航空装备事故的灰色时序组合预测模型

为提高航空装备事故预测水平,提出一种基于灰色和时间序列分析模型的航空装备事故组合预测模型。先构建灰色模型,提取历史数据中承载的趋势信息。然后进行模型选择、阶数识别和参数估计,建立灰色残差的时间序列分析模型,用以刻画历史数据中的随机波动特征。最后,将2个模型的预测值相加,得到所求的组合预测结果。实例中,以美国空军1996—1999年的A级飞行事故10万时率数据为基础,建立灰色时序组合模型,模型中短期预测精度优于单一灰色模型,平均相对误差控制在5%以内,预测结果能够反映航空装备安全的实际状况。

Grey and Time Series Combination Prediction Model for Aviation Equipment Accident

To improve the prediction level of aviation equipment accident,a combination prediction model,which is based on grey model and time series analysis model,was built.First,grey model was constructed to extract the trend information contained in historical data;then,through model selection,order number identification and parameter estimation,a time series analysis model for residual error of grey model was built in order to describe the stochastic fluctuation characteristics in historical data;finally,the combination prediction results can be obtained by adding the prediction values of two established models together.In actual example,a grey and time series combination model was set up on the basis of A class flight accident 100 000-hour-rate data of USAF from 1996 to 1999.The results show that compared to alone grey model the combination model has a better medium and short term prediction accuracy within average relative error 5%,which can reflect the actual situation of aviation equipment safety truthfully.