Basic Concepts

Basic Concepts in the Stability
Studies of Dynamic Systems

Description of a dynamic system
• The representation of the behavior of a dynamic system
𝑥𝑖 = 𝑓𝑖 𝑥1 , ⋯ , 𝑥𝑛 ; 𝑢1 , ⋯ , 𝑢𝑛 ; 𝑡 𝑖 = 1, ⋯ , 𝑛
or 𝑥 = 𝑓 𝑥, 𝑢, 𝑡
• Autonomous system
𝑥 = 𝑓 𝑥, 𝑢
𝑦 = 𝑔(𝑥, 𝑢)
The concept of state
• State: 𝑥(𝑡𝑜 )
The min amount of necessary info at any 𝑡𝑜 to determine future behavior
without the input before 𝑡𝑜 .
• State variables: 𝑥𝑖 𝑖 = 1, ⋯ , 𝑛 (not unique )
• State vector: 𝑥 = 𝑥1 ⋯ 𝑥𝑛 𝑇
• System behavior
𝑥 = 𝑓 𝑥, 𝑢 , 𝑥 𝑡𝑜 = 𝑥𝑜
State-space, state trajectory, equilibrium
• A state represented in an n-dimensional space, state-space
• State trajectory: points traced in the state-space
• Equilibrium points: points where all derivatives 𝑥1 , 𝑥2 , … , 𝑥𝑛 are zero.
For the system represented by 𝑥 = 𝑓(𝑥)
𝑥 ∗ 𝑖𝑠 𝑎𝑛 𝑒𝑞𝑢𝑖𝑙𝑖𝑏𝑟𝑖𝑢𝑚 𝑖𝑓 𝑎𝑛𝑑 𝑜𝑛𝑙𝑦 𝑖𝑓 𝑓 𝑥 ∗ = 0

Stability of an Equilibrium
Consider an autonomous continuous-time system
𝑥=𝑓 𝑥 , 𝑥 𝑡𝑜 = 𝑥𝑜
Assume that 𝑥 ∗ is an equilibrium point, i.e. 𝑓 𝑥 ∗ = 0
𝑥 ∗ is a stable equilibrium if for any given value 𝜀 > 0, there exists 𝛿 > 0
such that if 𝑥 𝑡𝑜 − 𝑥 ∗ < 𝛿 then 𝑥 𝑡 − 𝑥 ∗ < 𝜀 for all 𝑡 > 𝑡𝑜 .
(If all the solutions which are starting nearby the point stay nearby, the point
is stable. Otherwise unstable)
Asymptotically stable equilibrium
The equilibrium is asymptotically stable if all solutions also tend to the
equilibrium point:
𝑥 ∗ is asymptotically stable equilibrium if there exists 𝛿 > 0

such that if 𝑥 𝑡𝑜 − 𝑥 ∗ < 𝛿 ⇒ 𝑥 𝑡 − 𝑥 ∗ → 0 as 𝑡 → ∞ .
• Stable, unstable, asymptotically stable : local nature of the behavior of the

system.
• The set of all initial conditions (perturbations) that converge to a given
asymptotically stable equilibrium is called the region of attraction.
Stability in Linear Systems
A linear time-invariant continuous-time system
𝑥 = 𝐴𝑥
𝐴: system matrix
Only one equilibrium: 𝑥 ∗ = 0 (if A is nonsingular)
The (local) stability of 𝑥 ∗ = 0 can be determined by the eigenvalues of

the matrix 𝐴.
Stability criteria
• Eigenvalues of 𝐴: 𝜆𝑖 = 𝛼𝑖 ± 𝑗𝛽𝑖 𝑖 = 1, … , 𝑛
𝐼𝑓 𝛼𝑖 ≤ 0 𝑓𝑜𝑟 𝑎𝑙𝑙 𝑠𝑖𝑚𝑝𝑙𝑒 𝑟𝑜𝑜𝑡𝑠

𝑎𝑛𝑑
𝛼𝑖 < 0 𝑓𝑜𝑟 𝑎𝑙𝑙 𝑟𝑒𝑝𝑒𝑎𝑡𝑒𝑑 𝑟𝑜𝑜𝑡𝑠
𝒔𝒕𝒂𝒃𝒍𝒆
𝐼𝑓 𝛼𝑖 < 0
𝒂𝒔𝒚𝒎𝒑. 𝒔𝒕𝒂𝒃𝒍𝒆
𝑓𝑜𝑟 𝑎𝑙𝑙 𝑟𝑜𝑜𝑡𝑠
𝐼𝑓 𝛼𝑖 > 0 𝑓𝑜𝑟 𝑎𝑛𝑦 𝑠𝑖𝑚𝑝𝑙𝑒 𝑟𝑜𝑜𝑡
𝑜𝑟 𝒖𝒏𝒔𝒕𝒂𝒃𝒍𝒆
𝛼𝑖 ≥ 0 𝑓𝑜𝑟 𝑎𝑛𝑦 𝑟𝑒𝑝𝑒𝑎𝑡𝑒𝑑 𝑟𝑜𝑜𝑡
Nonlinear systems, Linearization
Consider a nonlinear system
𝑥 = 𝑓 𝑥, 𝑢
𝑦 = 𝑔 𝑥, 𝑢
Let 𝑥𝑜 be an equilibrium and 𝑢0 be the corresponding input vector:

