论文部分内容阅读
动态优化问题广泛存在于化工自动控制过程中,对其求解是化工过程工业发展的一个不可忽视的环节。群智能算法求解此类优化问题时不可避免地存在后期收敛速度慢、求解精度的不高等不足,这一直是一个研究热点。针对新兴的布谷鸟算法与以上问题,提出一种变步长自适应布谷鸟搜索算法(VSACS),将基本布谷鸟搜索(CS)算法中的随机步长改进成根据迭代次数自适应调整的步长。通过15个标准测试函数的测试,结果验证了改进的算法有较快的收敛速度和较高的求解精度。最后将改进的算法用于批示反应器、管式反应器、生物反应器等3个典型的化工动态优化问题中,获得了满意的实验结果,同时也进一步表明该算法的有效性。
Dynamic optimization problems exist widely in the process of automatic chemical control, and its solution is a process of chemical industry can not be ignored. Swarm intelligence algorithm to solve such optimization problems inevitably has a slow convergence rate after the end of the solution accuracy is not high enough, which has been a research hot spot. In view of the emerging cuckoo algorithm and the above problems, a variable step adaptive cuckoo search algorithm (VSACS) is proposed, which improves the random step size in the basic cuckoo search (CS) algorithm to a step adaptively adjusted according to the number of iterations long. Through the testing of 15 standard test functions, the results verify that the improved algorithm has faster convergence speed and higher solution accuracy. Finally, the improved algorithm is applied to three typical chemical dynamic optimization problems, such as batch reactor, tubular reactor, and bioreactor. The experimental results are satisfactory and the effectiveness of the proposed algorithm is further demonstrated.