Linear Multivariable Control (SCE-507)

Unit-3
State Space Modeling
Dr. Nezar ALyazidi
Contents (Unit-3):
1) State Space

2) Electrical Network State Space

3) Mechanical System State Space

SCE 507-241
State Space
System Solutions:
Approaches for Analysis and
Design of Control Systems

Classical Modern
OR OR
Frequency Domain Approach Time Domain Approach

Transfer Function State Space

SCE 507-241
Introduction to state space concepts
• The great progress in control science since about 1955 has changed the
basic concepts of analysis and synthesis of control systems. This progress
depended largely on mathematical study of optimal control systems.

• Modern control theory which is based on state space concepts is extremely

useful not only for designing a specific optimal control system but also for
improving the principle on which the systems will operate.
• By using the state space approach, the control engineer may be able to
design systems with performance characteristics that cannot be achieved
by the classical approach by means of the frequency response method (like
transfer function) and the root locus method.

SCE 507-241
STATE space Modeling and analysis

• State space analysis is a powerful and modern approach for designing and
analyzing control systems.
• The conventional or old methods for designing and analyzing control systems are
based on the transfer function method.
• The transfer function method for design and analysis had many drawbacks such
as:
– Transfer function is defined under zero initial conditions
– Applicable to LTI systems only.
– SISO systems only.
– Does not provide information regarding the internal state of the system
• Initial conditions can be incorporated into the system design
• State equations are highly compatible with simulation on analog or digital
computers

SCE 507-241
State Space:
➢ Frequency Domain Approach
✓ Disadvantage:
❖ Limited Applicability

❖ Only applied to LTI systems

✓ Advantage:
❖ Rapidly provide stability and transient response information

SCE 507-241
Advantages of state variable analysis.
• This can be applicable to
• Linear systems
• Non-linear system
• Time variant systems
• Time invariant systems
• Multiple input multiple output systems
• This gives idea about the internal state of the system

SCE 507-241
State Space:
➢ State Space Approach
✓ A unified method for Modeling, Analyzing and Designing a wide range of systems

✓ Used to represent non-linear systems that have Saturation, Backlash, and Dead Zone

✓ Handles, conveniently, systems with Non-Zero initial conditions

✓ Time-varying systems can be represented in the state space

SCE 507-241
 Selection of state variables
• The state variables of a system are not unique.

• There are many choices for a given system

Guidelines:
• 1. For a physical systems, the number of state variables (minimal state) needed to represent
the system must be equal to the number of energy storing elements present in the system.
• 2. If a system is represented by a linear constant coefficient differential equation, then the
number of state variables needed to represent the system must be equal to the order of the
differential equation
• 3. If a system is represented by a transfer function, then the number of state variables needed
to represent the system must be equal to the highest power of s in the denominator of the
transfer function.

SCE 507-241
Modelling of Nominal Systems
• Strictly speaking, all systems in practice are nonlinear.
• Fortunately, a large class of nonlinear systems can be approximated as linear
systems and thus can be analyzed and designed based on the linear system model.
Linear systems can be further classified as LTI systems and LTV systems.
• In this chapter, we focus on LTI dynamic systems. For LTI dynamic systems,
there are two commonly used representations: the state space model and the
transfer function model.
• The state-space model is an internal representation of dynamic systems. It consists
of two equations: the state equation and the output equation, where the state
equation describes the dynamics of system by a differential/difference equation
and the output equation describes the measurement process via an algebraic
equation.

SCE 507-241
State Model of Linear System

SCE 507-241
 Definitions
► System state: minimum information needed in order to completely determine the output of a system
from a given moment, provided the input is known from that moment
►System variable: any variable that responds to an input or initial conditions in a system
►State variables: the smallest set of linearly independent system variables such that the values of the set
members at time t0 along with known inputs completely determine the value of all system variables for
all t≥t0
►State vector: vector whose elements are the state variables
►State space: n-dimensional space whose axes represent the state variables
►State equations: set of n simultaneous, first order differential equations with n variables, where the n
variables to be solved are the state variables
►Output equation: algebraic expression of the output variables of a system as linear combinations of the
state variables and the inputs
SCE 507-241
State-space equations

SCE 507-241
SCE 507-241
State space model

SCE 507-241
State-Space Representation

SCE 507-241
Reduction of system equations
➢ Complex plant is usually characterized by multiple inputs, a few and more than ten state
variables, and a similar number of plant outputs. Without reduction, we may not be able to
study a complex system on a computer because of difficulties with algebraic loops, even if
the machine is large enough.

