In predicting the remaining lifespan of Proton Exchange Membrane Fuel Cells (PEMFC), it is crucial to accurately capture the multi-scale variations in cell performance. This study employs Variational Mode Decomposition (VMD) to decompose performance data into intrinsic modes, elucidating critical multi-scale dynamics vital for understanding the complex degradation processes in fuel cells. In addition to VMD, this research utilizes an Improved Whale Optimization Algorithm (IWOA) to optimize a Back Propagation (BP) Neural Network. The IWOA focuses on precise adjustments of weights and biases, enabling the BP network to effectively interpret complex nonlinear relationships within the dataset. This optimization enhances the predictive model’s reliability and stability. Extensive experimental evaluations demonstrate that the integration of VMD, and the learning capabilities of the IWOA-optimized BP network significantly improves the model’s accuracy and stability across multiple predictions, thereby increasing the reliability of lifespan predictions for PEMFCs. This methodology offers a robust framework for extending the operational life and efficiency of fuel cells.

This paper addresses the challenge of degradation prediction in proton-exchange membrane fuel cells (PEMFCs). Traditional methods often struggle to balance accuracy and complexity, particularly under dynamic operational conditions. To overcome these limitations, this study proposes a data-driven approach based on the gated recurrent unit (GRU) neural network, optimized by the grey wolf optimizer (GWO). The integration of the GWO automates the hyperparameter tuning process, enhancing the predictive performance of the GRU network. The proposed GWO-GRU method was validated utilizing actual PEMFC data under dynamic load conditions. The results demonstrate that the GWO-GRU method achieves superior accuracy compared to other standard methods. The method offers a practical solution for online PEMFC degradation prediction, providing stable and accurate forecasting for PEMFC systems in dynamic environments.

This paper presents an inductor current-based maximum power point tracking (IC-MPPT) strategy and a single-inductor multi-input single-output (SI-MISO) structure with energy storage battery for distributed photovoltaic (PV) systems. In this study framework, the duty cycle of each PV channel can be controlled independently based on the presented IC-MPPT strategy, and the components/sensors costs are reduced through the presented SI-MISO PV system structure. In addition, a model predictive control (MPC) method is presented to regulate DC bus voltage, by controlling the bidirectional converter in the battery circuit. The presented control strategies have been rigorously derived and experimentally validated, and the experimental results demonstrate that each PV module can rapidly and efficiently track to the maximum power point in less than 0.016 s, while the bus voltage is stabilized near the set value, with an overshoot of less than 2.6%.

The State of Charge (SOC) is vital for battery system management. Enhancing SOC estimation boosts system performance. This paper presents a combined framework that improves SOC estimation’s accuracy and stability for electric vehicles. The framework combines a Long Short-Term Memory (LSTM) network with an Adaptive Unscented Kalman Filter (AUKF). An Improved Arithmetic Optimization Algorithm (IAOA) fine-tunes the LSTM’s hyperparameters. Its novelty lies in its adaptive iteration algorithm, which adjusts iterations based on a threshold, optimizing computational efficiency. It also integrates a genetic mutation strategy into the AOA to overcome local optima by mutating iterations. Additionally, the AUKF’s adaptive noise algorithm updates noise covariance in real-time, enhancing SOC estimation precision. The inputs of the proposed method include battery current, voltage, and temperature, then producing an accurate SOC output. The predictions of LSTM are refined through AUKF to obtain reliable SOC estimation. The proposed framework is firstly evaluated utilizing a public dataset and then applied to battery packs on actual engineering vehicles. Results indicate that the Root Mean Square Errors (RMSEs) of the SOC estimations in practical applications are below 0.6%, and the Maximum Errors (MAX) are under 3.3%, demonstrating the accuracy and robustness of the proposed combined framework.