The thermal comfort indices are usually identified using empirical thermal models based on the human balanced equations and experimentations. In our paper, we propose a statistical regression method to predict these indices. To achieve this goal, first, the fuzzy support vector regression (FSVR) identification approach was integrated with the particle swarm optimization (PSO) algorithm. Then PSO was used as a global optimizer to optimize and select the hyper-parameters needed for the FSVR model. The radial basis function (RBF) kernel was used within the FSVR model. Afterward, these optimal hyper-parameters were used to forecast the thermal comfort indices: predicted mean vote (PMV), predicted percentage dissatisfied (PPD), new standard effective temperature (SET*), thermal discomfort (DISC), thermal sensation (TSENS) and predicted percent dissatisfied due to draft (PD). The application of the proposed approach on different data sets gave successful prediction and promising results. Moreover, the comparisons between the traditional Fanger model and the new model further demonstrate that the proposed model achieves even better identification performance than the original FSVR technique.
HwangCHongDHSeokKH. Support vector interval regression machine for crisp input and output data. Fuzzy Sets Syst2006; 157: 1114–1125.
2.
HongDHHwangC. Interval regression analysis using quadratic loss support vector machine. IEEE Trans Fuzzy Syst2005; 13: 229–237.
3.
VapnikVN. The nature of statistical learning theory, New York, NY: Springer-Verilog, 1995.
4.
VapnikVGolowichSESmolaA. Support vector method for function approximation, regression estimation, and signal processing”. Advances in Neural Information Processing Systems1996; 9: 281–287. .
5.
VapnikVN. Statistical learning theory, New York, NY: John Wiley and Sons, 1998.
6.
JengJ-TChuangbC-CSuS-F. Support vector interval regression networks for interval regression analysis. Fuzzy Sets Syst2003; 138: 283–300.
7.
HongW-C. Electric load forecasting by support vector model. Appl Math Modell2009; 33: 2444–2454.
8.
HongW-C. Chaotic particle swarm optimization algorithm in a support vector regression electric load forecasting model. Energy Conversion Manage2009; 50: 105–117.
9.
CheJWangJWangG. An adaptive fuzzy combination model based on self-organizing map and support vector regression for electric load forecasting. Energy2012; 37: 657–664.
Fanger PO and Toftum J. In: Proc. of moving thermal comfort standards into the 21st century, Windsor, 2001, pp.11–18.
15.
CortesCVapnikVN. Support vector network. Mach Learn1995; 20: 1–25.
16.
Forghani Y, Sadoghi Yazdi H, Sigari Tabrizi R and Reza Akbarzadeh Totonchi M. Fuzzy support regression. In; ICCKE2011, international conference on computer and knowledge engineering, Iran, 2011, pp.10–13.
17.
DiamondPKörnerR. Extended fuzzy linear models and least squares estimates. Comput Math Appl1997; 33: 15–32.
18.
KeerthiS. Efficient tuning of SVM hyperparameters using radius/margin bound and iterative algorithms. IEEE Trans Neutral Netw2002; 13: 1225–1229.
19.
Cawley GC. Leave-one-out cross-validation based model selection criteria for weighted LS-SVMs. In: Proceedings of the international joint conference on neural networks, Vancouver, BC, Canada, 2006. http://theoval.cmp.uea.ac.uk/~gcc/publications/pdf/ijcnn2006a.pdf.
20.
CherkasskyVMaY. Practical selection of SVM parameters and noise estimation for SVM regression. Neural Netw2004; 17: 113–126.
21.
CiweiGBompardENapoliRChengH. Price forecast in the competitive electricity market by support vector machine. Physica2007; 382: 98–113.
22.
Lin PT, Su SF and Lee TT. Support vector regression performance analysis and systematic parameter selection. In: Proceedings of international joint conference on neutral networks, Montreal, Canada, 31 July–4 August 2005, Vol. 2, pp.877–882.
23.
HongDHHwangC. Fuzzy nonlinear regression model based on LS-SVM in feature space, FSKD2006, LNAI 4223, Berlin, Germany: Springer-Verlag Berlin Heidelberg, 2006, pp. 208–216.
24.
BuckleyJFeuringTHayashiY. Multivariate non-linear fuzzy regression: an evolutionary algorithm approach. Int J Uncertain Fuzziness Knowl-Based Syst1999; 7: 83–98.
25.
BuckleyJFeuringT. Linear and non-linear fuzzy regression: evolutionary algorithm solutions. Fuzzy Sets Syst2000; 112: 381–394.
26.
HaoP-YChiangJ-H. Senior Member, IEEE. Fuzzy regression analysis by support vector learning approach. IEEE Trans Fuzzy Syst2008; 16: 428–441.
27.
Gunn S. Support vector machines for classification and regression, ISIS Tech. Report, University of Southampton, 1998.
28.
Smola AJ and Scholkopf B. A tutorial on support vector regression, NeuroCOLT2 Tech. Report, NeuroCOLT, 1998.
29.
ClercMKennedyJ. The particle swarm–explosion, stability, and convergence in a multidimensional complex space. IEEE Trans Evol Comput2002; 6: 58–73.
30.
Pahikkala T, Boberg J and Salakoski T. Fast n-fold cross-validation for regularized least-squares. In: Honkela T, Raiko T, Kortela J and Valpola H (eds) In: Proceedings of the ninth Scandinavian conference on artificial intelligence (SCAI 2006), Espoo, Finland, Otamedia Oy, 2006, pp.83–90.
31.
ASHRAE Standard 62-2013, Ventilation for acceptable indoor air quality, American Society of Heating, Refrigerating and Air-Conditioning Engineers, Inc., 2013.
32.
ASHRAE. 2013 ASHRAE Standard 55-2013, Thermal environment conditions for human occupancy. Atlanta: American Society of Heating, Refrigerating and Air-Conditioning Engineers, Inc. (ANSI approved), 2013.
33.
GaggeAP. Rational temperature indices of a man's thermal environment and their use with a 2-node model of his temperature regulation. Fed Proc1973; 32: 1572–1582.
34.
GaggeAPBerglundaLGF. Standard predictive index of human response to the thermal environment. ASHREA Trans1986; 92: 709–731.
35.
MegriACEl NaqaI. Prediction of the thermal comfort indices using improved support vector machine classifiers and non-linear kernel functions. Indoor and Built Environment, Epub ahead of print, 2014. 11 July. DOI: 10.1177/1420326X14539693.
