结合规则知识与神经网络于空袭武器的辅助选择

结合规则推理与神经网络,提出了一种新的空袭武器辅助选择方法。其根据空袭武器选择的基本原则以及问题的特点,构造T-S模糊逻辑系统及其实现的相应神经网络模型,然后通过已有的战例与试验例子训练神经网络,并对问题进行求解,得出空袭武器选择提供依据(一种量化的武器适用度)。另外,还利用专家提供的规则知识于学习实例的预处理,学习实例信息的扩充。

Coordinate Rule and Neural Network for Computer Aided Air-Raid-Weapon Selection

This paper presents a new method of aided air-raid weapon selection by coordinating rule-based reasoning and neural network.Based on the fundamental principle of air-raid-weapon selection and the characteristics of problem,this method constructs the T-S fuzzy logic system and the corresponding realiziation neural network,and then trains it with actual combat cases and experimental cases.At last solve the problem and get the selection basis a quantitative preference on weapon.Otherwise,it apply the rules got from expert to preprocessing the learning cases,expanding the information of learning cases and adjusting the result.