遗传双线性模型在海洋冰情预测中的应用

提出了一套建立双线性模型（ＢＭ）的简便方案。用作者研制的加速遗传算法可同时估计ＢＭ模型各参数,成功地解决了ＢＭ建模这一难题,为ＢＭ模型的广泛应用提供了强有力的工具。实例的计算结果说明,这套方法在海洋冰情预测中是可行而有效的,通过利用预测过程中产生的残差信息进行反馈矫正,保证了ＢＭ模型高的拟合粗度和稳健的预测性能,增强了ＢＭ对复杂非线性动态系统的适应性。该方案具有一般性,在各种非线性时序动态预测中具

Application of genetic bilinear time series model to predicting marine ice condition

A simple and practical scheme is presented for establishing the bilinear time series model (BM). With the accelerating genetic algorithm developed by the authors, all parameters of BM can be optimized at the same time, and the difficulty problem for the modeling BM is resolved, which gives a strong tool for widely using BM. The computational result shows that the scheme applied to predicting marine ice condition degree series is practical and efficient, and that BM can ensure high fitting precision, robust forecasting and good adaptability to complex nonlinear dynamic systems by using the feedback information of prediction residual errors. As a general way, the scheme has major theoretic value and wide ranging application for predicting the nonlinear time series.