CN116538654A - Self-adaptive space thermal environment intelligent control method - Google Patents
Self-adaptive space thermal environment intelligent control method Download PDFInfo
- Publication number
- CN116538654A CN116538654A CN202310819404.2A CN202310819404A CN116538654A CN 116538654 A CN116538654 A CN 116538654A CN 202310819404 A CN202310819404 A CN 202310819404A CN 116538654 A CN116538654 A CN 116538654A
- Authority
- CN
- China
- Prior art keywords
- human body
- space
- environment
- thermal
- model
- 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
- 238000000034 method Methods 0.000 title claims abstract description 54
- 230000007613 environmental effect Effects 0.000 claims abstract description 27
- 238000013528 artificial neural network Methods 0.000 claims abstract description 22
- 238000013507 mapping Methods 0.000 claims abstract description 17
- 238000004458 analytical method Methods 0.000 claims abstract description 15
- 239000011159 matrix material Substances 0.000 claims description 31
- 238000012549 training Methods 0.000 claims description 12
- 238000012545 processing Methods 0.000 claims description 10
- 238000012360 testing method Methods 0.000 claims description 8
- 230000007958 sleep Effects 0.000 claims description 7
- 238000000354 decomposition reaction Methods 0.000 claims description 5
- 230000003044 adaptive effect Effects 0.000 claims description 4
- 230000036760 body temperature Effects 0.000 claims description 4
- 230000014509 gene expression Effects 0.000 claims description 4
- 230000033001 locomotion Effects 0.000 claims description 4
- 230000037213 diet Effects 0.000 claims description 3
- 235000005911 diet Nutrition 0.000 claims description 3
- 230000003287 optical effect Effects 0.000 claims description 3
- 230000029058 respiratory gaseous exchange Effects 0.000 claims description 3
- 230000000007 visual effect Effects 0.000 claims description 3
- 238000012544 monitoring process Methods 0.000 abstract description 2
- 210000002569 neuron Anatomy 0.000 description 15
- 230000035807 sensation Effects 0.000 description 13
- 230000006399 behavior Effects 0.000 description 10
- 230000008859 change Effects 0.000 description 8
- 230000008569 process Effects 0.000 description 7
- 239000013598 vector Substances 0.000 description 7
- 238000004364 calculation method Methods 0.000 description 6
- 230000000694 effects Effects 0.000 description 6
- 238000004088 simulation Methods 0.000 description 5
- 230000009467 reduction Effects 0.000 description 4
- 238000004422 calculation algorithm Methods 0.000 description 3
- 230000006870 function Effects 0.000 description 3
- 238000012546 transfer Methods 0.000 description 3
- 230000006978 adaptation Effects 0.000 description 2
- 238000003062 neural network model Methods 0.000 description 2
- 230000004044 response Effects 0.000 description 2
- 230000003068 static effect Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000001816 cooling Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 230000004069 differentiation Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000010438 heat treatment Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 230000004060 metabolic process Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000036651 mood Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 238000003909 pattern recognition Methods 0.000 description 1
- 230000008447 perception Effects 0.000 description 1
- 230000037081 physical activity Effects 0.000 description 1
- 238000012706 support-vector machine 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
- F24F11/62—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
- F24F11/62—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
- F24F11/65—Electronic processing for selecting an operating mode
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Signal Processing (AREA)
- Physics & Mathematics (AREA)
- Fuzzy Systems (AREA)
- Mathematical Physics (AREA)
- Chemical & Material Sciences (AREA)
- Combustion & Propulsion (AREA)
- Mechanical Engineering (AREA)
- General Engineering & Computer Science (AREA)
- Air Conditioning Control Device (AREA)
Abstract
The invention relates to a self-adaptive space thermal environment intelligent control method, firstly, a space physical field three-dimensional environment model is established and reduced to obtain a space physical field real-time environment model; inputting a real-time environment model of the driving space physical field according to working conditions to perform prediction analysis; taking environmental parameters and human body parameters as input and real-time human body thermal comfort as output, and building a human body thermal comfort model by learning a mapping relation between an input sample and output feedback through a neural network; based on the human body thermal comfort model, predicting the human body thermal comfort of any position in space in real time; determining a spatial thermal environment control target based on the human thermal comfort and the spatial thermal load data; automatically performing system control based on the spatial thermal environment control target; the individual differences are identified through multiple monitoring means, the control targets are intelligently set, the predictive real-time control and adjustment of the environment control system are realized, the requirement of individual optimal environment comfort is met, and the power consumption cost is effectively reduced.
Description
Technical Field
The invention relates to the technical field of environmental control, in particular to a self-adaptive space thermal environment intelligent control method.
