航空电子技术
航空電子技術
항공전자기술
AVIONICS TECHNOLOGY
2014年
1期
46-51
,共6页
支持向量回归机(SVR )%QAR数据%燃油消耗模型
支持嚮量迴歸機(SVR )%QAR數據%燃油消耗模型
지지향량회귀궤(SVR )%QAR수거%연유소모모형
Support Vector Regression(SVR)%QAR data%Fuel Consumption Model
针对飞机巡航段燃油消耗量预测问题,提出一种基于支持向量回归机(SVR :Support Vector Regression)的预测建模方法,并应用Grid-Search 参数寻优法优化模型参数,基于真实QAR数据建立 SVR预测模型,并从平方相关系数和平均绝对百分误差两个不同指标与 BP 神经网络模型的预测结果进行比较,比较结果表明:SVR预测模型的预测结果精度高。
針對飛機巡航段燃油消耗量預測問題,提齣一種基于支持嚮量迴歸機(SVR :Support Vector Regression)的預測建模方法,併應用Grid-Search 參數尋優法優化模型參數,基于真實QAR數據建立 SVR預測模型,併從平方相關繫數和平均絕對百分誤差兩箇不同指標與 BP 神經網絡模型的預測結果進行比較,比較結果錶明:SVR預測模型的預測結果精度高。
침대비궤순항단연유소모량예측문제,제출일충기우지지향량회귀궤(SVR :Support Vector Regression)적예측건모방법,병응용Grid-Search 삼수심우법우화모형삼수,기우진실QAR수거건립 SVR예측모형,병종평방상관계수화평균절대백분오차량개불동지표여 BP 신경망락모형적예측결과진행비교,비교결과표명:SVR예측모형적예측결과정도고。
A support vector regression ( SVR) predicting method was proposed for predicting the fuel consumption in cruise phase. In this proposed method, the SVR parameters were optimized with Grid-Search algorithm. This paper established a SVR model based on QAR data. A comparison was made with the BPNN model from the two different indicators of squared correlation coefficient and mean absolute percent error, the results show that the accuracy of SVR model is better than that of BPNN model.