Introduction

Demand-side management (DSM) is one of the key components of the future smart grid to enable more efficient and reliable grid operation [1]. To achieve a high level of reliability and robustness in power systems, the grid is usually designed for peak demand rather than for average demand. This usually results in an under-utilized system. To remedy this problem, different programs have been proposed to shape the daily energy consumption pattern of the users in order to reduce the peak-to-average ratio in load demand and use the available generating capacity more efficiently, avoiding the installation of new generation and transmission infrastructures. However, the increasing expectations of the customers both in quantity and quality [2], emerging new types of demand such as plug-in hybrid electric vehicles (PHEVs), which can potentially double the average household energy consumption [3], the limited energy resources, and the lengthy and expensive process of exploiting new resources give rise to the need for developing some more advanced methods for DSM.

Since electricity cannot be stored economically, wholesale prices (i.e., prices set by competing generators to regional electricity retailers) vary drastically between the low-demand times of day and the high-demand periods. However, these changes are usually hidden from retail users. That is, end users are usually charged with some average price. To alleviate this problem, various time-differentiated pricing methods have been proposed in the literature. Some examples include day-ahead pricing, time-of-use pricing, critical-peak-load pricing, and adaptive pricing [4–7].

By equipping users with two-way communication capabilities in smart grid systems and by adopting real-time pricing (RTP) methods, it is possible to reflect the fluctuations of wholesale prices to retail prices.