➢ Model reduction techniques are often employed to manage this complexity. These
techniques simplify the system by reducing the number of state variables and inputs/outputs,
making it more feasible to analyze and simulate the system on a computer.

SCE 507-241
SCE 507-241
SCE 507-241
Procedure for Selecting State Variables

SCE 507-241
Electrical Network State Space
Electrical Network State Space:

SCE 507-241
Electrical Network State Space:
➢ A way to determine the number of state variables is to count the number of independent
energy storage elements in the system

➢ The number these energy storage elements equals the order of the differential equation
and the number of state variables

SCE 507-241
Electrical Network State Space:
𝑅 𝐿 𝐶

𝑉𝐿 𝑖𝐿 𝑉𝑐 𝑖𝑐

✓ In 𝑅𝐿𝐶 circuit, normally the state space variables are 𝑉𝑐 and 𝑖𝐿 .

✓ Although, these state space variables can be different.

SCE 507-241
Example

SCE 507-241
SCE 507-241
SCE 507-241
SCE 507-241
SCE 507-241
SCE 507-241
Example

SCE 507-241
SCE 507-241
SCE 507-241
Example: Find the state space model

SCE 507-241
 𝐿 ⅆ𝑖𝐿
𝑣𝐿 =
∂𝑡

𝐶 ⅆ𝑣𝑐
𝐼𝑐 =
∂𝑡

𝒗𝑳 = 𝒗𝒄 = 𝒗(𝒕)

SCE 507-241
SCE 507-241
Electrical Network State Space:
𝑢 𝑡 = 𝑖𝑐 + 𝑖𝐿

𝑢 𝑡 = 𝐶 𝑣ሶ𝑐 + 𝑖𝐿

𝐶 𝑣ሶ𝑐 = −𝑖𝐿 + 𝑢 𝑡

1 1
𝑣ሶ𝑐 = − 𝑖𝐿 + 𝑢(𝑡)
𝐶 𝐶
1
𝑢(𝑡) 𝑢(𝑡)
1 𝐶
𝑣ሶ𝑐 𝑣𝑐
𝐶

1
1 − 𝑖𝐿
𝐶
𝐶
𝑖𝐿

SCE 507-241
Electrical Network State Space:
𝑣𝑐 = 𝑣𝐿 + 𝑣𝑅

𝑣𝑐 = 𝐿𝑖′𝐿 + 𝑅𝑖𝐿

𝐿𝑖′𝐿 = 𝑣𝑐 − 𝑅𝑖𝐿

1 𝑅
𝑖′𝐿 = 𝑣𝑐 − 𝑖𝐿
𝐿 𝐿
1
𝑣𝑐 𝑣
1 𝐿 𝑐 𝑖𝐿
𝑖′𝐿
𝐿

𝑅
𝑅 − 𝑖
𝐿 𝐿
𝐿
𝑖𝐿

SCE 507-241
Electrical Network State Space:
1
𝑢(𝑡) 𝑢(𝑡) 1
1 𝐶
𝑣ሶ𝑐 𝑣𝑐 𝑣𝑐 𝑣
𝐶 1 𝐿 𝑐 𝑖𝐿
𝑖′𝐿
𝐿
1
1 − 𝑖𝐿 𝑅
𝐶 − 𝑖𝐿
𝐶 𝑅 𝐿
𝑖𝐿 𝐿
𝑖𝐿

SCE 507-241
Electrical Network State Space:
1
𝑢(𝑡) 𝑢(𝑡) 1
1 𝐶
𝑣ሶ𝑐 𝑣𝑐 𝑣
𝐶 1 𝐿 𝑐 𝑖𝐿
𝑖′𝐿
𝐿
1
1 − 𝑖𝐿 𝑅
𝐶 − 𝑖
𝐶 𝑅 𝐿 𝐿
𝑖𝐿 𝐿
𝑖𝐿

