论文部分内容阅读
风洞试验是空气动力学研究的一种重要手段。而风洞试验本身就是一项复杂的系统工程,试验结果受诸多因素的影响,不同状态、不同条件下风洞试验结果差别可能非常显著,一次风洞型号试验任务将会产生大量的试验结果。如何对风洞试验产生的海量数据结果有效、合理、快速的加以分析与利用,在型号设计中具有重要的现实意义。本项目将以一种新兴的机器学习算法——支持向量机(Support Vector Machine)为工具,建立以有限样本数据为支持的特征分析工具,对复杂数据进行有效的分类识别和数据回归分析,实现基于样本特征的非线性数据的自动判别与拓展,可以使数据分析人员在尽可能短的周期内对飞行器的气动性能进行快速的评估。本文将支持向量机算法引入到风洞试验数据分析领域,从而加快型号的研制进度,节约研制经费,为飞行器的气动性能研究提供有力的分析工具。
Wind tunnel test is an important means of aerodynamics research. The wind tunnel test itself is a complicated systematic project. The test results are affected by many factors. The differences between wind tunnel test results under different conditions and under different conditions may be very significant. A wind tunnel test task will produce a large number of test results. How to effectively, reasonably and quickly analyze the mass data produced by wind tunnel tests is of great practical significance in the design of models. This project will build a feature analysis tool based on Support Vector Machine (SVM), a new machine learning algorithm supported by limited sample data, to carry out effective classification and data regression analysis of complex data to achieve The automatic identification and expansion of nonlinear data based on sample features enables data analysts to quickly evaluate the aerodynamic performance of aircraft in the shortest possible period. In this paper, the support vector machine (SVM) algorithm is introduced into the field of wind tunnel test data analysis, so as to speed up the development of model and economize the research and development expenses, and provide powerful analytical tools for the aerodynamic performance research of aircraft.