of the air conditioning system terminal equipment through the judgment of the TSV feedback data set, and returning to the step S62

2. The method as claimed in claim 1, wherein the step S4 is performed by selecting different human skin temperature monitoring points for different users according to the gender and habit of the users.