用混沌理论提高潮汐预报的准确度

Enhancing tidal prediction precision based on the chaos theory

为提高潮位预报的准确性,在具有较长潮汐观测数据的站点,基于混沌理论,对观测值与潮汐模型预测值之差所构成的余水位序列(即误差序列),采用局域线性模型的分析方法,给出可能误差预测,修正模型的预报结果,提高潮汐预报的准确性。所给例子,对预测跨度T=2h,经局域法修正后,崇武站2007年12月份1个月预测水位统计的RMSE值减少74.7%,厦门站减少60.5%;对T=24h,崇武、厦门两站RMSE值减小都在50%左右。

To improve the tidal prediction accuracy,the error correction of a regional tidal model is carried out using a local model approach based on chaos theory where sufficient tidal measurement data were available.As the examples,for one month（December-2007）,the local model correction can remove as much as 74.7% of root mean square error（RMSE） at Chongwu measurement station and 60.5% at Xiamen Station for the prediction horizon T=2 h,and remove around 50% RMSE at above two stations for the prediction horizon T=24 h.