Predictive Maintenance by Risk Sensitive Particle Filtering

Risk Sensitive Particle Filtering in Support of Predictive Maintenance

Predictive Maintenance (PrM) exploits the estimation of the equipment Residual Useful Life (RUL) to identify the optimal time for carrying out the next maintenance action. Particle Filtering (PF) is widely used as prognostic tool in support of PrM, by reason of its capability of robustly estimating the equipment RUL without requiring strict modeling hypotheses. However, a precise estimate of th...

Risk Sensitive Particle Filters

We propose a new particle filter that incorporates a model of costs when generating particles. The approach is motivated by the observation that the costs of accidentally not tracking hypotheses might be significant in some areas of state space, and irrelevant in others. By incorporating a cost model into particle filtering, states that are more critical to the system performance are more likel...

Particle Filtering

Optimal filtering: The filtering problem involves the estimation of the state vector at time k, given all the measurements up to and including time k, which we denote by z1:k. In a Bayesian setting, this problem can be formalized as the computation of the distribution p(xk|z1:k), which can be done recursively in two steps. In the prediction step, p(xk|z1:k−1) is computed from the filtering dist...

Predictive Maintenance

Particle Filtering Optimized by Swarm Intelligence Algorithm

A new filtering algorithm — PSO-UPF was proposed for nonlinear dynamic systems. Basing on the concept of re-sampling, particles with bigger weights should be re-sampled more time, and in the PSO-UPF, after calculating the weight of particles, some particles will join in the refining process, which means that these particles will move to the region with higher weights. This process can be regard...