Background
Along with the improvement of life quality of people, the environmental comfort in the closed space directly influences the mood state of internal personnel, and different individuals have different heat sensation requirements. How to aim at individual variability, accurately and quickly adjust the environmental temperature of the closed space, and customize to realize optimal environmental control is an important requirement for improving life quality. In conventional temperature control systems, the temperature set point for temperature control is fixed, an equilibrium decision is made assuming that the space is static and that the internal personnel heat senses no difference from each other. However, the thermal sensation of each indoor person is individual variability and varies with external environmental conditions, physical characteristics of itself, and the state of personal activity, whereas current environmental control systems are not adaptive.
In addition, because of the large thermal inertia in the room, the conventional control parameters are manually given, excessive cooling air supply or heating may be used, energy waste is caused, and the time for reaching the stable control target value is long. And the stable output temperature of the temperature control system is reflected and is only fed back by a temperature sensor at an air outlet, the surface temperature of a human body after the propagation of the space environment cannot be directly represented, and the data cannot support the real-time judgment of the comfort of personnel, so that the environment control system is not timely and accurate.
Disclosure of Invention
The invention aims to overcome the defects of the technology, and provides a self-adaptive space environment intelligent control method, which is used for identifying individual differences through a plurality of monitoring means, intelligently setting control targets, realizing predictive real-time control and adjustment of an environment control system, meeting the requirement of individual optimal environment comfort and effectively reducing the power consumption cost.
In a first aspect, the present invention provides a method for intelligently controlling a self-adaptive space thermal environment, the method comprising the steps of: s1, establishing a space physical field three-dimensional environment model; s2, performing reduced-order processing on the three-dimensional environment model of the space physical field to obtain a real-time environment model of the space physical field; s3, driving the space physical field real-time environment model to conduct prediction analysis according to the working condition input, and obtaining space internal environment parameters and space thermal load data; s4, acquiring human parameters in the space; s5, taking the environmental parameters and the human body parameters as input and the real-time human body thermal comfort as output, and building a human body thermal comfort model by learning a mapping relation between an input sample and output feedback through a neural network; s6, predicting the human body thermal comfort level of any position in space in real time based on the human body thermal comfort level model; s7, determining a space thermal environment control target based on the human thermal comfort and space thermal load data; and S8, automatically performing system control based on the space thermal environment control target.
In some embodiments, the human body parameters are obtained by: and acquiring a human body image by using an optical camera, identifying the human body image, and evaluating to obtain the body appearance, age, wearing, sex, expression, sleep, diet and movement characteristics of the human body.
In some embodiments, the human body parameters are obtained by: and acquiring a thermal infrared image of a human body by using an infrared camera, and performing key point visual identification on the thermal infrared image by using a deep neural network to obtain the skin temperature of the human body.
In some embodiments, the human body parameters are obtained by: the temperature, heartbeat, respiration and sleep characteristics of the human body are acquired by using a wearable device or a sensing device on the seat armrest.
In some embodiments, the human body parameters include a human body temperature parameter and a human body characteristic parameter.
In some embodiments, the step S2 of reducing the order of the three-dimensional environmental model of the spatial physical field to obtain a real-time environmental model of the spatial physical field includes: s21, a space data sample set is established, and a data matrix is obtained based on the space data sample set; s22, carrying out principal component decomposition on the data matrix to obtain a principal component matrix and a residual error matrix; s23, processing and clearing the residual matrix, and establishing a mapping model of the principal element matrix and the physical field output to obtain a space physical field real-time environment model.
In some embodiments, the step S3 drives the spatial physical field real-time environment model to perform prediction analysis according to the working condition input, to obtain spatial environment parameters and spatial heat load data, including: s31, inputting different boundary conditions, obtaining the output of a real-time environment model of a space physical field, and establishing a training sample set; s32, establishing a multi-layer perceptron model according to the training sample set; s33, inputting actual working condition parameters, and obtaining space internal environment parameters and space heat load data through predictive analysis.
In some embodiments, the step S5 takes the environmental parameter and the human parameter as input and the real-time thermal comfort of the human body as output, learns a mapping relationship between an input sample and output feedback through a neural network, and establishes a model of the thermal comfort of the human body, including: s51, under a space environment test, collecting thermal feedback data of different people under different environments to construct a data sample set; s52, based on the data sample set, establishing a neural network by adopting a GRNN or SVM method; and S53, taking the environmental parameters and the human body parameters as input and the real-time human body thermal comfort as output, and building a human body thermal comfort model by learning the mapping relation between the input sample and the output feedback through a neural network.
In some embodiments, the human thermal comfort is categorized into 7 classes of very cold, slightly cold, moderate, slightly hot, very hot.
In some embodiments, the spatial environment control target comprises: the working mode selection, the selection of air outlets, and the switch, the temperature, the wind speed and the wind direction of each air outlet.
The technical scheme provided by the invention has the following beneficial effects:
1. according to the invention, a new acquisition mode and an analysis mode are adopted according to the perception data related to the comfort level of the human body, so that modeling is more in line with the real situation, and further a more accurate analysis result is obtained.
2. According to the invention, a three-dimensional field modeling method based on model price reduction and neural network mapping is adopted, so that real environment parameters of the position of a person can be accurately obtained while the real-time performance of simulation is ensured, and the environment parameters of the fixed air outlet in the past are not obtained.
