Study On Optimization Control Of Batch Crystallization Process

The batch crystallization process is one of the key operational processes in the food,chemical,pharmaceutical and other industries,especially in the production of small-scale,high value-added chemicals.The operation of the crystallization process not only affects the quality of the crystal,but also affects the subsequent process operations.The current challenge of the batch crystallization process is to improve product quality and consistency of crystal quality between batches to meet the demand for high quality chemicals and pharmaceuticals.The optimization control research of batch crystallization process has important application value for improving crystal production and quality of crystallization industry.Aiming at the optimization control problem of batch crystallization process,this paper established a batch crystallization process model of cooling and anti-solvent crystallization,and proposed a method to optimize the particle size distribution and improve the consistency of crystal quality in different batches.The specific research work of this paper is as follows.Firstly,a batch crystallization process model of y-aminobutyric acid combined with cooling and anti-solvent crystallization was constructed.The relationship between the operating variables and system state and crystal quality was determined,and the batch crystallization process was accurately simulated.The sequential quadratic programming algorithm was used to optimize the single-objective optimization problem of batch crystallization process.The simulation results showed that the crystal quality of seed crystal addition batch crystallization process is significantly better than that of amorphous seed addition crystallization process.The multi-objective particle swarm optimization algorithm was used to obtain the Pareto optimal solution for multi-objective optimization problems in batch crystallization process,which solved the problem of multi-objective conflict of particle size distribution.Finally,the state estimation method was used to adjust the model parameters to accurately estimate the system state.Combined with the nonlinear model predictive control method based on the endpoint characteristics,the problems of system interference and model mismatch were solved,the crystal quality deviation was reduced,and the consistency of crystal quality of different batches in the batch crystallization process was improved.