Neural-Network-Based Nonlinear Model Predictive Control of Multiscale Crystallization Process

The purpose of this study was to develop an integrated control strategy for multiscale crystallization processes. An image analysis method using a deep learning neural network is used measure the fine-scale information process, and mathematical statistical adopted obtain mean size crystal population. A feedforward subsequently trained employed in nonlinear model predictive formulation optimal profile manipulated variable. effectiveness proposed evaluated alum cooling experiments. Experimental results demonstrate benefits combination process.