基于互信息操作变量曲线参数化的间歇过程批内修正优化

针对间歇过程批次内扰动影响最终优化效果的问题,提出一种基于互信息操作变量曲线参数化的间歇过程批内修正优化方法.首先根据操作变量与指标变量间互信息和相关系数划分出操作变量曲线上对指标变量作用近似的时段;然后,结合操作变量曲线的形态特征选择有代表性参数建立优化模型,以降低优化模型求解的复杂度.考虑到生产过程噪声干扰影响最终优化效果,在批次内设置决策点,并根据当前工况信息对决策点后未实施的操作变量曲线进行调整,以减弱批次内扰动对最终优化效果的影响.最后,将所提出方法用于某一化工厂双酚A结晶过程的优化研究,通过仿真结果验证了该方法的有效性.

Intra-batch correction optimization of batch process with manipulated variable trajectory parameterization based on mutual information

Aiming at the problem that the intra-batch disturbance of batch processes affects the end-point optimization effect, a method of intra-batch correction optimization of batch process with manipulated variable trajectory parameterization based on mutual information is proposed. According to the mutual information and correlation coefficient between the manipulated variable and the index variable, the time period on the manipulated trajectory that has a similar effect on index variable is divided. Then, combine with the morphological characteristics of the manipulated variable trajectory, fewer parameters are selected to establish an optimization model to reduce the complexity of optimization model solution. Considering that noise interference in the production process affects the final optimization effect, the decision point is set in the batch and the unimplemented manipulated variable trajectory after the decision point is adjusted according to the current working condition information to reduce the impact of intra-batch disturbances on the final optimization effect. Finally, the proposed method is applied to the optimization of crystallization process of bisphenol A in a chemical plant. The simulation results validate the effectiveness of the method.