Demand side management of small scale loads in a smart grid using glow-worm swarm optimization technique

Demand Side Management (DSM) is one of the most important parts of future smart power grid. With the rise in global energy awareness, smart grids enhance the potency and peak levelling of power systems. DSM is the controlling scheme in such grids and it aims to optimize several characteristics of load demand. This smart grids comprises energy storage (battery) and distributed solar photovoltaic generation storage. In this proposed methodology the combination of Glow-worm Swarm Optimization (GSO) and Support Vector Machine (SVM) is used for decision making process in battery storage to reduce the electricity tariff. GSO is a powerful technique to obtain near optimal solution which is used for this load rescheduling problem for a sample test system to minimize the cost of end user. Especially, the electricity expenditures of the end user can be reduced by responding to pricing which changes with different hours of a day. Then optimized range of the battery's energy storage is extracted from the GSO. Here, the SVM is trained based on the optimized data from the GSO. This combination is used for finding the amount of energy is transferred in/out the battery which aims the minimal electricity bill value. The electricity tariff of the proposed methodology of Average gosc is 2.27 for residential load, while considering it is less when compared to the existing method of 2.3 at the consumed load of 8.2 kWh/day. The proposed GSO-SVM method reduces 11.2% of energy cost which helps decision makers to take best demand-side actions for balancing the stability.