摘要人工神经网络具有并行处理能力、自学习能力、自适应能力和以任意精度逼近任意非线性函数的特点，在模式识别、系统辨识、控制等领域都得到了广泛的应用。本文就神经网络参数的自适应辨识做了基本的设计与研究。65242
本文综合了解目前自适应辨识和神经网络系统同步的相关仿真方法和实现方法，利用Matlab建立仿真平台，比较系统参数的变化对系统辨识性能的影响，调节仿真中参数，说明系统参数不确定性以及外部干扰对系统辨识性能的影响，并尝试优化辨识策略。通过我们的仿真实验我们验证了论文中的自适应控制算法，并且我们发现原论文中的某些结论在一定的条件下不成立，通过仿真验证了我们的结论。
毕业论文关键词 系统辨识 自适应辨识 系统同步 神经网络
毕业设计说明书（论文）外文摘要
Title Design and implementation of adaptive identification of neural networks
Abstract Many functions are possessed by the neural network, such as, parallel processing, self-learning and self-adapting, and so on. It is reported that the neural networks can approximate any nonlinear function with any precision. Therefore, the neural network is widely used in many aspects, for instance, pattern recognition, system identification and control fields. A simple adaptive identification of neural network parameter is considered in this article.
In this article, we focus mainly on the basic design and theory analysis of adaptive identification of neural network. The comprehensive understandings of the current related simulation methods and the realization methods of synchronization system are achieved, which are based on the Matlab simulation platform. The comparative changes in system parameters which affect the performance of system identification are performed. By adjusting the parameters in the simulation, describing system parameters uncertainties and external disturbance effect on the performance of system identification, some simulation results are proposed. In addition, some numerical results show that the criterion proposed in one article is incorrect. We verify the results by some simulation.
Keywords system identification; adaptive identification;Synchronization; neural networks
目 次
1 引言 1
1.1 人工神经网络发展与现状及理论简介 1
1.2 神经网络的模型结构及功能简介 2
1.3 系统辨识理论综述 4
1.4 神经网络系统辨识理论综述 5
1.4.1 神经网络系统在线性系统辨识中的应用 5
1.4.2 神经网络系统在非线性系统辨识中的应用 6
1.5 本文的主要工作 7
2 软件开发平台介绍 9
3 神经网络模型和准备工作 10
4 神经网络的同步 11
4.1 具有已知参数的神经网络的同步 11
4.2 具有未知参数的神经网络的同步 Matlab神经网络参数自适应辨识仿真算法设计与实现:http://www.751com.cn/zidonghua/lunwen_72779.html