A novel support vector forecasting model based on chaos theory is presented in this paper.It makes use of support vector machines' strongly nonlinear mapping ability, and network's structure is optimally auto-created.It adopts support vector machines as nonlinear forecaster and determines network's input variable number through computing reconstruct phase space's saturated embedding dimension.The maximum effective forecasting steps is determined by computing chaos time series' largest lyapunov exponent.Application results in aeroengine compressor show that this presented method possesses much better precision, which proves that the method is feasible and effective.This method is contributive and instructional for nonlinear time series forecasting via support vector machines for chaos time series.%提出了一种基于混沌理论的通用支持向量预测方法.该方法通过重构相空间的饱和嵌入维数确定支持向量机的最佳输入变量的选取;通过计算混沌序列的最大Lyapunov指数确定支持向量机预测模型的最大有效预测步数;利用支持向量机强大非线性映射能力、网络结构的自动最优化特性,实现时间序列的非线性预测.最后,应用于压气机的试车数据序列建模与分析,结果证明该方法具有较高的预测精度,实验结果与理论计算基本相符,该方法对于采用支持向量机进行混沌时间序列预测具有一定的参考价值和理论指导意义.
展开▼