3. According to the invention, the human body thermal comfort level prediction model based on machine learning is adopted, the self-adaptive network method is based on the improved control target setting method, the result of the method is more real-time and accurate, the behavior habit of a user can be learned, and the self-adaptive updating of the model is completed.
4. According to the invention, the human body thermal comfort level at the moment is calculated according to the human body surface temperature at the moment and the space environment temperature, the target environment temperature of the human surface in the three-dimensional field is calculated according to the human body thermal comfort level at the moment, then the set target temperature of the control system is obtained according to the target environment temperature calculation of the human body position, the control system performs work adjustment according to the set target temperature, and the new human body thermal comfort level after temperature adjustment is calculated. The state change of personnel can be tracked in real time, so that the control system target is automatically changed, and quick response is achieved. The instant heat sensation of the human body is used as a key ring to be added into the control link, the traditional constant value is not used as a setting target, the optimal control is adjusted in real time according to the internal and external changes, and the comfort of personnel is ensured.
It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the invention as claimed.
Detailed Description
The disclosure will now be discussed with reference to several exemplary embodiments. It should be understood that these embodiments are discussed only to enable those of ordinary skill in the art to better understand and thus practice the present disclosure, and are not meant to imply any limitation on the scope of the present disclosure.
As used herein, the term "comprising" and variants thereof are to be interpreted as meaning "including but not limited to" open-ended terms. The term "based on" is to be interpreted as "based at least in part on". The terms "one embodiment" and "an embodiment" are to be interpreted as "at least one embodiment. The term "another embodiment" is to be interpreted as "at least one other embodiment".
In conventional temperature control systems, the temperature set point for temperature control is fixed, an equilibrium decision is made assuming that the space is static and that the internal personnel heat senses no difference from each other. However, the thermal sensation of each indoor person is individual variability and varies with external environmental conditions, physical characteristics of itself, and the state of personal activity, whereas current environmental control systems are not adaptive.
Based on the above, the embodiment of the invention discloses a self-adaptive space thermal environment intelligent control method, which comprises the following steps: s1, establishing a space physical field three-dimensional environment model; s2, performing reduced-order processing on the space physical field three-dimensional environment model to obtain a space physical field real-time environment model; s3, inputting a real-time environment model of the driving space physical field according to working conditions to conduct prediction analysis, and obtaining space internal environment parameters and space thermal load data; s4, acquiring human parameters in the space; s5, taking environmental parameters and human body parameters as input and real-time human body thermal comfort as output, and building a human body thermal comfort model by learning a mapping relation between an input sample and output feedback through a neural network; s6, predicting the human body thermal comfort level of any position in the space in real time based on the human body thermal comfort level model; s7, determining a space thermal environment control target based on the human body thermal comfort level and the space thermal load data; s8, automatically performing system control based on the space thermal environment control target. The thermal sensation of each individual may vary and vary with various environmental conditions and their own state. Therefore, the key of intelligent control is to directly acquire characteristic key parameters from an individual instead of acquiring integral information (such as temperature, humidity and the like of air outlet and return air) in a space, so that the real-time thermal comfort of the individual is improved.
In some embodiments, the human body parameters are obtained by: the human body image is acquired by using the optical camera, the human body image is identified, and the body appearance, age, wearing, sex, expression, sleep, diet and movement characteristics of the human body are evaluated.
In some embodiments, the human body parameters are obtained by: and acquiring a thermal infrared image of a human body by using an infrared camera, and performing key point visual identification on the thermal infrared image by using a deep neural network to obtain the skin temperature of the human body.
In some embodiments, the human body parameters are obtained by: the temperature, heartbeat, respiration and sleep characteristics of the human body are acquired by using a wearable device or a sensing device on the seat armrest.
In some embodiments, the human body parameters are obtained by: other channel information such as bill information, mobile phone intelligent APP, personnel manually input personal characteristic parameters, are customized for configuration, and remain as preference setting.
In the above embodiments, one or more combinations of the above may be used in different scenarios to obtain key parameters of the human body characteristics. For example, an intelligent bracelet or a camera device can be used in a house to acquire information such as movement or sleep state, body heart rate, wearing and the like of a person; the characteristics of sex, appearance, temperature and the like of the personnel on the seat can be obtained through analysis of passenger ticket checking information and an infrared camera device in the cabin environment. The method can be further expanded to application scenes of service halls such as banks, sales offices and the like.
In some embodiments, the body parameters include a body temperature parameter and a body characteristic parameter.
