神经网络在复杂系统建模中的应用

 分析了神经网络模型的特点、复杂系统的特征与建模方法论要求，表明神经网络应用于复杂系统建模可行且适用。探讨了当前神经网络在复杂系统建模中应用的文本语言处理、语音识别、图像处理、故障诊断等，针对所存在的不能很好地描述系统结构与演化、不能同时处理多个任务、不能很好解决非重复性问题等主要不足，提出了解决这些问题的初步思路。

Applications of neural network in complex system modeling

By analyzing and comparing the characteristics of neural network models and complex systems and the requirements of their modeling methodology, it is shown that the applications of the neural network in the complex system modeling are feasible. The main application fields of the neural network in the complex system modeling are discussed, such as the natural language processing, the speech recognition, the visual processing, and the fault diagnosis; and the main problems are pointed out, including the facts that the neural network can not describe the structure and the evolution of the system, it can not handle multiple tasks at the same time, and it can not solve the non repetitive problems well. Also the solutions of these problems are discussed preliminarily.