Enhanced Hybrid Energy Storage Management Using MPC with Real- Time Battery Aging

Abstract

In order to dynamically distribute power between the hybrid energy storage system's battery and ultracapacitor, the suggested energy management approach makes use of an Adaptive Model Predictive Control (MPC) architecture. In contrast to traditional MPC methods that use static parameters, adaptive MPC constantly adjusts its internal model and control goals in response to actual system conditions. Online data on battery aging, in particular the instantaneous, time-varying C-rate calculated from the observed battery current, drives the adaptive process. The controller may modify power-split decisions in reaction to anticipated high-stress operating situations by including this real-time degradation indicator into the MPC cost function and limitations. In particular, the adaptive MPC moves more high-frequency power demand to the ultracapacitor and increases the weighting associated with battery deterioration when aggressive load transients or increased C-rates are expected within the prediction horizon. On the other side, when things aren't too hectic, the controller lowers degradation penalties to make the system work better and use less energy. An adaptive trade-off between performance, efficiency, and battery lifespan is made possible by the suggested technique by adaptively changing the MPC weighting factors and control actions. When contrasted with fixed-parameter MPC techniques, this adaptive structure increases resilience against driving-cycle variability, decreases stress on the battery, and considerably lessens capacity fading.