In some embodiments, step S2, performing reduced-order simplification on the spatial physical field three-dimensional environment model to obtain a spatial physical field real-time environment model, includes: s21, a space data sample set is established, and a data matrix is obtained based on the space data sample set; s22, carrying out principal component decomposition on the data matrix to obtain a principal component matrix and a residual error matrix; s23, processing and clearing the residual matrix, and establishing a mapping model of the principal element matrix and the physical field output to obtain a space physical field real-time environment model. The real-time environment of the closed space is a target for temperature control and is also the most direct factor influencing the heat sensation of internal personnel, and the environment in the space is influenced by the change of the external environment besides the regulation and control of the temperature control system, and comprises the combination of various boundary condition factors such as the time of day, the place, the weather and the like. Such as the instantaneous value of the thermal load in the aircraft cabin, is linked to the flight location, altitude, state, off-cabin weather conditions, the operating state of the personnel and equipment on board, etc. of the aircraft at that moment; the instantaneous value of the heat load in the home or office is related to factors such as outdoor weather change, time change, personnel activities and the like, however, as the temperature in the space is not consistent and consistent with each position, the temperature of the air outlet of the temperature control system cannot represent the temperature of the surface of the personnel in the space, so that a three-dimensional detailed flow field model of the space is required to be established, and a physical field is simulated to obtain the space-time information of the physical quantity of the whole field. However, the traditional method needs to perform discretization on the space domain and the time domain at the same time, and the discretization system order is usually very large, so that the calculated amount is huge, the calculation time cost is high, and the real-time performance is poor, and therefore, a model reduction method is needed to perform model simplification.
In this embodiment, the high-dimensional system is projected onto a low-dimensional subspace, and the original data features are converted by researching the correlation between the features, so that a few features are linear combinations of the original data features, and the new features can represent the original data structure features to the greatest extent without losing information. The flow field data under different working conditions can be represented by a small-order approximate system model. The high-precision simulation model can be simplified through model reduction, so that the requirements of real-time simulation and rapid optimization of the system are met; meanwhile, the calculation time and hardware resources of the processing chip can be reduced; and finally, the method can be used for a part of controller software and used for model prediction, state estimation and other online adaptive models.
In the data sample set u=r m*n Where m is the number of samples and n is the variable dimension. The observed variables for their composition are: x= (x 1 ,x 2 , ... ,x n ). The m observations of the vector form an observation matrix X m*n 。
Principal componentThe analytical model building process is essentially a eigenvalue decomposition process of the data matrix X covariance matrix. For matrix X m*n Its covariance matrix can be expressed as:
，
and (3) performing eigenvalue decomposition:
，
obtaining n eigenvalues of S (X)And n feature vectors->. The data matrix can be decomposed into principal components:
，
wherein,,，/>respectively a principal component score matrix and a residual score matrix, E is called a residual matrix and represents that X is in a non-principal component +.>The change in the above is that the general characteristic value sequence decays rapidly, get +.>Thereby realizing dimension reduction processing of X. But->，/>Respectively a principal element load matrix andand a residual load matrix, wherein k is the number of principal elements. E is removed to obtain a principal component model of PCA:
，
in this way, the original data space is reduced in dimension and decomposed into a principal element subspace and a residual subspace, the two subspaces are mutually orthogonal, and the principal element subspace X is also the feature matrix extracted after the data matrix is reduced in dimension. In addition, the spatial flow field distribution under different boundary conditions is characterized differently, so that the three-dimensional space is modeled repeatedly when the external working condition is changed, and the process is complicated and time-consuming. The multi-layer perceptron model can be used for establishing the mapping relation of the dimension-reducing characteristic data under different boundary input conditions, and the rapid reconstruction of the space field is realized based on the predicted characteristic data, so that the effect of real-time simulation is achieved. Therefore, the method realizes the reduced order processing of the optimized three-dimensional simulation model, builds the prediction model under different boundary input conditions, obtains the space environment model capable of realizing real-time response, and can acquire the environment parameters of any position in space based on the space environment model.
In some embodiments, step S3, performing prediction analysis according to a working condition input driving spatial physical field real-time environment model to obtain spatial environment parameters and spatial heat load data, includes: s31, inputting different boundary conditions, obtaining the output of a real-time environment model of a space physical field, and establishing a training sample set; s32, establishing a multi-layer perceptron model according to the training sample set; s33, inputting actual working condition parameters, and obtaining space internal environment parameters and space heat load data through predictive analysis.
In some embodiments, step S5, taking the environmental parameter and the human parameter as input and the real-time thermal comfort of the human body as output, learns the mapping relationship between the input sample and the output feedback through the neural network, and establishes the thermal comfort model of the human body, including: s51, under a space environment test, collecting thermal feedback data of different people under different environments to construct a data sample set; s52, based on the data sample set, establishing a neural network by adopting a GRNN or SVM method; s53, taking the environmental parameters and the human body parameters as input and the real-time human body thermal comfort as output, and building a human body thermal comfort model by learning the mapping relation between the input sample and the output feedback through a neural network.
