依据二滩水电站地下厂房和机组的原型观测数据对机组和厂房结构振动的相关性进行分析,据此建立基于粒子群优化最小二乘支持向量计算法的厂房振动响应预测模型,预测结果与实测资料吻合。在此基础上将运行水头作为输入因子引入到智能预测模型中,扩大了该智能预测模型的适用范围,取得了很好的效果。
The vibration of powerhouse structures is mainly induced by hydraulics factors, mechanical and electromagnetic factors of the generating unit. It nonlinearly couples with the generating unit. Based on prototype observation data of Ertan Hydropower Station, the paper analyzes the coupling effect between vibration of units and powerhouse, and then the vibration response forecasting model of the powerhouse is built based on LS-SVM opti- mized by particle swarm optimization algorithm, and the prediction results are coincide with the observed data. Fur- ther, the paper introduces the running water head as an input divisor into the intelligent prediction model while the forecasting range is extended, and the result is satisfactory.