אפקה - המכללה האקדמית להנדסה בתל אביב | תערוכת פרויקטי גמר 2014 - קטלוג פרויקטים - page 168

Student name:
Ilana Gitlins
Department: Industrial engineering
andmanagement
Advisor name:
Dr. Yevgenia Apartsin
Dataminingmethods for prediction
of electricity loads
2014
---
1
Theproblem:
The world population is growing rapidly, and along it, the technological and
industrial usage of electricity. Both of these caused the electricity load to grow in a
nonlinear
way, making it
hard topredict future loads
.
ProjectGoal:
Finding thebest dataminingmethods, based on the smallest error in comparison to
the benchmark results, out of all of the examinedmodels, to
predict aweek ahead
electricity load.
Project description:
Implementation of dataminingmodels to forecast short termelectricity loads, based
on
historical data
.
The data (electrical loads and temperature) was supplied by an
external company.
Descriptive parameters
were added to the data. The models
Neural network
,
Regression tree
and
GARCH
are employed, the forecast error is
calculated and a recommendation is given to user based on the lowest error in
comparison to the actual values.
Neural network
Regression tree
1...,169,170,171,172,173,174,175,176,177,178 158,159,160,161,162,163,164,165,166,167,...276