Simulation Modeling in Farm Machinery and Power

Engineering
Dimensional Analysis

for Ph.D. in Agricultural Mechanization

Reference:
H.L. Langhaar, Dimensional Analysis and Theory of Models-Chapters 1 to 4.
Simulation Modeling in Farm
Machinery and Power
Engineering
For Ph.D. in Agricultural Mechanization
 Modeling Methodologies

• Dimensional Analysis
It helps in the formulation of mathematical relations between different variables in a large scale
phenomenon where a systematic derivation of an analytical expression may not be possible.
Examples: Hydraulic structures, Rivers and Harbors
Hydraulic machines, Airplanes, Structures, Ships.
Application Areas:
Fluid motion
Heat transfer
Elasticity problems
Electromagnetic fields
 Preliminaries

Quantity Units Dimensions

Mass kg, g , lb [M]
Length meter, ft, inch [L]
Time seconds [T]
oC  
Temperature , oK  

Angle radians non-dimensional

Different variables are expressed as products of dimensions with powers.

Force: [M][L][T]-2
Acceleration: [L][T]-2
Velocity: [L][T]-1
Kinetic energy: [M][L]2[T]-2
Potential energy: [M][L]2[T]-2
Work done : [M][L]2[T]-2
Viscosity: [M][L]-1[T]-1
Mass density: [M][L]-3

• Transformation of units of measurement- numerical value of the quantity changes with change of
units of measurement. However, the dimensions remain same.

1 ft / s   112 2.54  m / s  0.3048m / s

 

 100 
[L][T]-1
=1.09728 kmph [L][T]-1

T  2 L T 1.107
g g=32.2 ft/s2 for different numerical values of g , L

= 9.81 m/s2 2 L

= 981 cm/s2  0.2 L

 Steps in Dimensional Analysis

• A philosophical insight into the natural phenomenon is essentially required.

Step 1. Identification of independent variables which affect the phenomenon.

We should have clear idea about the theory of the problem to decide about the variables that really
influence the phenomenon.
The governing differential equations if available, they will show directly which variables are
significant.
Step 2. Formation of a complete set of dimensionless products of the variables.
Step 3. Formation of a set of homogeneous linear algebraic equations from the exponents of the
set of dimensionless products.
Step 4. Solution of the homogeneous equations to obtain the exponents of the dimensionless variables.

For each set of exponents, the products of non-dimensional terms is equated with the relation as
1, 2 , 3,....etc.
The phenomenon can be explained through selection of appropriate equation of i
 Buckingham’s Theorem
• The total number of dimensionless products in a complete set is equal to the
total number of variables minus the number of fundamental dimensions in the
problem.
• Leibniz suggested a small change considering the system of homogeneous linear equations
involved as:
The number of dimensionless products in a complete set is equal to the total
number of variables minus the rank of their dimensional matrix.
Rank of a matrix : The highest order of the nonzero determinant ‘r’ that a matrix
contains and all other determinants of higher order are zero.
There are 6 variables, the rank of the matrix is 3
0 0 k k k 0 The number of non-dimensional groups:
3 4 5 1 1 0
k k k 3k k k 3 1 1  3 6-3=3
1 2 3 4 5 6
k 0 2k 0 k 2k 0 1 2
1 3 5 6
• The variables are: V, L, F ,  ,  , g
• The non-dimensional numbers are : Pressure coefficient, P  F / V 2L2
• Reynold’s number, R  VL / 
• Froude number, F  V 2 / Lg

• Linearly dependent matrix- If a matrix has two rows; the first row is multiplied
with a number; the second row is multiplied by another number. The resulting
rows are added together to get the third row. Then the rows of the matrix
formed are called a Linearly dependent.

• If the rows of a matrix are linearly dependent, then the rank of the matrix is
less than the number of rows.
 Formation of Linear Equations

• Suppose the variables are: V, L, F,  ,  , g

• The non-dimensional equation,   V 1 L 2 F 3  4  5 g
k k k k k k6

• Writing the above in dimensional notation:   = [LT-1]k1 [L]k2 [MLT-2]k3 [ML-3] k4 [ML-1T-1]k5[LT-2]k6

• [ ]  [M ]k k k [ L]k k k 3k k k [T ]k 2k k 2k

3 4 5 1 2 3 4 5 6 1 3 5 6

• If  is required to be dimensionless, then equating the exponents on either side, the exponents
of M,L, T must be equal to zero.
k3  k4  k5  0
k1k2 k3 3k4 k5 k6 0
k12k3k5 2k6 0

• Any solution of these equations is a set of exponents in a dimensionless product.

