红外与激光工程
紅外與激光工程
홍외여격광공정
INFRARED AND LASER ENGINEERING
2015年
8期
2334-2338
,共5页
反向蒙特卡洛%航空发动机%红外辐射%效率
反嚮矇特卡洛%航空髮動機%紅外輻射%效率
반향몽특잡락%항공발동궤%홍외복사%효솔
BMC%aero-engine%infrared radiation%efficiency
反向蒙特卡罗(BMC)法的计算精度高、适应性强,是计算目标红外辐射特征的常用方法。基于CFD 和IR 综合计算分析的方法,研究了射线步长、射线数、射线携带能量方式等因素对BMC法计算航空发动机红外辐射强度的效率的影响。结果表明:射线步长与喷管出口当量直径比在0.05~0.1的范围内,射线数达到105量级时的计算效率较高;采用射线携带多个波段能量的方法,计算时间可缩短数十倍;通过MPI平台实现了并行计算,进一步缩短BMC法的计算时间,但并行效率随着计算核心的增加而下降。在保证计算精度的前提下,通过合理地选择参数、算法改进和并行计算,提高了计算效率。
反嚮矇特卡囉(BMC)法的計算精度高、適應性彊,是計算目標紅外輻射特徵的常用方法。基于CFD 和IR 綜閤計算分析的方法,研究瞭射線步長、射線數、射線攜帶能量方式等因素對BMC法計算航空髮動機紅外輻射彊度的效率的影響。結果錶明:射線步長與噴管齣口噹量直徑比在0.05~0.1的範圍內,射線數達到105量級時的計算效率較高;採用射線攜帶多箇波段能量的方法,計算時間可縮短數十倍;通過MPI平檯實現瞭併行計算,進一步縮短BMC法的計算時間,但併行效率隨著計算覈心的增加而下降。在保證計算精度的前提下,通過閤理地選擇參數、算法改進和併行計算,提高瞭計算效率。
반향몽특잡라(BMC)법적계산정도고、괄응성강,시계산목표홍외복사특정적상용방법。기우CFD 화IR 종합계산분석적방법,연구료사선보장、사선수、사선휴대능량방식등인소대BMC법계산항공발동궤홍외복사강도적효솔적영향。결과표명:사선보장여분관출구당량직경비재0.05~0.1적범위내,사선수체도105량급시적계산효솔교고;채용사선휴대다개파단능량적방법,계산시간가축단수십배;통과MPI평태실현료병행계산,진일보축단BMC법적계산시간,단병행효솔수착계산핵심적증가이하강。재보증계산정도적전제하,통과합리지선택삼수、산법개진화병행계산,제고료계산효솔。
The Backward Monte Carlo (BMC) method is a common method for calculating the target’s infrared radiation signature because of its high accuracy and strong adaptability. Based on general CFD/IR numerical calculation method, the impact of ray discrete step size, ray number, energy carrying method on aero-engine infrared radiation intensity computation efficiency was investigated by using BMC method. The results show that the computation efficiency is high when the ratio of ray discrete step size to equivalent diameter of nozzle exit is between 0.05 to 0.1 and ray’s number is more than 105; the computation time can be shortened by dozens of times through ray carrying multiple bands energy method; the computation time of BMC method can be reduced further by parallel computation with MPI platform, but the parallel efficiently will be decreased with increasing the computation core number. In the premise of ensuring the accuracy of the results, the efficiency of infrared radiation computation was improved by reasonable selection of parameters, algorithm improvement and parallel computation.