Transcript datasheet
Novel load forecasting technique for three timehorizons in electric power system Swasti R. Khuntia Challenge the future 1 Thesis Factsheet • Title of thesis: Novel load forecasting technique for three time-horizons in electric power system • Name of student: Swasti R. Khuntia • Start Date / End Date: 10/04/2014 – 09/04/2018 (Expected) • Departments involved: Electrical Sustainable Energy (formal) and Applied Probability (informal) • Supervisor(s): Dr. –Ing. Jose L. Rueda • Promotor(s): Prof. Mart A. M. M. van der Meijden • Companies and funding agencies involved: EU FP7 Project comprising of 20 partners (8 industries and 12 universities) • Associated Project Title: GARPUR (Generally Accepted Reliability Principle with Uncertainty modelling and through probabilistic Risk assessment) [www.garpur-project.eu] Challenge the future 2 Thesis content • Motivation • European power system reliability management surviving on ‘N1’ criterion • Time to upgrade the reliability criterion • Demand-side uncertainty • Intermittent energy sources • Part of the project involved, it aims at socio-economic benefits • Research question • “Designing, developing, assessing, and evaluating new reliability criteria” • My contribution: Load forecast model for three time-horizons using novel data mining technique Challenge the future 3 Thesis content • Leader: T5.2 and T6.2 • Contributor: • T2.3, T2.4 • T4.2, T4.3 • T5.1, T5.3, T5.4 • T6.3, T6.4 • T7.2, T7.3 • T9.1 Challenge the future 4 Thesis challenge • • • • Difficulty in modeling for three time-horizons Data uncertainty Upgrading the existing reliability criterion Driving factor: Project implementation in pan-European level in next decade Challenge the future 5 Methodology • Literature study on “Understanding the three time-horizons” in electrical transmission system • Literature study on “Asset management in electrical transmission system” • Literature study on “Existing methodology used in participating TSOs” [continuing] • Power flow study on Bayesian Network • One of the most efficient probabilistic graphical models to represent uncertainty and inferences thereof • Nothing much done in load forecast area for three time-horizons • Trying to assess the usability and planning to design a “hybrid” model Challenge the future 6 Project Plan Literature Test-case study WP5, 6 Implementation/Validation Writing Thesis 04/2014 2015 2016 2017 04/2018 Challenge the future 7 Challenge the future 8