04-Jan-23

INDUSTRIAL
INSTRUMENTATION
Lecturer: Nguyen Duc Hoang
Department of Control & Automation
Faculty of Electrical & Electronics Engineering
Ho Chi Minh City University of Technology
Email: [email protected]

Chapter 3: Measurement Uncertainty

• Introduction

• Sources of Systematic Error

• Reduction of Systematic Errors

• Sources of Random Errors

• Statistical Analysis of Measurements Subject to Random Errors

• Aggregation of Measurement System Errors

1
 04-Jan-23

Introduction

• The existence of measurement uncertainty means that we

would be entirely wrong to assume that the output of a
measuring instrument or larger measurement system gives the
exact value of the measured quantity.
• Measurement errors are impossible to avoid, although we can
minimize their magnitude by good measurement system
design accompanied by appropriate analysis and processing of
measurement data.

Introduction

• Errors can be divided into two groups, known as systematic

errors and random errors.
• Systematic errors describe errors in the output readings of a
measurement system that are consistently on one side of the
correct reading, that is, either all errors are positive or are all
negative.
• Random errors are perturbations of the measurement either side
of the true value caused by random and unpredictable effects.
4

2
 04-Jan-23

Sources of Systematic Error

• Effect of environmental disturbances, often called modifying inputs.

• Disturbance of the measured system by the act of measurement.

• Changes in characteristics due to wear in instrument components

over a period of time.

• Resistance of connecting leads.

Sources of Systematic Error

• Example :

R1 = 400Ω R2 = 600Ω

R3 = 1000Ω R4 = 500Ω

R5 = 1000Ω Rm = 9500Ω

What is the measurement error caused by the resistance of the measuring

instrument?

3
 04-Jan-23

Reduction of Systematic Errors

• Calibration
• …

Sources of Random Errors

• Measurements taken by human observation of an
analogue meter, especially where this involves
interpolation between scale points.
• Electrical noise.

• Random environmental changes, for example,

sudden draught of air.

4
 04-Jan-23

Statistical Analysis of Measurements

Subject to Random Errors

Mean and Median Values

• For any set of n measurements x1, x2,…,xn of a constant
quantity, the most likely true value is the mean given by

𝑥 + 𝑥 + ⋯+ 𝑥
𝑥 =
𝑛

𝑥 =𝑥 /

10

5
 04-Jan-23

Mean and Median Values

Example: the length of a steel bar is measured as follows:

Set A:
398 420 394 416 404 408 400 420 396 413 430

Set B:
409 406 402 407 405 404 407 404 407 407 408

11

Standard Deviation and Variance

• Examining how measurement values are distributed about the mean value.
• Variance
𝑑 +𝑑 +⋯+𝑑
𝑉 =
𝑛
• Standard deviation
𝑑 + 𝑑 + ⋯+ 𝑑
σ= 𝑉 =
𝑛
Where: 𝑑 = 𝑥 − 𝑥
Note: these formal definitions for the variance and standard deviation of
data are made with respect to an infinite population of data values .
12

6
 04-Jan-23

Standard Deviation and Variance

• In all practical situations, we can only have a finite set of measurements.
• Applying the Bessel correction factor (n/n-1) to the formula for Vs

𝑛 𝑑 + 𝑑 + ⋯+ 𝑑
𝑉= 𝑉 =
𝑛−1 𝑛−1
• Standard deviation
𝑑 + 𝑑 + ⋯+ 𝑑
σ= 𝑉=
𝑛−1
Where: Vs is the variance of the finite set of measurements
V is the variance of the infinite population of measurements.
13

Standard Deviation and Variance

Example: Calculate σ and V for measurement sets A, B.

Set A:
398 420 394 416 404 408 400 420 396 413 430

Set B:
409 406 402 407 405 404 407 404 407 407 408

14

7
 04-Jan-23

Gaussian (Normal) Distribution

• Measurement sets that only contain random errors usually conform to a distribution with a
particular shape that is called Gaussian
• A Gaussian curve:
1 /
𝐹 𝑥 = 𝑒
𝜎 2𝜋

• If the standard deviation is used as a unit of error, the Gaussian curve can be used to determine
the probability that the deviation in any particular measurement in a Gaussian data set is greater
than a certain value.
• The probability that the error lies in a band between error levels D1 and D2 can be expressed as:

