PROJECT TITLE :

Statistical Estimation of the Residential Baseline

ABSTRACT:

Demand response on the residential market is changing into a answer to adapt client consumption to the supply out there and therefore lower the electricity peak costs. Tariff incentives and direct load control of residential air-conditioners and electrical heaters are versatile solutions to scale back the peak demand. To include residential demand response resources in designing operators, quantifying the demand reduction is becoming a major issue for all electrical stakeholders. Current ways are primarily based on day or weather matching, regressions and control cluster approaches. In general, strategies using obtainable information from a control group give a lot of correct results. With the introduction of sensible meters, the electrical utilities generate a giant quantity of quality data, offered almost in real time. In this paper, we tend to recommend using these on the market residential load curves to pick a management cluster primarily based on individual load curves. One amongst the benefits of our method is that the chosen control cluster might adapt at anytime to the number of people belonging to the demand reduction program, as this variety evolves with customers getting into and leaving the program. Constrained regression methods and an algorithm are developed and evaluated on real information, providing a reliable solution for an operational use.