SCE 507-241
Electrical Network State Space:
1 1
𝑣ሶ𝑐 = − 𝑖𝐿 + 𝑢 𝑡
𝐶 𝐶

1 𝑅
𝑖′𝐿 = 𝑣𝑐 − 𝑖𝐿
𝐿 𝐿

1 1
𝑣ሶ𝑐 = 0 𝑣𝑐 − 𝑖𝐿 + 𝑢(𝑡)
𝐶 𝐶
1 𝑅
𝑖′𝐿 = 𝑣𝑐 − 𝑖𝐿 + 0 𝑢 𝑡
𝐿 𝐿
𝑥1 = 𝑣𝑐
𝑆𝑡𝑎𝑡𝑒 𝑆𝑝𝑎𝑐𝑒 𝑉𝑎𝑟𝑖𝑎𝑏𝑙𝑒𝑠 𝑥2 = 𝑖𝐿
1 1
𝑣ሶ𝑐 0 −𝑣𝑐 1 0 − 1
= 𝐶 + 𝐶 𝑢(𝑡)
𝑥ሶ 1
= 𝐶 𝑥1 +
𝑖′𝐿 1 𝑅 𝑖𝐿 𝑥ሶ 2 1 𝑅 𝑥2 𝐶 𝑢 𝑡
− 0 − 0
𝐿 𝐿 𝐿 𝐿
SCE 507-241 𝒙ሶ = 𝑨𝒙 + 𝑩𝒖
Electrical Network State Space:
𝑣0 = 𝑅𝑖𝑅
𝑣0 = 𝑅𝑖𝐿

𝑖𝐿 𝑣0
𝑅

𝑦 = 0 𝑣𝑐 + 𝑅𝑖𝐿 + 0 𝑢 𝑡
𝑣𝑐
𝑦 = 0 𝑅 𝑖 + 0 𝑢(𝑡)
𝐿

𝑥1
𝑦 = 0 𝑅 𝑥 + 0 𝑢(𝑡)
2

𝒚 = 𝑪𝒙 + 𝑫𝒖

SCE 507-241
Electrical Network State Space:
1
𝑢(𝑡) 𝑢(𝑡) 1
1 𝐶
𝑣ሶ𝑐 𝑣𝑐 𝑣
𝐶 1 𝐿 𝑐 𝑖𝐿
𝑖′𝐿 𝑣0
𝐿 𝑅
1
1 − 𝑖𝐿 𝑅
𝐶 − 𝑖
𝐶 𝑅 𝐿 𝐿
𝑖𝐿 𝐿
𝑖𝐿

SCE 507-241
 Electrical Network State Space:
𝑣𝑖 = 𝑣𝑐 1 + 𝑣𝐿 𝑖𝑐 1 = 𝑖𝐿 + 𝑖𝑅 1
𝑖𝑐 2 = 𝑖𝑅 = (𝑣𝐿 − 𝑣𝑐 2 )
𝑣𝑖 = 𝑣𝑐 1 + 𝐿𝑖′𝐿 𝑅
1
𝐶1 𝑣ሶ𝑐 1 = 𝑖𝐿 + (𝑣 − 𝑣𝑐 2 )
1 1 𝑅 𝐿 𝐿 1
𝐶2 𝑣ሶ𝑐 2 = 𝑖′𝐿 − 𝑣𝑐 2
𝑖′𝐿 = − 𝑣𝑐 1 + 𝑣𝑖 𝑅 𝑅
𝐿 𝐿 𝐿 1
𝐶1 𝑣ሶ𝑐 1 = 𝑖𝐿 + 𝑖′𝐿 − 𝑣𝑐 2 𝐿 1 1 1
𝑅 𝑅
𝐶2 𝑣ሶ𝑐 2 = (− 𝑣𝑐 1 + 𝑣𝑖 ) − 𝑣𝑐 2
𝑅 𝐿 𝐿 𝑅
𝐿 1 1 1
𝐶1 𝑣ሶ𝑐 1 = 𝑖𝐿 + ( 𝑣𝑖 − 𝑣𝑐 1 ) − 𝑣𝑐 2
𝑅 𝐿 𝐿 𝑅 1 1 1
𝑣ሶ𝑐 2 =− 𝑣 − 𝑣 + 𝑣
𝑅𝐶2 𝑐 1 𝑅𝐶2 𝑐 2 𝑅𝐶2 𝑖
1 1 1 1
𝑣ሶ𝑐 1 = − 𝑣𝑐 1 + 𝑖𝐿 − 𝑣𝑐 2 + 𝑣
𝑅𝐶1 𝐶1 𝑅𝐶1 𝑅𝐶1 𝑖

