Gas Concentration Forecasting Based on Support Vector Regression in Correlation Space via KPCA

Authors:
Jian Cheng;Jian-Sheng Qian;Guang-Dong Niu;Yi-Nan Guo
Affiliations:
School of Information and Electrical Engineering, China University of Mining and Technology, Xu Zhou, China 221116;School of Information and Electrical Engineering, China University of Mining and Technology, Xu Zhou, China 221116;School of Information and Electrical Engineering, China University of Mining and Technology, Xu Zhou, China 221116;School of Information and Electrical Engineering, China University of Mining and Technology, Xu Zhou, China 221116
Venue:
ISNN '09 Proceedings of the 6th International Symposium on Neural Networks on Advances in Neural Networks
Year:
2009

Citing 6
Cited 0

Prediction of chaotic time series based on the recurrent predictor neural network

IEEE Transactions on Signal Processing
Fuzzy Weighted Support Vector Regression With a Fuzzy Partition

IEEE Transactions on Systems, Man, and Cybernetics, Part B: Cybernetics
An overview of statistical learning theory

IEEE Transactions on Neural Networks
Prediction of noisy chaotic time series using an optimal radial basis function neural network

IEEE Transactions on Neural Networks
Support vector machine with adaptive parameters in financial time series forecasting

IEEE Transactions on Neural Networks
Blind Image Deconvolution Through Support Vector Regression

IEEE Transactions on Neural Networks

Quantified Score

Hi-index	0.00

Visualization

Abstract

Gas concentration is one of key factors which influence safe production of coal mine, and it is very important to forecast gas concentration accurately for ensuring the coal mine safety. A novel approach is presented to forecast gas concentration based on support vector regression (SVR) in correlation space reconstructed by kernel principal component analysis (KPCA). A two-stage architecture is proposed to improve its prediction accuracy and generalization performance for gas concentration forecasting. In the first stage, KPCA is adopted to extract features and obtain kernel principal components, so the correlation space of gas concentration is reconstructed according to the accumulative contribution ratio. Then, in the second stage, support vector regression (SVR) is employed for forecasting gas concentration, which hyperparameters are selected by adaptive chaotic cultural algorithm (ACCA). The approach is compared with the forecasting model in whole space of gas concentration. The simulation shows that SVR model in correlation space using KPCA performs much better than that without correlation analysis.