Confidence Interval Estimation for Distribution Systems Power Consumption by using the Bootstrap Method

The objective of this thesis is to estimate, for adistribution network, confidence intervals containing thevalues of nodal hourly power consumption and nodal maximumpower consumption per customer where they are not measured.The values of nodal hourly power consumption are needed inoperational as well as in planning stages to carry out loadflow studies. As for the values of nodal maximum powerconsumption per customer, they are used to solve planningproblems such as transformer sizing. Confidence intervalestimation was preferred to point estimation because ittakes into consideration the large variability of theconsumption values. A computationally intensive statisticaltechnique, namely the bootstrap method, is utilized toestimate these intervals. It allows us to replace idealizedmodel assumptions for the load distributions by model freeanalyses. Two studies have been executed. The first one is based onthe original nonparametric bootstrap method to calculate a95% confidence interval for nodal hourly power consumption.This estimation is carried out for a given node and a givenhour of the year. The second one makes use of the parametricbootstrap method in order to infer a 95% confidence intervalfor nodal maximum power consumption per customer. Thisestimation is realized for a given node and a given month.Simulation results carried out on a real data set arepresented and discussed.