随机声载荷基于相空间重构的GMDH方法预测

GMDH prediction in view of phase space reconstruction of random acoustice loading

把数据处理的组合法(GMDH)与相空间重构理论相结合,产生基于相空间重构的GMDH方法应用于随机声载荷的状态预测中。根据此法所建立的模式系统具有较好的客观性,预测结果与时间序列自回归模型ARMA(17,16)相比,预报均方误差有明显改进。

Combine the Grop Method of Data Handing（GMDH） with phase space reconstruction theory produce GMDH in view of phase space reconstruction applied to state prediction of random acoustice loading. The modulating system established according to the method have better objectivity and the prediction mean square error have distinct improvement in comparison with time series autoregression model ARMA（17,16）.