| 神经网络预测控制及其在二级倒立摆中的仿真 |
| |
| 引用本文： | 葛锁良,陈学军,邱丹. 神经网络预测控制及其在二级倒立摆中的仿真[J]. 自动化技术与应用, 2005, 24(6): 4-6,12 |
| |
| 作者姓名： | 葛锁良 陈学军 邱丹 |
| |
| 作者单位： | 1. 合肥工业大学,电气与自动化工程学院,安徽,合肥,230009
2. 淮北矿业集团,安徽,淮北,235006 |
| |
| 摘 要： | 本文介绍了一种基于神经网络的预测控制算法,考虑到算法中复杂性能指标的寻优难以找到全局最优解,提出了一种在工作点附近线性化得到对象的瞬时ARIX模型,再利用GPC(广义预测控制)算法求得近似解,最后以其为初值进行寻优获得全局最优解的方法,提高了算法的性能。最后的二级倒立摆仿真验证了算法的可靠性。
|
| 关 键 词： | 预测控制 神经网络 非线性规划 遗传算法 倒立摆 |
| 文章编号： | 1003-7241(2005)06-0004-04 |
Neural Network-Based Predictive Control of a Double Inverted Pendulum |
| |
| GE Suo-liang,CHEN Xue-jun,QIU Dan. Neural Network-Based Predictive Control of a Double Inverted Pendulum[J]. Techniques of Automation and Applications, 2005, 24(6): 4-6,12 |
| |
| Authors: | GE Suo-liang CHEN Xue-jun QIU Dan |
| |
| Affiliation: | GE Suo-liang~1,CHEN Xue-jun~1,QIU Dan~2 |
| |
| Abstract: | This paper introduces a neural network based predictive control algorithm. Because of the trouble of finding the global optimal solution in the optimization of complex cost function, a new method is proposed to improve the performance of the algorithm that linearizes the predictive model round the operation point to get the instantaneous ARIX model. The GPC(Generalized Predictive Control) algorithm is used to get an approximate solution first the final global optimal solution can then been found by optimization of the cost function. A simulation study of a double inverted pendulum is also given. |
| |
| Keywords: | Predictive control Neural network Nonlinear programming Genetic algorithm Inverted pendulum |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
|
