GARPUR
Project Description
Load forecasts can be divided into three categories: short-term forecasts which are usually from one hour to one week, medium forecasts which are usually from a week to a year, and long-term forecasts which are longer than a year. The forecasts for different time horizons are important for different operations within a utility company. Also, the time-frame of different time-horizons as set by GARPUR is be taken into consideration. For modelling in the different time-horizons, a large amount of data recorded from SCADA/WAMS devices (good and/or bad) is involved. In such cases, data-mining is useful for analysing data in a sense that the feature variables are identified in a large database. It is one of the important intelligent systems in handling data classification/clustering and evaluating based on some rules and knowledge.
In my proposed study, Bayesian Network (BN) is to be considered. BN is one of the most efficient probabilistic graphical models to represent uncertain information and inferences thereof. BNs have found wide applications in many fields, and it’s use in power systems can be found in various literature. Although BNs have been applied in reliability assessment, fault detection, prediction of weather-related failures of overhead lines and fault location finding, but little or no work is done in the area of load forecasting in different time-horizons.
Literature survey suggests that Bayesian approach has been applied for short-term load forecasting. With my proposed approach, a platform will be constructed in which it would be feasible for load forecasting in all time-horizons. The tool would be validated with real data and against the GARPUR Quantification Platform (GQP), planned to be developed under GARPUR. And, aligning the work with the project, I believe the work will deliver fruitful results which will eventually be helpful for future academic research and industrial implementation.
Project Team:
S. R. Khuntia
Swasti R. Khuntia was born in Rourkela, India. Currently he is a guest researcher in the Intelligent Electrical Power Grids (IEPG) research group of Dept. Electrical Sustainable Energy after finishing his PhD. His PhD research was supervised by Dr. -Ing. José L. Rueda and Prof. Mart A.M.M. van der Meijden. Prior to his PhD, Swasti obtained his Masters of Science degree from Illinois Institute of Technology, Chicago in Electrical Engineering and his undergraduate degree from National Institute of Science and Technology, India in Electrical and Electronics Engineering. He is a DAAD-RISE Pro scholarship awardee, and actively involved in IEEE and CIGRE.Swasti takes keen interest in learning the role of data in power system applications. In the past four years, he worked on statistical modeling mainly focusing on high-dimensional dependence studies, forecasting in short-term and long-term horizon and trying to gasp the importance of addressing spatio-temporal correlation in power system security. Swasti was heavily involved in EU-funded FP7 project named ”GARPUR” which aimed at developing a new probabilistic reliability criteria at pan-European level.