Supervisor:
Southwest Ordnance Industry Bureau
Organizer:
Chongqing Ordnance Industry Society
Chongqing University of Technology
Chongqing University of Technology
Structure optimization of DC EMP based on BP neural network and NSGA Ⅱ algorithm
| DOI: | 10.11809/bqzbgcxb2025.01.024 |
| Keywords: | DC EMP; pressure; efficiency; BP neural network; NSGA Ⅱ |
| Abstract: | To improve the pressure of DC EMP to drive the liquid metal flow for heat dissipation and improve the transmission efficiency of the pump, the structural parameters and input current of the pump were optimized by using the optimization design method combining numerical simulation and intelligent algorithm. With liquid metal gallium indium tin alloy as the working fluid, and targeting a flow condition of 1.5 L/min, the electromagnetic pump was numerically simulated using COMSOL software. The static pressure difference of the pump was obtained and compared with experimental values to verify the accuracy of the simulation model. The Box Behnken experimental design method was used for data collection, and based on the experimental data, a BP neural network surrogate model was constructed. With the objectives of maximizing pressure and efficiency, structural optimization was carried out using the NSGA Ⅱ algorithm. The results show that under the same flow rate, the input current is reduced by 20 A after optimization, which is more conducive to heat dissipation;the pressure and efficiency are increased by 2 577.6 Pa and 11.8% respectively, the loss along the way is reduced, and the overall performance is improved. Compared with the flow field analysis, the flow velocity inside the pump is more uniform and the streamline distribution is more stable. |
| Issue: | Vol. 46 No. 1 (2025) |
| Published: | 2025-01-31 |