| 基于正则化与遗忘因子的极限学习机及其在故障预测中的应用 |
| |
| 引用本文： | 杜占龙,李小民,郑宗贵,张国荣,毛 琼.基于正则化与遗忘因子的极限学习机及其在故障预测中的应用[J].仪器仪表学报,2015,36(7):1546-1553. |
| |
| 作者姓名： | 杜占龙 李小民 郑宗贵 张国荣 毛 琼 |
| |
| 作者单位： | 1.军械工程学院无人机工程系石家庄050003；2.第二炮兵研究院北京100085；
3.厦门警备区厦门361003 |
| |
| 基金项目： | 总装院校科技创新工程（ZYX12080008）项目资助 |
| |
| 摘 要： | 为了解决在线贯序极限学习机(OS-ELM)算法容易产生奇异矩阵、算法贯序更新过程中没有考虑训练样本时效性的问题,提出基于l2-正则化和自适应遗忘因子的OS-ELM(RFOS-ELM)算法。RFOS-ELM在初始阶段加入正则化机制,克服因矩阵奇异而降低OS-ELM泛化能力的缺点。在贯序更新阶段,RFOS-ELM通过引入自适应遗忘因子实时调整新旧训练样本所占比重,推导正则化条件下带遗忘因子RFOS-ELM的递推更新算法,提高其对动态变化系统的跟踪能力。某型无人机机载发射机故障预测实例表明,相比于传统OS-ELM和正则化OS-ELM算法,本文提出方法具有更高的预测精度。
|
| 关 键 词： | 故障预测 时间序列 在线贯序极限学习机 l2-正则化 遗忘机制 |
Extreme learning machine based on regularization and forgetting factor
and its application in fault prediction |
| |
| Du Zhanlong,Li Xiaomin,Zheng Zonggui,Zhang Guorong,Mao Qiong.Extreme learning machine based on regularization and forgetting factor
and its application in fault prediction[J].Chinese Journal of Scientific Instrument,2015,36(7):1546-1553. |
| |
| Authors: | Du Zhanlong Li Xiaomin Zheng Zonggui Zhang Guorong Mao Qiong |
| |
| Affiliation: | 1.Department of UAV Engineering, Ordnance Engineering College, Shijiazhuang 050003, China;
2.Academe of Second Artillerist, Beijing 100085, China; 3. Xiamen Garrison, Xiamen 361003, China |
| |
| Abstract: | On line sequential extreme learning machine (OS ELM) algorithm is prone to generate singularity matrix, and the OS ELM has no consideration about the training sample timeliness during the sequential updating process. To solve the problems mentioned above, an improved OS ELM algorithm (RFOS ELM) is presented based on l2 regularization and adaptive forgetting factor. Regularization mechanism is utilized in the RFOS ELM initialization phase, which avoids the shortcoming of worsening the OS ELM generalization ability due to the matrix singularity. During sequential updating phase, the RFOS ELM introduces an adaptive forgetting factor to adjust the proportion between new training sample and the old one in real time. The RFOS ELM recursive updating algorithm with forgetting factor is deduced under the regularization condition, which improves its tracking ability for dynamic varying system. The fault prediction case study on a certain unmanned aerial vehicle transmitter indicates that compared with conventional OS ELM and regularized OS ELM, the proposed method achieves higher prognostic accuracy. |
| |
| Keywords: | failure prognosis time series on line sequential extreme learning machine l2 regularization forgetting mechanism |
| 本文献已被 CNKI 等数据库收录！ |
| 点击此处可从《仪器仪表学报》浏览原始摘要信息 |
|
点击此处可从《仪器仪表学报》下载全文 |
|
