摘要
电子系统的状态预测是利用其历史信息来实现系统未来状态和趋势的估计,以防止灾难性故障的发生,对于推动视情维修具有重要意义。针对典型模拟滤波电路,通过分析其关键测试信号的特点,研究了基于灰色理论的状态预测方法,并针对该预测模型的不足,设计粒子群算法选择最佳预测维数,设计新陈代谢法使该模型参数在线改变,从而建立符合电子系统信号特点的灰色预测模型。将该模型与ARAM模型比较,实验结果验证了该模型具有较好的状态预测精度和预测性能。
The state prediction of electronic system uually makes full use of historical information to estimate its future state and tendency aiming at avoiding disastrous faults, which is very significant to the development of condition based maintenance. This thesis puts an analog filter circuit as an example and the state prediction technology based on grey theory is studied through analyzing the characters of its key testing signals, where particle swarm optimization algorithm is used to obtain the best forecast dimension and the metabolism method is presented to make the model parameters on-line change. Compared with the ARAM model, the experiment results show that the improved model has good precision and performance for state prediction.
出处
《火力与指挥控制》
CSCD
北大核心
2012年第5期52-55,59,共5页
Fire Control & Command Control
基金
国家自然科学基金资助项目
关键词
状态预测
灰色模型
粒子群算法
新陈代谢法
state prediction,grey model ,particle swarm optimization algorithm, metabolism method