• 3 equations, 6 unknowns: this system of equations is undetermined. It may have infinite solutions.
• By assigning any values to k1, k2, k3, the equations may be solved for the remaining three
unknowns.
• First set: assume k1=-2, k2=-2, k3=1; we get k4=-1, k5=0, k6=0
2  2
we have 1  V L F 
1
This is the Pressure coefficient, P  F / V 2 L2 

• Second set: assume k1=1, k2=1, k3=0, we have k4=1, k5=-1, k6=0
we have  2  V 1L1 1 This is the Reynold’s number, R  VL / 

• Third set: assume k1=2, k2=-1, k3=0, we have k4=0, k5=0, k6=-1
2
we have  3  V / Lg This is Froude’s number, F  V / Lg
2
•

• Fourth set: assume k1=10, k2=-5, k3=8, we have k4=5, k5=-10, k6=-5
We have   V L F   g which is the product of P, R, F:   P8 R16 F 5
10 5 8 8 16 5
•
• In conclusion, we can say that if we want to form a dimensionless product of ‘n’ variables, the
exponents k1, k2, k3, ….kn of these variables must for a set of linear algebraic equations.
a1k1  a2k2  .....  ankn  0 (a)
b1k1  b2k 2 ......  bnkn  0

ai , bi (i=1, …n) are the coefficients of the rows in the dimensional matrix. The equations are said to
be homogeneous.
• Disregarding the trivial solution ko  0 , equations (a) possess exactly (n-r) linearly independent
solutions in which ‘ r ‘ is the rank of the matrix of the coefficients in equations (a).
• The set of (n-r) linearly independent solutions is called a fundamental system of solutions.
• Any solution is a linear combination of the solutions in any fundamental system.

• Another example for construction of a set of dimensionless products is illustrated in the following.
 Example for Construction of a Set of Dimensionless Products

• P, Q, R, S, T, U, V are seven variables. M, L, T are the fundamental dimensions. The

dimensional matrix is given below. It is required to compute the dimensionless products of
the variables.
1 2 3 4 5 6 7
P Q R S T U V
M 2 -1 3 0 0 -2 1
L 1 0 -1 0 2 1 2
T 0 1 0 3 1 -1 2
0 2 1
• The first step is to compute the rank of the matrix 2 1 2 1
• We have a nonzero determinant of the order of matrix 3 , the rank is: 3 1 1 2
• We have a set of linear homogeneous equations as below
2k1  k2  3k3  2k6  k7  0
k1  k3  2k5  k 6 2k7  0
k2  3k4  k5  k6  2k7  0
 Solution of Equations

• By assigning k1=1, k2=k3=k4=0, we get k5=-11, k6=5, k7=8

• Taking k2=1, k1=k3=k4=0, we get k5=9, k6=-4, k7=-7
• Taking k3=1, k1=k2=k4=0, we get k5=-9, k6=5, k7=7
• Taking k4=1, k1=k2=k3=0, we get k5=15, k6=-6, k7=-12

k1 k2 k3 k4 k5 k6 k7
P Q R S T U V
• 1 1 0 0 0 -11 5 8 we can note that 5, 6, 7 columns are the
•  2 0 1 0 0 9 -4 -7 solutions of the equations.
 3 0 0 1 0 -9 5 7
• 1-4 columns are all zero except for those in the
 4 0 0 0 1 15 -6 -12

• principal diagonal.
1  PT 11U 5V 8
 2  QT 9U 4V 7
• Now we can write the products of non-dimensional variables as :  3  RT 9U 5V 7
 4  ST 15U 6V 12
 PROBLEMS

1. Calculate a complete set of dimensionless products of the following variables Volume Q,

acceleration A, velocity V, power P, momentum M, angular velocity N.

2. The dimensions of the five variables are given by the following matrix.

y x1 x2 x3 x4
 
M 1 1 -1 2 0
L 3 -2 4 1 -1
T 2 -1 3 3 -1

Derive a dimensionally homogeneous product of the following type as below.

k k k k
y  x11 x2 2 x3 3 x44