20.500.12556/DKUM-77142 Napovedovanje temperature vodnikov in obremenitev na stebrih daljnovodov s pomočjo genetskih algoritmov diplomsko delo Predictions of line temperature and stress at towers of overhead transmission lines using genetic algorithms Diplomsko delo zajema reševanje problema v sistemu prenosa in distribucije električne energije. Proučevali smo atmosferske vplive in vplive odvzema električne energije iz sistema na temperaturo in poves električnih vodnikov ter napetosti in vibracije v podporni konstrukciji. Za modeliranje smo uporabili metodo genetskih algoritmov, ki temeljijo na darvinističnih izsledkih o razvoju vrst in posnemajo principe naravne selekcije. Razviti modeli omogočajo študij teže posameznega vpliva. Namen dobljenih modelov je določanje najmočnejših vplivov in možnost napovedovanja vrednosti odzivov za daljše časovno obdobje in s tem ocenjevanje življenjske dobe daljnovodov. Rezultat našega dela so napovedni modeli z visoko stopnjo prilagojenosti na učni in testni množici. Modeli podajajo natančne napovedi, predvsem za temperaturo električnih vodnikov in njihov poves. Nižjo stopnjo natančnosti napovedi pri napetostih in vibracijah v stebrih pripisujemo dejstvu, da na te odzive vplivajo tudi obremenitve na konstrukciji in lastnosti materiala. This thesis is focused on solving a problem inside the systems for electric transfer and distribution. We have studied the influence of various atmospheric effects and the effect of electric current usage on the temperature and angle of overhead transmission lines as well as stress in the legs of the transmission towers and vibrations on the tower. For modelling, we used a method of genetic algorithms that derive from the results of the research conducted by Charles Darwin and mimic natural selection and species evolution. Developed models allow us to determine the weight of influence for individual effect. The goal of the developed models is to determine the most influential effect and to predict individual responses for a longer period and with that the life span of overhead transmission lines and their supporting construction. The results of our research are prediction models with a high rate of adaptiveness on learning and test sets. Models are especially accurate in their predictions of line temperature and line angle. Lower rate of accuracy for the stress in the legs of the towers and the vibrations on the tower is attributed to the fact that the effects of material and load are not included in the research. električni vodniki stebri napetosti vibracije genetski algoritmi modeliranje in optimizacija. overhead transmission lines transmission towers stress vibrations genetic algorithms modelling and optimization. true true false [K. Šket] Slovenski jezik Angleški jezik Diplomsko delo/naloga 2020-08-18 23:30:29 2020-09-10 18:51:21 2022-08-09 14:25:16 2020-11-29 03:12:16 2020 0 Maribor 0 Maribor XII, 55 str. 0000-00-00 NiDoloceno NiDoloceno NiDoloceno 0000-00-00 0000-00-00 0000-00-00 621.384.658(043.2) 39819267 URN:SI:UM:DK:OJWXGFEV UN_Sket_Kristijan_2020.pdf UN_Sket_Kristijan_2020.pdf 1 913221685B33EE17EC8E74901BD6D14F bff9d95ed0bd91fb0526f4b45504cd3a474d7b214def8984f972e574420fdae5 d4a3fe22-7c11-11eb-bb7a-00155d0001ca 20.500.12556/dkum/21db75f6-97f2-4144-a16c-4d6ed92f8743 https://dk.um.si/Dokument.php?lang=slv&id=145818 Fakulteta za strojništvo 0 0 0