Unscented Kalman Filtering for Prognostics Under Varying Operational and Environmental Conditions

Abstract

Prognostics gained a lot of research attention over the last decade, not the least due to the rise of data-driven prediction models. Also hybrid approaches are being developed that combine physics-based and data-driven models for better performance. However, limited attention is given to prognostics for varying operational and environmental conditions. In fact, varying operational and environmental conditions can significantly influence the remaining useful life of assets. A powerful hybrid tool for prognostics is Bayesian filtering, where a physical degradation model is updated based on realtime data. Although these types of filters are widely studied for prognostics, application for assets in varying conditions is rarely considered in literature. In this paper, it is proposed to apply an unscented Kalman filter for prognostics under varying operational conditions. Four scenarios are described in which a distinction is made between the level in which real-time and future loads are known and between short-term and long-term prognostics. The method is demonstrated on an artificial crack growth case study with frequently changing stress ranges in two different stress profiles. After this specific case, the generic application of the method is discussed. A positioning diagram is presented, indicating in which situations the proposed filter is useful and feasible. It is demonstrated that incorporation of physical knowledge can lead to highly accurate prognostics due to a degradation model in which uncertainty in model parameters is reduced. It is also demonstrated that in case of limited physical knowledge, data can compensate for missing physics to yield reasonable predictions.

Atmospheric Corrosion Prognostics Using a Particle Filter

“…This is due to the fact that the corrosion process is highly uncertain and noisy. Parameter convergence in Bayesian filters is only obtained when the degradation model and its governing loads are fully defined (Keizers et al, 2021). Because of the uncertain corrosion model which does not include all governing loads, the degradation model is not fully defined and adap- tivity remains.…”

Smart Maintenance for Modern Naval Ships

“…A limited amount of failure data (caused by a limited number of failures but also poor failure registrations) makes it hard to perform accurate remaining useful life analyses. Therefore, others work on developing hybrid prognostic models that combine physics-of-failure methods with a data driven approach, as done in (Keizers et al, 2021), and apply these to diesel engine data sets.…”

Integration of multiple failure mechanisms in a life assessment method for centrifugal pump impellers

“…Another way to assess the predictive performance of the proposed method would be to compare with periodic condition measurements or inspections, that define the momentary condition of the impeller. In scientific literature, several hybrid approaches are available to use Bayesian updating to tune (physical) models with this kind of data, see for example Keizers et al 69 However, in industrial practice, especially for relatively simple systems like pumps, these condition measurements or inspection results are typically not available, and can thus not be used to update or validate the prediction models. Due to this lack of realistic validation data, the authors decided to normalize data to the available case.…”