𝑓 𝑥𝑜 , 𝑢𝑜 = 0
Let us perturb the system

𝑥 = 𝑥𝑜 + Δ𝑥
𝑢 = 𝑢𝑜 + Δ𝑢
𝜕𝑓 𝜕𝑓
Δ𝑥 ≅ Δ𝑥 + Δ𝑢 = 𝐴 Δ𝑥 + 𝐵 Δ𝑢
𝜕𝑥 𝜕𝑢
𝜕𝑔 𝜕𝑔
Δ𝑦 ≅ ∆𝑥 + ∆𝑢 = 𝐶 Δ𝑥 + 𝐷 Δ𝑢
𝜕𝑥 𝜕𝑥
𝜕𝑓
The system matrix: 𝐴 =
𝜕𝑥
Eigen-properties of the System Matrix
• 𝐴𝑣 = 𝜆𝑣 , 𝑣 ≠ 0
• 𝐴 − 𝜆𝐼 𝑣 = 0
• Characteristic equation: det 𝐴 − 𝜆𝐼 = 0
• Roots = Eigenvalues
• Real or complex
Free Motion of a Dynamic System
Δ𝑥 = 𝐴Δ𝑥
𝐴 is diagonalizable ⇒ 𝐴 = 𝑆Λ𝑆 −1
Let 𝑧 ≜ 𝑆 −1 Δ𝑥 ⇒ 𝑧 = Λ𝑧
𝑧𝑖 𝑡 = 𝑧𝑖 0 𝑒 𝜆𝑖 𝑡
The free response given as a linear combination of n modes:
𝑛
Δ𝑥 𝑡 = 𝑐𝑖 𝑣𝑖 𝑒 𝜆𝑖 𝑡
𝑖=1
Dynamic modes
Mode associated with the eigenvalue 𝜆𝑖 = 𝛼𝑖 ± 𝑗𝛽𝑖
Real eigenvalue (𝛽𝑖 =0)⇒ a non-oscillatory mode

negative real (𝛼𝑖 < 0) ⇒ decaying mode
positive real (𝛼𝑖 > 0) ⇒ aperiodic mode
Complex conjugate pair of eigenvalues (𝛽𝑖 ≠0) ⇒oscillatory mode

Frequency of oscillation : 𝑓 = 𝛽𝑖 (2𝜋)
−𝛼𝑖
Damping ratio: 𝜁 = 2
𝛼𝑖 +𝛽𝑖
Phase portraits
Generating unit as an oscillating system
Swing Equation
Per-unit Swing Equation
𝑑2 𝛿
2𝐻 2 = 𝑃𝑚(𝑝𝑢) − 𝑃𝑒 𝑝𝑢 − 𝑃𝐷 𝑝𝑢
𝑑𝑡
𝑑𝛿
= 𝜔 − 𝜔𝑠 = 𝜔𝑠 𝜔 𝑝𝑢 − 1
𝑑𝑡
𝑑𝜔(𝑝𝑢)
2𝐻 = 𝑃𝑚(𝑝𝑢) − 𝑃𝑒 𝑝𝑢 − 𝐷(𝑝𝑢) (𝜔 𝑝𝑢 − 1)
𝑑𝑡
EQULIBRIUM POINTS
• Balance between mechanical torque and electrical torque
• Steady-state model of generator-infinite-bus system
• Power delivered to the system
• Power angle / rotor angle

At Equilibrium:
Air-gap power (round-rotor machine):

Steady-state Stability of Unregulated System
• Small-distrubance / small-signal stability
• Small disturbance and linearization
• Affected by automatic control of the generator and turbine
• Unregulated sytem = no control

• Steady-state synchronizing power coefficient
• Pull-out power (steady-state stability limit):
• Steady-state stability margin:

Basic Concepts

Uploaded by

Copyright:

Available Formats

Basic Concepts

Uploaded by

Document Information

Original Description:

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Basic Concepts

Uploaded by

Copyright:

Available Formats

Basic Concepts in the Stability

Studies of Dynamic Systems

• State variables: 𝑥𝑖 𝑖 = 1, ⋯ , 𝑛 (not unique )

• State trajectory: points traced in the state-space

• Equilibrium points: points where all derivatives 𝑥1 , 𝑥2 , … , 𝑥𝑛 are zero.

For the system represented by 𝑥 = 𝑓(𝑥)

𝑥 ∗ 𝑖𝑠 𝑎𝑛 𝑒𝑞𝑢𝑖𝑙𝑖𝑏𝑟𝑖𝑢𝑚 𝑖𝑓 𝑎𝑛𝑑 𝑜𝑛𝑙𝑦 𝑖𝑓 𝑓 𝑥 ∗ = 0

Assume that 𝑥 ∗ is an equilibrium point, i.e. 𝑓 𝑥 ∗ = 0

𝑥 ∗ is asymptotically stable equilibrium if there exists 𝛿 > 0

• Stable, unstable, asymptotically stable : local nature of the behavior of the

Only one equilibrium: 𝑥 ∗ = 0 (if A is nonsingular)

The (local) stability of 𝑥 ∗ = 0 can be determined by the eigenvalues of

𝐼𝑓 𝛼𝑖 ≤ 0 𝑓𝑜𝑟 𝑎𝑙𝑙 𝑠𝑖𝑚𝑝𝑙𝑒 𝑟𝑜𝑜𝑡𝑠

Let 𝑥𝑜 be an equilibrium and 𝑢0 be the corresponding input vector:

Let us perturb the system

Real eigenvalue (𝛽𝑖 =0)⇒ a non-oscillatory mode

Complex conjugate pair of eigenvalues (𝛽𝑖 ≠0) ⇒oscillatory mode

• Power angle / rotor angle

Air-gap power (round-rotor machine):

• Small disturbance and linearization

• Affected by automatic control of the generator and turbine

• Unregulated sytem = no control

• Pull-out power (steady-state stability limit):

• Steady-state stability margin:

You might also like