基于EEMD-HW-PSO-ELM耦合模型的排土场边坡位移预测模型

为了准确预测小样本、非线性特点的排土场边坡位移，提出了一种基于经验模态分解法、三次指数平滑法和粒子群优化极限学习机的EEMD-HW-PSO-ELM边坡位移组合预测模型。以伊敏露天矿排土场GPS位移监测数据为例，验证该模型的有效性。研究结果表明：EEMD模型分解后的边坡位移时间序列包括4个IMF分量和1个余量，将分解后的数据重构为趋势项和波动项，物理意义明确。分别选择三次指数平滑法和粒子群优化极限学习机预测趋势项和波动项位移，将分项预测结果的等权叠加值作为最终预测结果，预测值的平均相对误差为0.38%，均方根误差为1.15。选择了BP模型和Elman模型进行对比预测，结果表明组合预测模型的预测效果较好，能够为边坡安全管理提供理论依据。

Displacement Prediction of Dump Slope Based on EEMD-HW-PSO-ELM Coupling Model

In order to accurately predict the displacement of waste dump slope with small samples and nonlinear characteristics，an EEMD-HW-PSO-ELM slope displacement combined prediction model based on empirical mode decomposition method，cubic exponential smoothing method and particle swarm optimization limit learning machine was proposed.Taking the GPS displacement monitoring data of Yiminhe open pit waste dump as an example to verify the effectiveness of the model，the research results show that the time series of slope displacement decomposed by EEMD model includes four IMF components and one margin.The decomposed data is reconstructed into trend term and fluctuation term with clear physical meaning.The cubic exponential smoothing method and particle swarm optimization limit learning machine were selected to predict the displacement of trend term and fluctuation term respectively.The equal weight superposition value is the final prediction result.The average relative error of the prediction value is 0.38% and the root mean square error is 1.15.BP model and Elman model were selected for comparative prediction.The results show that the com-bined prediction model has good prediction effect and can provide a theoretical basis for safety management.