摘要
针对燃气轮机在数学建模过程中难以确定其部件特性方程最优拟合次数的问题,提出了采用统计学检验指标确定特性方程最优拟合次数的方法。最优拟合次数的选择,使得拟合方程既防止由于拟合次数过高,将测量数据中的噪声也纳入拟合模型,又避免了拟合次数太低所引起的结果粗糙。结果表明:综合运用拟合检验指标,避免了由经验选择拟合次数的局限性,并为特性方程最优拟合次数的选取提供了理论依据。
Aiming at determining gas turbine components characteristics equation＇s optimal fitting order times in the process of establishing the mathematical model of gas turbine, the statistics evaluation indexes was proposed to evaluate the optimal fitting order times. The approach was adopted for avoiding the noise of measurement data, which was brought into fitting model for the fitting order times too high, and it caused imprecise result for fitting order times too low. The results indicate that the limitations of the experience selecting fitting order times can be avoided by using synthetically fitting evaluation indexs. It is provided a theoretical foundation for selecting optimal fitting order times of characteristic equation.
出处
《航空发动机》
2011年第5期7-10,共4页
Aeroengine
基金
国防预研基金(40101020102)
"十一五"国防预研基金(4010102010504)资助
关键词
燃气轮机
特性曲线
曲线拟合
特性方程
检验指标
最优拟合次数
gas turbine
characteristic curve
curve fitting
characteristic equation
fitting evaluation indexs
optimal fitting order times
