Optimal batch scheduling of suspension polyvinyl chloride production

Abstract

This work studies optimal batch scheduling of suspension polyvinyl chloride production process, which is a hybrid process consising of batch and continuous operation. The batch operation consists of two polymerization reactors to convert raw material of inyl chloride monomer to polyvinylchlorie, where raw material feed stream and utility supply unist are shared. These two reactors are connected in parallel fashion, and cannot be started simultaneously. The outputs from these reactors are then serially connected with the continuous production section. In our study, the computer program called Matlab, is used to develop a model of production process which is dynamic in nature. We use a discretization technique to convert time-dependent variables to series of variables of a short time period. Thus, the resulting model is in the form of a system of algebraic equations. The model formulated is then validated with the data from real production. The obtained results well agreement with the data obtained from real production. The optimal batch schedule under given mass balance, utility supply unit, and other equipments mechanical constraints is then determined, where the objective of the search is to maximize the production of polyvinylchloride under given circumstances. The algorithm used to optimize the operation schedule is differential evolutionary algorithm. It was found that the solution obtained results in 19% increasin of production compared with average normal operation. The results obtained from this work also contribute to decreasing of loss that may be occur during experiment in real production process, decreasing of safety and environment impacts.