Aristotle University Of Thessaloniki (AUTH)
Εφαρμογή νευρωνικών δικτύων στην βραχυπρόθεσμη πρόβλεψη φορτίου
Abstract
dc:descriptionThis work studies the applicability of this kind of models and offers some extra models for electric load forecasting. The work is intended to be a basis for a real forecasting application. In the first chapters, a literature survey was conducted on the subject. Most of the reported models are based on the so-called Multi-Layer Perceptron (MLP) network. There are numerous model suggestions, but the large variation and lack of comparisons make it difficult to directly apply proposed methods. It was concluded that a comparative study of different model types seems necessary. The aim of this thesis is to find and apply more adequate models for electric power load forecasting. Earlier works are based on the assumption of weakly stationary process after 24 hour and 168 hour differentiation of the input signals. It is clear, however, that the process displays different properties depending on the time of day and day of week. The models were divided into two classes. First, forecasting the load for a whole day at once was studied. Then hourly models, which are able to update the forecasts as new data arrives, were considered. Furthermore some models use external variables (temperature) and some of them do not need external variables. Depending on what kind of information is available, one can chose the appropriate model for concrete application. This work studies the applicability of different neural network models on short-term load forecasting. The approach is comparative. The models are divided into two classes: models forecasting the load for one whole day at a time, and models forecasting ahead hour by hour. Testing is carried out on the real load data of a Greek power system for 1995. The objective is to compare different models and to develop and apply more adequate models for short term electric power load forecasting. The load data are taken from the National Center of Energy Management of Greece. The temperature data has been taken from The National Meteorological Company. The thesis does not study the forecasting for special days, such as religious and legal holidays. Special days have different consumption profiles from ordinary days, which makes forecasting very difficult for them. When implementing a real application, a means to take these days into account has to be found. The most common approach, but not necessarily the best one, is to treat them as Sundays. The thesis is organized in six chapters and three appendixes.
Degree
thesis:*- Grantor dc:publisher
- Aristotle University Of Thessaloniki (AUTH)
- Year dc:date
- 2003
Author and committee
dc:creator, dc:contributor.*- Authors dc:creator
-
- Τσακουμής, Αθανάσιος
- Tsakoumis, Athanasios
Subjects
dc:subject × 20- Βραχυπρόθεσμη πρόβλεψη φορτίου
- Χρονοσειρές
- Δίκτυα πεπερασμένης απόκρισης
- Δίκτυα Elman
- Πολυστρωματικά δίκτυα
- Χάος
- Ελκυστής
- Πείραμα Lyapunov
- Short-term luad forecasting
- Time series
- Finite impulse response network
- Elman network
- Multilayer perreptron (MLP)
- Chaos
- Attractors
- Lyapunov exponents
- Επιστήμες Μηχανικού και Τεχνολογία
- Επιστήμη Ηλεκτρολόγου Μηχανικού, Ηλεκτρονικού Μηχανικού, Μηχανικού Η/Υ
- Engineering and Technology
- Electrical Engineering, Electronic Engineering, Information Engineering
Rights
- Language dc:language
- gre
Identifiers
dc:identifier.*- Identifier
- 10.12681/eadd/14706