Grid loss reduction with Demand response
Examensarbete för masterexamen
Sustainable energy systems (MPSES), MSc
In order to achieve more efficient use of the electricity grid, the Swedish market inspectorate has introduced an incentive for the grid companies in Sweden to reduce electricity grid losses and reduce peak demand. The incentive provides an increase in the revenue cap (the maximum allowed income) for the grid companies. One way to achieve reduced losses is to introduce demand response where the idea is to reduce or shift demand during peak hours to off-peak hours. This reduces losses because losses are proportional to the square of the load power. The aim of this thesis is to investigate how much grid loss reduction can be achieved in an electricity distribution grid by using demand response and what the quantitative economic benefits are. The methodology of this report consists of creating two cases of demand response; one with a conservative estimation of its potential to reduce and shift load, and one optimistic estimation. These are compared with a base case, depicting the current situation without demand response. The consumption data and grid data of the grid in Ulricehamn, Sweden, are analyzed using a data analyzation program called Lavastorm, to extract the necessary data in order to perform loss calculations. The loss calculations are based on the extracted grid data and performed using power flow analysis in MATLAB. The results show a reduction of losses compared to the base case of 0.6 % for the conservative estimation of demand response and 13.4 % for the optimistic estimation of demand response. There is an increase of the revenue cap compared to the base case for the conservative estimation of 10 237 SEK and 201 698 SEK for the optimistic estimation of demand response. The achievable cost reduction by reducing losses for the two cases of demand response are 20 126 SEK and 449 484 SEK respectively. The load reduction results in a cost reduction of 0 SEK and 983 531 SEK respectively. These results should be taken as indications rather than absolute truths. Due to the quality of the measurement data, the analysis was only made on a relatively small part of the grid. The economic results above were obtained by scaling the results from the small part of the grid to the entire grid. Another conclusion from the study is that the quality of the grid data and consumption data are relatively low for the purposes of this study. Furthermore, the method of depicting the grid using spreadsheet data instead of grid maps should prove useful when analyzing hourly measurements, but the data needs to have sufficient quality for it to work.
Elkraftteknik , Electric power engineering