SCE 507-241
 Electrical Network State Space:
1 1 1 1 1 1 1 1
𝑣ሶ𝑐 1 = − 𝑣𝑐 1 + 𝑖𝐿 − 𝑣𝑐 2 + 𝑣 𝑥1 = 𝑣𝑐1 𝑥ሶ 1 = − 𝑥1 + 𝑥2 − 𝑥3 + 𝑣
𝑅𝐶1 𝐶1 𝑅𝐶1 𝑅𝐶1 𝑖 𝑅𝐶1 𝐶1 𝑅𝐶1 𝑅𝐶1 𝑖
𝑆𝑡𝑎𝑡𝑒 𝑆𝑝𝑎𝑐𝑒 𝑉𝑎𝑟𝑖𝑎𝑏𝑙𝑒𝑠 𝑥2 = 𝑖𝐿
1 1 𝑥3 = 𝑣𝑐2
𝑖′𝐿 = − 𝑣𝑐 1 + 𝑣𝑖 1 1
𝐿 𝐿 𝑥ሶ 2 = − 𝑥1 + (0)𝑥2 + (0)𝑥3 + 𝑣𝑖
𝐿 𝐿
1 1 1
𝑣ሶ𝑐 2 = − 𝑣𝑐 1 − 𝑣𝑐 2 + 𝑣 1 1 1
𝑅𝐶2 𝑅𝐶2 𝑅𝐶2 𝑖 𝑥ሶ 3 = − 𝑥1 + (0)𝑥2 − 𝑥3 + 𝑣
𝑅𝐶2 𝑅𝐶2 𝑅𝐶2 𝑖

𝑦 = 𝑣𝑐 2
1 1 1 1
𝑦 = 0 𝑥1 + 0 𝑥2 + 𝑥3 + (0)𝑣𝑖 − −
𝑅𝐶1 𝐶1 𝑅𝐶1 𝑅𝐶1
𝑥ሶ 1 𝑥1
𝑥1 1 1
𝑥ሶ 2 = − 0 0 𝑥2 + 𝑣𝑖
𝑦 = 0 0 1 𝑥2 + 0 𝑣𝑖 𝐿 𝑥3 𝐿
𝑥3 𝑥ሶ 3
1 1 1
− 0 −
𝒚 = 𝑪𝒙 + 𝑫𝒖 𝑅𝐶2 𝑅𝐶2 𝑅𝐶2

SCE 507-241 𝒙ሶ = 𝑨𝒙 + 𝑩𝒖
Mechanical Network State Space
SCE 507-241
Example

SCE 507-241
SCE 507-241
SCE 507-241
Two-masses

SCE 507-241
Mechanical Network State Space:
(𝑀1 𝑠 2 + 𝐷𝑠 + 𝑘)𝑥1 (𝑠) − 𝑘 𝑥2 (𝑠) = 0

− 𝑘 𝑥1 (𝑠) + 𝑀2 𝑠 2 + 𝑘 𝑥2 (𝑠) = 𝐹(𝑠)

Inverse Laplace

𝑀1 𝑥ሷ 1 (𝑡) + 𝐷 𝑥ሶ 1 (𝑡) + 𝑘𝑥1 (𝑡) − 𝑘𝑥2 (𝑡) = 0

−𝑘𝑥1 (𝑡) + 𝑀2 𝑥ሷ 2 (𝑡) + 𝑘𝑥2 (𝑡) = 𝑓(𝑡)

SCE 507-241
Mechanical Network State Space:

𝑥ഺ1 𝑣ሶሷ 1
𝑥ሶሷ 2 𝑣ሶ 2 𝑥ሶሷ 1 𝑣ሶ 1
𝑣2 𝑣1
𝑣3
𝑥ሶ 3 𝑥3 𝑥ሶ 2 𝑥2 𝑥ሶ 1 𝑥1

Phase Variables (Set of Variables):

