A PSO-SVM based model for alpha particle activity prediction inside decommissioned channels

Authors:
Mingzhe Liu;Xianguo Tuo;Jun Ren;Zhe Li;Lei Wang;Jianbo Yang
Affiliations:
State Key Laboratory of Geohazard Prevention and Geoenvironment Protection, Chengdu University of Technology, Chengdu, China,College of Nuclear Technology and Automation Engineering, Chengdu Unive ...;State Key Laboratory of Geohazard Prevention and Geoenvironment Protection, Chengdu University of Technology, Chengdu, China,College of Nuclear Technology and Automation Engineering, Chengdu Unive ...;College of Nuclear Technology and Automation Engineering, Chengdu University of Technology, Chengdu, China;College of Nuclear Technology and Automation Engineering, Chengdu University of Technology, Chengdu, China;College of Nuclear Technology and Automation Engineering, Chengdu University of Technology, Chengdu, China;College of Nuclear Technology and Automation Engineering, Chengdu University of Technology, Chengdu, China
Venue:
ISNN'12 Proceedings of the 9th international conference on Advances in Neural Networks - Volume Part I
Year:
2012

Citing 5
Cited 0

An introduction to support Vector Machines: and other kernel-based learning methods

An introduction to support Vector Machines: and other kernel-based learning methods
Choosing Multiple Parameters for Support Vector Machines

Machine Learning
A novel LS-SVMs hyper-parameter selection based on particle swarm optimization

Neurocomputing
Support vector-based feature selection using Fisher's linear discriminant and Support Vector Machine

Expert Systems with Applications: An International Journal
The particle swarm - explosion, stability, and convergence in amultidimensional complex space

IEEE Transactions on Evolutionary Computation

Quantified Score

Hi-index	0.00

Visualization

Abstract

This paper presents a hybrid Support Vector Machine (SVM) and Particle Swarm Optimization (PSO) model for predicting alpha particles emitting contamination on the internal surfaces of decommissioned channels. Six measuring parameters (channel diameter, channel length, distance to radioactive source, radioactive strength, wind speed and flux) and one ionizing value have been obtained via experiments. These parameters show complex linear and nonlinear relationships to measuring results. The model used PSO to optimize SVM parameters. The comparison of computational results of the hybrid approach with normal BP networks confirms its clear advantage for dealing with this complex nonlinear prediction.