Stiff-PINN: Physics-Informed Neural Network for Stiff Chemical Kinetics

Recently developed physics-informed neural network (PINN) has achieved success in many science and engineering disciplines by encoding physics laws into the loss functions of network, such that not only conforms to measurements, initial boundary conditions but also satisfies governing equations. This work first investigates performance PINN solving stiff chemical kinetic problems with equations ordinary differential (ODEs). The results elucidate challenges utilizing ODE systems. Consequently, we employ Quasi-Steady-State-Assumptions (QSSA) reduce stiffness systems, then can be successfully applied converted non/mild-stiff Therefore, suggest could major reason for failure regular studied Stiff-PINN approach utilizes QSSA enable solve kinetics shall open possibility applying various reaction-diffusion systems involving dynamics.