您的位置：首页 > 论文页面
基于贝叶斯网络的风机偏航系统声学检测方法
发表时间：2019-12-31 浏览量：940 下载量：162
| 全部作者： | 谢磊，高宝成，夏政 |
| 作者单位： | 北京邮电大学自动化学院；广东德风科技有限公司工程部 |
| 摘 要： | 针对目前风电机组偏航系统故障诊断方法的不足，研究提出一种基于贝叶斯网络的声学诊断算法。首先，分析偏航系统声信号的特点，提取无量纲的倍频程能量比来表征偏航系统的状态。然后，对所提特征进行离散化，将连续特征值转化为离散属性值。在此基础上，设计基于自组织映射（self-organizing map，SOM）算法的贝叶斯网络故障诊断模型。最后，基于实测数据验证了算法的有效性。 |
| 关 键 词： | 信息处理技术；故障诊断；偏航系统；贝叶斯网络；倍频能量比；离散化 |
| Title： | Acoustic detection method of wind turbine yaw systems based on Bayesian networks |
| Author： | XIE Lei, GAO Baocheng, XIA Zheng |
| Organization： | School of Automation, Beijing University of Posts and Telecommunications; Engineering Department, Guangdong Define Energy System Co. Ltd. |
| Abstract： | In view of the shortcomings of current fault diagnosis methods for wind turbine yaw systems, an acoustic detection method based on Bayesian network is proposed in this paper. Firstly, the characteristics of the acoustic signal of the yaw system are analyzed, the dimensionless octave energy ratios are extracted for characterizing the state of yaw system. Subsequently, the proposed feature is discretized and the eigenvalue is converted into a discrete attribute value. On this basis, a Bayesian network fault diagnosis model based on the self-organizing map (SOM) is designed. Finally, the effectiveness of the proposed algorithm is validated based on the dataset. |
| Key words： | information processing technology; fault diagnosis; yaw system; Bayesian network; octave energy ratios; discretization |
| 发表期数： | 2019年12月第6期 |
| 引用格式： | 谢磊，高宝成，夏政. 基于贝叶斯网络的风机偏航系统声学检测方法[J]. 中国科技论文在线精品论文，2019，12（6）：1008-1013. |
4
评论数 0请您登录
暂无评论
下载全文
多维论文