基于混沌神经网络模型的模糊预测控制及应用  

Fuzzy Forecast Controller Based on Chaos Neural Network Model Research and Its Application

作　　者：窦春霞[1] 

机构地区：[1]燕山大学电气工程学院,秦皇岛066004

出　　处：《系统仿真学报》2002年第10期1372-1375,共4页Journal of System Simulation

摘　　要：由于混沌时间序列内部确定的规律性,其重构出混沌吸引子的相空间具有高精度短期预测性。为此,本文根据非线性、大时滞系统的时间序列及所得的Lyapunov指数规律,计算出系统的饱和嵌入维数和可预报尺度。并以此为指导,采用混沌神经网络重构混沌时间序列相空间,该混沌神经网络即便在网络输入不完整或发生变异的情况下,仍能对系统作高精度预测。在此基础上,又将预测模型与模糊控制相结合,提出了一种新型的模糊预测控制决策,实现了对非线性、大时滞系统高精度的自适应控制。将该控制应用在单元机组负荷控制中,仿真表明该控制具有实时性、容错性和鲁棒性。Because of chaotic time series internal certain regularity, its reconstructing chaotic attractors space has high precision short-term forecast. Therefore, systemic embed dimension and forecast measure are calculated by time series of nonlinear big-lagged system and Lyapunov exponent characteristics in this paper. By above all, a chaos neural network is presented to reconstruct chaotic time series space, and the systemic outputs are forecasted even by imperfect and variation inputs. On the basis of this, the neural network forecast model is combined with fuzzy control; a new fuzzy forecast controller is presented. It realizes high precision adaptive control to the nonlinear big-lagged system. The controller is applied to the units load system; simulation result shows the controller抯 adaptability and robustness.

关 键 词：混沌神经网络模型 模糊预测控制 模糊控制 混沌吸引子 LYAPUNOV指数 鲁棒性 

分 类 号：TP273.4[自动化与计算机技术—检测技术与自动化装置] TP183[自动化与计算机技术—控制科学与工程]