Evaluation of Electrical Baseline Load Profile Models for Estimating Residential Demand Response Load Residential Demand Response Load
| dc.contributor.author | John, Dileep | |
| dc.date.accessioned | 2012-09-10T13:42:02Z | |
| dc.date.available | 2012-09-10T13:42:02Z | |
| dc.date.issued | 2012-09-10 | |
| dc.description | MBA thesis (WBS) | en_ZA |
| dc.description.abstract | In light of the electricity supply constraint that South Africa is facing, electricity distribution utilities are looking towards residential demand response (DR) as a demand side management solution. The estimation of residential demand response load impacts requires the use of standardized and consistent methods to meet both demand response programme administrator and participant objectives. For load impacts assessment, an essential part of the analysis is the estimation of the baseline load profile. In this research, an evaluation of a select set of baseline load profile models is performed using data obtained from a residential demand response Pilot in Gauteng. The evaluation is performed by comparing the accuracy and precision of models in estimating baseline loads for a large number of residential customers for a select set of proxy event days. Measures of the accuracy and precision of the different models, the importance of load variability and weather effects, and the effect of applying adjustment factors are presented. Our results suggest that (1) both average based and temperature based regression models evaluated tend to perform equally well from an accuracy and precision point of view, with the simple average based models performing slightly better than the temperature based models for residential customers,(2) the accuracy of all models deteriorate with increasing load variability and hence the characterization of residential loads by variability is a useful indicator in reconciling demand response evaluation expectations and outcomes, (3) residential customers are weakly temperature sensitive and hence temperature based regression models do not outperform simple average based models, and (4) owing to the high load variability for residential customers, the inclusion of an adjustment factor worsens the performance of the models and hence should be avoided | en_ZA |
| dc.identifier.uri | http://hdl.handle.net/10539/11923 | |
| dc.language.iso | en | en_ZA |
| dc.subject | Electricity supply | en_ZA |
| dc.subject | Residential demand response models | en_ZA |
| dc.title | Evaluation of Electrical Baseline Load Profile Models for Estimating Residential Demand Response Load Residential Demand Response Load | en_ZA |
| dc.type | Thesis | en_ZA |