汽轮发电机组故障诊断专家系统的研究

Study an Expert System for Turbogenerator

建立一个基于知识的汽轮发电机组故障诊断专家系统ＫＢＦＤＥＳ，该系统采用“框架＋规则”知识表示技术，并将框架驻留在内存，而将规则驻留在虚拟盘上；在诊断过程中采用广义的不精确推理策略，并对重要信息用一个组合神经网络进行智能识别；系统将神经网络技术和ＩＤ３算法结合，可以实现从诊断实例自动获取知识。

knowledge-based fault diagnostic expert system for turbogenerator isdeveloped in this paper. The knowledge representation of the system adopts 'Frame + Rule' representation technology. Frames and rules respectively store in memory and ramdriver.In diagnosis, the system uses fuzzy-inference strategy and applies a hybrid neural networkto recognize some important information of turbogenerator. With combination of neuralnetwork technique and ID3 algorithm, the system can realize knowledge automatic acquisition from diagnostic examples.