Search In this Thesis
   Search In this Thesis  
العنوان
Neural Networks Approaches to Determine Electrical and Rheological Properties of Polymer High Voltage Insulators/
المؤلف
Ahmed,Sherif Hussein Haggag
هيئة الاعداد
باحث / شريف حسين حجاج احمد
مشرف / حنفي محمود اسماعيل
مناقش / مازن محمد شفيق عبد السلام
مناقش / عادل صدقى سيد عماره
تاريخ النشر
2022
عدد الصفحات
124p.:
اللغة
الإنجليزية
الدرجة
الدكتوراه
التخصص
الهندسة الكهربائية والالكترونية
تاريخ الإجازة
1/1/2022
مكان الإجازة
جامعة عين شمس - كلية الهندسة - الهندسة الكهربائية والالكترونية
الفهرس
Only 14 pages are availabe for public view

from 148

from 148

Abstract

Cables are an essential component in the power system. Cables
insulation should have basic electrical parameters to give it the capability of
performing its function successfully. High dielectric strength, significant
resistivity, besides low permittivity and loss factors, are important electric
features that should be possessed in the cable insulation materials; however,
thermal and mechanical aspects such as thermal withstand capacity and
adequate degree of elasticity are substantial as well.
Due to their chain structure, polymers have low dielectric constant and
high electric breakdown; In addition, they are characterized with multiple
valuable mechanical parameters that make them suitable for cable insulation
applications.
Polymer composites and blends in dielectric applications have lately
garnered significant interest from researchers. The applicability of customizing
polymers features to suit particular objectives met approbation by the industrial
sector.
This research introduces the changes of a comprehensive
electromechanical properties bundle for low density polyethylene compounded
to nano and micro scale magnesia (LDPE/MgO) to obtain electrical cables
insulating material with advanced performance. Composites of various filler
loading weight ratios were prepared by melt intercalation technique; multiple
samples were produced in sets as they were cut with definite dimensions as per
recommendations of the related testing standard then electrically and
mechanically examined following the instruction dictated by the code while
preserving typical test condition for all sets. Dielectric strength, volume
resistivity, capacitance, and loss angle were the tests of the electrical test pack,
while elongation, tensile strength, and melt flow rate were the mechanical and
rheological tests applied. Water absorption is another test that has been
conducted on the proposed composite.
Artificial Neural Network was designed and trained using acquired
practical results in order to predict dielectric strength of composites that have
typical contents but with different filler loading or impacted by other media
salinity degrees.
The research findings and comparisons showed that the Low-Density
Polyethylene compounded to surface characterized nano magnesia is better in
most electrical and mechanical properties than the pristine host, nanocomposite
with non-modified filler surface and micro composite, best results obtained
with low percentages of filler loading of almost 1.5 wt%.