Abstract:

Smart grids coulpled with effiicient demand side management (DSM) is an important step for greener cities of the 

future. DSM has the potential to significantly improve smart grid operations by reducing the peak to average 

ratio. Current DSM schemes are able to reduce peak load by as much as 30% which can translate to significant 

cost savings and reduction in green house emissions. But for realistic deployment of DSM systems in the grid 

there are two very important aspects which need to be considered: scalability and user acceptability. Since the 

current DSM algorithms are required to control potentially hundreds of thousands of devices, they have to be 

scalable and tractable for such myriad numbers. On the other hand DSM affects the life style of the consumer 

and this should be as less disruptive as possible. The various DSM techniques proposed in the literature attempt 

to first reduce the cost and then attempt to resolve one of the two aforementioned aspects. The result is that 

the techniques are either scalable or are only considerate of the deadlines of the consumers. An ideal system 

should cater to both of these aspects. Our system Deli2P is user centric and scalable thus catering to both of 

these aspects. Essentially we provide to the consumer a deadline centric interface. The deadlines solutions are 

generally not scalable. But instead of solving this problem as a scheduling for deadline problem we transform the 

problem to a priority-based problem thus making it scalable for large number of devices. Our results show that with 

this scheme we can reduce peak power by upto 30% but without violating consumers' deadlines.