Lithium-Ion Battery RUL Prediction with KRML and Kalman Filtering
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing lithium-ion battery prediction models struggle to accurately predict remaining useful life (RUL) due to the inability to incorporate prior knowledge and handle uncertainty in real-time data, leading to challenges in maintaining battery health and safety.
Innovation Solution
A self-adaptive lithium-ion battery method using knowledge-reinforced machine learning (KRML) integrates an artificial neural network (ANN) with dual extended Kalman filters (DEKFs) and Gaussian process regression to capture battery capacity data, training a stochastic capacity degradation model for precise RUL prediction and temperature adjustment.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If data-driven machine learning methods are used to predict battery RUL, then prediction capability is improved, but prediction accuracy and uncertainty reduction remain challenging due to lack of prior knowledge incorporation
Solution Approach 1:
The patent applies preliminary action by incorporating prior knowledge about battery capacity fade into the machine learning model before actual RUL prediction. The knowledge-reinforced Gaussian process regression integrates established battery degradation patterns and physical constraints as prior information, which guides the prediction process and reduces uncertainty from the outset rather than attempting to correct inaccuracies after prediction.
Solution Approach 2:
The patent uses knowledge-reinforced Gaussian process regression as an intermediary between raw monitoring data and RUL prediction. This intermediary layer incorporates prior knowledge about battery degradation mechanisms, acting as a bridge that filters and interprets data through the lens of established scientific understanding, thereby improving both accuracy and reducing uncertainty.
2Device complexity
If traditional battery models are used for predicting battery aging, then model simplicity is maintained, but accurate RUL prognostics are difficult due to inability to incorporate large amount of real-time data while handling various sources of uncertainty
Solution Approach 1:
The patent applies dynamics by using a sequential updating approach where the Gaussian process regression model continuously incorporates new real-time monitoring data while maintaining prior knowledge. The model dynamically adapts to changing battery conditions and uncertainty levels, allowing the structure to remain relatively simple while processing large amounts of data through incremental updates rather than requiring complex batch processing architectures.
Solution Approach 2:
The patent changes parameters by incorporating uncertainty quantification into the RUL prediction framework. The Gaussian process regression provides not only point estimates but also confidence intervals, transforming the prediction output from a single value to a probability distribution. This parameter change allows the model to handle various sources of uncertainty explicitly while maintaining computational tractability.
Data Source
AI summary
The present disclosure provides a self-adaptive lithium-ion battery method using knowledge-reinforced machine learning and Kalman filtering, an electronic device, and a storage medium. The method includes training and synchronizing an artificial neural network with dual extended Kalman filters to capture battery capacity data of each of lithium-ion batteries; integrating prior knowledge with Gaussian process regression to form an integrated knowledge-reinforced Gaussian process regression; training a stochastic capacity degradation model by employing integrated knowledge-reinforced Gaussian process regression with captured battery capacity data to obtain a trained stochastic capacity degradation model; performing capacity prediction using trained stochastic capacity degradation model to obtain remaining useful life of one or more testing lithium-ion batteries; generating an air mass flow rate and a charging/discharging rate by a controller; and inputting the air mass flow rate and the charging/discharging rate into a battery thermal management system to improve battery RULs by adjusting lithium-ion battery temperature.