In the embodiment, a data sample set is firstly constructed, and because different individual reactions exist in personnel to a thermal environment, the temperature control and regulation behaviors of the personnel show larger differentiation, so that the thermal feedback data sample collection of different people under different environments is carried out by adopting a spatial environment test. Firstly, acquiring indoor and outdoor environment data to record the environment state of the temperature control system at the moment of occurrence of the use behavior, and obtaining the environment parameters at the moment; meanwhile, key parameters of human body characteristics are obtained, and the people in different forms, wearing, sex and activity states are classified, so that indoor environment comfort feedback corresponding to each person at the moment is recorded respectively, and thermal comfort feeling of different people under different input conditions can be obtained. After a large number of subject data samples are obtained, output thermal sensation feedback labels of the personnel are obtained, and the labels at the moment are discrete values. And then, a neural network model is built, the real-time thermal sensation behaviors of the individuals are affected by the indoor and outdoor environment parameters and the personal difference, and the nonlinear relation is formed, and the prediction model based on the neural network can well solve the complex problem comprising a plurality of independent parameters and the nonlinear relation, so that the thermal comfort model of the individuals is built by adopting a neural network method GRNN or SVM.
In the embodiment, the GRNN algorithm has a relatively high convergence rate, small calculation amount in the process and can well process the condition of few training samples.
The model of the GRNN neural network contains four parts: an input layer, a mode layer, a summation layer, and an output layer. (a) The number of neurons of the input layer is equal to the dimension of input vectors in the test sample, namely the dimension of sample characteristics, and the input of the input layer is environment parameters, human body temperature parameters and human body sign parameters for the individual thermal comfort model to be constructed. The input vector is passed directly to the mode layer. (b) A pattern layer, the number of the neurons of the pattern layer is equal to the number n of training samples (the number of the feedback samples of personnel participating in the space environment test), and one neuron corresponds to one trainingSample, mode layer transfer function isThe output of the ith neuron of the pattern layer is the input vector and the corresponding sample vector X of the neuron i Squared Euclidean distance of (2)Is an exponential form of (c). Wherein X is an input vector, X i Is a training sample corresponding to the neuron. (c) A summation layer that sums using two types of neurons. The calculation formula of the first class is->The output of all mode layer neurons is arithmetically summed, the connection weight of the mode layer and each neuron is 1, and the transfer function is +.>The method comprises the steps of carrying out a first treatment on the surface of the The other type of calculation formula is +.>It performs weighted summation on all neurons of the mode layer, and the connection weight between the ith neuron of the mode layer and the jth neuron of the summation layer is the ith output sample Y i The j-th element of (2) has a transfer function of +.>The method comprises the steps of carrying out a first treatment on the surface of the (d) An output layer, the number of neurons in the output layer is equal to the dimension k of the output vector in the training sample (the output is the real-time thermal comfort of the personnel), each neuron divides the output of the summation layer, and the output of the neuron j corresponds to the estimated result->Is the j-th element of->。
The SVM algorithm can effectively solve the pattern recognition problem under the conditions of small samples, nonlinearity, high-dimensional characteristics and the like, and under the condition of limited training sample number, the method can also compromise the model complexity and the learning capacity, and can overcome the over-learning problem easily occurring in the learning method of a neural network and the like. The SVM is proposed for the two classification problem. It is not directly applicable to solve multi-class classification problems, which are involved in individual thermal comfort models, and therefore must be extended to accommodate multi-class requirements for multi-class problems by combining multiple two-class support vector machines. Training a neural network model through a data sample with thermal feedback obtained through a test, obtaining through parameter conditions, and continuously optimizing a learning result and realizing a better prediction effect along with the use behavior feedback of personnel and model updating in actual operation.
In some embodiments, human thermal comfort is categorized into 7 classes of very cold, slightly cold, moderate, slightly hot, very hot.
In some embodiments, it is desirable to set control targets for the temperature control system to generate a control program when the individual thermal comfort at that time and the room ambient temperature at that time are obtained. The traditional control based on the set point temperature can lead the space to be constant at a certain temperature (the actual human body surface temperature is not the set point temperature due to the distribution of the space physical field), and when the external environment is disturbed (such as external time weather change, human body activity metabolism change and the like), the instant state of the human body is not reflected in the space, and the comfort of the human body can not be well satisfied. The control program requires personnel to continuously adjust, and is a fluctuating process: when people feel very cold, the temperature control temperature is habitually adjusted to the highest, after a period of time, the temperature rise can feel overheated, and then the temperature is reduced, so that the people are in an uncomfortable state for a long time easily, and the method is unfavorable for the efficient utilization of energy sources. The self-adaptive space environment intelligent control method can predict an optimal control target (such as air outlet temperature, air speed and air direction) by sensing parameters of personnel and environment parameters in real time based on real-time human body thermal sensation prediction, or output air in different control modes at different air outlets according to a plurality of air outlet positions to respectively regulate temperature in different areas, or control multi-layer sectional regulation and control of partial area closing and partial area opening. When disturbance occurs, the state or environmental change of indoor personnel can be quickly and effectively identified, the indoor personnel can be quickly restored to a comfortable state, and the personnel is not required to actively adjust.
