标题：Forecast of small faults extending length based on support vector machine with particle swarm optimization
作者：Shi, Long Qing ;Qiu, Mei ;Han, Jin ;Teng, Chao
作者机构：[Shi, Long Qing ;Qiu, Mei ;Han, Jin ;Teng, Chao ] Shandong Provincial Key Laboratory of Depositional Mineralization and Sedimentary Minerals, College 更多
会议名称：International Conference on Frontier of Energy and Environment Engineering, ICFEEE 2014
会议日期：December 6, 2014 - December 7, 2014
来源：Environment, Energy and Applied Technology - Proceedings of the 2014 3rd International Conference on Frontier of Energy and Environment Engineering, ICFEEE 2014
摘要：Take samples data of small faults in No. 4 coal seam of Ezhaung minefield in Laiwu coal field as research subjects. Applying the method of grey correlative analysis, strike, dip, dip angle and throw are selected as predictors of small faults extending length. Prediction of extending length is an intricate task affected by many factors. The proposed PSO—SVM method is applied to predict extending length in the paper, among which Particle Swarm Optimization (PSO) is used to optimize the critical parameters of Support Vector Machine (SVM) so as to avoid artificial arbitrariness and enhance the forecast accuracy. The results achieved indicate that the model has a higher precision and is suitable for prediction of extending length.
© 2015 Taylor & Francis Group, London.