About this set. These are original practice questions written
in GATE style for the 2018 Control Systems syllabus, with fully worked
solutions. They are not reproductions of the official 2018 question paper.
[1 mark] The response of a linear time-invariant system to a unit step applied at \(t=0\) is
\(y(t)=2\left(1-e^{-4t}\right)\) for \(t\ge 0\). The transfer function of the system is
- \(\dfrac{2}{s+4}\)
- \(\dfrac{8}{s+4}\)
- \(\dfrac{2}{s+2}\)
- \(\dfrac{8}{s+8}\)
Solution
Take the Laplace transform of the measured output (Chapter 7):
Equation
\[Y(s)=2\left(\frac{1}{s}-\frac{1}{s+4}\right)=2\cdot\frac{(s+4)-s}{s(s+4)}=\frac{8}{s(s+4)}\]
The input was a unit step, so \(R(s)=1/s\) and
Equation
\[G(s)=\frac{Y(s)}{R(s)}=\frac{8}{s(s+4)}\cdot s=\frac{8}{s+4}\]
The check is consistent: the DC gain is \(8/4=2\), matching the final value of \(y(t)\), and the pole at \(s=-4\) gives the time constant \(0.25\) s seen in the exponential.
B
Final Answer
Correct answer: (B) \(8/(s+4)\).
[1 mark] A unity-negative-feedback system has the open-loop transfer function
Equation
\[G(s)=\frac{50}{(s+1)(s+5)}\]
The steady-state error for a unit-step input is _____ (round to four decimal places).
Solution
There is no pole at the origin, so the system is Type 0 and the position error constant applies (Chapter 9):
Equation
\[K_p=\lim_{s\to 0}G(s)=\frac{50}{(1)(5)}=10\]
Equation
\[e_{ss}=\frac{1}{1+K_p}=\frac{1}{1+10}=\frac{1}{11}=0.0909\]
The closed loop is stable, since its characteristic equation \(s^{2}+6s+55=0\) has both coefficients positive with roots \(-3\pm j6.78\).
✓
Final Answer
Correct answer: 0.0909.
[2 marks] In a feedback arrangement, the reference \(R\) enters an outer summing junction. Its
output feeds a block \(G_1(s)=10/s\), whose output enters a second summing junction. That
second junction drives a block \(G_2(s)=1/(s+2)\), and the output of \(G_2\) is the system output
\(C\). Two negative feedback paths exist: the output of \(G_2\) returns through a block
\(H_1=4\) to the second summing junction, and \(C\) returns directly (unity gain) to the outer
summing junction. The closed-loop transfer function \(C/R\) is
- \(\dfrac{10}{s^{2}+2s+10}\)
- \(\dfrac{10}{s^{2}+4s+10}\)
- \(\dfrac{10}{s^{2}+6s+10}\)
- \(\dfrac{10}{s^{2}+6s+14}\)
Solution
Reduce the inner loop first (Chapter 4). The inner loop is \(G_2\) with negative feedback \(H_1\):
Equation
\[G_{inner}(s)=\frac{G_2}{1+G_2H_1}=\frac{\frac{1}{s+2}}{1+\frac{4}{s+2}}=\frac{1}{s+6}\]
The forward path is now \(G_1G_{inner}\), closed by unity negative feedback:
Equation
\[\frac{C}{R}=\frac{G_1G_{inner}}{1+G_1G_{inner}}=\frac{\frac{10}{s(s+6)}}{1+\frac{10}{s(s+6)}}=\frac{10}{s(s+6)+10}\]
Equation
\[\frac{C}{R}=\frac{10}{s^{2}+6s+10}\]
Equivalently, in the general form \(G_1G_2/(1+G_2H_1+G_1G_2)\), clearing \(s(s+2)\) gives \(10/\left(s(s+2)+4s+10\right)=10/(s^{2}+6s+10)\), the same result.
C
Final Answer
Correct answer: (C) \(10/(s^{2}+6s+10)\).
[2 marks] A closed-loop system has the transfer function
Equation
\[T(s)=\frac{25}{s^{2}+8s+25}\]
For a unit-step input, the time at which the output reaches its first peak is _____ seconds
(round to three decimal places).
Solution
Match the standard second-order form (Chapter 8):
Equation
\[\omega_n^{2}=25 \Rightarrow \omega_n=5~\text{rad/s},\qquad 2\zeta\omega_n=8 \Rightarrow \zeta=\frac{8}{10}=0.8\]
Since \(\zeta \lt 1\) the response is underdamped and a peak exists. The damped natural frequency is
Equation
\[\omega_d=\omega_n\sqrt{1-\zeta^{2}}=5\sqrt{1-0.64}=5(0.6)=3~\text{rad/s}\]
Equation
\[t_p=\frac{\pi}{\omega_d}=\frac{3.14159}{3}=1.047~\text{s}\]
For reference, the overshoot at that instant is \(\exp\!\left(-\pi(0.8)/0.6\right)=e^{-4.189}=0.0152\), i.e. 1.52 %, and the 2 % settling time is \(4/(\zeta\omega_n)=4/4=1\) s.
