Neural network based predictive control of personalized heating systems

被引:53
作者
Katic, Katarina [1 ]
Li, Rongling [2 ]
Verhaart, Jacob [1 ]
Zeiler, Wim [1 ]
机构
[1] Eindhoven Univ Technol, Dept Built Environm, POB 513, NL-5600 MB Eindhoven, Netherlands
[2] Tech Univ Denmark, Dept Civil Engn, Nils Koppels Alle, DK-2800 Lyngby, Denmark
关键词
Personalized heating; Automatic control; Predictive control; Machine learning; Artificial neural network; NARK; SUPPORT VECTOR MACHINE; THERMAL COMFORT; ENERGY-CONSUMPTION; SKIN TEMPERATURE; BODY-COMPOSITION; BUILDINGS; MODEL; REGRESSION; CLASSIFICATION; ENVIRONMENT;
D O I
10.1016/j.enbuild.2018.06.033
中图分类号
TU [建筑科学];
学科分类号
0813 ;
摘要
The aim of a personalized heating system is to provide a desirable microclimate for each individual when heating is needed. In this paper, we present a method based on machine learning algorithms for generation of predictive models for use in control of personalized heating systems. Data was collected from two individual test subjects in an experiment that consisted of 14 sessions per test subject with each session lasting 4 h. A dynamic recurrent nonlinear autoregressive neural network with exogenous inputs (NARX) was used for developing the models for the prediction of personalized heating settings. The models for subjects A and B were tested with the data that was not used in creating the neural network (unseen data) to evaluate the accuracy of the prediction. Trained NARX showed good performance when tested with the unseen data, with no sign of overfitting. For model A, the optimal network was with 12 hidden neurons with root mean square error equal to 0.043 and Pearson correlation coefficient equal to 0.994. The best result for model B was obtained with a neural network with 16 hidden neurons with root mean square error equal to 0.049 and Pearson correlation coefficient equal to 0.966. In addition to the neural network models, several other machine learning algorithms were tested. Furthermore, the models were on-line tested and the results showed that the test subjects were satisfied with the heating settings that were automatically controlled using the models. Tests with automatic control showed that both test subjects felt comfortable throughout the tests and test subjects expressed their satisfaction with the automatic control. (C) 2018 The Authors. Published by Elsevier B.V.
引用
收藏
页码:199 / 213
页数:15
相关论文
共 68 条
[1]  
[Anonymous], 2008, The Elements of Statistical Learning
[2]  
[Anonymous], MANUF SYST
[3]  
[Anonymous], INTRO STAT LEARNING
[4]   NA Neural Network Modelling of Mushroom Dynamic Vapour Sorption Kinetics [J].
Argyropoulos, Dimitrios ;
Paraforos, Dimitris S. ;
Alex, Rainer ;
Griepentrog, Hans W. ;
Mueller, Joachim .
IFAC PAPERSONLINE, 2016, 49 (16) :305-310
[5]  
Beale M.H., 2015, MathWorks, DOI DOI 10.1002/0471221546
[6]   Impact of available and perceived control on comfort and health in European offices [J].
Boerstra, Atze ;
Beuker, Tim ;
Loomans, Marcel ;
Hensen, Jan .
ARCHITECTURAL SCIENCE REVIEW, 2013, 56 (01) :30-41
[7]   Comfort and performance impact of personal control over thermal environment in summer: Results from a laboratory study [J].
Boerstra, Atze C. ;
te Kulve, Marije ;
Toftum, Jorn ;
Loomans, Marcel G. L. C. ;
Olesen, Bjarne W. ;
Hensen, Jan L. M. .
BUILDING AND ENVIRONMENT, 2015, 87 :315-326
[8]  
Brager GS, 2004, ASHRAE T, V110, P1
[9]   Thermal comfort prediction using normalized skin temperature in a uniform built environment [J].
Chaudhuri, Tanaya ;
Zhai, Deqing ;
Soh, Yeng Chai ;
Li, Hua ;
Xie, Lihua .
ENERGY AND BUILDINGS, 2018, 159 :426-440
[10]   NONLINEAR-SYSTEM IDENTIFICATION USING NEURAL NETWORKS [J].
CHEN, S ;
BILLINGS, SA ;
GRANT, PM .
INTERNATIONAL JOURNAL OF CONTROL, 1990, 51 (06) :1191-1214