极值-I型风速预测的Bayes方法
发布时间：2018-10-17 07:41
【摘要】：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.
[Abstract]:In order to improve the accuracy of extreme value-I wind speed prediction, the expression of extreme value-I wind speed prediction is derived by using Lindley approximation method in Bayes estimation on the basis of Jeffreys criterion. The Monte Carlo method is used to generate the pseudo wind speed master sample with the distribution of extreme value-I, and the extreme value-I model wind speed prediction method based on the Bayes theory and the maximum likelihood estimation theory is used to predict the wind speed, respectively, based on the pseudo wind speed master sample. And compared with the theoretical value of pseudo wind speed master sample. The results show that compared with the maximum likelihood estimation method, the maximum value I wind speed prediction model based on Bayes theory is more accurate than the maximum likelihood estimation method, and the accuracy increases with the increase of the sample size of the pseudo-wind speed sample. The increase of the number of a priori sample and the increase of the prior variance have no effect on the accuracy of the calculation.
【作者单位】： 土木工程防灾国家重点实验室(同济大学);
【基金】：科技部国家重点实验室基金(SLDRCE14-B-08)
【分类号】：U442.59
本文编号：2275986
[Abstract]:In order to improve the accuracy of extreme value-I wind speed prediction, the expression of extreme value-I wind speed prediction is derived by using Lindley approximation method in Bayes estimation on the basis of Jeffreys criterion. The Monte Carlo method is used to generate the pseudo wind speed master sample with the distribution of extreme value-I, and the extreme value-I model wind speed prediction method based on the Bayes theory and the maximum likelihood estimation theory is used to predict the wind speed, respectively, based on the pseudo wind speed master sample. And compared with the theoretical value of pseudo wind speed master sample. The results show that compared with the maximum likelihood estimation method, the maximum value I wind speed prediction model based on Bayes theory is more accurate than the maximum likelihood estimation method, and the accuracy increases with the increase of the sample size of the pseudo-wind speed sample. The increase of the number of a priori sample and the increase of the prior variance have no effect on the accuracy of the calculation.
【作者单位】： 土木工程防灾国家重点实验室(同济大学);
【基金】：科技部国家重点实验室基金(SLDRCE14-B-08)
【分类号】：U442.59
【相似文献】
相关期刊论文 前9条
1 周源泉,郭建英;指数分布单样预测的Bayes方法[J];质量与可靠性;2001年03期
2 周源泉,郭建英;指数分布双样预测的Bayes方法[J];强度与环境;2002年01期
3 陈曜;王红芳;李东风;;SPA和Bayes方法在城市可持续发展评价中的应用[J];四川水利;2009年03期
4 于捷;石耀霖;张海龙;申桂香;贾亚洲;;基于Bayes方法的数控机床可靠性研究[J];机床与液压;2009年08期
5 席战伟;;Bayes方法在鱼雷系统可靠性试验与评定分析中的应用[J];电子制作;2014年03期
6 赵飞;王华伟;王岩峰;;基于Bayes方法的民机冗余部件随机可靠性评估[J];机械科学与技术;2011年01期
7 潘文庚,童雅湘,潘火荣;弹药可靠性鉴定中的Bayes方法应用[J];弹箭与制导学报;2000年01期
8 伏洪勇;林宝军;陈福恩;;一种基于通用失效率的航天设备可靠性评估算法[J];空间科学学报;2007年03期
9 ;[J];;年期
相关硕士学位论文 前3条
1 周枫;基于Bayes方法的海岛环境弹药储存可靠性研究[D];国防科学技术大学;2010年
2 刘丽平;基于Bayes方法的航天继电器可靠性试验研究[D];哈尔滨工业大学;2007年
3 李文哲;分层Bayes方法在安徽省气象数据格点化研究中的应用[D];中国科学技术大学;2011年
，本文编号：2275986
本文链接：https://www.wllwen.com/kejilunwen/daoluqiaoliang/2275986.html
