Genetic algorithm based optimized load frequency control for storageless photo voltaic generation in a two area multi-agent system

In this paper, a multi-agent load frequency balancing control algorithm based on Genetic Algorithm for a storage-less Photo Voltaic (PV) generation is proposed. For maximum deployment of available renewable energy, the PV generation is tracked at its maximum power point however through the use of a virtual synchronous generator converter control, the PV generator is able to follow the load demand so far its possible maximum power remains in excess of the load demand. When the maximum power falls below demand, the conventional synchronous generator in the second area covers the power deficit. Through Genetic Algorithm based Optimization, the virtual inertia parameter is determined for optimum performance of the PV load following capability. Through the Universal Multi Agent Platform (UMAP), we are able to communicate the frequencies of the two areas (representing two agents) to a central agent which computes the Area Control Error (ACE) and therefore take frequency bias into account in our system model. By implementing load following without any energy storage, we significantly reduce costs accrued from energy storage installations.