Advanced Vehicle Dynamics (AUE 8306)

Introduction to Dynamics
of MBS
Mechanical Systems and Vehicle Engineering Program
Adama Science and Technology University
2018-19 A.Y Semester II
 Multibody System
• A system is defined to be a collection of entities that act and interact together
towards the accomplishment of some logical end.
• Multibody system is defined to be a collection of subsystems called bodies,
components, or substructures.
• A multibody dynamic (MBD) system is made of two or more rigid or flexible body,
joined together through kinematic pairs, which allow relative movement between
adjacent bodies
• The study of MBD is the analysis of how mechanism systems move under the
influence of forces
• The motion of bodies is described by their kinematic behavior. The dynamic behavior
results from the equilibrium of applied forces and the rate of change of momentum
• Many mechanical and structural systems such as vehicles, space structures, robotics,
mechanisms and aircraft consist of interconnected components that undergo large
translational and rotational displacements.
 Dynamic System
• Definition
• A physical system undergoing a time varying interchange or dissipation of energy among or within its
elementary storage or dissipative device.
• Composition of dynamic system
Mathematical
Input Output Interpretation
Model

Braking distance Response

Acceleration, Braking Laws of Mechanics, Mass of
Velocity & acceleration Effect
Steering, Road conditions, Vehicle, Forces & Moments,
Safety and comfort
Aerodynamic inputs Stiffness & damping

Differential Method X
Example Applying laws of nature
Force F(t) Laplace method
Vibration Newton-Euler Equ. State space
Xሶ
Excitations
System ODE or PDE Numerical Xሷ
Dynamic System Classification

System
Experiment with Experiment with a
The actual system model of a system
Physical
Mathematical
Model
Model

Analytical Simulation
Solution

Dynamic Lumped Constant

Deterministic Linear Discrete
System Parameter Coefficient

Distributed Variable
Stochastic Non-Linear Continuous
Parameter Coefficient
Approaches to Solve MBD System
• Multibody system analysis has been developed based on the achievements of classical mechanics,
which is generally divided into two branches
• Vector Mechanics (Direct Approach to Dynamics)
• Consider the individual particles on the system
• Connecting force and motion on vectors
• Force and momentum are considered as the primary parameters in differential equations of motion.
• can be directly derived by employing the approach of Newton and Euler 2nd law of motion
• Analytical (Lagrangean) Mechanics (Indirect Approach to Dynamics)
• Treats the system as a whole.
• Deals with scalar quantities such as K.E & P.E
• Introducing the concept of virtual work.
• In order to make the concept mathematically consistent, Lagrange utilized the results of D’Alembert, making
possible the systematic analysis of a constrained particle system.
 Elements of MBD system
• Mass Spring Damper
a
X1 x2
F V1 V2
fc fc

1 • stores potential energy (𝑓𝑘 = −k(x2 − X1) • Is a mechanical element that

• Stores kinetic energy (K.E=2mv2 )
dissipates energy
• A force fk to generate a deflection in spring
• Stored K.E in a mass m is proportional to is proportional to relative displacement of • Element relating force to
the square of its velocity, v2 its ends (element relating force to velocities
• The velocity v = 𝑥ሶ may be a function of displacement) • is considered to have neither
position & time • A spring is characterized by its stiffness k. mass nor elasticity

• The required force fm to move a mass m is • If k is constant then, the value of stored P.E • 3 form of Damping : Viscous,
is equal to the work done by the spring Coulomb & Hysteretic
proportional to its acceleration a=𝑥ሷ
(element relating force to acceleration) force fk f =Cv (fs)max=µsN if Vrel=0
𝑥2 c rel
1
𝑉 = − න 𝑓𝑘 𝑑𝑥 = න 𝑘𝑥𝑑𝑥 = 𝑘𝑥 2 fk=µkN, if Vrel>0
2
 Elements of MBD System
• Example
x x
k

kx
F t − cxሶ − kx = mxሷ
F(t)
m F(t) m
C𝑥ሶ
C mxሷ + cx+
ሶ kx=F t

C𝒙ሶ kx
C k

F t + 𝑚𝑔 − cxሶ − kx = mxሷ
m x
m
x mxሷ + cx+
ሶ kx-mg=F t
F(t) F(t) mg
Vector Mechanics- Kinematics of Rigid Bodies
• Kinematics of Rigid Bodies y

