Mastering the Nyquist Stability Criterion for Engineering Control

Introduction to Nyquist Stability

In the study of control theory, determining whether a closed-loop system is stable is a fundamental requirement for any engineering design. While tools like the Routh-Hurwitz criterion provide algebraic solutions, the Nyquist stability criterion offers a powerful graphical approach. By mapping the frequency response of an open-loop system, you can determine the stability of the closed-loop system without needing to solve for the roots of the characteristic equation directly.

For undergraduate engineering students, mastering this topic is essential for both exam success and practical system design. You will learn how to interpret the Nyquist plot, understand the relationship between open-loop poles and closed-loop stability, and apply the Cauchy Argument Principle to predict system behaviour. This article provides the theoretical foundation and step-by-step worked examples to ensure you are exam-ready.

The Nyquist Contour and Mapping

The Nyquist criterion relies on the Cauchy Argument Principle, which relates the number of zeros and poles of a complex function inside a closed contour to the number of encirclements of the origin. In control theory, we define a specific path in the s-plane known as the Nyquist D-contour. This contour encompasses the entire right-half plane (RHP) by travelling up the imaginary axis ($s = j\omega$) from $-j\infty$ to $+j\infty$ and closing with a large semi-circle in the RHP.

When we map this contour through the open-loop transfer function $L(s) = G(s)H(s)$, the resulting plot in the complex plane is the Nyquist plot. The stability of the closed-loop system is determined by the number of clockwise encirclements ($N$) of the critical point $(-1, 0)$ in the $L(s)$-plane.

The Nyquist Stability Criterion Formula

The core of the criterion is expressed by the equation:

$$Z = P - N$$

Where:

• $Z$ is the number of closed-loop poles in the RHP (for stability, we require $Z = 0$).
• $P$ is the number of open-loop poles in the RHP.
• $N$ is the number of clockwise encirclements of the point $(-1, 0)$.

If $N$ is negative, it represents counter-clockwise encirclements. To ensure a stable system, the number of counter-clockwise encirclements must exactly match the number of unstable open-loop poles.

Worked Example 1: First-Order System

Consider an open-loop system $L(s) = \frac{K}{s+a}$ where $K, a > 0$.

1. Identify Poles: The pole is at $s = -a$. Since $-a$ is in the left-half plane, $P = 0$.
2. Frequency Response: Substitute $s = j\omega$: $L(j\omega) = \frac{K}{j\omega + a}$.
3. Plotting: As $\omega$ goes from $-\infty$ to $+\infty$, the plot starts at $0$, moves through the real axis at $K/a$ (when $\omega=0$), and returns to $0$.
4. Stability: Since $P=0$, we need $N=0$ for stability ($Z=0$). The plot does not encircle $(-1, 0)$, so $N=0$. Thus, $Z = 0 - 0 = 0$. The system is stable for all $K > 0$.

Worked Example 2: Determining Gain for Stability

Consider $L(s) = \frac{K}{s(s+1)(s+2)}$.

1. Poles: $P = 0$ (all poles are at $0, -1, -2$).
2. Frequency Response: $L(j\omega) = \frac{K}{j\omega(j\omega+1)(j\omega+2)} = \frac{K}{-3\omega^2 + j\omega(2 - \omega^2)}$.
3. Crossing the Real Axis: Set the imaginary part to zero: $2 - \omega^2 = 0 \implies \omega = \sqrt{2}$.
4. Real Part at Crossing: $L(j\sqrt{2}) = \frac{K}{-3(2)} = -K/6$.
5. Stability Condition: For stability, we need $N=0$. The point $(-1, 0)$ must not be encircled. This happens if the crossing point $-K/6$ is to the right of $-1$. $$-K/6 > -1 \implies K < 6$$ The system is stable for $0 < K < 6$.

Common Mistakes

• Ignoring the D-contour: Students often forget that the Nyquist plot must include the return path from $+j\infty$ to $-j\infty$ via the large semi-circle.
• Misinterpreting Encirclements: Remember that $N$ is the net number of clockwise encirclements. If the plot loops around $(-1, 0)$ once clockwise and once counter-clockwise, $N=0$.
• Sign Errors: Always ensure you are using the negative feedback convention. If the system is positive feedback, the critical point shifts to $(+1, 0)$.
• Poles on the Imaginary Axis: If a pole exists at $s=0$, the D-contour must be indented around the origin, which changes the shape of the plot significantly.

Frequently Asked Questions

What is the difference between Nyquist and Bode plots? Bode plots show magnitude and phase separately against frequency, while Nyquist plots show the complex frequency response on a single polar plot. Nyquist is more general for systems with RHP poles.

How do I find the Gain Margin from a Nyquist plot? It is the reciprocal of the magnitude at the frequency where the phase is $-180^\circ$. If the plot crosses the negative real axis at $-x$, the gain margin is $1/x$.

Does the Nyquist criterion work for non-linear systems? No, the standard Nyquist criterion is strictly for linear time-invariant (LTI) systems.

Conclusion

The Nyquist stability criterion is a cornerstone of control engineering, bridging the gap between frequency response and closed-loop stability. By visualising the mapping of the D-contour, you can confidently assess the robustness of any feedback system. To see these concepts in action with interactive visualisations, visit MathInstructor AI to generate a free animated lesson on this topic.