基于经验模式分解和自适应神经模糊推理的风速短期智能预测混合方法

发布时间：2018-08-05 15:18

【摘要】：为实现风速的超前多步高精度预测,提出一种基于经验模式分解与自适应神经模糊推理的混合方法。该方法利用经验模式分解法对铁路风速进行多层分解计算以降低风速的强随机性,对分解后的各层风速数据分别建立自适应神经模糊推理预测模型并完成预测计算,最终加权各层预测值获得原实测数据的对应步数的预测结果。运用所提出的方法对青藏铁路某监控点的风速进行预测。研究结果表明:所提出的混合方法有效融合了经验模式分解法的信号细分性能和自适应神经模糊推理法的非线性追踪能力,混合模型的超前1步、2步、3步预测的平均相对误差分别为6.24%,11.11%和14.30%,体现出良好的非平稳信号预测性能。
[Abstract]:A hybrid method based on empirical mode decomposition (EMD) and adaptive neural fuzzy reasoning (ANFR) is proposed to predict the wind speed with multiple steps and high accuracy. In order to reduce the strong randomness of wind speed, the empirical mode decomposition method is used to calculate the railway wind speed. The adaptive neural fuzzy inference prediction model is established for each layer of wind speed data after decomposition and the prediction calculation is completed. Finally, the prediction results of the corresponding steps of the original measured data are obtained by weighted prediction values of each layer. The proposed method is used to predict the wind speed at a monitoring point of the Qinghai-Tibet Railway. The results show that the proposed hybrid method effectively combines the signal subdivision performance of the empirical mode decomposition method and the nonlinear tracking ability of the adaptive neuro-fuzzy reasoning method. The average relative error of the mixed model is 6.2411% and 14.30% respectively, which shows good prediction performance of non-stationary signals.
【作者单位】：中南大学交通运输工程学院轨道交通安全教育部重点实验室;德国Rostock大学信息与电子工程学院;
【基金】：国家自然科学基金资助项目(51308553) 国家高铁联合基金资助项目(U1134203,U1334205) 湖南省教育厅科学研究项目(省优秀博士学位论文奖励专项) 中南大学升华育英人才计划项目(502034011) 中南大学研究生自主探索创新项目基金资助项目(2013zzts041)~~
【分类号】：U298

【参考文献】