An ameliorated particle swarm optimizerfor solving numerical optimization problems
Ke Chen, Fengyu Zhou*, Yugang Wang, and LeiYin
Abstract:Although theparticle swarm optimizer (PSO) has been widely used to addressvarious complicated engineering problems, it is likely to suffer lack ofdiversity and ineffectiveness of balance between the global search ability and thelocal search ability in the search process. In this paper, we report an innovative and improved optimization method calledameliorated particle swarm optimizer (A-PSO), which is different from theoriginal PSO algorithm and its variants in parameter update and the position generation of each particle。 InA-PSO, the nonlinear dynamic acceleration coefficients, logistic map and amodified particle position update approach are introduced in PSO to improve thesolution quality and accelerate the global convergence rate。 Twenty well-knownnumerical optimization functions are adopted to evaluate the effectiveness ofthe proposed method and it is illustrated that, for most numerical optimizationproblems, the convergence performance and search accuracy of the A-PSO methodare superior to the similar heuristic optimization algorithms and otherwell-known PSO variants。 Namely, the proposed A-PSO technique has afaster convergence rate and is more stable than other PSO variants and similarpopulation-based methods for almost all numerical optimization problems。Therefore, the A-PSO method is successfully used as a new optimizationtechnique for solving numerical optimization problems。
Keywords: Particleswarm optimizer; Nonlinear dynamic acceleration coefficients; Logistic mapsequence; Numerical optimization problems; Optimization.
1。 The nonlinear dynamic accelerationcoefficients (NDAC) as a new parameter update mechanism for the cognitivecomponentc1 andthe social componentc2,respectively.
2。 The logistic map sequence is adopted to tunethe inertia weight ω。
3. Dynamic weight, correction factor andbest-so-far position introduced to update the new position with original updateformula.
Applied SoftComputing (JCR一区, SCI二区, IF=3.907)