Automated synthesis of steady-state continuous processes using reinforcement learning

Abstract Automated flowsheet synthesis is an important field in computer-aided process engineering. The present work demonstrates how reinforcement learning can be used for automated without any heuristics or prior knowledge of conceptual design. environment consists a steady-state simulator that contains all physical knowledge. An agent trained to take discrete actions and sequentially build up flowsheets solve given problem. A novel method named SynGameZero developed ensure good exploration schemes the complex Therein, modelled as game two competing players. plays this against itself during training artificial neural network tree search forward planning. applied successfully reaction-distillation quaternary system.