论文部分内容阅读
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。
In order to improve the prediction level of aviation equipment accidents, a combination forecasting model of aviation equipment accidents based on gray and time series analysis model is proposed. First build a gray model, extract the historical data carried in the trend of information. Then, model selection, order identification and parameter estimation are carried out. A time series analysis model of gray residuals is established to characterize the random fluctuations in historical data. Finally, the predicted values of the two models are added to obtain the combined forecast result. In the example, based on the rate data of A-level flight accident of USAF from 1996 to 1999, a gray time series combination model is established. The accuracy of short-term prediction is better than single gray model, the average relative error is less than 5% The results reflect the actual state of aviation equipment safety.