摘要
为了对工业大系统进行稳态递阶优化，必须首先获得系统的稳态模型．从神经网络的分析入手，给出了工业大系统稳态模型的动态辨识方法及基于神经网络模型的推导方法．为了提高算法的收敛速度，引入Ｌａｇｒａｎｇｅ函数解决大系统优化问题中的各种约束，并用Ｈｏｐｆｉｅｌｄ网络实现了大系统稳态递阶优化的网络算法，最后给出了某一大系统辨识及优化的仿真结果．
In order to do steady-state hierarchical optimization for large-scale industrial systems, the steady-state model of the system must be obtained. By means of neural network, this paper presents a dynamic identification method for steady-state models of large-scale industrial systems with neural network, and proposes a way for modelling. For improving convergence, this paper firstly introduces Lagrange function to solve constraint problem in large-scale system optimization, secondly constructs the hierarchical optimization networks for large-scale industrial systems with Hopfield network.
出处
《自动化学报》
EI
CSCD
北大核心
1996年第2期175-181,共7页
Acta Automatica Sinica
基金
国家重点实验室基金
国家自然科学基金
关键词
稳态递阶优化
神经网络
工业大系统
辨识
Steady-state hierarchical optimization, Hopfield neural network,feedforward neural network, large-scale industrial system.
