Location via proxy:   [ UP ]  
[Report a bug]   [Manage cookies]                
Scribd Logo
Upload

Welcome to Scribd!

  • 30 views

    08 Curve Fit Mechanical Measurements

    Uploaded by

    Fidha
    The document discusses calibration techniques for mechanical measurements. It describes static calibration and least squares analysis to determine the best fit line for calibration data. It also discusses hysteresis error, coefficient of determination, uncertainty analysis, and elements of instrument error.

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd

    08 Curve Fit Mechanical Measurements

    Uploaded by

    Fidha
    30 views10 pages
    The document discusses calibration techniques for mechanical measurements. It describes static calibration and least squares analysis to determine the best fit line for calibration data. It also discusses hysteresis error, coefficient of determination, uncertainty analysis, and elements of instrument error.

    Original Description:

    mechanical measurements

    Copyright

    © © All Rights Reserved

    Available Formats

    PDF, TXT or read online from Scribd

    Did you find this document useful?

    Is this content inappropriate?

    The document discusses calibration techniques for mechanical measurements. It describes static calibration and least squares analysis to determine the best fit line for calibration data. It also discusses hysteresis error, coefficient of determination, uncertainty analysis, and elements of instrument error.

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd
    Download as pdf or txt
    30 views10 pages

    08 Curve Fit Mechanical Measurements

    Uploaded by

    Fidha
    The document discusses calibration techniques for mechanical measurements. It describes static calibration and least squares analysis to determine the best fit line for calibration data. It also discusses hysteresis error, coefficient of determination, uncertainty analysis, and elements of instrument error.

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd
    Download as pdf or txt
    of 10

    Mechanical

    Measurements
    Static Calibration
    Actual Calibration Practice

    • A given true value is not repeated several times.


    • True value is varied over a range – measured value, therefore, varies over a range.

    Least Squares – Determine the fit

    2
    Least Squares Equation – Straight Line
    𝑦 = 𝑚𝑥 + 𝐶 is the fit and (𝑥𝑘 , 𝑦𝑘 ) are the experimental points (𝑘 = 1 to 𝑁)

    &

    Error = 𝐸 = 2 𝑚𝑥# + 𝐶 − 𝑦# "

    #$%

    𝜕𝐸 𝜕𝐸
    Suffix 𝑘 is dropped from 𝑥# and 𝑦# for easiness = = 0 to minimize this error
    𝜕𝑚 𝜕𝐶

    𝜕𝐸
    = 0 ⇒ 2 2 𝑚𝑥 + 𝐶 − 𝑦 𝑥 = 0
    𝜕𝑚

    𝜕𝐸
    = 0 ⇒ 2 2 𝑚𝑥 + 𝐶 − 𝑦 1 = 0
    𝜕𝐶

    𝑚 2 𝑥 " + 𝐶 2 𝑥 − 2 𝑥𝑦 = 0

    𝑚 2 𝑥 + 𝐶𝑁 − 2 𝑦 = 0
    3
    Least Squares Equation – Straight Line
    𝑦 = 𝑚𝑥 + 𝐶 is the fit and (𝑥𝑘 , 𝑦𝑘 ) are the experimental points (𝑘 = 1 to 𝑁)

    𝑁 ∑ 𝑥𝑦 − ∑ 𝑥 ∑ 𝑦
    𝑚=
    𝑁 ∑ 𝑥" − ∑ 𝑥 "

    ∑𝑦 − 𝑚∑𝑥
    𝐶=
    𝑁
    &
    1
    Standard deviation of data 𝜎 " = 2 𝑚𝑥# + 𝐶 − 𝑦# "
    N−1
    #$%

    This standard deviation would possibly be the same if one repeated the “actual point – measured point” calibration over and over again.

    Assume same standard deviation would occur even if a particular reading were to be repeated over and over again.

    Use 99.7% limits (3s); \ Accurate value = 𝑚𝑥 + 𝐶 ± 3𝜎


    4
    Esther found how many hours of sunshine vs how many ice creams were sold at the shop from Monday to Friday:

    "x" "y"
    Hours of Sunshine Ice Creams Sold Let us find the best 𝑚 (slope) and 𝐶 (intercept)
    that suits the data with 𝑦 = 𝑚𝑥 + 𝐶
    2 4
    3 5
    5 7
    7 10 𝑁 ∑ 𝑥𝑦 − ∑ 𝑥 ∑ 𝑦
    𝑚= = 1.5183
    𝑁 ∑ 𝑥" − ∑ 𝑥 "
    9 15
    ∑𝑦 − 𝑚∑𝑥
    𝐶= = 0.3049
    x y x2 xy 𝑁

    2 4 4 8 𝑦 = 1.518𝑥 + 0.305
    3 5 9 15
    1
    5 7 25 35 𝜎" = 2 𝑦'() − 𝑦# " = 0.7973
    N−1
    7 10 49 70
    9 15 81 135
    Σx: 26 Σy: 41 Σx2: 168 Σxy: 263

    5
    https://www.mathsisfun.com/data/least-squares-regression.html
    Esther found how many hours of sunshine vs how many ice creams were sold at the shop from Monday to Friday:

    𝑦 = 1.518𝑥 + 0.305 ± 0.548

    Esther hears that the weather forecast which says “we expect 8 hours of sun
    tomorrow”, so she uses the above equation to estimate that she will see:

    𝑦 = 1.518×8 + 0.305 ± 0.548 = 12.449 ± 0.548

    Esther makes fresh waffle cone mixture for 14 ice creams just in case.

    Least squares is sensitive to outliers 6


    https://www.mathsisfun.com/data/least-squares-regression.html
    Coefficient of Determination, 𝑹𝟐

    𝑦 = 1.518𝑥 + 0.305

    Total sum of squares, 𝑆𝑆)*)+, = 2 𝑦'() − 𝑦Q# "

    Sum of squares, 𝑆𝑆-./(01+,/ = 2 𝑦'() − 𝑦# "


    = 2 𝑒#"

    "
    𝑆𝑆-./(01+,/
    𝑅 =1− = 0.9595
    𝑆𝑆)*)+,

    7
    Summary of the Uncertainty

    Uncertainty = Bias ! + Precision !


    8
    R.S. Figliola and D.E. Beasley, Theory and Design for Mechanical Measurements, Fifth Edition, John Wiley and Sons, 2011.
    Sequential Test and Hysteresis

    • A sequential test applies a sequential variation


    in the input value over the desired input range Hysteresis
    (upscale or downscale direction)
    Downscale
    • Hysteresis error is the difference in the values

    Output value
    measured in upscale and downscale tests

    𝑒U = 𝑦VWXYZ[\ − 𝑦]^_`XYZ[\
    Upscale
    𝑒U!"#
    %𝑒U!"# = ×100 Input value
    𝑟a 9
    Elements of Instrument Error

    Overall Instrument Error

    𝒆𝒄 = 𝒆𝟐𝒉 + 𝒆𝟐𝒍 + 𝒆𝟐𝒔 + 𝒆𝟐𝒛 + 𝒆𝟐𝒓

    R.S. Figliola and D.E. Beasley, Theory and Design for


    Mechanical Measurements, Fifth Edition, John Wiley and
    Sons, 2011.
    10

    You might also like