Gaussian process regression for forecasting battery state of health

By Robert R. Richardson, Michael A. Osborne, David A. Howey. 2017

627 citations recorded by the source corpus.

Abstract Accurately predicting the future capacity and remaining useful life of batteries is necessary to ensure reliable system operation and to minimise maintenance costs. The complex nature of battery degradation has meant that mechanistic modelling of capacity fade has thus far remained intractable; however, with the advent of cloud-connected devices, data from cells in various applications is becoming increasingly available, and the feasibility of data-driven methods for battery prognostics is increasing. Here we propose Gaussian process (GP) regression for forecasting battery state of health, and highlight various advantages of GPs over other data-driven and mechanistic approaches. GPs are a type of Bayesian non-parametric method, and hence can model complex systems whilst handling uncertainty in a principled manner. Prior information can be exploited by GPs in a variety of ways: explicit mean functions can be used if the functional form of the underlying degradation model is available, and multiple-output GPs can effectively exploit correlations between data from different cells. We demonstrate the predictive capability of GPs for short-term and long-term (remaining useful life) forecasting on a selection of capacity vs. cycle datasets from lithium-ion cells.

Subfields: Automotive Engineering, Control and Systems Engineering, Electrical and Electronic Engineering

View Paper: https://www.semanticscholar.org/paper/98bc6779e4fb4451f24a56f76bfbd15fc581bcf1

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Automotive Engineering

In Control and Systems Engineering

In Electrical and Electronic Engineering

Citations in the corpus

Gaussian Process Regression With Automatic Relevance Determination Kernel for Calendar Aging Prediction of Lithium-Ion Batteries (2020, background): “Supposing that the same Gaussian distribution exists between the training set i and the new dataset i′, then the corresponding output y′ can be calculated by the conditional distribution as [24], [33]”
The challenge and opportunity of battery lifetime prediction from field data (2021, methodology): “For prognostics, methods based on incremental updates of capacity fade [35, 36, 45, 50, 51] may work well with field data since they are able to handle rapidly varying operating conditions, usually using some”
Identifying degradation patterns of lithium ion batteries from impedance spectroscopy using machine learning (2020, methodology): “We use a zero mean function and a linear (LIN) covariance function Although GPR has been used in the literature in the context of Li-ion batteries 17,33,34 , we depart from those pioneering works by employing impedance spectra as input, as well employing ARD to shed light on salient frequencies.”
Gaussian Process Regression for In Situ Capacity Estimation of Lithium-Ion Batteries (2017, background): “have been previously used in relatively few studies on battery diagnostics/prognostics [25]–[27]; however, they possess a variety of desirable attributes.”
A Data-Driven Approach With Uncertainty Quantification for Predicting Future Capacities and Remaining Useful Life of Lithium-ion Battery (2021, background): “According to [21], these short-term capacity regenerations commonly occur in real-world applications.”
Physics-informed neural network for lithium-ion battery degradation stable modeling and prognosis (2024, positive): “Commonly used data-drivenmethodsincludelinearregression 25 ,support vectormachines 26 , Gaussian process regression 27 , deep neural networks 28,29 , etc. Xia et al. 30 extracted features from incremental capacity (IC) curves and differential voltage (DV) curves to estimate SOH.”