Remaining useful life estimation of Lithium-ion battery based on interacting multiple model particle filter and support vector regression

Lithium-ion batteries have become an integral part of our lives, and it is important to find a reliable and accurate long-term prognostic scheme to supervise the performance degradation and predict the remaining useful life of batteries. In the perspective of information fusion methodology, an interacting multiple model framework with particle filter and support vector regression is developed to realize multi-step-ahead estimation of the capacity and remaining useful life of batteries. During the multi-step-ahead prediction period, the support vector regression model with sliding windows is used to compensate the future measurements online. Thus, the interacting multiple model with particle filter can relocate the particles and update the capacity estimation. The probability distribution of the remaining useful life is also obtained. Finally, the proposed method is compared and validated with particle filter model using the benchmark data. The experimental results prove that the proposed model yields stable forecasting performance and narrows the uncertainty in remaining useful life estimation.

Automated summary

In this study, a battery remaining useful life prediction model based on information fusion methodology is proposed, which combines particle filter and support vector regression to achieve accurate multi-step-ahead estimation of battery capacity and remaining useful life.