论文部分内容阅读
内部热耦合精馏(ITCDIC)是迄今为止所提出的四大精馏节能技术中节能效果最高,但唯一没有商业化的节能技术,比常规精馏要节能40%以上,没有商业化的主要种类基于减聚类、K-means原因之一在于该过程具有较强的非线性、复杂动态特性以及耦合性,给控制方案的设计带来了诸多困难。由于径向基(RBF)神经网络具有快速学习并能逼近任意非线性函数的优点,本文提出了一种基于RBF神经网络内模控制的混合优化算法,是一种粒子群优化的混合优化算法,以苯-甲苯物系作为研究实例,并与国际公开报道的结果进行了详细比较,研究结果表明基于混合优化算法的RBF神经网络内模控制相比于传统的PID、常规RBF算法和国际公开报道有着更好的控制效果。
Internal Thermal Coupled Distillation (ITCDIC) is the most energy-saving technology in the four distillation technologies proposed so far, but the only commercial technology that is not commercialized, more than 40% more energy efficient than conventional rectification, and is not commercialized One of the reasons for K-means based on subtractive clustering is that the process has strong nonlinearity, complex dynamic characteristics and coupling, which brings many difficulties to the design of control schemes. Because radial basis function (RBF) neural network has the advantages of fast learning and approximating any non-linear function, this paper presents a hybrid optimization algorithm based on RBF neural network internal model control, which is a hybrid optimization algorithm of particle swarm optimization, Taking the benzene-toluene system as an example, the results are compared with those reported in the international public. The results show that the internal model control of the RBF neural network based on the hybrid optimization algorithm is better than the traditional PID, the conventional RBF algorithm and the international report Have a better control effect.