Optimal Power Flow Solution Using Self-Evolving Brain-Storming Inclusive Teaching-Learning-Based Algorithm

In this paper, a new hybrid self-evolving algorithm is presented with its application to a highly nonlinear problem in electrical engineering. The optimal power flow problem described here focuses on the minimization of the fuel costs of the thermal units while maintaining the voltage stability at each of the load buses. There are various restrictions on acceptable voltage levels, capacitance levels of shunt compensation devices and transformer taps making it highly complex and nonlinear. The hybrid algorithm discussed here is a combination of the learning principles from Brain Storming Optimization algorithm and Teaching-Learning-Based Optimization algorithm, along with a selfevolving principle applied to the control parameter. The strategies used in the proposed algorithm makes it self-adaptive in performing the search over the multi-dimensional problem domain. The results on an IEEE 30 Bus system indicate that the proposed algorithm is an excellent candidate in dealing with the optimal power flow problems.