摘要
在微细电火花加工中 ,由于加工信号的频率高、加工波形的畸变 ,使得常规放电状态检测法已不再适用。本文描述了用于目标识别与分类的基于模型的多传感器系统 ,该系统选用以决策层为主的方法 ,以模糊神经网络作为某信息融合的工具。通过实验 ,该系统在提高正确识别的前提下 ,降低了整个Mi cro EDM系统的成本 ,提高了检测的可靠性 ,体现了多传感器信息融合的优越性。
The discharge condition detection method sused in conventional EDM (Electirc Discharge Manufacturing) are not suitable for the Micro EDM due to the high frequency and distortion of voltage wave shape.In this paper,We describe a multi sensor system which is based on model tool and applied for target recognition and classification. In the data fusion process, a fuzzy neural network (f NN)is selected and used for the data fusion at report level.By experiment result,the proposed method can work correctly, the cost decreases and the reliability of the recognition increases.The superiority of this method has been shown.
出处
《电加工与模具》
2000年第3期6-9,共4页
Electromachining & Mould
基金
航空科学基金资助项目! ( 98H52 0 62 )
江苏省应用技术基金资助项目! (BJ970 57)
