机载软件无线电系统故障预测软件设计与实现

通过分析机载软件无线电系统的结构特性,为满足其高可靠性高灵活性等要求,设计并实现了针对机载软件无线电系统的故障预测软件平台.该软件平台在MFC框架基础上进行开发,集成了自回归预测模型(AR)、灰色预测模型(GM(1,1))等预测算法,并对预测算法关键参数进行优化,使得该软件能够根据不同预测对象,以及不同预测精度要求自动选择最佳预测算法进行故障预测.同时,可根据需要在线升级已有预测算法或动态加载新的预测模型,提出并实现了一种适用可靠的开放式故障预测系统.

Design and Implementation of Failure Prediction Software for Onboard Software Radio System

By analyzing the structural characteristics of onboard software radio system,in order to meet its high flexibility and high reliability requirements,failure prediction software platform for onboard software radio system is designed and implemented.The software platform is developed based on the framework in MFC,it integrates multiple prediction algorithms,such as auto-regressive prediction model (AR),grey prediction model(GM(1,1)),and optimizes the key parameters of these predictive algorithms,so that the software can automatically select the best prediction algorithm according to different forecast target and different prediction accuracy requirements to carry out fault prediction.At the same time,according to the need to upgrade existing prediction algorithm online or dynamically loading new prediction model,a suitable and reliable open type fault prediction system is proposed and implemented.