基于ANN求导的DSFT中故障概率变化趋势研究
为了在遇到不利工作环境之前,提前采取措施控制元件故障发生,提出基于ANN求导的元件故障概率变化趋势的确定方法。该方法可在不了解系统或元件构成和性质的情况下,仅利用实际故障监测数据分析不同工作环境下元件故障概率变化的趋势和程度。同时该方法也充实了空间故障树(SFT)下的离散型空间故障树(DSFT)理论。论文给出了ANN求导法处理问题的理论基础和公式推导。结合了一个元件进行了方法的应用,并最终得到了该元件的故障概率变化趋势。为实际生产中“先知先觉”的故障预防控制措施提供参考。...
Saved in:
| Published in | 计算机应用研究 Vol. 34; no. 2; pp. 449 - 452 |
|---|---|
| Main Author | |
| Format | Journal Article |
| Language | Chinese |
| Published |
辽宁工程技术大学安全科学与工程学院,辽宁阜新123000
2017
大连交通大学辽宁省隧道与地下结构工程技术研究中心,辽宁大连116028%辽宁工程技术大学安全科学与工程学院,辽宁阜新,123000%大连交通大学辽宁省隧道与地下结构工程技术研究中心,辽宁大连,116028%辽宁工程技术大学力学与工程学院,辽宁阜新,123000 |
| Subjects | |
| Online Access | Get full text |
| ISSN | 1001-3695 |
| DOI | 10.3969/j.issn.1001-3695.2017.02.029 |
Cover
| Summary: | 为了在遇到不利工作环境之前,提前采取措施控制元件故障发生,提出基于ANN求导的元件故障概率变化趋势的确定方法。该方法可在不了解系统或元件构成和性质的情况下,仅利用实际故障监测数据分析不同工作环境下元件故障概率变化的趋势和程度。同时该方法也充实了空间故障树(SFT)下的离散型空间故障树(DSFT)理论。论文给出了ANN求导法处理问题的理论基础和公式推导。结合了一个元件进行了方法的应用,并最终得到了该元件的故障概率变化趋势。为实际生产中“先知先觉”的故障预防控制措施提供参考。 |
|---|---|
| Bibliography: | 51-1196/TP In order to take measures in advance control element failure before coming bad work environment, this paper put forward the method of determining the changing trend of component failure probability based on ANN derivation. Without the composition and nature of system or component, and only with the actual fault monitoring data, this method could understand the trend and changing degree of system or component in different working conditions. The method also enriched the theory of the discrete space fault tree(DSFT) in the framework of space fault tree. It presented the theoretical basis and the derived formula with ANN derivation. Combined with a component for the application of'the proposed method, it obtained the change tendency of the failure probability distribution finally. It provided "the reference to adopt measures to prevent failure in advance in the process of production. Cui Tiejun1,3, Li Shasha1, Ma Yundong2, Wang Laigui2(1. a. College of Safety Science & Engineering, b. College of Mechani |
| ISSN: | 1001-3695 |
| DOI: | 10.3969/j.issn.1001-3695.2017.02.029 |