دورية أكاديمية

ESTIMATION METHOD OF LOAD SPECTRUM DISTRIBUTION BASED ON EM ALGORITHM

التفاصيل البيبلوغرافية
العنوان: ESTIMATION METHOD OF LOAD SPECTRUM DISTRIBUTION BASED ON EM ALGORITHM
المؤلفون: SU Fang, CHEN XiaoNan, WANG ChenSheng, ZHAO HaiYan
المصدر: Jixie qiangdu, Vol 42, Pp 1119-1124 (2020)
بيانات النشر: Editorial Office of Journal of Mechanical Strength, 2020.
سنة النشر: 2020
المجموعة: LCC:Mechanical engineering and machinery
LCC:Materials of engineering and construction. Mechanics of materials
مصطلحات موضوعية: Harvester, Mixture of Gaussian function, Parameter estimate, Rainflow method, Distribution estimation, Mechanical engineering and machinery, TJ1-1570, Materials of engineering and construction. Mechanics of materials, TA401-492
الوصف: As the real load is very complicated during the process of harvester operation,it is of great significance for the prediction of fatigue life to describe the load characteristics and distribution form accurately. In order to obtain the probability density distribution model of the load cycles. Take RF-half shaft torque as the research object,the mixture Gaussian fitting method based on EM was suggested. n Code software was used for rain flow counting and the load cycle was described with the vector S.The cluster number and cluster center loading were got from the clustering of the loading cycle by the SPSS software,and using EM algorithm to resolve the mean and covariance matrix of Gaussian component. The index value of SSE and RSME are close to0,but R and Raare both up to 0. 85 based on the index model. It showed that the Gaussian mixture indeed fitting out of the original data of load-cycle. The results of research can provide a theory evidence for the fatigue life analysis and load spectrum extrapolation.
نوع الوثيقة: article
وصف الملف: electronic resource
اللغة: Chinese
تدمد: 1001-9669
Relation: http://www.jxqd.net.cn/thesisDetails#10.16579/j.issn.1001.9669.2020.05.016; https://doaj.org/toc/1001-9669
DOI: 10.16579/j.issn.1001.9669.2020.05.016
URL الوصول: https://doaj.org/article/59534aa3f004412eba5da782316e8590
رقم الأكسشن: edsdoj.59534aa3f004412eba5da782316e8590
قاعدة البيانات: Directory of Open Access Journals
الوصف
تدمد:10019669
DOI:10.16579/j.issn.1001.9669.2020.05.016