Health assessment and prognostics based on higher‐order hidden semi‐Markov models
Health assessment and prognostics based on higher‐order hidden semi‐Markov models
复制标题
DOI：
10.1002/nav.21947
复制
发表时间：
2020-02
期刊：
影响因子：
--
通讯作者：
Ying Liao;Yisha Xiang;Min Wang
中科院分区：
文献类型：
--
作者：
Ying Liao;Yisha Xiang;Min Wang
本文提出了一种新的灵活的基于高阶隐半马尔可夫模型（HOHSMM）的系统或组件与不可观测的健康状态和复杂的过渡动力学框架。HOHSMM扩展了基本的隐马尔可夫模型（HMM），允许隐藏状态依赖于其更遥远的历史，并假设一般分布状态持续时间。设计了一种有效的Gibbs抽样算法用于HOHSMM的统计推断。我们进行了模拟研究，以评估拟议的HOHSMM采样器的性能，并检查遥远的历史依赖性的影响。我们设计了一个解码算法来估计隐藏的健康状态使用学习模型。剩余使用寿命预测使用模拟方法给定的解码隐藏状态。通过对美国国家航空航天局（NASA）涡扇发动机的实例研究，证明了所提出的发动机动力学框架的实用性。我们进一步比较了建议HOHSMM和基准混合高斯HMM预测方法之间的RUL预测性能。结果表明，基于HOHSMM的故障诊断框架为复杂系统提供了良好的隐藏健康状态评估和RUL估计。
This paper presents a new and flexible prognostics framework based on a higher‐order hidden semi‐Markov model (HOHSMM) for systems or components with unobservable health states and complex transition dynamics. The HOHSMM extends the basic hidden Markov model (HMM) by allowing the hidden state to depend on its more distant history and assuming generally distributed state duration. An effective Gibbs sampling algorithm is designed for statistical inference of the HOHSMM. We conduct a simulation study to evaluate the performance of the proposed HOHSMM sampler and examine the impacts of the distant‐history dependency. We design a decoding algorithm to estimate the hidden health states using the learned model. Remaining useful life is predicted using a simulation approach given the decoded hidden states. The practical utility of the proposed prognostics framework is demonstrated by a case study on National Aeronautics and Space Administration (NASA) turbofan engines. We further compare the RUL prediction performance between the proposed HOHSMM and a benchmark mixture of Gaussians HMM prognostics method. The results show that the HOHSMM‐based prognostics framework provides good hidden health‐state assessment and RUL estimation for complex systems.