The development of a demand profile forecasting model for Eskom, with particular emphasis on the estimation of the demand impact of time differentiated tariffs

• Main Author: Berrisford, Andrew John
• Format: Others
• Language: en
• Published: 2016
• Subjects: Electric power distribution > Mathematical models; Electric power distribution > Economic aspects
• Online Access: http://hdl.handle.net/10539/20671

A dissertation submitted to the Faculty of Engineering. University of the Witwatersrand. Johannesburg, in fulfilment of the requirements for the degree of Master of Science in Engineering. D.tT~.awart •• w1t~4i.tinotion on , Dec ••ber I,e7 Johannesburg. ItN7 === Accurate forecastinu of system Maximum Demand (MDl is vital ~o Esk011l. Under-estimating the MD could re-ul: in a generation capacity shortage. with devastating consequences for the economy. Similarly. a high MD forecast would result in overcapacity. with expensive generating plant standing idle The traditional method of MD forecasting in Eskorn has become unreliable due to a changing relationship between forecast energy sales and expected maximum demand. The reasons for the changing system demand profile were isolated and analysed. Alternative MD forecasting techniques are evaluated and end-use hourly aggregation modets were ;d'~ntifieJ as a method suitable for Eskorn, An experimental demand profile forecasting moue! was developed, using data from a previous project. The model was tested and proved able to cope with the structural changes in the system demand profile. This resulted in the adoption of this technique by Eskorn and approval for the devclopn-ent of a f\.JJ1 scale de manu profile forecasting model.