Data-driven femtosecond optical soliton excitations and parameters discovery of the high-order NLSE using the PINN

We use the physics-informed neural network to solve a variety of femtosecond optical soliton solutions high-order nonlinear Schrödinger equation, including one-soliton solution, two-soliton rogue wave W-soliton solution and M-soliton solution. The prediction error for one-soliton, is smaller. As distance increases, will gradually increase. unknown physical parameters equation are studied by using as data sets. optimized from three aspects number layers network, neurons, sampling points. Compared with previous research, our greatly reduced. This not replacement traditional numerical method, but hopefully open up new ideas.