Optimization of the Novel Three-Stage Throttling Groove of Proportional Spool Valve Based on the Multi-Objective Particle Swarm Optimization Algorithm

The proportional spool valve is a kind of extremely widely used flow control valve in hydraulic servo control systems, and the structure of its spool is one of the core factors that determine its operating characteristics. The throttling grooves of the spool are normally designed and optimized to meet desired flow characteristics. To obtain superior flow control accuracy, this paper is not subject to the traditional shapes of the throttling groove and proposes a novel three-stage combined groove shape. The flow characteristics of the new shape are analyzed by computational fluid dynamics (CFD) simulation. Paving the way for optimization, the Latin hypercube method is adopted to design the experiment. Based on the simulation datasets, the deep neural network (DNN) is used to establish the response surface of throttling groove parameters. With the response surface, the multi-objective particle swarm optimization (MPSO) algorithm is applied for parameter optimization. Moreover, not only the flow rate and flow force are used as the optimization objectives, but also the pressure gradient of the throttle groove is introduced as one of the optimization objectives. Finally, the flow characteristic curves with Pareto optimal parameters are imported into AMEsim software, and the improvement of the flow control accuracy of the proposed structure is verified by comparing simulation results with U-shape groove shape.