基于q-高斯的模糊神经网络在飞机作战效能评估中的应用  

Fuzzy Neural Network Based on q-Gaussian and Its Application in Operational Effectiveness Evaluation of Planes

作　　者：赵伟[1] 伞冶[1] 

机构地区：[1]哈尔滨工业大学控制与仿真中心,黑龙江哈尔滨150001

出　　处：《北京理工大学学报》2010年第6期674-677,682,共5页Transactions of Beijing Institute of Technology

基　　金：国家自然科学基金资助项目(60474069)

摘　　要：为了增强模糊神经网络的自学习和自适应能力,提出基于q-高斯的模糊神经网络评估飞机作战效能.采用q-高斯函数作为模糊神经网络的模糊隶属度函数,利用量子粒子群算法优化基于q-高斯的模糊神经网络参数,将非广延熵指数q编码为粒子并随着种群的进化自适应地调整.通过评估飞机作战效能,结果表明,基于q-高斯的模糊神经网络作战效能评估的结果更准确,自学习和自适应能力更强.In order to enhance the self-learning and adaptive ability of fuzzy neural network,fuzzy neural network based on q-Gaussian was proposed for operational effectiveness evaluation of the planes. q-Gaussian function was taken as fuzzy membership function of fuzzy neural network and quantum-behaved particle swarm optimization algorithm was employed to optimize the parameters of fuzzy neural network based on q-Gaussian. The nonextensive entropic index q was encoded in the particle and was adjusted adaptively in the evolution of population. The simulation result of operational effectiveness evaluation of planes shows that fuzzy neural network based on q-Gaussian can obtain more accurate results and has better self-learning and adaptive ability.

关 键 词：作战效能评估 模糊神经网络 q-高斯 量子粒子群算法 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程] V271.4[自动化与计算机技术—控制科学与工程]