Physics-informed deep learning for modelling particle aggregation and breakage processes

Particle aggregation and breakage phenomena are widely found in various industries such as chemical, agricultural pharmaceutical processes. In this study, a physics-informed neural network is developed for solving both the forward inverse problems of particle method, population balance equation directly embedded loss function so that can be trained efficiently fulfil physical constraints. For problems, solutions equations obtained through optimization where predictions well match analytical solutions. modelling, data-driven discovery model parameters investigated. The sensitivity regarding selection different structures also with approach promising processes noisy observation data.