• The position of a particle P on a rigid body at any instant in time can be P(x,y,z)
referred to a fixed rectangular or Cartesian coordinate system j
rP/A
Position Velocity Acceleration rP A
𝑟𝑝 = 𝑥𝑖 + 𝑦𝑗 + 𝑧𝑘 𝑉 = 𝑥𝑖ሶ + 𝑦𝑗ሶ + 𝑧𝑘
ሶ 𝑎 = 𝑥𝑖ሷ + 𝑦𝑗ሷ + 𝑧𝑘
ሷ rA
O
• The position, Velocity & Acceleration. of particle P can be give k i x
Position Velocity Acceleration
𝑟𝑝 = 𝑟𝐴 + 𝑟𝑃/𝐴 𝑉𝑝 = 𝑉𝐴 + 𝑉𝑃/𝐴 𝑎𝑝 = 𝑎𝐴 + 𝑎𝑃/𝐴
• If a rigid body is rotating about an axis . The angular Velocity and acceleration
given
Angular Velocity Angular Acceleration
𝜔 = 𝜔𝑘 𝛼 = 𝑑𝜔Τ𝑑𝑡
• Assuming particle P and A are fixed to a rigid body, the relative velocity of particle P w. r. t. A
Relative Velocity Relative Acceleration
𝑉𝑃/𝐴 = 𝜔 × 𝑟𝑃/𝐴 𝑎𝑃/𝐴 = 𝛼 × 𝑟𝑃/𝐴
 Vector Mechanics- Linear & Angular Momentum
• Linear momentum
• the resultant of the forces acting on the particle is equal to the rate of change of linear momentum of
the particle mv
∑F=0
dv d d
∑𝐹Ԧ = m = (m𝑉) ∑F = 𝑃 = Pሶ m
dt dt dt
• The vector P = mv is called the Linear Momentum
• Angular Momentum or Moment of Momentum v
• The moment about O of the vector mv y
V
𝐿 = r × 𝑃 = rm𝑉 sinϕ Ho φ
• Differentiating w.r.t time
=0, Coz, Cross Product of 2 parallel vector is zero r m

𝐿ሶ = 𝑟ሶ × 𝑃 + r × 𝑃ሶ = 𝑟 × σ 𝐹 = σ 𝑀 = σ 𝜏Ԧ O i
x
• The sum of the moments (Torque) about O of the forces acting on the particle is equal to the rate of change of
moment of momentum, or angular momentum of the particle about O.
Vector Mechanics- Equation of Motion
• Equations of Motion for a Rigid Body
• The motion of the mass center G of the body w.r.t the Newtonian frame of Y’ F4
F1
reference Oxyz is given by the rate of change of linear momentum y F3
∑F= ma = 𝑃ሶ Newton’s Equation of Motion
• The motion of the body relative to the centroidal frame of reference j Δmi
Gx’y’z’ is given by the rate change of angular momentum Go X’
F2

∑MG=Iα = Lሶ G Euler Equation of Motion Z’

O
i x
Angular momentum, L can be written in terms of Moment of Inertia k
𝐿G = σ𝑛𝑖=1 𝑟 × 𝑣𝑖 ∆𝑚𝑖 𝑟×𝑣𝑖 Z
Where 𝜔 =
𝑟2
𝐿G = σ𝑛𝑖=1 ∆𝑚𝑖 𝑟 × 𝑣𝑖
𝑟×𝑣𝑖 where − r the postion & vi ∆mi linear mom of particle pi w.r.t Gx’y’ z’ axis
𝐿G = σ𝑛𝑖=1 ∆𝑚𝑖 𝑟 2
𝑟2
ҧ
𝐿G = 𝐼𝜔 where 𝐼 ҧ = σ𝑛𝑖=1 𝑟𝑖2 ∆𝑚𝑖 = 𝐼 ҧ is M.I about Centroidal Axis
Mass moment of Inertia
• In analyzing the motion of rigid bodies, two types of integrals arise that belong to the geometry of
the body.
• The first type defines
• the center of mass and is important when the translation motion of the body is considered.
• The second type defines
• the moment of inertia that appears when the rotational motion of the body is considered.
𝐼𝑥𝑥 𝐼𝑥𝑦 𝐼𝑥𝑧
• Every rigid body has a 3 × 3 moment of inertia matrix I, which is denoted by 𝐼 = 𝐼𝑦𝑥 𝐼𝑦𝑦 𝐼𝑦𝑧
• The diagonal elements 𝐼𝑖𝑗 , 𝑖 = 𝑗 are called polar moments of inertia 𝐼𝑧𝑥 𝐼𝑧𝑦 𝐼𝑧𝑧

