دانلود مقاله ISI انگلیسی شماره 52534
ترجمه فارسی عنوان مقاله

یک مطالعه بر روی مدل خستگی برای فرآیند پهنای باند گاوسی از دو دوره اوج توسط شبکه عصبی مصنوعی

عنوان انگلیسی
A study on the fatigue damage model for Gaussian wideband process of two peaks by an artificial neural network
کد مقاله سال انتشار تعداد صفحات مقاله انگلیسی
52534 2016 13 صفحه PDF
منبع

Publisher : Elsevier - Science Direct (الزویر - ساینس دایرکت)

Journal : Ocean Engineering, Volume 111, 1 January 2016, Pages 310–322

ترجمه کلمات کلیدی
تجزیه و تحلیل طیفی خستگی - روند پهنای باند - تابع چگالی احتمال - شبکه عصبی مصنوعیشمارش کمترین مربعات
کلمات کلیدی انگلیسی
Spectral fatigue analysis; Wide-band process; Probability density function; Artificial neural network; Rainflow counting; Least squares
پیش نمایش مقاله
پیش نمایش مقاله  یک مطالعه بر روی مدل خستگی برای فرآیند پهنای باند گاوسی از دو دوره اوج توسط شبکه عصبی مصنوعی

چکیده انگلیسی

Calculations of the fatigue damage on marine structures with a wideband nature are difficult to be done in spectral approach point of view because the link between the spectrum of stress and the probability distribution is difficult to define. This paper addresses the methodology through which the functional relationship between the probability density function and the response spectrum of a bimodal wide-band process by using the artificial neural network technique. An artificial neural network scheme was used to identify the multivariate functional relationship between the two continuously varying functions. For this, the spectra were idealized as the superposition of two triangles with an arbitrary location, height and width and the probability density functions were represented by the linear combination of equally spaced Gaussian basis functions. To train the network under supervision, a variety of different wide-band spectra were assumed and the converged probability density function of the stress range was derived using the rainflow counting method and all these data sets were fed into the three layer perceptron model. It turned out that the network trained using the given data set could reproduce the probability density function of an arbitrary wide-band spectrum of two triangles with great success.