一种间歇反应过程迭代动态规划方法

An approach to iterative dynamic programming for batch reactor systems

随着工业过程对降低产品成本、改进产品质量、满足安全要求和环境规范,间歇反应过程的优化变得越来越重要。本文因此给出了一种有效的基于随机选点的间歇反应过程迭代动态规划算法,并给出了算法实现的详细步骤,能够有效实现间歇反应过程中温度、浓度等变量的动态优化问题。所述的迭代动态规划算法通过调节分段数P和离散点数(N和M)可以有效的避免计算量激增的问题,具有稳定可靠、易寻找到全局最优解的优点。以典型的间歇反应动态优化问题作为实例进行了研究,并与国际上公开报道结果进行了详细的比较研究,结果表明了所述方法的可靠有效性。

With the requirement of reducing in production costs, improving product quality, safety and environmental regulations in industrial process, dynamic optimization of batch reactor has attracted attention in recent decades. An efficient iterative dynamic programming （IDP） approach of batch reactor is therefore explored based on random admissible values, and the detailed procedures of the presented approach are then given, which can efficiently achieve the dynamic optimization of variables （temperature, concentration, etc. ） in the batch reaction processes, and avoid the problem of proliferating in the computational time by adjusting the number of subsection P and discrete points （N and M） simultaneously. Moreover, it is reliable and easy to find global optimum of the model. A classical dynamic optimization problem in the batch reactor is then carried out as a illustration example. The detailed comparative research with the results reported in the open literatures is further carried out. The research＇s results proved the efficiency of the presented approach.