Recurrent Neural Network-Based Nonlinear Optimization for Braking Control of Electric Vehicles

In this paper, electro-hydraulic braking (EHB) force allocation for electric vehicles (EVs) is modeled as a constrained nonlinear optimization problem (NOP). Recurrent neural networks (RNNs) are advantageous in many folds solving NOPs, yet existing RNNs’ convergence usually requires convexity with calculation of second-order partial derivatives. recurrent network-based NOP solver (RNN-NOPS) developed. It seen that the RNN-NOPS designed to drive all state variables asymptotically converge feasible region, loose requirement on NOP’s first-order derivative. addition, RNN-NOPS’s equilibria proved meet Karush–Kuhn–Tucker (KKT) conditions, and behaves strong robustness against violation constraints. The comparative studies conducted show advantages EHB problem, it reported overall regenerative energy 15.39% more than method comparison under SC03 cycle.