Improved Regression Analysis of Temperature-Dependent Strain-Gage Balance Calibration Data

摘要

An improved approach is discussed that may be used to directly include first and second order temperature effects in the load prediction algorithm of a wind tunnel strain-gage balance. The improved approach was designed for the Iterative Method that fits strain-gage outputs as a function of calibration loads and uses a load iteration scheme during the wind tunnel test to predict loads from measured gage outputs. The improved approach assumes that the strain-gage balance is at a constant uniform temperature when it is calibrated and used. First, the method introduces a new independent variable for the regression analysis of the balance calibration data. The new variable is defined as the difference between the uniform temperature of the balance and a global reference temperature. This reference temperature should be the primary calibration temperature of the balance so that, if needed, a tare load iteration can be performed. Then, two temperature-dependent terms are included in the regression models of the gage outputs. They are the temperature difference itself and the square of the temperature difference. Simulated temperature-dependent data obtained from Triumph Aerospace's 2013 calibration of NASA's ARC-30K five-component semi-span balance is used to illustrate the application of the improved approach.