摘要
为了对锚泊状态下未来0.5 h的船面风速和风向进行估算,提出了一种基于小波变换和最小二乘支持向量机的估算方法。首先对原始风速数据进行正交分解,得到X轴和Y轴风速;然后分别对X轴和Y轴风速进行小波分解,提取出低频和高频数据序列;其次利用最小二乘向量机方法,分别对低频、高频序列进行估算,分别将X轴和Y轴各序列估算结果进行叠加得到X轴和Y轴估算风速;最后利用矢量法则,同时实现了风速和风向未来0.5 h的估算。以试验船在东海海域锚泊状态下船舶气象仪所测的风速风向数据进行建模与实例验证分析,结果表明该方法具有较高的估算精度。
In order to estimate the wind speed and direction on ship surface in the next half hour under the anchoring condition, an estimation method is proposed in this paper based on wavelet transform and least squares support vector machine. Firstly, the X-axis and Y-axis wind speeds are obtained by orthogonal decomposition from the original wind data series. Then the low and high frequency data series are extracted by wavelet decomposition. Secondly, based on the theory of least squares support vector machines, the prediction is made for the low and high frequency series, and the wind estimation values of X-axis and Y-axis are obtained respectively by the superposition of the predicted results of all series. Finally, the estimation values of wind speed and direction in the next half hour are obtained simultaneously by using the vector algorithm. The proposed method is evaluated with the wind data measured by the ship-borne meteorological instrument on the anchoring ship in the East China Sea. The results show that this method has high estimation precision.
出处
《海洋技术学报》
2016年第2期66-70,共5页
Journal of Ocean Technology
基金
山东省自主创新及成果转化专项资助项目(2014CGZH1203)
关键词
小波变换
最小二乘支持向量机
短时风速风向
估算方法
锚泊状态
wavelet transform
least squares support vector machine
short-term wind speed and direction
estimation method
anchoring condition
