摘要
概述了BP人工神经网的原理,并利用其对声发射(AE)腐蚀信号进行识别。识别过程中,以破茬圆盘刀点腐蚀过程中产生的气泡和膜破裂为源,利用改进的BP人工神经网进行识别,来判断破茬圆盘刀点蚀所处的阶段。实验证明,人工神经网能够显著提高AE信号的处理速度,并能同时降低破茬圆盘刀点腐蚀的误判率。
The principle of BP artificial neural network was introduced,and it was used to identify the acoustic emission （AE） corrosion signal. During the recognition,bubble and film rupture caused by disc knife corrosion were taken as the AE sources,the improved BP artificial neural network was used to decide the corrosion stage of disc sickle. The experiment proved that artificial neural network could significantly improve the speed of AE signal processing, and could also reduce the misjudgment rate.
出处
《黑龙江八一农垦大学学报》
2013年第4期22-24,共3页
journal of heilongjiang bayi agricultural university
基金
教育部高等学校博士点基金(20092305110002)
黑龙江省农垦总局"十二五"重点科技攻关项目(HNK11A-05-11)
黑龙江省农垦总局科技计划项目(HNK10A-09-02-04)
黑龙江省教育厅科研课题(12511359)
关键词
破茬圆盘刀
声发射
人工神经网络
模式识别
disc sickle
acoustic emission
artificial neural network
mode identification
