关键词: 天鹰优化算法, 折射反向学习, 种内互助, Bernoulli序列, 自适应权重, 柯西-高斯变异

Abstract:

In order to solve the problem that aquila optimization algorithm(AO) is easy to fall into local optimum and slow convergence, this paper proposes an improved aquila optimization algorithm with multi-strategy integration(MSIAO). In this algorithm, the refracted opposition-based learning combined with Tent chaotic map is used to initialize the population to improve the early search efficiency of the algorithm, and intraspecific and mutual assistance and optimization strategy are used to solve the problem of optimization stagnation of the algorithm. The convergence speed of the algorithm is improved by an adaptive weighting strategy based on Bernoulli chaotic sequences. Cauchy-Gaussian mutation operator is introduced to enhance the ability of the algorithm to escape local extremum in the later iteration. This paper conducts experiments on 10 benchmark functions and some CEC2014 test function sets, and the proposed MSIAO is applied to 2 engineering design optimization problems. The results show that MSIAO has higher convergence accuracy and faster convergence speed than AO for high-dimensional single-peak, high-dimensional multi-peak and fixed-dimensional complex multimode functions. Compared with AO, MSIAO saves 4.62% and 0.77% in economic cost of optimal design of pressure vessel and welding beam, which verifies the practicability and superiority of MSIAO in dealing with mechanical engineering problems.

Key words: aquila optimization, refracted opposition-based learning, intraspecific and mutual assistance, Bernoulli sequence, adaptive weight, Cauchy-Gaussian mutation