噪声与振动控制
譟聲與振動控製
조성여진동공제
NOISE AND VIBRATION CONTROL
2014年
2期
158-163
,共6页
张星辉%李凤学%赵劲松%曹端超%滕红智
張星輝%李鳳學%趙勁鬆%曹耑超%滕紅智
장성휘%리봉학%조경송%조단초%등홍지
振动与波%混合高斯输出贝叶斯信念网络%状态识别%剩余寿命预测
振動與波%混閤高斯輸齣貝葉斯信唸網絡%狀態識彆%剩餘壽命預測
진동여파%혼합고사수출패협사신념망락%상태식별%잉여수명예측
vibration and wave%mixture of Gaussians Bayesian belief network%state recognition%residual life
基于混合高斯输出贝叶斯信念网络模型的齿轮箱退化状态识别与剩余寿命预测新方法,应用聚类评价指标对全寿命过程退化状态数进行优化,通过计算待识别故障特征向量的概率值来确定齿轮箱退化状态,在退化状态识别的基础上,提出了剩余寿命计算方法。再利用齿轮箱全寿命实验数据对此进行验证。结果表明,该方法可以有效地识别齿轮箱故障状态并实现剩余寿命预测,平均预测正确率为96.47%,为齿轮箱的健康管理提供参考。
基于混閤高斯輸齣貝葉斯信唸網絡模型的齒輪箱退化狀態識彆與剩餘壽命預測新方法,應用聚類評價指標對全壽命過程退化狀態數進行優化,通過計算待識彆故障特徵嚮量的概率值來確定齒輪箱退化狀態,在退化狀態識彆的基礎上,提齣瞭剩餘壽命計算方法。再利用齒輪箱全壽命實驗數據對此進行驗證。結果錶明,該方法可以有效地識彆齒輪箱故障狀態併實現剩餘壽命預測,平均預測正確率為96.47%,為齒輪箱的健康管理提供參攷。
기우혼합고사수출패협사신념망락모형적치륜상퇴화상태식별여잉여수명예측신방법,응용취류평개지표대전수명과정퇴화상태수진행우화,통과계산대식별고장특정향량적개솔치래학정치륜상퇴화상태,재퇴화상태식별적기출상,제출료잉여수명계산방법。재이용치륜상전수명실험수거대차진행험증。결과표명,해방법가이유효지식별치륜상고장상태병실현잉여수명예측,평균예측정학솔위96.47%,위치륜상적건강관리제공삼고。
A new approach for gearbox’s fault diagnosis and residual life prediction based on Mixture of Gaussian-Bayesian Belief Network (MoG-BBN) is presented. State number optimization method is established based on cluster validity measures. The gearbox’s degradation state can be recognized by calculating the probability of the fault characteristic vectors. On this basis, the method for residual life prediction is presented. Finally, results of this prediction method are verified by the data of gearbox’s full life cycle test. The results show that the mean accuracy can reach 96.47%.