𝐼𝑥𝑥 = 𝐼𝑥 = 𝐼1 = න 𝑦 2 + 𝑧 2 𝑑𝑚 𝐼𝑦𝑦 = 𝐼𝑦 = 𝐼2 = න 𝑧 2 + 𝑥 2 𝑑𝑚 𝐼𝑧𝑧 = 𝐼𝑧 = 𝐼3 = න 𝑥 2 + 𝑦 2 𝑑𝑚

𝐵 𝐵 𝐵
• The off-diagonal element 𝐼𝑖𝑗 , 𝑖 ≠ 𝑗 are called product of inertia

𝐼𝑥𝑦 = 𝐼𝑦𝑥 = − න 𝑥𝑦 𝑑𝑚 𝐼𝑧𝑥 = 𝐼𝑥𝑧 = − න 𝑧𝑥 𝑑𝑚 𝐼𝑦𝑧 = 𝐼𝑧𝑦 = − න 𝑦𝑧 𝑑𝑚

𝐵 𝐵 𝐵
• The elements 𝐼 are calculated about a body coordinate frame attached to the mass center C of the body
Mass moment of Inertia
𝐵
• 𝐼 is a frame-dependent quantity and must be written like 𝐼 to show the frame it is computed in
𝐼𝑥𝑥 𝐼𝑥𝑦 𝐼𝑥𝑧 𝑦2 + 𝑧2 −𝑥𝑦 −𝑧𝑥
𝐵
𝐼 = 𝐼𝑦𝑥 𝐼𝑦𝑦 𝐼𝑦𝑧 𝐼=න −𝑥𝑦 𝑧2 + 𝑥2 −𝑦𝑧 𝑑𝑚
𝐼𝑧𝑥 𝐼𝑧𝑦 𝐼𝑧𝑧 𝐵
−𝑧𝑥 −𝑦𝑧 𝑥2 + 𝑦2
• Principal moment of Inertia
• If the local coordinate frame Oxyz is located such that the products of inertia vanish, the local
coordinate frame is called principal coordinate associated M.I is called Principal M.I
• Principal axes and principal M.I can be found by solving
𝐼𝑥𝑥 − 𝐼 𝐼𝑥𝑦 𝐼𝑥𝑧 characteristic eq will be a 𝐼1
𝑑𝑒𝑡 𝐼𝑖𝑗 − 𝐼 𝛿𝑖𝑗 =0 𝐼𝑦𝑥 𝐼𝑦𝑦 − 𝐼 𝐼𝑦𝑧 = 0 cubic eq which has 3 Eigen 𝐼= 𝐼2
𝐼𝑧𝑥 𝐼𝑧𝑦 𝐼𝑧𝑧 − 𝐼 values (𝐼1 , 𝐼2 and 𝐼3 ) 𝐼3

• Example: Consider the inertia matrix I

characteristic equation:
20 − 𝜆 30 − 𝜆 40 − 𝜆
− 4 40 − 𝜆 = 0 30.38
𝐼= 19.62
Three roots (Eigen Values)
40
𝐼1 = 30.385, 𝐼2 = 19.615, 𝐼3 = 40
 𝑄𝑗 is related Virtual work 𝛿𝑊𝑗𝑁𝐶 = 𝛿𝑞𝑗 where
Lagrangian Mechanics 𝛿𝑞𝑗 is virtual displacement and
𝑄𝑗 = 0 for conservative forces
• Lagrangian Function 𝐿 = 𝐾. 𝐸 − 𝑃. 𝐸 = 𝑇 − 𝑉
• The generalized coordinates, 𝑞𝑗
• Equation of Motion can be determined by • Not necessarily to be Cartesian
𝑑 𝜕𝐿 𝜕𝐿 • Not even Inertial
− = 𝑄𝑗
𝑑𝑡 𝜕𝑞ሶ 𝑗 𝜕𝑞𝑗
• Must be independent and complete
• Where : 𝑞𝑗 is the generalized coordinates and • The system must be Holonomic
• System required No. of DOF and is equal to
: 𝑄𝑗 is the generalized non conservative forces No. coordinate

