An Improved Black Widow Optimization Algorithm for Engineering Constrained Optimization Problems

In solving engineering constrained optimization problems, the conventional black widow optimization algorithm (BWOA) has some shortcomings such as insufficient robustness and slow convergence speed. Therefore, an improved black widow optimization algorithm (IBWOA) is proposed by combining methods of double chaotic map, Cauchy center of gravity inverse difference mutation and golden sine guidance strategy. Firstly, the quality of the initial population of the BWOA is improved based on the double chaotic map; Secondly, in order to make full use of the difference information between the current and the optimal position thus improve optimization accuracy, the golden sine algorithm (Gold-SA) is introduced to update the position of the black widow individuals; Finally, the Cauchy barycenter reverse differential mutation operator is employed to increase the diversity of the population, avoid local optimization thus improve the global search ability of the algorithm. In addition, the global convergence characteristics of the IBWOA are analyzed based on the Markov process and the convergence probability reaches 1 for the globally optimal solution. The performance of the proposed IBWOA was evaluated based on eight continuous / discrete hybrid engineering optimization problems and typical benchmark functions. The results show that the improved BWOA can improve the search accuracy, convergence speed and robustness effectively comparing with some other conventional optimization algorithms.