Hybrid Cuckoo Search Algorithm for Optimal Placement and Sizing of Static VAR Compensator

Introduction

This chapter presents a hybrid version of the Cuckoo search algorithm (HCSA) and its application for determining optimal placement and sizing of Static VAR compensator (SVC) devices in the electric power system. The Cuckoo search algorithm is a nature-inspired optimization technique, which works on the parasitic behavior of Cuckoo species. The conventional method includes two probability-generating stages. In the first stage, The Lévy flight generates random Cuckoo eggs, and then these eggs will be laid into the neighbors’ nests to create new solutions. The better solutions will carry over the next generation. The second stage describes the action of the host birds to abandon Cuckoo eggs in their nests. In this proposed Hybrid Cuckoo search algorithm, we combine the learner stage of the Teaching-learning-based optimization to improve performances of Cuckoo eggs instead of abandoning them. The proposed method is applied for optimizing location and sizing of SVC devices to enhance the voltage profile of the power system. This problem combines discrete and continuous unknowns with many equal and unequal constraints. The discrete unknowns are natural numbers representing the location of installed SVC devices, and the continuous unknowns represent injected reactive power of SVC devices. The proposed method is evaluated on three IEEE standard systems and compared with both of conventional Cuckoo search algorithm and Teaching-learning-based optimization. Numerical results show that the proposed Hybrid Cuckoo search algorithm gives better solutions and has higher performance than other methods.

This chapter divides into seven sections. Section 2 gives the literature review of the Cuckoo search algorithm and the problem of optimal placement and sizing of SVC devices. Section 3 proposes the Hybrid Cuckoo search algorithm. Section 4 describes the structure of a SVC device and the objective function. In Section 5, the implement of Hybrid Cuckoo search algorithm for the problem has been discussed. Section 6 shows numerical results on three benchmark problems. Finally, conclusions are made.