基于相空间重构与最小二乘支持向量机的时延预测

针对网络控制系统的时延预测问题，提出一种基于相空间重构与最小二乘支持向量机的时延预测方法.首先利用0-1测试法确定时延序列具有混沌特性，引入相空间重构技术提高预测精度.对实际采集的时延序列进行Hurst指数分析，选择最小二乘支持向量机作为预测模型.然后利用C-C方法确定时延序列相空间重构参数，通过递归图确定时延序列的局部可预测性，利用遗传算法对最小二乘支持向量机的参数进行离线优化.最后通过优化后的最小二乘支持向量机并结合相空间重构对时延序列进行在线预测.与其它预测方法进行了仿真对比，结果表明本文方法具有更高的预测精度与更小的预测误差，同时并未降低预测算法的实时性.

Time-Delay Prediction Based on Phase Space Reconstruction and Least Squares Support Vector Machine

In order to solve time-delay prediction problem of networked control system,a time-delay prediction method based on phase space reconstruction and least squares support vector machine is proposed in this paper.Firstly,0-1 test algorithm for chaos is used to determine the chaotic characteristics of the time-delay sequence,and the phase space reconstruction technique is introduced to improve the prediction accuracy.The Hurst exponent of the real time-delay sequence is analyzed,and least squares support vector machine is selected as the prediction model.Then,C-C method is used to determine the parameters of phase space reconstruction.The partial predictability of time-delay is determined by recurrence plot.The parameters of least squares support vector machine are off-line optimized by genetic algorithm.Finally,the time-delay sequence is predicted by the optimized least squares support vector machine combined with the phase space reconstruction.Compared with other prediction methods,the simulation results show that the proposed method has higher prediction accuracy and smaller prediction error,and does not reduce the real-time performance of the algorithm.