دراسة حول أداء طرائق أشباه نيوتن الجديدة لإيجاد القيمة الصغرى لمسائل الأمثلية غير المقيدة
Study on the Performance of New Quasi-Newton Methods to Find Minimum for Unconstrained Optimization Problems
- الباحث
- عبدالرحمن راغب أيوب الخاتوني
- المشرف
- باسم عباس حسن
- الجامعة
- جامعة الموصل
- الكلية / القسم
- كلية علوم الحاسوب والرياضيات / الرياضيات
- الشهادة
- ماجستير
- السنة
- 2022
- لغة الرسالة
- عربي
- الكلمات المفتاحية
- طرائق أشباه نيوتن ; الأمثلية غير المقيدة ; القيمة الصغرى ; الأداء العددي
Abstract
We tried in this thesis studying on the performance of a new quasi- Newton methods to find minimum for unconstrained optimization Problems it is one of the most common classes of nonlinear numerical optimization methods. The effect of the quasi-Newton equation using function information can be improved by quasi - Newtonian methods. This is why the new quasi-Newtonian equations were derived Based on the second and third order Tyler series and its derivative so as to get on more accurate approximations for the Hessian matrix. under hypotheses occasion, we establish convergence Global for the suggested method and retain most of the properties of the usual quasi-Newton methods. The numerical results were calculated with the effect that the new algorithms are more efficient compared to the similar algorithms in the family. Numerical experiments have proven that the new algorithms Efficient comparison with the BFGS algorithm based on scales for calculating functions and frequencies. We concluded the message by discussing the new conclusions and giving some suggestions for the purpose of working on them in the future.
طلب نسخة من الملف
الملفات غير منشورة على الموقع. اترك بياناتك وسيتواصل معك موظفو قسم الشؤون الثقافية والفنون لإرسال النسخة مجاناً.