论文部分内容阅读
20世纪80年代中后期以来,国际上以美国、日本为首的工业发达国家,掀起了竞相研究开发神经网络,设计构造神经计算机的热潮。主要应用在模式识别、经济管理、优化控制等方面;同时,人工神经网络在自动控制领域得到了广泛应用,为自学习、自适应模糊控制提供了一种新的有效途径。本文利用人工神经网络的优点,设计出一种神经网络的在线学习方法。通过神经网络的离线训练和在线自学习,使控制器具有自调整和自适应的性能,从而进一步改进实时控制效果,以便应用于温度过程控制中。
Since the mid and late 1980s, industrialized countries, led by the United States and Japan, set off a wave of research and development on neural networks and the design and construction of neural computers. Mainly used in pattern recognition, economic management, optimal control, etc .; the same time, artificial neural network has been widely used in the field of automatic control, which provides a new effective way for self-learning and adaptive fuzzy control. In this paper, the advantages of artificial neural network, to design a neural network online learning method. Through the neural network’s offline training and on-line self-learning, the controller has self-adjusting and adaptive performance, so as to further improve the real-time control effect for temperature process control.