The invention discloses a multi-objective optimization design method for a high-
voltage nanosecond pulse
adder, and aims to solve the problem that the prior art depends on experience
trial and error and is difficult to consider multiple objectives such as fast
leading edge, high efficiency, portability and reliability. The method comprises the following steps: firstly, building a
simulation and experiment platform, and acquiring output waveform data for changing topological parameters; constructing a
physical model of the influence rule of the parasitic parameters on the pulse rising edge time and the
voltage gain, and establishing a power supply efficiency prediction model based on a deep neural network; taking the minimization of the pulse rising edge time and the maximization of the
voltage gain and the energy efficiency as targets, constructing a
particle swarm optimization multi-target optimization model in combination with the model, and obtaining a
Pareto optimal solution set under the constraints of portability and parasitic parameters; and finally, carrying out experimental
verification iteration to obtain an optimal scheme. According to the invention, the design period is obviously shortened, the multi-target conflict is effectively balanced, the rising edge time is shortened, the energy efficiency is improved, the voltage
gain is improved, the portability is enhanced through the
modular design, and the wearable application is compatible.