Sepia - A self-organizing electricity pricing system

The demand for electrical power is not constant. There are certain times of the day where the demand levels are much higher than the rest of the day. The demand can often exceed the generation capacity and when that happens, the utility companies can either shed loads or buy additional electrical energy from wholesale electricity markets to close the gap between demand and supply. These markets clear energy at spot prices that fluctuate widely and can be much higher when the demand is high than when the demand is low. When the market rate for electricity rises above the approved retail rate, utilities are caught in the middle, which can be financially disastrous for them. As such, utility companies, to protect themselves from widely fluctuating costs and to reduce peak demands, are introducing Advanced Metering Infrastructure (AMI) and considering various dynamic pricing mechanisms such as Time Of Use (TOU) and Critical Peak Pricing (CPP). However, in these mechanisms, there can be both a significant delay in information reaching consumers and gaps in consumption data. These delays and gaps can undercut the premise of how smart meter technologies will empower consumers to make decisions about their electricity use based on real-time prices. Moreover, these pricing schemes are centralized, in the sense that, meters at customer premises connect to the utility systems to obtain the current price. Such a centralized systems are inefficient because they require substantial communication and computation resources. To address these shortcomings, we propose Sepia, a self-organizing real-time electricity-pricing scheme, that computes the price of a kilowatt-hour of electricity as a function of consumption history, grid load and the type (hospital/commercial/industrial etc.) of the customer. In this paper, we describe the details of this pricing scheme and demonstrate, using a simulator, how this scheme could potentially alter the consumption patterns.