1 /
𝑃 𝐷 ≤𝐷≤𝐷 = 𝑒 𝑑𝐷
𝜎 2𝜋
Where: 𝐷 = 𝑥 − 𝑚

15

Standard Gaussian curve (z distribution)

• Substitute 𝑧 = 𝐷/𝜎

1 /
𝑃 𝐷 ≤𝐷≤𝐷 =𝑃 𝑧 ≤𝑧≤𝑧 = 𝑒 𝑑𝑧
𝜎 2𝜋

16

8
 04-Jan-23

Standard Gaussian Tables (z Distribution)

17

Standard Gaussian Tables (z Distribution)

• Example : How many measurements in a data set subject to random errors lie outside
deviation boundaries of + σ and - σ , that is, how many measurements have a deviation
greater than |σ|?

• P(E< -σ or E> σ) = P(E< -σ)+P(E> σ)

• P(E<-σ)=F(-1)=1-F(1)=1-0.8413=0.1587
P(E> σ)=1-P(E< σ)=1-F(1)=1-0.8413=0.1587
• P(E< -σ)+P(E> σ)=0.1587+0.1587 = 0.3174  32%

18

9
 04-Jan-23

Standard Gaussian Tables (z

Distribution)
• Example: The statistics of a well-defined varying voltage signal are given by
𝑥 = 8.5𝑉, 𝜎 = 2.25𝑉

If a single measurement of the voltage signal is made, determine the probability

that the measured value indicated will be between 10.0 and 11.5 V.

19

Standard Error of the Mean

• The standard deviation of mean values of a series of finite sets of

measurements relative to the true mean (the mean of the infinite
population that the finite set of measurements is drawn from) is defined
as the standard error of the mean, .

𝛼 = 𝜎/ 𝑛

• If we use the mean value of a finite set of measurements to predict the

true value of the measured quantity, what is the likely error in this
prediction?

20

10
 04-Jan-23

Standard Error of the Mean

• The measurement value obtained by calculating the mean of a set of n
measurements, x1,x2,…,xn, can be expressed as:

𝑥= 𝑥 ±𝛼
with 68% certainty that the magnitude of the error does not exceed ||.

Question:
𝑥= 𝑥 ± 2𝛼
?% certainty that the magnitude of the error does not exceed |2|.

21

Standard Error of the Mean

• Example:compute the measurement value with 68% certainty that the
magnitude of the error does not exceed || and with 95.4% certainty that
the magnitude of the error does not exceed |2|

409 406 402 407 405 404 407 404 407 407 408 410

406 405 408 406 409 406 405 409 406 407 406

22

11
 04-Jan-23

Estimation of Random Error in a Single

Measurement
• In many situations where measurements are subject to random errors, it is not practical
to take repeated measurements and find the average value.
• Also, the averaging process becomes invalid if the measured quantity does not remain
at a constant value, as is usually the case when process variables are being measured.
• Thus, if only one measurement can be made, need to estimate the likely magnitude of
error.
• Normal approach to this is to calculate the error within 95% confidence limits
(deviation of ±1.96σ)
• The maximum likely error in a single measurement can be expressed as:

𝑬𝒓𝒓𝒐𝒓 = ±𝟏. 𝟗𝟔 𝝈 + 𝜶

23

Estimation of Random Error in a Single

Measurement

• Example: Suppose that a standard mass is measured 30 times with the same
instrument to create a reference data set, and the calculated values of σ and 
are σ = 0.46 and  = 0.08. If the instrument is then used to measure an
unknown mass and the reading is 105.6 kg, how should the mass value be
expressed?

24

12
 04-Jan-23

Aggregation of Measurement System

Errors
• Combined Effect of Systematic and Random Errors

• Aggregation of Errors from Separate Measurement

System Components

25

Đo điện trở sử dụng Vôn kế và Ampe kế.

Biết Vôn kế có độ chính xác 1%, Ampe kế có
độ chính xác 2%.
Biểu diễn giá trị điện trở đo được nếu
Vôn kế chỉ 5V, Ampe kế chỉ 1A.

26

13
 04-Jan-23

Biết một thiết bị đo chiều dài có độ chính xác 1%

được sử dụng để đo kích thước của một vật thể hình
khối.
Hãy biểu diễn khối lượng đo được nếu kích thước vật
thể đo được lần lượt là 1m, 2m, 3m.
Tỉ trọng vật 100g/cm3.

27

28

14
 04-Jan-23

29

BT4
BT1

BT2

30

15