Design and analysis of direct action solenoid valve based on computational intelligence

Control Rod Hydraulic Drive Mechanism (CRHDM) is a newly invented patent of the Institute of Nuclear and New Energy Technology Tsinghua University which owns CRHDM's independent intellectual property rights while the integrated valve made up of three direct action solenoid valves is the key part of this mechanism. Therefore, the performance of the solenoid valve affects the integrated valve and the CRHDM directly. In this paper, we present a method to design the parameters of the direct action solenoid valve based on orthogonal experiment design, back propagation (BP) neural network and particle swarm optimization (PSO). The result proves that the method is feasible and accurate to design the parameters in order to obtain the biggest electromagnetic force. Besides, the result also shows that it is the current which influences the electromagnetic force of the direct action solenoid valve most.