Modeliranje vlečne sile pri oranju z nevronsko mrežo

diplomsko delo

Martin Kraner (Author), Aleksander Zidanšek (Mentor)

Abstract

V uvodu sem opisal, zakaj sem se odločil za izdelavo nevronske mreže za izračun sil pri oranju. V nadaljevanju sem predstavil fizikalni proces pri oranju. Opisal sem enostaven približek pluga ter vlečne sile, ki se pojavljajo pri oranju, torej vlečno silo na plužno telo, lemež in rezilo. Nato sem opisal zgradbo nevronske mreže in uporabo modela nevronske mreže. Da bi lahko izračunali sile pri danih parametrih, sem nevronsko mrežo učil iz podatkov pridobljenih iz literature. Prav tako sem v nadaljevanju naredil preizkus s podatki oziroma meritvami iz literature. Nato sem primerjal rešitve in dobil zanimive rezultate, ki se jih lahko še dodela z vnovičnim učenjem nevronske mreže! Diplomska naloga tako predstavlja teoretični uvod k izdelavi učinkovite aplikacije za kmetovalce in proizvajalce plužnih teles za izračun sil na plužna telesa pri izbranih parametrih!

Keywords

diplomska dela;plug;plužno telo;nevronska mreža;sile na plužno telo;

Data

Language:	Slovenian
Year of publishing:	2015
Typology:	2.11 - Undergraduate Thesis
Organization:	UM FNM - Faculty of Natural Sciences and Mathematics
Publisher:	[M. Kraner]
UDC:	631.431.3:004.032.26(043.2
COBISS:	21500168
Views:	1283
Downloads:	120
Average score:	0 (0 votes)
Metadata:

Other data

Secondary language:	English
Secondary title:	Modeling of plowing force with neural network
Secondary abstract:	The purpose of the thesis was to create a neural network and to train it to calculate forces in plowing depending on the speed and depth of plowing as well as in relation to the type of soil. In the introduction, I described my decision to create a neural network for calculation of forces in plowing. I then described the physical process of plowing, simple approximation of the plow, and forces during plowing including the traction force on the plowing component, coulter and the blade. I further described the construction of the neural network and the use of the neural network model. In order to calculate the forces at the given parameters, the neural network was trained using data obtained from the literature. After training the neural network was tested with further data and measurements from literature. Interesting results were obtained from comparison of the neural network solutions, which could still be refined by retraining of the neural network! The thesis therefore represents a theoretical intruduction to manufacturing of an efficient application for farmers and producers of the ploughs for the calculation of forces on the plow body in different conditions!
Secondary keywords:	theses;plow;plowing component;neural network;plowing forces;
URN:	URN:SI:UM:
Type (COBISS):	Undergraduate thesis
Thesis comment:	Univ. v Mariboru, Fak. za naravoslovje in matematiko, Oddelek za fiziko
Pages:	V, 42 f.
ID:	8773125