基于神经网络和规则的专家系统的应用研究

本文对专家系统在材料加工领域的应用特点及现状进行了分析与研究，采用基于人工神经网络和规则的知识表示和获取方法，实现了一个混合型的专家系统；并结合实例，讨论了神经网络与专家系统的集成，以及系统的程序框架与功能设计、各模块的主要实现思路；最后，通过实验数据验证，系统的泛化结果误差小于6％。

Application Research of the ESs Based on Neural Networks and Rules

On the basis of analyzing and summarizing the characteristics and status quo of the application of expert systems in material processing, the paper presents an integrated expert system by using knowledge representation and knowledge acquisition based on artificial neural networks and rules. With an example, the paper discusses the integration of ES and ANN, system frame and function design, and the main implementation idea of each module. Finally experimental data show that the system generalization error is less than 6%.