GWO-LSSVM耦合模型在变形预测中的应用

Application of GWO-LSSVM Coupling Model in Deformation Prediction

针对原始监测数据中包含随机噪声,导致预测结果不理想,以及单一预测模型的局限性,本文提出一种基于经验模态分解(EMD)与灰狼算法(GWO)优化最小二乘支持向量机(LSSVM)耦合的EMDGWO-LSSVM变形预测新模型。通过工程实例表明,新模型与LSSVM、GWO-LSSVM模型进行对比,预测精度最高,稳定性最好,能够为变形预测提供一定的参考价值。

In view of the random noise in the original monitoring data, the prediction results are not satisfactory and the limitation of a single prediction model, this paper proposes an empirical modal decomposition (EMD) and grey wolf algorithm (GWO) optimized least squares support vector machine (LSSVM) coupled EMD-GWO-LSSVM deformation prediction new model. The engineering example shows that the new model is compared with the LSSVM and GWO-LSSVM models. The prediction accuracy is the highest and the stability is the best, which can provide some reference value for deformation prediction.