论文部分内容阅读
模拟退火(SA)系寻找全局最优并能跨越局部最优的随机优化算法,它源于对高温物质的退火过程的模拟即在给定温度下对微观粒子(如原子)平衡的统计力学模拟.SA采用模拟算法及随机抽样;Kirkpatrick等深入研究了组合优化问题;Bohackevsky等提出了通用模拟退火(GSA)法;Kalivas等研究了GSA用于多元校正.本文将SA与GSA用于多组分分析,获得良好效果.
Simulated annealing (SA) is a stochastic optimization algorithm that searches for the global optimum and crosses the local optimum. It stems from the simulation of the annealing process of high-temperature materials, that is, the statistical mechanical simulation of the equilibrium of microscopic particles (such as atoms) at a given temperature .SA uses the simulation algorithm and the random sampling; Kirkpatrick et al. Studied the combinatorial optimization problem in depth; Bohackevsky et al proposed the general simulated annealing (GSA) method; Kalivas et al. Studied GSA for multivariate calibration. Analysis, get good results.