1 June 2021 in IEEE Systems Journal
In this article, we propose a practical prognosis approach. It estimates online the remaining duration of the system before the system performance requirements are no longer met to fulfill a specific mission. The systems targeted by this approach have the particularity of a lack of measuring instruments capable of providing indications of potential degradation. The approach is based on the estimated model of system behavior. The approach is in two phases. In the first phase, an observer is used to estimate unmeasured states and relevant parameters that can characterize system performance. In the second phase, in the beginning, the historical parameters obtained are used to identify models describing their dynamics using learning time-series prediction methods. In this article, the support vector regression and the adaptive neuro-fuzzy inference system are illustrated. Then, the model is simulated to estimate the future performance of the system and compare it to the desired performance. A comparison of the results obtained using learning time-series prediction methods with those obtained using the maximum likelihood method is carried out. The results show that the proposed approach makes it possible to estimate the remaining lifetime with particularly good quality. To illustrate the proposed failure prognosis approach, a Li-ion battery was used.
11 October 2018 in IFAC-PapersOnLine
24 February 2014 in Proceedings of the Institution of Mechanical Engineers, Part O: Journal of Risk and Reliability
This article deals with modelization of industrial process by using hidden Markov model. The process is seen as a discrete event system. We propose different structures based on Markov automata, called topologies. A synthetic hidden Markov model is designed in order to match to a real industrial process. The models are intended to decode industrial maintenance observations (also called “symbol”). Symbols are produced with a corresponding degradation level (also called “state”). These 2-tuple (symbol, state) are known as Markov chains, also called “a signature.” Hence, these various 2-tuple are implemented in the proposed topologies by using the Baum–Welch learning algorithm (decoding by forward variable) and the segmental K-means learning (decoding by Viterbi). We assess different frameworks (topology, learning and decoding algorithm, distribution) by relevancy measurements on model outputs. Then, we determine the most relevant framework for use in maintenance activities. Afterward, we try to minimize the size of the learning data. Thus, we could evaluate the model by using “sliding windows” of data. Finally, an industrial application is developed and compared with this framework. Our goal is to improve worker safety, maintenance policy, process reliability and reduce CO 2 emissions in the industrial sector.
1 May 2012 in IFAC Proceedings Volumes