【作者】 朱珍德； 李红波； 尚剑飞； 刘金辉；

【Author】 ZHU Zhende1,2, LI Hongbo1,2, SHANG Jianfei1,2, LIU Jinhui1,3 (1.Research Institute of Geotechnical Engineering,Hohai University,Nanjing 210098,China;2. Key Laboratory of Ministry of Education for Geomechanics and Embankment Engineering, Hohai University, Nanjing 210098,China;3. Fucun Coal Mine of Zaozhuang Mining Group, Zaozhuang 277605, China)

【机构】 河海大学岩土工程科学研究所； 河海大学岩土力学与堤坝工程教育部重点实验室； 枣庄矿业集团付村煤矿；

【摘要】 建立了基于粒子群算法的支持向量机预测模型,采用粒子群算法对核函数和模型参数进行了优化。将建立的模型应用于付村煤矿3上411运顺的变形预测,结果表明预测结果与实际监测数据误差很小。对基于粒子群算法的支持向量机PSO—SVM模型与灰色GM(1,1)预测模型和人工神经网络L-MBP模型的预测结果进行了比较,结果显示PSO—SVM预测方法具有更高的预测精度,表明采用PSO—SVM模型进行巷道变形预测是可行的,有一定的推广应用价值。更多还原

【Abstract】 This paper puts forward a forecasting model based on particle swarm optimization. The kernel function and model parameters are optimized using particle swarm optimization. The forecasting model is applied to predict the surrounding deformations of the roadway called 3-411 and the result shows that the forecast result is very close to the real monitoring data. At the same time, the PSO-SVM model is compared with the GM(1,1) model and L-M BP network model. The results show that PSO-SVM method is better in the aspect of prediction accuracy and the PSO-SVM roadway deformation prediction model is feasible.更多还原