摘要
提出分维指数加权一阶局域算法。该算法用最大Lyapunov指数和重构相点各分量所对应的延迟时间的乘积作为幂,构造一个指数形式的衰减因子,对加权一阶局域法的向量距离公式进行修正。修正后的距离公式不仅体现了各相点与中心点的相关性,还表示了相点各分量与中心点第一分量的关联程度。利用该算法对Logistic混沌时间序列进行预测的结果表明,相对于现有算法,本文所提算法明显提高了预测精度,而且序列的混沌性愈强,嵌入维数愈大,改进效果愈明显。
A novel dimension-exponent adding-weight one-rank local-region method is introduced in this paper. An index form attenuation factor composed of the product of the largest Lyapunov exponent and the delay time corresponding to each dimension of the adjacent point, is applied to amend the vector distance formula of original method. The revised distance formula not only expresses different relevance of each phase points and the center point, but also the correlation between each dimension of this phase points and the first dimension of the center point. The Logistic chaotic time series are forecasted using this improved method, the results show that the prediction accuracy is improved in the proposed method compared to the original one. Besides, the more chaotic the time series is, and the greater the embedding dimension is, the more obvious the improving effectiveness is.
出处
《电测与仪表》
北大核心
2010年第5期12-15,7,共5页
Electrical Measurement & Instrumentation
基金
河北省自然科学基金资助项目(F2009000224)
河北省科技攻关计划资助项目(072135190)
关键词
加权一阶局域法
最大LYAPUNOV指数
混沌时间序列
预测
adding-weight one-rank local-region method, the largest Lyapunov exponent, chaotic time series, prediction