In some embodiments, there are two ways to implement optimal control of target temperature: the method is a simplified nonlinear regulation model, and temperature and wind power regulation with fixed rules are carried out under different environment temperatures and comfortableness. The core idea is as follows: during the use of the temperature control system, for the temperature control adjustment behavior, personnel can set according to own preference and the current indoor environment. In general, when a person is hot, the lower the temperature is set, the larger the wind power is set, the more direct the wind direction is, and the faster the temperature of the body is reduced. And vice versa. Therefore, the behavior pattern of the temperature control regulation can be considered to be related to the real-time thermal sensation of the human body, the outdoor climate, the indoor environment temperature, the set temperature and the difference between the two, and the behavior pattern sectional regulation method of the temperature control regulation behavior is established according to the parameters. The individual thermal comfort level obtained through the previous analysis is 7-gear discretization value, and continuous fitting can be carried out on the individual thermal comfort level according to analysis requirements, so that the regulation and control value is more accurate and stable. For example, when the person feels that the heat is very hot at the moment (namely, when the output of the individual thermal comfort model is 3), the temperature control set temperature is directly reduced to the minimum, the wind power is maximum, and the like; the rate of modulation may be reduced when the person feels the heat between very hot and very cold at this time (i.e., when the individual thermal comfort model output is 0-3). Therefore, the temperature can be calculated by a nonlinear formula as a whole:
，
，
wherein x is according to the body heatThermal sensation feedback calculated by moderate model, T now T is the current set temperature min For the lowest set temperature of the temperature control system, T max For the highest set temperature of the temperature control system, T set-new Is a new set target. Other wind directions, wind power and coordination between different air outlets are similar to the adjusting process.
The second is the regulation and control based on the individual thermal comfort data network learning mapping relation, the method is to predict and obtain the optimal control target at the current moment through a neural network method according to the regulation rule obtained by learning the user characteristics and the historical samples of the user habit, and the aim of optimal regulation is achieved. In the space environment test, each tested person can automatically control and regulate the temperature according to the current heat sensation. The user portrait is constructed based on GRNN/SVM algorithm by learning different environmental parameters, human body physical sign parameters and temperature control and regulation behaviors under thermal sensation feedback, and a mapping network with the environmental parameters, the human body physical sign parameters and the individual thermal comfort degree as inputs and the optimal control target as output is formed. And the method is similar to the method based on incremental learning, the regulation habit of a user is continuously and adaptively learned, and the temperature control self-learning control method based on the individual thermal comfort model and personnel preference is formed.
In some embodiments, the spatial environment control target comprises: the working mode selection, the selection of air outlets, and the switch, the temperature, the wind speed and the wind direction of each air outlet.
It is understood that the term "plurality" in this disclosure means two or more, and other adjectives are similar thereto. "and/or", describes an association relationship of an association object, and indicates that there may be three relationships, for example, a and/or B, and may indicate: a exists alone, A and B exist together, and B exists alone. The character "/" generally indicates that the context-dependent object is an "or" relationship. The singular forms "a," "an," and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise.
It is further understood that the terms "first," "second," and the like are used to describe various information, but such information should not be limited to these terms. These terms are only used to distinguish one type of information from another and do not denote a particular order or importance. Indeed, the expressions "first", "second", etc. may be used entirely interchangeably. For example, first information may also be referred to as second information, and similarly, second information may also be referred to as first information, without departing from the scope of the present disclosure.
It will further be appreciated that the terms "center", "longitudinal", "transverse", "front", "rear", "upper", "lower", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", etc. refer to an orientation or positional relationship that is merely for convenience in describing the present embodiments and simplifying the description, and do not indicate or imply that the devices or elements referred to must have a particular orientation, be constructed and operate in a particular orientation.
It will be further understood that "connected" includes both direct connection where no other member is present and indirect connection where other element is present, unless specifically stated otherwise.
It will further be appreciated that although operations are described in a particular order in the disclosed embodiments, it should not be construed as requiring that such operations be performed in the particular order shown or in sequential order, or that all illustrated operations be performed, to achieve desirable results. In certain circumstances, multitasking and parallel processing may be advantageous.
Other embodiments of the disclosure will be apparent to those skilled in the art from consideration of the specification and practice of the disclosure disclosed herein. This disclosure is intended to cover any adaptations, uses, or adaptations of the disclosure following the general principles of the disclosure and including such departures from the present disclosure as come within known or customary practice within the art to which the disclosure pertains. It is intended that the specification and examples be considered as exemplary only, with a true scope and spirit of the disclosure being indicated by the following claims.
It is to be understood that the present disclosure is not limited to the precise construction that has been described above, and that various modifications and changes may be made without departing from the scope thereof. The scope of the present disclosure is limited only by the appended claims.
