Utvärdering och uppdatering av typkurvor

• Main Authors: Dahlenborg, Marcus, Grahn, Carl
• Format: Others
• Language: Swedish
• Published: Karlstads universitet, Avdelningen för fysik och elektroteknik 2014
• Subjects: typkurvor
• Online Access: http://urn.kb.se/resolve?urn=urn:nbn:se:kau:diva-32938

The purpose of this thesis is to evaluate and update load curves for Fortum Distribution AB. Theload curve method is a tool used to forecast the strain of an electric grid. The method is based oncustomer annual energy consumption, outdoors temperature as well as customer category. Theload curve method was established by Svenska Elverksföreningen at the start of the nineties andfocus was put on standardizing annual energyconsumption, in order to make the method usableacross the nation. Present consumption patterns have changed since the original load curves wereproduced, which leads to a need to update the load curves.The work began with an update of present load curves according to recommendation taken fromanother earlier thesis. Thereafter a forecast model was produced in Excel to calculate the strainfor one customer, using the new load curves. The updated load curves were computed into theFortum network calculation software to be used in the calculations of the outgoing feeders indistribution stations in four geographically different areas in Sweden. The calculations were madethe third Wednesday in January of both 2013 and 2014. The day is arbitrarily chosen. Calculatedvalues were compared to measured hourly values of feeders in distribution stations of the sameday. An acceptable border for calculated values is determined by Fortum to plus/minus 10% inrelation to observed values. Network calculations with the updated load curves showed ofaverage an error marginal of more than 30 % in all four distribution channels. We discovered nogeographical influence of the error marginal.Due to the new average error marginal being far above Fortums permit limitation we believe thisnew categorization should not be used. Additional adjustments with related evaluations of loadcurves should be made before the categorization can be applied.