Marginal stability is a classification of the dynamical behavior of a system in which the system is neither asymptotically stable nor unstable. In such a state, small perturbations neither decay to zero nor grow without bound; instead, they persist indefinitely, often manifesting as sustained oscillations or constant-amplitude motions.
Definition and Formal Criteria
In linear time‑invariant (LTI) systems, marginal stability is determined by the location of the system’s eigenvalues (continuous‑time) or poles (discrete‑time) in the complex plane:
-
Continuous‑time systems:
- All eigenvalues must have non‑positive real parts.
- Any eigenvalues with zero real part (purely imaginary) must be simple (i.e., have algebraic multiplicity one).
- The presence of any eigenvalue with a positive real part indicates instability, while repeated eigenvalues on the imaginary axis also lead to instability.
-
Discrete‑time systems:
- All poles must lie inside or on the unit circle in the complex z‑plane.
- Poles on the unit circle must be simple; repeated poles on the unit circle cause instability.
When these conditions are satisfied, the system’s response to bounded inputs remains bounded (BIBO stability), but the system does not return to equilibrium after a disturbance.
Typical Manifestations
- Sustained oscillations: A classic example is an undamped harmonic oscillator, where energy is conserved and the system oscillates forever at its natural frequency.
- Constant‑amplitude modes: In electrical circuits, an LC tank with ideal components exhibits marginal stability, maintaining a sinusoidal voltage indefinitely.
- Neutral equilibrium: In mechanical systems, a perfectly balanced inverted pendulum without friction can remain in its upright position after a small displacement, neither falling nor returning.
Contexts of Use
- Control Theory: Marginal stability informs controller design, as designers often aim to shift marginally stable poles into the left half‑plane (continuous) or inside the unit circle (discrete) to achieve asymptotic stability.
- Dynamical Systems: In the study of bifurcations, a system may pass through a marginally stable configuration (e.g., Hopf bifurcation) where a pair of complex conjugate eigenvalues cross the imaginary axis.
- Fluid Dynamics: Certain laminar flow configurations are marginally stable, meaning infinitesimal disturbances neither decay nor amplify, a condition relevant to transition to turbulence.
- Population Biology: Linearized models of population dynamics may exhibit marginal stability when growth rates are exactly balanced by mortality, resulting in constant population size.
Implications
- Sensitivity to Perturbations: Because marginally stable systems do not attenuate disturbances, they can be highly sensitive to modeling errors, parameter variations, and external noise.
- Design Considerations: Engineers typically avoid marginally stable designs in safety‑critical applications, preferring asymptotically stable configurations that guarantee convergence to equilibrium.
- Energy Conservation: In idealized physical models, marginal stability reflects perfect energy conservation; any real‑world dissipation typically renders the system asymptotically stable.
Related Concepts
- Asymptotic stability: All eigenvalues have strictly negative real parts (continuous) or lie strictly inside the unit circle (discrete).
- Instability: At least one eigenvalue has a positive real part (continuous) or a pole lies outside the unit circle (discrete).
- Neutral stability: Often used synonymously with marginal stability, though some literature distinguishes neutral stability as a broader category that includes certain non‑linear behaviors.
Mathematical Example
Consider the continuous‑time LTI system $\dot{x}=Ax$ with
$$ A = \begin{bmatrix} 0 & -\omega \ \omega & 0 \end{bmatrix}, $$
where $\omega>0$. The eigenvalues are $\lambda_{1,2} = \pm j\omega$, purely imaginary and simple. The system’s state evolves as
$$ x(t) = \begin{bmatrix} \cos(\omega t) & -\sin(\omega t)\ \sin(\omega t) & \cos(\omega t) \end{bmatrix} x(0), $$
producing perpetual sinusoidal motion—a hallmark of marginal stability.
References
- Ogata, K., Modern Control Engineering, 5th ed., Prentice Hall, 2010.
- Khalil, H. K., Nonlinear Systems, 3rd ed., Prentice Hall, 2002.
- Strogatz, S. H., Nonlinear Dynamics and Chaos, Westview Press, 2015.