Techno-economical optimization of hybrid pv/wind/battery system using Neuro-Fuzzy

High cost of renewable energy systems has led to its slow adoption in many countries. Hence, it is vital to select an appropriate size of the system in order to reduce the cost and excess energy produced as well as to maximize the available resources. The sizing of hybrid system must satisfy the LPSP (Loss of Power Supply Probability) which determines the ability of the system to meet the load requirements. Once the lowest configurations are determined, the cost of the system must then be taken into consideration to determine the system with the lowest cost. The optimization methodology proposed in this paper uses the ANFIS (Adaptive Neuro-Fuzzy Inference System) to model the PV and wind sources. The algorithm developed is compared to HOMER (Hybrid Optimization Model for Electric Renewables) and HOGA (Hybrid Optimization by Genetic Algorithms) software and the results demonstrate an accuracy of 96% for PV and wind. The optimized system is simulated in PSCAD/EMTDC and the results show that low excess energy is achieved. The optimized system is also able to supply power to the load without any renewable sources for a longer period, while conforming to the desired LPSP. ► Using ANFIS to model PV and wind sources from real weather data which can learn, predict and linearized the output. ► Optimized using loss of power supply probability and cost of energy to obtain the lowest configuration of hybrid system. ► Configured system has high autonomy, less excess with high efficiency. ► System is compared to HOMER and HOGA with good results up to 96% accuracy. ► PSCAD simulation of the system is used to validate the system performance to prove the parameters.