Computational protocol: Uncertainly Analysis of Two Types of Humidity Sensors by a Humidity Generator with a Divided-Flow System

Similar protocols

Protocol publication

[…] The establishment of the calibration equation builds the relationship between the standard humidity values and the reading values of sensors. There are two statistical ways to establish the calibration equations: the classical method and inverse method. Because the predicted uncertainty of the inverse method is significantly lower than that of the classical method [], the inverse method was used in this study. (9)yi=f(xi) where yi is the standard RH values and xi is the reading values of sensors.If yi and xi have a linear relationship, then (10)y=b0+b1×x where b0 and b1 are constants.If yi and xi have a polynomial relationship, then (11)y=c0+c1×x+c2×x2+…+cn×xn where c0, c1, …,cn are constants.The calibration equation was analyzed by using of SigmaPlot v12.2 (SPSS Inc., Chicago, IL, USA). The quantitative criteria to compare the fitting performance of calibration equations were the coefficient of determination R2 and standard of the estimated value s. The suitable order of the polynomial equation was tested by the t-value of the parameter values of each equation. The qualitative criteria are the residual plots. For a suitable equation, the data distribution of residual values should be a horizontal band centered on zero. […]

Pipeline specifications

Software tools SigmaPlot, SPSS
Application Miscellaneous