摘要
将人工神经网络与规则推理相结合 ,建立了薄板裂纹诊断的专家系统。该系统充分利用神经网络与规则推理的优点 ,采用正反向混合推理方式调用知识库中的各种知识进行薄板结构的裂纹诊断。采用基于实验的计算机仿真技术建立裂纹诊断的标准谱 ,基于模拟退火算法训练神经网络。实例表明 ,本文建立的专家系统克服了基于规则专家系统的自学习困难问题和基于神经网络诊断系统系统控制能力弱的缺点 。
By combining the artificial neural network with the rule reasoning expert system, an expert diagnosing system for crack in thin plate was established. This expert system takes advantages of both neural network and role reasoning expert system, it can make use of all kinds of knowledge in the repository to diagnose the crack in thin plate with the positive and negative mixing reasoning mode. The standard crack diagnosing spectrum was founded with the computer imitating technology based on experiment. The neural networks were trained with simulated annealing algorithm. Examples indicate that: the expert system overcomes the self learning difficulty of the rule reasoning expert system and the shortcoming of the poor system control ability of the neural network. The expert system founded in this paper has powerful diagnosing ability.
出处
《机械强度》
EI
CAS
CSCD
北大核心
2001年第2期141-143,205,共4页
Journal of Mechanical Strength
关键词
裂纹诊断
专家系统
知识库
薄板
神经网络
Crack diagnosis
Expert system
Repository
Thin plate
Neural network
