- 期刊
An Efficient Sine Cosine Algorithm for Global Complex Optimization Problems
摘要
Optimization algorithms are widely used in the field of network security optimization. The Sine Cosine Algorithm (SCA) is an effective algorithm for solving global complex optimization problems. However, problems remain, such as insufficient solution accuracy, slow convergence speed, and difficulty jumping out of local extreme values. To improve the optimization performance and application ability of the SCA and apply it better to solve complex optimization problems in network information security, three improved strategies, namely Elite leadership, Quadratic interpolation optimization, and Self-feedback memory refresh, are introduced into the Sine Cosine Algorithm (EQSSCA). The elite leadership strategy first coordinates the algorithm's global exploration and local mining capabilities. Then, the quadratic interpolation optimization strategy is adopted to improve the algorithm's solution accuracy and enrich the population's diversity. Finally, a self-feedback memory refresh strategy is introduced to enhance the population's capability to evade local extremes and improve the algorithm's convergence rate. In addition, the EQSSCA and SCA are proven consistent in terms of time complexity by theoretical analysis. To evaluate the proposed algorithm's optimization capability, the optimization accuracy, difference significance, and convergence curves of EQSSCA and four high-performance comparison algorithms are tested and analyzed on various dimensions of the CEC2017 test suite. The test results indicate that the proposed three strategies can effectively enhance SCA's solution accuracy, robustness, adaptability, and effectiveness in solving global optimization problems. And the proposed algorithm is superior to the other four comparison algorithms.