极值-I型风速预测的Bayes方法

为提高极值-I型风速预测精度,在Jeffreys准则的基础上,采用Bayes估计中的Lindley近似方法推导极值-I型风速预测表达式.采用Monte Carlo法产生服从极值-I型分布的伪风速母样,基于伪风速母样分别采用基于Bayes理论和最大似然估计理论的极值-I型风速预测方法进行风速预测,并与伪风速母样的理论值进行对比分析.结果表明:与最大似然估计法相比,采用基于Bayes理论建立的极值-I型风速预测模型进行风速预测的精度更高,且精度随着伪风速母样样本量的增加而提高,位置参数先验样本数量的增加以及先验方差的增大对计算精度没有影响.

Wind speed prediction of extreme value type I distribution based on the Bayes method

In order to improve prediction accuracy of wind speed of extreme value type I distribution, the wind speed prediction model was proposed based on Jeffreys criterion and the Lindley approximation method of Bayesian theory. Monte Carl method was used to generate the pseudo wind speed samples, and the maximum likelihood parameter estimation method and Bayes statistical theory were used to estimate the wind prediction value of the extreme value type I distribution, then the prediction value was compared with the theoretical extreme value. The result indicates that the wind speed prediction model of extreme value type I distribution is more accurate than the maximum likelihood estimation. The accuracy increases with the increasing of pseudo wind speed sample numbers, but is not affected by the numbers of prior samples and prior variance for location parameter.