Claims (10)
1. The intelligent control method for the self-adaptive space thermal environment is characterized by comprising the following steps of:
s1, establishing a space physical field three-dimensional environment model;
s2, performing reduced-order processing on the three-dimensional environment model of the space physical field to obtain a real-time environment model of the space physical field;
s3, driving the space physical field real-time environment model to conduct prediction analysis according to the working condition input, and obtaining space internal environment parameters and space thermal load data;
s4, acquiring human parameters in the space;
s5, taking the environmental parameters and the human body parameters as input and the real-time human body thermal comfort as output, and building a human body thermal comfort model by learning a mapping relation between an input sample and output feedback through a neural network;
s6, predicting the human body thermal comfort level of any position in space in real time based on the human body thermal comfort level model;
s7, determining a space thermal environment control target based on the human thermal comfort and space thermal load data;
and S8, automatically performing system control based on the space thermal environment control target.
2. The intelligent control method for the self-adaptive space thermal environment according to claim 1, wherein the human body parameters are obtained by the following method: and acquiring a human body image by using an optical camera, identifying the human body image, and evaluating to obtain the body appearance, age, wearing, sex, expression, sleep, diet and movement characteristics of the human body.
3. The intelligent control method for the self-adaptive space thermal environment according to claim 1, wherein the human body parameters are obtained by the following method: and acquiring a thermal infrared image of a human body by using an infrared camera, and performing key point visual identification on the thermal infrared image by using a deep neural network to obtain the skin temperature of the human body.
4. The intelligent control method for the self-adaptive space thermal environment according to claim 1, wherein the human body parameters are obtained by the following method: the temperature, heartbeat, respiration and sleep characteristics of the human body are acquired by using a wearable device or a sensing device on the seat armrest.
5. The method of claim 1, wherein the human body parameters include a human body temperature parameter and a human body characteristic parameter.
6. The method according to claim 1, wherein the step S2 of performing reduced-order simplification on the three-dimensional environmental model of the space physical field to obtain the real-time environmental model of the space physical field includes: s21, a space data sample set is established, and a data matrix is obtained based on the space data sample set; s22, carrying out principal component decomposition on the data matrix to obtain a principal component matrix and a residual error matrix; s23, processing and clearing the residual matrix, and establishing a mapping model of the principal element matrix and the physical field output to obtain a space physical field real-time environment model.
7. The method according to claim 1, wherein the step S3 of driving the spatial physical field real-time environment model to perform predictive analysis according to the working condition input to obtain the spatial environmental parameter and the spatial thermal load data comprises: s31, inputting different boundary conditions, obtaining the output of a real-time environment model of a space physical field, and establishing a training sample set; s32, establishing a multi-layer perceptron model according to the training sample set; s33, inputting actual working condition parameters, and obtaining space internal environment parameters and space heat load data through predictive analysis.
8. The method according to claim 1, wherein the step S5 takes the environmental parameter and the human parameter as input and the real-time thermal comfort of the human body as output, learns the mapping relationship between the input sample and the output feedback through a neural network, and establishes the model of the thermal comfort of the human body, and includes: s51, under a space environment test, collecting thermal feedback data of different people under different environments to construct a data sample set; s52, based on the data sample set, establishing a neural network by adopting a GRNN or SVM method; and S53, taking the environmental parameters and the human body parameters as input and the real-time human body thermal comfort as output, and building a human body thermal comfort model by learning the mapping relation between the input sample and the output feedback through a neural network.
9. The intelligent control method for the thermal environment of the self-adaptive space according to claim 1, wherein the human body thermal comfort level is classified into 7 grades of very cold, micro cold, moderate, slightly hot, hot and very hot.
10. The method for intelligently controlling the thermal environment of the adaptive space according to claim 1, wherein the spatial environment control target comprises: the working mode selection, the selection of air outlets, and the switch, the temperature, the wind speed and the wind direction of each air outlet.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310819404.2A CN116538654A (en) | 2023-07-06 | 2023-07-06 | Self-adaptive space thermal environment intelligent control method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310819404.2A CN116538654A (en) | 2023-07-06 | 2023-07-06 | Self-adaptive space thermal environment intelligent control method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116538654A true CN116538654A (en) | 2023-08-04 |
Family
ID=87456343