• Plug in L=T -V in the above Equation • Steps

=0
𝑑 𝜕𝑇 𝑑 𝜕𝑉 𝜕𝑇 𝜕𝑉 • Determine the number of DOF &
− − + = 𝑄𝑗 Chose the coordinates, 𝑞𝑗
𝑑𝑡 𝜕𝑞ሶ 𝑗 𝑑𝑡 𝜕𝑞ሶ 𝑗 𝜕𝑞𝑗 𝜕𝑞𝑗
• Verify :complete independent and
• For mechanical system the P.E= V is not a function of time & velocity
• the kinetic energy is only a function of velocity
Holonomicty
1 2 3 4 • Compute T & V
𝑑 𝜕𝑇 𝜕𝑇 𝜕𝑉
− + = 𝑄𝑗 • Compute 1 2 3 4 for each 𝑞𝑗
𝑑𝑡 𝜕𝑞ሶ 𝑗 𝜕𝑞𝑗 𝜕𝑞𝑗
 Lagrangian Mechanics
• Lagrangian Function 𝐿 = 𝐾. 𝐸 − 𝑃. 𝐸 = 𝑇 − 𝑉

• Let 𝑞𝑗 is the generalized coordinates and 𝑄𝑗 is the generalized non conservative force
𝑑 𝜕𝑇 𝜕𝑇 𝜕𝑉
• Equation of Motion for conservative Forces − + =0
𝑑𝑡 𝜕𝑞ሶ 𝑗 𝜕𝑞𝑗 𝜕𝑞𝑗
• Example- Consider a mass of particle fall from y distance above the ground
m
• Let the generalize coordinate 𝑞𝑗 = 𝑦 = 𝑥 and 𝑞ሶ 𝑗 = 𝑉 = 𝑥ሶ
1 1 y V
• Kinetic Energy 𝐾. 𝐸 = 𝑇 = 𝑚𝑉 2 = 𝑚𝑥ሶ 2 Potential Energy 𝑃. 𝐸 = 𝑉 = 𝑚𝑔𝑥 = ∆𝑈
2 2
1
𝐿 = 𝐾. 𝐸 − 𝑃. 𝐸 = 𝑚𝑉 2 − 𝑚𝑔𝑥
2
𝜕𝐿 𝜕 𝜕 1
For conservative forces 𝑄𝑗 =0
= 𝑚𝑔𝑥 = 𝑚𝑔 but 𝑚𝑥ሶ 2 =0
𝜕𝑥 𝜕𝑥 𝜕𝑥 2
𝜕𝐿 𝜕 1 2 𝜕 𝑑 𝜕𝑇 𝜕𝑉
= 𝜕𝑥ሶ 𝑚 𝑥ሶ = 𝑚𝑥ሶ but −𝑚𝑔𝑥 = 0 − =0
𝜕𝑥ሶ 2 𝜕𝑥ሶ 𝑚𝑥ሷ = −𝑚𝑔
𝑑𝑡 𝜕 𝑥ሶ 𝜕𝑥
𝑑 𝜕𝐿 𝑑
= 𝑚𝑥ሶ = 𝑚𝑥ሷ 𝑚𝑥ሷ + 𝑚𝑔 = 0 𝐹 = 𝑚𝑥ሷ = 𝑚𝑔
𝑑𝑡 𝜕𝑥ሶ 𝑑𝑡
 Lagrangian Mechanics x
k
• Example-2 Find Equation of Motion Using Lagrangian Method
m F(t)
𝑑 𝜕𝐿 𝜕𝐿
− = 𝑄𝑗 For Non Conservative Forces
𝑑𝑡 𝜕𝑞ሶ 𝑗 𝜕𝑞𝑗
C

