| 基于ARIMA与SVM的飞行事故组合预测方法 |
| |
| 引用本文： | 甘旭升,端木京顺,丛伟,高建国.基于ARIMA与SVM的飞行事故组合预测方法[J].中国安全科学学报,2011,21(7). |
| |
| 作者姓名： | 甘旭升 端木京顺 丛伟 高建国 |
| |
| 作者单位： | 1. 西京学院汽车学院,陕西西安,710123
2. 空军工程大学工程学院,陕西西安,710038 |
| |
| 摘 要： | 飞行事故预测的目的在于预防事故。为提高事故预防的针对性和有效性,必须加强预测,以增强预防飞行事故的主动性。在ARIMA和SVM基础上,提出一种飞行事故组合预测方法。首先建立ARIMA模型,用以描述历史数据中的线性关系;然后,对ARIMA模型的残差构建SVM模型,用以模拟数据中的非线性规律,两者预测值之和就是最后的预测结果。美国空军1954—1993年飞行事故损坏飞机万时率的实证分析结果表明:利用该方法所建立的模型,能够对飞行事故作出较为准确的预测,模型精度总体优于单一的ARIMA或SVM模型。
|
| 关 键 词： | 差分自回归滑动平均(ARIMA) 单位根检验 支持向量机(SVM) 飞行事故 组合预测 |
Flight Accident Prediction Method Based on the Combination of ARIMA and SVM |
| |
| GAN Xu-sheng,DUANMU Jing-shun,CONG Wei,GAO Jian-guo.Flight Accident Prediction Method Based on the Combination of ARIMA and SVM[J].China Safety Science Journal,2011,21(7). |
| |
| Authors: | GAN Xu-sheng DUANMU Jing-shun CONG Wei GAO Jian-guo |
| |
| Institution: | GAN Xu-sheng1 DUANMU Jing-shun2 CONG Wei2 GAO Jian-guo2(1 Automobile College,Xijing College,Xi'an Shaanxi 710123,China 2 Engineering College,Air Force Engineering University,Xi'an Shaanxi 710038,China) |
| |
| Abstract: | The aim of flight accident prediction is to prevent accidents.To improve its pertinence and efficiency,the prediction must be strengthened to enhance the initiative of accident prevention.A hybrid prediction method,based on ARIMA and SVM,is proposed.First,ARIMA model is established to describe the linear relation of historical data,and then SVM model is built on the residual error of ARIMA model in order to simulate the nonlinear law of historical data.The sum of the two values predicted by the two models i... |
| |
| Keywords: | auto-regressive integrating moving average(ARIMA) unit root test support vector machine(SVM) flight accident hybrid prediction |
| 本文献已被 CNKI 万方数据 等数据库收录！ |
|
