عنوان مقاله
بهینه سازی توالی پشته سازی لامینات های کامپوزیت با شبکه عصبی و الگوریتم ژنتیکی
فهرست مطالب
مقدمه
شبکه های عصبی مصنوعی
تقریب تابع کمانش استوانه کامپوزیت با روش MLP
الگوریتم های ژنتیکی
بهینه سازی پوسته لامینات کمپوزیت با لامینات های 10 لایه
نتیجه گیری
بخشی از مقاله
خصوصیات شبکه عصبی
در این مطالعه ، شبکه عصبی پرسپترون سه لایه با 6 نورون (5 نورون در 2 لایه مخفی و یک نورون در لایه خروجی) برای مدلسازی بار کمانش بکاربرده شد. نماهای عمومی این شبکه در شکل 1 نشان داده شده است.
مدلسازی و بحث
برای مدلسازی بار کمانش، از مجموعه داده های آزمایشی متشکل ازW1 و W2 و مقادیر آزمایشی بار کمانش استفاده گریدد. این مجموعه داده به عنوان داده های آموزشی برای شبکه عصبی پرسپترون سه لایه با خصوصیات بیان شد.
کلمات کلیدی:
Optimization of Stacking Sequence of Composite Laminates for optimizing buckling load by Neural Network and genetic algorithm A.R. Ghasemi1 , M.H. Hajmohammad , A. Qaderi Abstract Composite beams, plates and shells are widely used in the aerospace industry because of their advantages over commonly used isotropic Structures especially when it comes to weight savings. Buckling analyses of composite structural components must be performed in order to ensure, for instance, that a composite panel designed to be part of a control surface does not buckle thereby compromising its aerodynamic shape. Optimization of composite structures has been performed in this paper using Genetic algorithm. Genetic algorithm (GA) approaches are successfully implemented to the TSP. The buckling load of composite plate, which is obtained by the Artificial Neural Networks, was used as the fitness function in the GA to find its optimized value by arranges the ply stacking sequence. Keywords: stacking sequence, genetic algorithms, composite laminate, buckling load, Neural Networks 1. Introduction composite laminates have widespread applications in aerospace structures, and optimizations of the corresponding stacking sequences are indispensable. Many of the structures manufactured from these materials such as spars in aircraft wing boxes constitute thin walled shells and as such are prone to failure by buckling. It is essential therefore that the onset of this buckling can be predicted, and the effect of buckling on the load carrying capacity of the structure is understood and quantified. Until now, the problem has mainly been addressed by using gradient-based optimization techniques. Recently, it has been studied by using evolutionary algorithms, which are optimization tools that do not make use of the gradient of the objective function. Genetic algorithm (GA) is the most common choices treated in literature. GA is an implementation of the rules stated by the Darwin's theory. This simple algorithm deals with discrete optimization problems. Genetic algorithms (GAs) are adopted for optimization of the stacking sequences [3–9]. Since GAs are one of the stochastic search approaches, they require several parameter tuning processes to avoid any reduction in the computational performance. In the evaluation of chromosomes, these GAs involve high computational costs. Researchers applied various optimization methods to a number of different design problems involving composite materials using design variables such as fiber direction, ply thickness, or stacking sequence. As one of the methods, enumeration, evaluating all possible designs, has a restricted applicability; for most problems it is computationally too expensive, for many impracticable. Genetic algorithms (GA) are well suited for stacking sequence optimization, and because of their random nature, they can produce alternative optima in repeated runs. The
