广义Gauss模型及其模拟退火算法
Extended Gauss Model and Its Simulated Annealing Algorithm
-
摘要: 研究了求解优化问题全局解的随机神经网络方法,将Gauss模型拓展为广义Gauss模型,使之能求解一 般优化问题的全局解。进而引入全局性较好的模拟退火算法的思想,提出了广义Gauss模型的模拟退火算法。 通过算例比较了几种计算智能算法的全局性。广义Gauss模型的模拟退火运行全局性最好,但它付出了时间的 代价;广义Gauss模型兼顾了计算效率和全局性;广义Hopfield网络的全局性最不理想。Abstract: The method for global optimization by random artificial neural networks (AN2) is studied. The Gauss model is modified to an extended Gauss model, which can be used to obtain the global optimum. With the simulated annealing algorithm being taken into consideration, the simulated algorithm for the extended Gauss model is proposed. Simulation shows that the extended Gauss model has the best comprehensive behavior, the extended Gauss model running with the simulated algorithm gives the best global performance with much longer operation time than other two, and the extendedHopfield model is the worst among them.
-
Key words:
- neural networks /
- optimization /
- algorithms /
- simulated annealing
点击查看大图
计量
- 文章访问数: 1682
- HTML全文浏览量: 75
- PDF下载量: 184
- 被引次数: 0