Abstract: To address the significant recoverable energy generated during braking of electric drive mining truck and the multi-objective coupled challenge of optimizing energy storage system configuration for energy recovery efficiency, economy, lightweight design, and spatial compactness, this study proposes and validates a parameter collaborative optimization method based on a dynamic adaptive genetic algorithm. First, the Entropy Weight Method determines weights for key indicators. Combined with the TOPSIS comprehensive evaluation model, ternary lithium batteries (closeness coefficient 0.82) are selected as the energy storage device. Subsequently, a quad-objective coupled optimization function maximizing energy recovery efficiency while minimizing economic cost and jointly minimizing mass-volume is constructed. An improved dynamic adaptive genetic algorithm solves this problem, incorporating a nonlinear dynamic adaptive mechanism to adjust crossover and mutation probabilities for enhanced search efficiency. Combined with NSGA-II non-dominated sorting for Pareto frontier analysis, the global optimal solution set is obtained. To validate this optimized configuration, an AMESim-based co-simulation model coupling vehicle dynamics with ternary lithium battery nonlinear characteristics is built and simulated under typical mining conditions. Results show: 19.6% overall performance improvement, 13.2% energy storage cost reduction, 13.1% volume decrease, and 13.2% mass reduction. Key performance validation confirms: battery SOC fluctuation stably controlled below 5%; DC bus voltage precisely maintained at 1 300 V±2% with effective braking transient voltage spike suppression; motor torque response error below 3%. This study resolves the multi-objective optimization challenge for mining truck energy storage systems, demonstrating significantly enhanced performance with reduced cost, volume, and weight while exhibiting excellent robustness and engineering applicability under complex mining conditions.