 Let the generalize coordinate 𝑞𝑗 = 𝑥 and 𝑞ሶ 𝑗 = 𝑉 = 𝑥ሶ

1 1
 Kinetic Energy 𝐾. 𝐸 = 𝑇 = 𝑚𝑉 = 𝑚𝑥ሶ 2
2
𝑑 𝜕𝐿 𝑑 𝜕𝑇 𝑑
2 2 = = 𝑚𝑥ሶ = 𝑚𝑥ሷ
𝑑𝑡 𝜕𝑞ሶ 𝑗 𝑑𝑡 𝜕 𝑥ሶ 𝑑𝑡
1 2
 Potential Energy 𝑃. 𝐸 = 𝑉 = 𝑘𝑥
2
 Now Substitute all in the general Lagrange Equation
 Frictional Viscous damping Force 𝑄𝑗 = 𝐹 𝑡 − 𝐶 𝑥ሶ 𝑑 𝜕𝑇 𝜕𝑉
+ = 𝑄𝑗
𝜕𝐿 𝜕𝑉 𝜕 1 2 𝑑𝑡 𝜕𝑞ሶ 𝑗 𝜕𝑞𝑗
 Now Find = = 𝑘𝑥 = 𝑘𝑥
𝜕𝑞𝑗 𝜕𝑥 𝜕𝑥 2
𝑚𝑥ሷ 𝑘𝑥 𝐹 𝑡 − 𝐶 𝑥ሶ
𝜕𝐿 𝜕𝑇 𝜕 1
= = 𝑚𝑥ሶ 2 = 𝑚𝑥ሶ
𝜕𝑞ሶ 𝑗 𝜕𝑥ሶ 𝜕𝑥ሶ 2 𝑚𝑥ሷ + 𝑘𝑥 = 𝐹 𝑡 − 𝐶 𝑥ሶ 𝒎𝒙ሷ + 𝑪𝒙ሶ + 𝒌𝒙 = 𝑭 𝒕
 ODE
State Space Approach
• ODE
• More fundamental
• Can be extended for nonlinear & time varying system
• Need intensive calculus operation
• Laplace Transformation Solving Linear
• Has algebraic Operation and More convenient Dynamic
System
• Not suitable for computer application
• State Space
• Overcome the drawback of the above two method Laplace
State Space
• Popular computer implementation Transform
• Important Aspect of State Space
𝑢1 𝑌1
• Identify the state of the system
𝑢2 Dynamic System 𝑌2
• The output of the system can be deduced algebraically once 𝑢𝑘
𝑌𝑘
the state of the system & knowing input quantity
Input Sate Output
Simple Reparations State Equation 𝑥ሶ = 𝐴𝑥 + 𝐵𝑢 𝑥ሶ 1
𝑥ሶ 2
Of State Space Out-put Equation Y= 𝐶𝑥 + 𝐷𝑢 𝑥ሶ 𝑛
State Space Approach
• State Equation Matrix
• A set of n coupled ODE
𝑥ሶ 1 𝑥1 𝑢1
𝑥ሶ 2 = 𝐴 𝑥2 + 𝐵 𝑢2
Simple Reparations 𝑥ሶ = 𝐴𝑥 + 𝐵𝑢
𝑥ሶ 𝑛 𝑛×𝑛 𝑥𝑛 𝑛×𝑘 𝑢𝑘

• Output Equation Matrix

𝑦1 𝑥1 𝑢1
𝑦2 = 𝐶 𝑥2 + 𝐷 𝑢2 Simple Reparations 𝑦 = 𝐶𝑥 + 𝐷𝑢
𝑦𝑚 𝑚×𝑛 𝑥𝑛 𝑚×𝑘 𝑢𝑛
 State Space Approach X (t)
k
• Example Find Equation of motion in State Space Approach F(t)
• Let 𝑥 𝑡 = 𝑥1 m
• 𝑥ሶ 𝑡 = 𝑥2 = 𝑥ሶ 1
C
• 𝑢 𝑡 = 𝑢1
• State variables are 𝑥ሶ 1 and 𝑥ሶ 2 𝒎𝒙ሷ + 𝑪𝒙ሶ + 𝒌𝒙 = 𝑭 𝒕

