Improved Multi-Fusion Black-Winged Kite Algorithm for Optimizing Stochastic Configuration Networks for Lithium Battery Remaining Life Prediction

The accurate estimation of lithium battery state of health (SOH) plays an important role in the health management of battery systems. In order to improve the prediction accuracy of SOH, this paper proposes a stochastic configuration network based on a multi-converged black-winged kite search algorithm, called SBKA-CLSCN. Firstly, the indirect health index (HI) of the battery is extracted by combining it with Person correlation coefficients in the battery charging and discharging cycle point data. Secondly, to address the problem that the black-winged kite optimization algorithm (BKA) falls into the local optimum problem and improve the convergence speed, the Sine chaotic black-winged kite search algorithm (SBKA) is designed, which mainly utilizes the Sine mapping and the golden-sine strategy to enhance the algorithm’s global optimality search ability; secondly, the Cauchy distribution and Laplace regularization techniques are used in the SCN model, which is referred to as CLSCN, thereby improving the model’s overall search capability and generalization ability. Finally, the performance of SBKA and SBKA-CLSCN is evaluated using eight benchmark functions and the CALCE battery dataset, respectively, and compared in comparison with the Long Short-Term Memory (LSTM) model and the Gated Recurrent Unit (GRU) model, and the experimental results demonstrate the feasibility and effectiveness of the SBKA-CLSCN algorithm.