基于人工神经网络的风洞动态测压试验结果预测  

Prediction of Dynamic Pressure Measurement Results in Wind Tunnel Based on Artificial Neural Network

作　　者：刘苗鑫 雷杰[1] 原正庭[1] LIU Miaoxin;LEI Jie;YUAN Zhengting

机构地区：[1]中国飞行试验研究院,西安710000

出　　处：《科技创新与应用》2023年第36期70-73,共4页Technology Innovation and Application

摘　　要：通过机翼的风洞动态测压试验,可以得到机翼的动态性能参数并了解动态失速的特点,但在实际风洞动态测压试验中,由于资源和现场条件所限仅能够得到典型试验状态下的数据。为在有限的试验资源下得到更加全面的压力数据,该文采用BP人工神经网络,利用180组试验数据建立试验工况与压力值之间的预测模型,并采用18组试验数据用来对预测结果进行验证。将预测结果与实际结果进行对比发现,BP神经网络可以有效地对压力值及波动值进行预测,结果误差小。因此在有限的试验状态下,BP神经网络可以对大的分离流动进行较为准确的预测,能够符合实际的流动特性,可以为风洞测压试验提供全面、可靠的预测数据。Through the dynamic pressure measurement test of the wing in the wind tunnel,the dynamic performance parameters of the wing can be obtained and the characteristics of dynamic stall can be understood,but in the actual dynamic pressure measurement test in the wind tunnel,due to the limitation of resources and field conditions,the data in the typical test state can only be obtained.In order to obtain more comprehensive pressure data under the limited test resources,this paper uses BP artificial neural network to establish a prediction model between test conditions and pressure values,and 18 groups of test data are used to verify the prediction results.Comparing the predicted results with the actual results,it is found that the BP neural network can effectively predict the pressure value and fluctuation value,and the error of the result is small.Therefore,in the limited test state,the BP neural network can accurately predict the large separated flow,can accord with the actual flow characteristics,and can provide comprehensive and reliable prediction data for the wind tunnel pressure test.

关 键 词：风洞试验 动态测压 人工神经网络 预测 流动特性 

分 类 号：V211.746[航空宇航科学与技术—航空宇航推进理论与工程]