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310819404.2A Pending CN116538654A (en) | 2023-07-06 | 2023-07-06 | Self-adaptive space thermal environment intelligent control method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116538654A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117387172A (en) * | 2023-12-11 | 2024-01-12 | 江苏中江数字建设技术有限公司 | Terminal air conditioner energy saving method and system based on accurate recommended equipment control parameters |
| CN118605207A (en) * | 2024-08-08 | 2024-09-06 | 济南千里马电子科技有限公司 | Intelligent control method and system for hotel guest room environment |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109164707A (en) * | 2018-09-28 | 2019-01-08 | 苏州市建筑科学研究院集团股份有限公司 | A kind of indoor environment negative-feedback regu- lation system based on artificial neural network algorithm |
| US20200073347A1 (en) * | 2018-09-05 | 2020-03-05 | Guangdong University Of Technology | Multi-mode and low-energy indoor thermal conditioning method |
| CN111639462A (en) * | 2020-05-29 | 2020-09-08 | 桂林电子科技大学 | Building indoor thermal comfort prediction method in natural ventilation environment based on deep belief neural network |
| US20220171356A1 (en) * | 2020-11-30 | 2022-06-02 | Xi'an University Of Architecture And Technology | Control system and control method for individual thermal comfort based on computer visual monitoring |
| CN115131854A (en) * | 2022-06-13 | 2022-09-30 | 西北工业大学 | Global subspace face image clustering method based on fuzzy clustering |
| CN116049658A (en) * | 2023-03-30 | 2023-05-02 | 西安热工研究院有限公司 | Wind turbine generator abnormal data identification method, system, equipment and medium |
-
2023
- 2023-07-06 CN CN202310819404.2A patent/CN116538654A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20200073347A1 (en) * | 2018-09-05 | 2020-03-05 | Guangdong University Of Technology | Multi-mode and low-energy indoor thermal conditioning method |
| CN109164707A (en) * | 2018-09-28 | 2019-01-08 | 苏州市建筑科学研究院集团股份有限公司 | A kind of indoor environment negative-feedback regu- lation system based on artificial neural network algorithm |
| CN111639462A (en) * | 2020-05-29 | 2020-09-08 | 桂林电子科技大学 | Building indoor thermal comfort prediction method in natural ventilation environment based on deep belief neural network |
| US20220171356A1 (en) * | 2020-11-30 | 2022-06-02 | Xi'an University Of Architecture And Technology | Control system and control method for individual thermal comfort based on computer visual monitoring |
| CN115131854A (en) * | 2022-06-13 | 2022-09-30 | 西北工业大学 | Global subspace face image clustering method based on fuzzy clustering |
| CN116049658A (en) * | 2023-03-30 | 2023-05-02 | 西安热工研究院有限公司 | Wind turbine generator abnormal data identification method, system, equipment and medium |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117387172A (en) * | 2023-12-11 | 2024-01-12 | 江苏中江数字建设技术有限公司 | Terminal air conditioner energy saving method and system based on accurate recommended equipment control parameters |
| CN117387172B (en) * | 2023-12-11 | 2024-05-17 | 江苏中江数字建设技术有限公司 | Terminal air conditioner energy saving method and system based on accurate recommended equipment control parameters |
| CN118605207A (en) * | 2024-08-08 | 2024-09-06 | 济南千里马电子科技有限公司 | Intelligent control method and system for hotel guest room environment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Lu et al. | Data-driven simulation of a thermal comfort-based temperature set-point control with ASHRAE RP884 | |
| Merabet et al. | Intelligent building control systems for thermal comfort and energy-efficiency: A systematic review of artificial intelligence-assisted techniques | |
| Peng et al. | Temperature-preference learning with neural networks for occupant-centric building indoor climate controls | |
| CN116538654A (en) | Self-adaptive space thermal environment intelligent control method | |
| André et al. | User-centered environmental control: a review of current findings on personal conditioning systems and personal comfort models | |
| Mofidi et al. | Intelligent buildings: An overview | |
| JP6925536B2 (en) | Systems and methods to control operation | |
| Shah et al. | Dynamic user preference parameters selection and energy consumption optimization for smart homes using deep extreme learning machine and bat algorithm | |
| Aryal et al. | Intelligent agents to improve thermal satisfaction by controlling personal comfort systems under different levels of automation | |
| KR102439453B1 (en) | Optimizing hvac system operarion using interconnected neural networks and online learning and operatiing method thereof | |
| Wang et al. | An occupant-centric adaptive façade based on real-time and contactless glare and thermal discomfort estimation using deep learning algorithm | |
| CN112613232A (en) | Indoor human body thermal comfort prediction and evaluation method under winter heating condition | |
| US20220019186A1 (en) | Method and system for smart environment management | |
| Suman et al. | Potential impacts of smart homes on human behavior: A reinforcement learning approach | |
| Wu et al. | A systematic review of research on personal thermal comfort using infrared technology | |
| Deng et al. | Room match: Achieving thermal comfort through smart space allocation and environmental control in buildings | |
| Aparicio-Ruiz et al. | KNN and adaptive comfort applied in decision making for HVAC systems | |
| Li et al. | Indoor temperature preference setting control method for thermal comfort and energy saving based on reinforcement learning | |
| CN110991478A (en) | Method for establishing thermal comfort model and method and system for setting user preference temperature | |
| Huang et al. | State of the art review on the HVAC occupant-centric control in different commercial buildings | |
| Khalil et al. | An IoT environment for estimating occupants’ thermal comfort | |
| Laftchiev et al. | Personalizing individual comfort in the group setting | |
| Javed et al. | Personalized thermal comfort modeling based on Support Vector Classification | |
| EP4051968B1 (en) | System and method for thermal control based on invertible causation relationship | |
| Zhang | A bio-sensing and reinforcement learning control system for personalized thermal comfort and energy efficiency |
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 |