Computational performance of artificial immune system-based sizing technique for grid-connected photovoltaic system / Ismail Musirin … [et al.]

In grid-connected photovoltaic system (GCPV) operation, the solar energy generated by the photovoltaic (PV) array is directly used to supply the load demand and the excess energy is exported to the utility grid via an inverter. Although the GCPV systems have become popular in Malaysian residential power systems, the optimal design of GCPV system is still an issue to the local GCPV system integrator. The design process commonly known as sizing process is a very important aspect in GCPV system design as the performance of the sizing components are heavily influenced by inconsistent solar radiation, ambient temperature and wind speed [A. Mellit et. al., 2009]. An undersized or oversized GCPV system could either increase the overall system cost or degrade the performance of the GCPV system (Kil et. al., 1994). Previously, J.D. Mondol et. al. (2009), S. Islam et. al. (2003) and Van der Borg (2003) and J.D. Mondol (2006) had investigated the optimal GCPV system design using both empirical and procedural techniques. As these conventional sizing techniques are tedious and time-consuming, an Artificial Immune System (AIS)- based sizing algorithm is proposed to improve the conventional sizing technique by improving the computation time and the flexibility of the design process. As AIS is proven be a better optimization technique compared to other evolutionary techniques due to its clonal operation, the proposed AlS-based sizing algorithm is expected to produce faster computation time and able to handle more design options design decision is made. It is aimed to produce a GCPV design that could maximize the inverter-to-PV array sizing ratio by selecting the optimal PV module and inverter from a pre-developed databases as well as to determine the optimal PV module mounting orientation for a given roof space.