Where each subsequent state variable is defined to be the derivative of the previous state variable
SCE 507-241
 Mechanical Network State Space:
𝑀1 𝑥ሷ 1 (𝑡) + 𝐷 𝑥ሶ 1 (𝑡) + 𝑘𝑥1 (𝑡) − 𝑘𝑥2 (𝑡) = 0

−𝑘𝑥1 𝑡 + 𝑀2 𝑥ሷ 2 𝑡 + 𝑘𝑥2 𝑡 = 𝑓 𝑡

𝑀1 𝑣ሶ 1 (𝑡) + 𝐷𝑣1 (𝑡) + 𝑘𝑥1 (𝑡) − 𝑘𝑥2 (𝑡) = 0

−𝑘𝑥1 (𝑡) + 𝑀2 𝑣ሶ 2 (𝑡) + 𝑘𝑥2 (𝑡) = 𝑓(𝑡) 𝑥ሶ = 𝐴𝑥 + 𝐵𝑢

𝑘 𝐷 𝑘 𝑦 = 𝐶𝑥 = 𝐷𝑢
𝑣ሶ 1 (𝑡) = − 𝑥1 (𝑡) − 𝑣1 (𝑡) + 𝑥 (𝑡) + (0)𝑣2 (𝑡) + (0)𝑓(𝑡)
𝑀1 𝑀1 𝑀1 2

𝑘 𝑘 1
𝑣ሶ 2 (𝑡) = 𝑥 (𝑡) + (0)𝑣1 (𝑡) − 𝑥 (𝑡) + (0)𝑣2 (𝑡) + 𝑓(𝑡)
𝑀2 1 𝑀2 2 𝑀2

SCE 507-241
 Mechanical Network State Space:
𝑘 𝐷 𝑘 𝑥ሶ 1 𝑡 = 0 𝑥1 𝑡 + 𝑣1 (𝑡) + (0)𝑥2 (𝑡) + (0)𝑣2 (𝑡) + 0 𝑓(𝑡)
𝑣ሶ 1 (𝑡) = − 𝑥1 (𝑡) − 𝑣1 (𝑡) + 𝑥 (𝑡) + (0)𝑣2 (𝑡) + (0)𝑓(𝑡)
𝑀1 𝑀1 𝑀1 2 𝑘 𝐷 𝑘
𝑣ሶ 1 (𝑡) = − 𝑥 (𝑡) − 𝑣 (𝑡) + 𝑥 (𝑡) + (0)𝑣2 (𝑡) + (0)𝑓(𝑡)
𝑀1 1 𝑀1 1 𝑀1 2
𝑘 𝑘 1 𝑥ሶ 2 𝑡 = 0 𝑥1 𝑡 + (0)𝑣1 (𝑡) + (0)𝑥2 (𝑡) + 𝑣2 (𝑡) + 0 𝑓(𝑡)
𝑣ሶ 2 (𝑡) = 𝑥1 (𝑡) + (0)𝑣1 (𝑡) − 𝑥2 (𝑡) + (0)𝑣2 (𝑡) + 𝑓(𝑡)
𝑀2 𝑀2 𝑀2 𝑘 𝑘 1
𝑣ሶ 2 (𝑡) = 𝑥 (𝑡) + (0)𝑣1 (𝑡) − 𝑥 (𝑡) + (0)𝑣2 (𝑡) + 𝑓(𝑡)
𝑀2 1 𝑀2 2 𝑀2

𝑦 = 𝑥2 0 1 0 0
𝑘 D 𝑘 𝑥1 0
𝑦 = 0 𝑥1 + 0 𝑣1 + 𝑥2 + 0 𝑣2 + 0 𝑓(𝑡) 𝑥ሶ 1
− − 0 0
𝑣ሶ 1 𝑀1 𝑀1 𝑀1 𝑣1
𝑥1 = + 0 𝑓 𝑡
𝑥ሶ 2 0 0 0 1 𝑥2 1
𝑣1 𝑘 𝑘
𝑦 = 0 0 1 0 𝑥 + 0 𝑓(𝑡) 𝑣ሶ 2 𝑣2
2 0 − 0 𝑀2
𝑣2 𝑀2 𝑀2
𝒙ሶ = 𝑨𝒙 + 𝑩𝒖
𝒚 = 𝑪𝒙 + 𝑫𝒖

SCE 507-241
SCE 507-241