Particle swarm optimization based ultra fast renewable energy source optimization tool design

Purpose: The aim of this study is to design an alternative rapid optimization tool that eliminates the sensitivity, difficult search space and speed disadvantages of the HOMER software, which is widely used in the optimization of renewable energy resources. Thanks to this tool, it will also be easier to produce a large number of data by obtaining the necessary optimization outputs to train surrogate models, machine learning or deep learning -based systems very quickly. Theory and Methods: PSO algorithm is preferred as an optimization algorithm because it is fast and easy. The capacity shortage parameter, which is not used much in the literature, is used as a reliability parameter. The capacity shortage parameter was used for the first time in the optimization of renewable energy sources with the swarm -based algorithm. Optimization with the capacity shortage parameter is more advantageous and provides more accurate system sizing. Because, when determining the capacity shortage, the simulation is made as if enough energy to meet the predetermined extra instant loads and even a part of the production is reserved for unpredictable loads. Cost of energy is used as the cost function. Battery charge -discharge processes are simulated realistically. Detailed information about the renewable energy source, parameters such as battery life, excess energy, unmet energy, served energy, battery autonomy are calculated for the user. Results: The results were compared with the HOMER commercial hybrid system optimization program, and it was seen that both results were almost equivalent to each other. However, it is seen that the simulation time is much shorter than the HOMER in the proposed structure. These results show that the designed optimization system is superior to HOMER in terms of speed. Conclusion: Comparing the tool designed with HOMER, it has proven that it can be used in optimization processes alone and is much faster than HOMER. However, if it is desired to work with the commercial software HOMER or to benefit from the plug -ins of HOMER, the search space of the HOMER program can be created with frequent values around this optimum by quickly finding the optimum values with the tool designed in this study. Thus, the solution is reached in a much shorter way.