Reducing Electricity Demand Peaks by Scheduling Home Appliances Usage

Nowadays there is a tendency to consume electricity during the same period of the day leading to demand peaks. Regular energy consumption habits lead to demand peaks at specific temporal intervals, because users consume power at the same time. In order to avoid demand peaks, users’ appliances should consume electricity in a more temporarily distributed way.

A new methodology to schedule the usage of home appliances is proposed and analyzed in this paper. The main concept behind this approach is the aggregation of home appliances into priority classes and the definition of a maximum power consumption limit, which is not allowed to be exceeded during peak hours.

The scenario simulated describes a modern household, where the electrical devices are classified in low and high priority groups. The high priority devices are always granted power in order to operate without temporal restrictions. On the contrary, the low priority devices have to pause their operation, when the algorithm dictates it, and resume it in the future.

This can become beneficial for both energy companies and users. The electricity suppliers companies will be capable of regulating power generation during demand peaks periods. Moreover, users can be granted lower electricity bill rates for accepting delaying the operation of some of their appliances.

In order to analyze this scenario, teletraffic engineering theory, which is used in evaluating the performance of telecommunication networks, is used. A reversible fair scheduling (RFS) algorithm, which was originally developed for telecommunication networks, is applied. The purpose is to analyze how a power consumption limit and priorities for home appliances will affect the demand peak and the users’ everyday life.

Verification of the effectiveness of the RFS algorithm is done by means of simulation and by using real data for power consumption and operation hours. The defined maximum power limit of 750 and 1000 Watt was not exceeded during peak demand hours. The trade‐off was an average delay of 36,1 and 12,36 minutes, respectively, for the aggregated low priority class.