Conference proceeding
Power Distribution Networks Load Forecasting Using Deep Belief Networks: The South African Case
2019 IEEE JORDAN INTERNATIONAL JOINT CONFERENCE ON ELECTRICAL ENGINEERING AND INFORMATION TECHNOLOGY (JEEIT), pp 507-512
01 Jan 2019
Featured in Collection : UN Sustainable Development Goals @ Drexel
Abstract
Load forecasting is considered a time series problem, whose accuracy is important in operations and planning of micro and large power systems. South Africa is a developing country that is in its 25th year of democracy. The study of load forecasting using artificial intelligence techniques in South Africa is limited. The application of deep learning techniques in South African load forecasting is non-existent. This paper overcomes these shortfalls by introducing deep learning techniques in South African load forecasting. This was conducted using a real South African distribution sub-stations loading data and a sophisticated deep learning technique, deep belief networks. The substation is an 80 MVA, 88/11 kV distribution substation. Weather parameters, and more specifically temperature, have been seen to improve the accuracy of load forecasts. It has, however, been shown that this is not always the case. This paper investigated the impact of temperature on the load forecasting error when using a sophisticated deep learning technique, deep belief networks. The impact of using cleaned data and uncleaned data was also investigated. The lowest error was obtained with non-cleaned data with temperature as an input parameter. The obtained errors were around 4%. Hence, deep learning techniques can be applied in South African distribution networks for load forecasting, and therefore cost reduction through improved planning.
Metrics
Details
- Title
- Power Distribution Networks Load Forecasting Using Deep Belief Networks: The South African Case
- Creators
- Sibonelo Motepe - University of JohannesburgAli N. Hasan - University of JohannesburgBhekisipho Twala - University of South AfricaRiaan Stopforth - University of South Africa
- Contributors
- K M Jaber (Editor)
- Publication Details
- 2019 IEEE JORDAN INTERNATIONAL JOINT CONFERENCE ON ELECTRICAL ENGINEERING AND INFORMATION TECHNOLOGY (JEEIT), pp 507-512
- Publisher
- IEEE
- Number of pages
- 6
- Resource Type
- Conference proceeding
- Language
- English
- Academic Unit
- Engineering Leadership and Society/Engineering Technology
- Web of Science ID
- WOS:000470894100095
- Scopus ID
- 2-s2.0-85067111415
- Other Identifier
- 991022004621604721
UN Sustainable Development Goals (SDGs)
This publication has contributed to the advancement of the following goals:
InCites Highlights
Data related to this publication, from InCites Benchmarking & Analytics tool:
- Collaboration types
- Domestic collaboration
- Web of Science research areas
- Computer Science, Information Systems
- Engineering, Electrical & Electronic