Regularised Iterative Generalised Least Squares with Optimal Selection of the Hyper-Parameter for Identifying Nonlinear Phenomenological Models

The paper proposes a ridge regression scheme to address parameter confounding in phenomenological models, such as those used for state‑of‑health prediction in lithium‑ion batteries. An automated, information‑theoretic method optimises the ridge hyper‑parameter at each iteration, converging rapidly via fixed‑point iteration. The resulting regularised iterative generalised least squares framework can fit heteroscedastic and serially correlated data, with simulations confirming its effectiveness.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv Machine Learning.