计算机工程
計算機工程
계산궤공정
COMPUTER ENGINEERING
2010年
8期
12-14
,共3页
模糊神经网络%可靠度预计%电火花线切割机床%隶属函数
模糊神經網絡%可靠度預計%電火花線切割機床%隸屬函數
모호신경망락%가고도예계%전화화선절할궤상%대속함수
fuzzy neural network%reliability prediction%Wire cut Electric Discharge Machining(WEDM)%membership function
为了准确预计电火花线切割机床(WEDM)的可靠度,建立基于自适应模糊神经网络的可靠度预计模型.该模型以平均无故障时间为输入,以可靠度为输出,通过神经网络自适应训练获得适合WEDM可靠度预计的平均无故障间隔时间隶属函数.仿真结果表明,该模型的预计精度较高,与应用神经网络的WEDM可靠度预计结果相比,提高了96.4%.
為瞭準確預計電火花線切割機床(WEDM)的可靠度,建立基于自適應模糊神經網絡的可靠度預計模型.該模型以平均無故障時間為輸入,以可靠度為輸齣,通過神經網絡自適應訓練穫得適閤WEDM可靠度預計的平均無故障間隔時間隸屬函數.倣真結果錶明,該模型的預計精度較高,與應用神經網絡的WEDM可靠度預計結果相比,提高瞭96.4%.
위료준학예계전화화선절할궤상(WEDM)적가고도,건립기우자괄응모호신경망락적가고도예계모형.해모형이평균무고장시간위수입,이가고도위수출,통과신경망락자괄응훈련획득괄합WEDM가고도예계적평균무고장간격시간대속함수.방진결과표명,해모형적예계정도교고,여응용신경망락적WEDM가고도예계결과상비,제고료96.4%.
In order to predict the reliability of Wire cut Electric Discharge Machining(WEDM) accurately, this paper establishes a reliability prediction model based on self-adaptive fuzzy neural network. This model takes the Mean Time Between Failure(MTBF) as the input and takes the reliability as the output. The membership function of the MTBF is achieved by the neural network self-adaptive training. Simulation results show that this model has high prediction precision which is improved by 96.4% compared with that of WEDM reliability by using neural network.