• 𝑥ሶ 1 = 𝑥ሶ =1 𝒙𝟐
𝟏 𝒌 𝑪
𝑭 𝒕 − 𝒌𝒙 − 𝑪𝒙ሶ = 𝒎𝒙ሷ
• 𝑥ሶ 2 = 𝒙ሷ = 𝒖 − 𝒙 − 𝒙
𝒎 𝟏 𝒎 𝟏 𝒎 𝟐

• Rewriting in the matrix Form Input Spring Fic. Acc

• 𝑥ሶ 1 = +0𝒙𝟏 + 1 𝒙𝟐 + 0 𝒖𝟏 U(t) Force Damp Force
Force
𝒌 𝑪 𝟏
• 𝑥ሶ 2 = − 𝒙 − 𝒙 + 𝒖
𝒎 𝟏 𝒎 𝟐 𝒎 𝟏

𝑼 𝒕 − 𝒙𝟏 𝒌 − 𝒙𝟐 𝑪 = 𝒙ሶ 𝟐 𝒎
State Space Approach
• State Equation becomes

𝑥ሶ 1 0 1 𝑥1 0
= + 𝑢1 𝑥ሶ = 𝐴𝑥 + 𝐵𝑢
𝑘 𝐶 1
𝑥ሶ 2 − − 𝑥2
𝑚 𝑚 𝑚

• Now to determine the output function

Let 𝑦 = 𝑥1 = 𝑥(𝑡) Let 𝑦 = 𝑥2 = 𝑥(𝑡)
ሶ
=1 𝑥1 + 0 𝑥2 + 0 𝑈1 =0 𝑥1 + 1 𝑥2 + 0 𝑈1

𝑥1 𝑥1
𝑦 = 1 0 𝑥 + 0 𝑢1 𝑦 = 0 1 𝑥 + 0 𝑢1
2 2
State Space to Transfer Function
• State Equation 𝑥ሶ = 𝐴𝑥 + 𝐵𝑢 Output Equation 𝑦 = 𝐶𝑥 + 𝐷𝑢

𝑦(𝑠)
• Transfer Function 𝐺𝑝 𝑆 =
𝑢(𝑠)

• Laplace of state and output Equations

• ℒ 𝑥ሶ = 𝑠𝑋 𝑠 = 𝐴𝑋 𝑠 + 𝐵𝑈 𝑠 𝑋 𝑠 = 𝐼𝑠 − 𝐴 −1
𝐵𝑈 𝑠

• ℒ 𝑦 = 𝑦 𝑠 = 𝐶𝑋 𝑠 + 𝐷𝑈(𝑠) 𝑦 𝑠 = 𝐶 𝐼𝑠 − 𝐴 −1 𝐵𝑈 𝑠 + 𝐷𝑈(𝑠)

• Transfer Function
y(s) −1
Gp S = = C Is − A B+D
u(s)
State Space to Transfer Function
• For Spring- Mass System Previous Problem
𝐶
1 𝑆+ 1 0
0 1 0 𝑚 1
𝑘 𝐶 Gp S = 1 0
𝐴= 𝐵= 1 C= 1 0 𝐷= 0 𝑐 𝑘 𝑘
− − 𝑆 𝑠+𝑚 +𝑚 − 𝑆 𝑚
𝑚 𝑚 𝑚 𝑚
• Transfer Function
1
1 𝑚
y(s) Gp S = 1 0
Gp S = = C Is − A −1
B+D 𝑐 𝑘 𝑠
𝑆2 + 𝑆 𝑚 + 𝑚
u(s) 𝑚
−1 1 1
𝑠 −1 0
Gp S = 1 0 𝑘 𝐶 1 +0 Gp S = +0
𝑆+𝑚 2 𝑐 𝑘 𝑚
𝑚 𝑚 𝑆 +𝑆 +
𝑚 𝑚
• Inverse of a matrix 1
Gp S =
𝑎 𝑏 1 𝑚𝑆 2 + 𝑆𝑐 + 𝑘
𝐸= 𝑑 −𝑏
𝑐 𝑑 𝐸 −1 =
𝑑𝑒𝑡 𝐸 −𝑐 𝑎
Advanced Vehicle Dynamics (AUE 8306)

End of Lecture-2

Mechanical Systems and Vehicle Engineering Program

Adama Science and Technology University
2018-19 A.Y Semester II