Developing Metaheuristic Optimization Techniques for Optimal Power Flow Considering Renewable Generators Controllable Loads and Electric Vehicles

Abstract

The demand of energy is increasing due to the growing population of the world and improvements of technology. One of the best significant solution techniques to fulfil this energy demand is utilization of renewable energy sources (RESs). Modern power systems, which integrate RESs, such as wind, small hydro or solar energy sources need to carry out the uncertainty by the accessibility of demanded or injected power. Therefore, it is necessary to consider uncertainty costs in optimal power flow (OPF) problems. newlineThe drawback of large penetration of different types RESs in system is that the optimal power flow problem turns into a Mixed-Integer Non- Linear Programming (MINLP) problem due to large number of continuous, discrete and binary variables and network non-linear equations. Therefore, it is sometimes very difficult to solve such problems using classical optimization methods, because in these techniques the inverse of hessian may become ill conditioned for this type of large-scale MINLP problem. Sometimes in classical methods, the non-linear constraints are assumed as linearized, which creates the accuracy related problem to get the sub-optimal solution. Another drawback of classical optimization method is that, these techniques may take a large execution time, often requiring more than one day to provide the sub-optimal solution. This work proposed a novel hybrid meta-heuristic algorithm entitled Cross Entropy Cuckoo Search Algorithm (CE-CSA). The application of levy flights in the Cuckoo Search Algorithm (CSA) improves the local exploitation capability while the Cross Entropy (CE) method is used in the initial stage for global exploration due to its fast convergence. The effectiveness of the proposed hybrid algorithm has been demonstrated in solving the OPF problem, considering RESs and controllable loads for different stochastic scenarios in a benchmark system to minimize the total operation cost. To verify its effectiveness, its performance is compared with the most advanced and recently proposed hybrid met