基于递归神经网络的C3I系统故障预报系统

阐述了C3I系统故障预测系统的设计要求,分析了带有偏差单元的递归神经网络的相关原理,建立了C3I系统故障的神经网络预测模型并提出了相应的训练算法.提出了把神经网络、预测理论和专家系统有机地结合起来建立故障预报系统的初步设想.并以防空C3I系统的典型故障为例,实现了故障的预测.预测结果证明,给出的神经网络预测模型和训练算法是可行的,且具有较高的精度.

On C3I System Fault Predication System based on Neureal Network Model

Design requirement of fault prediction system and the advantage of recursive neural network technique are studied.Recursive neural network technique of C~3I fault diagnostic system is presented,and the algorithm based on the neural network model in the prediction processes is given.The neural network and the fault prediction technique with experts are combined to form a fault prediction system.Finally,the air defense C~3I system as an example is used to realize fault prediction.And the result proves that the forecast model and algorithm based on the neural network are feasible.