Model Surrogate-Assisted Multi-Objective Optimization of Distribution Structure for a Single-Piston Two-Dimensional Electro-Hydraulic Pump

Under high-frequency commutation conditions, the Single-Piston Two-Dimensional Electro-Hydraulic Pump suffers from severe reverse flow and pressure pulsation, which limit its volumetric efficiency and operational stability. To address this issue, this study proposes a surrogate-assisted multi-objective optimization framework for the pump distribution structure. First, a dynamic model is established to analyze the influence of triangular damping groove geometry on flow and pressure characteristics, and four key parameters are selected as design variables. Then, sample data generated from AMESim simulations are used to train a Genetic Algorithm-optimized Backpropagation neural network surrogate model. Finally, the surrogate model is integrated with NSGA-II to minimize the peak reverse flow and pressure pulsation amplitude simultaneously. The results show that the GA-BP model predicts reverse flow and pressure pulsation with mean relative errors of 2.72% and 2.99%, respectively. Compared with the initial design, the optimized structure reduces the peak reverse flow by 27.6% and decreases the pressure pulsation amplitude from 0.78 MPa to 0.41 MPa. These results indicate that, within the parameter ranges and operating conditions considered in this study, the proposed framework provides an effective tool for the coordinated optimization of damping groove parameters for the Single-Piston Two-Dimensional Electro-Hydraulic Pump.