摘要：
An instantaneous self-healing method of embedded software components based on cooperative game is proposed.The hidden layer center of the fault diagnosis model is determined by genetic algorithm.Particle swarm optimization is used to optimize the adaptive weights in the fault diagnosis network model.According to the adjustment of weights,the instantaneous fault diagnosis model of the best component is obtained,and the high-precision diagnosis of instantaneous fault is realized.The participants involved in fault self-healing are analyzed,and the schemes obtained by the main agents are combined to obtain the self-healing of non-cooperative game Then all the subjects are used to complete the cooperative game decision-making,and the objective aggregation function of the approach to the ideal solution is taken as the objective function of each participant so as to select the satisfactory solution to the instantaneous self-healing of the component.The simulation results show that the method proposed in this paper is superior to the current method in the instantaneous fault diagnosis of components and the instantaneous self-healing coefficient of software components is controlled at about 0.5,which makes the instantaneous self-healing effect of components reach the best.%为了提高软件组件使用寿命,增强组件利用率,需要对嵌入式软件组件瞬时故障自愈合进行研究.针对当前基于CBR的软件组件瞬时故障自愈合方法,在3GPP自愈合流程的基础上实现组件瞬时故障自愈合的分析,故障诊断准确性差,自愈合效果不理想的问题.提出一种基于合作博弈的嵌入式软件组件瞬时故障自愈合方法,通过遗传法对故障诊断模型隐层中心进行确定.采用粒子群法对故障诊断网络模型中的自适应权重进行寻优,根据权重的调整,得到最佳组件瞬时故障诊断模型,实现瞬时故障的高精度诊断.对故障自愈合的参与主体进行分析,把各主体获得的方案进行组合,获得非合作博弈的自愈合方案均衡解的集合,接着利用所有主体完成合作博弈的决策,并将与理想方案的接近程度目标集结函数,当作各个参与主体目标函数,以选出组件瞬时故障自愈合的满意解.仿真表明,所提方法对组件瞬时故障诊断的精度要高于当前方法,同时将软件组件瞬时故障自愈合系数控制在0.5左右,使得组件瞬时故障自愈合效果达到最佳.
