A new online fuzzy modelling method considering prior information with its application in PHM
Zhaoqiang Wang, Changhua Hu, Qi Zhang, Tao Zhou · International Journal of Advancements in Computing Technology · 2013
Prognostic modelling methods play a key role in prognostics and health management (PHM). Over recent years, a number of researches have been undertaken to establish the prognostic models, roughly divided into physical models, knowledge-based models, and data-driven models. Currently, data-driven modelling methods have received considerable attention with the development of high technologies, especially digital computer and advanced sensor techniques. Adaptive fuzzy systems belong to the data-driven modelling methods, but the fuzzy model is different from many other data-driven modelling methods owing to its advantage of using not only quantitative data but also qualitative information with fuzzy uncertainty. Compared with the prediction from an offline fuzzy model, an online prediction is more desired, since we can monitor the health condition of a system and predict its trend in real time. However, most of the existing online fuzzy modelling methods neglect the full utilization of the offline information. In this paper, a new online fuzzy modelling method considering prior information is proposed for prognosis, which is briefly called off-online fuzzy modelling method here. The results of the case studies show that our developed method gains a significant improvement over the existing fuzzy modelling methods in terms of accuracy and the corresponding prognosis algorithm is effective.