Iteration control of particle size distribution in cobalt oxalate process based on moment generation network

Particle size distribution (PSD) control is very complicated to realize due to the absence of effective measuring methods for PSD. In order to solve this problem, an iteration learning control strategy based on the moment generation network model is proposed. The tracking control of PSD which cannot be measured online is converted into the measurable moment control. There are two steps to develop the control strategy. Firstly, a moment generation network is built to construct the relationship between the moment and process measurements. Then an iteration learning control strategy is designed to drive the process to achieve a target PSD according to the repetitive nature of the process. The cobalt oxalate synthesis process was selected to test the performance of the control strategy. The experimental results demonstrated that the approach had a strong ability in convergence and resisting disturbance.