✓
Final Answer
Correct answer: 1.047 s.
[2 marks] The number of roots of the polynomial
Equation
\[P(s)=s^{4}+2s^{3}+3s^{2}+4s+5\]
that lie in the right half of the \(s\)-plane is
- 0
- 1
- 2
- 3
Solution
Construct the Routh array (Chapter 11). The first two rows are the alternating coefficients:
Equation
\[\begin{array}{c|ccc} s^{4} & 1 & 3 & 5\\ s^{3} & 2 & 4 & 0 \end{array}\]
The \(s^{2}\) row:
Equation
\[b_1=\frac{2(3)-1(4)}{2}=\frac{6-4}{2}=1,\qquad b_2=\frac{2(5)-1(0)}{2}=5\]
The \(s^{1}\) row:
Equation
\[c_1=\frac{1(4)-2(5)}{1}=4-10=-6\]
The \(s^{0}\) row is \(5\). The first column reads \(1,\,2,\,1,\,-6,\,5\). The sign changes from \(+1\) to
\(-6\) and back from \(-6\) to \(+5\), giving two sign changes, hence two roots in the right half-plane.
Direct root extraction agrees: the roots are \(-1.288\pm j0.858\) and \(+0.288\pm j1.416\).
C
Final Answer
Correct answer: (C) 2.
[2 marks] A plant \(G(s)=1/(s+2)\) is controlled by a PI controller
\(G_c(s)=K_p+K_i/s\) in a unity-negative-feedback loop. The gains are chosen so that the
closed-loop poles have \(\omega_n=4\) rad/s and \(\zeta=0.5\). The value of \(K_p\) is _____.
Solution
The loop transfer function is (Chapter 19)
Equation
\[G_c(s)G(s)=\frac{K_ps+K_i}{s}\cdot\frac{1}{s+2}=\frac{K_ps+K_i}{s(s+2)}\]
The characteristic equation \(1+G_cG=0\) becomes
Equation
\[s(s+2)+K_ps+K_i=0 \Rightarrow s^{2}+(2+K_p)s+K_i=0\]
Match this against \(s^{2}+2\zeta\omega_n s+\omega_n^{2}=0\):
Equation
\[K_i=\omega_n^{2}=16,\qquad 2+K_p=2\zeta\omega_n=2(0.5)(4)=4\]
The integral action also raises the system to Type 1, so the steady-state error to a step becomes zero.
✓
Final Answer
Correct answer: \(K_p=2\) (with \(K_i=16\)).
[2 marks] A unity-negative-feedback system has the open-loop transfer function
Equation
\[G(s)H(s)=\frac{K(s+1)}{s(s-1)},\qquad K \gt 0\]
The closed-loop system is stable for
- all \(K \gt 0\)
- \(K \gt 1\)
- \(0 \lt K \lt 1\)
- no value of \(K\)
Solution
The open loop has one pole in the right half-plane, at \(s=+1\), so \(P=1\). By the Nyquist
criterion (Chapter 16) stability requires \(N=-P=-1\), that is one
anticlockwise encirclement of the \(-1\) point. The algebraic form of the same condition is
simplest here. The characteristic equation is
Equation
\[s(s-1)+K(s+1)=0 \Rightarrow s^{2}+(K-1)s+K=0\]
For a quadratic, all roots lie in the left half-plane exactly when every coefficient is positive:
Equation
\[K-1 \gt 0 \ \text{and}\ K \gt 0 \Rightarrow K \gt 1\]
Checking two values confirms it: at \(K=2\) the roots are \(-0.5\pm j1.323\), stable; at \(K=0.5\)
they are \(+0.25\pm j0.661\), unstable. The gain must therefore be pushed above unity for the
Nyquist plot to encircle \(-1\) once in the anticlockwise sense.
B
Final Answer
Correct answer: (B) \(K \gt 1\).
[2 marks] A system is described in state-space form by
Equation
\[A=\begin{bmatrix}0 & 1\\ -3 & -4\end{bmatrix},\quad B=\begin{bmatrix}0\\ 1\end{bmatrix},\quad C=\begin{bmatrix}2 & 0\end{bmatrix},\quad D=0\]
The DC gain of the system, that is the value of its transfer function at \(s=0\), is _____
(round to three decimal places).
Solution
The transfer function is \(G(s)=C(sI-A)^{-1}B+D\) (Chapter 23).
Equation
\[sI-A=\begin{bmatrix}s & -1\\ 3 & s+4\end{bmatrix},\qquad \det(sI-A)=s(s+4)+3=s^{2}+4s+3\]
Equation
\[(sI-A)^{-1}=\frac{1}{s^{2}+4s+3}\begin{bmatrix}s+4 & 1\\ -3 & s\end{bmatrix}\]
Equation
\[(sI-A)^{-1}B=\frac{1}{s^{2}+4s+3}\begin{bmatrix}1\\ s\end{bmatrix}\]
Equation
\[G(s)=\begin{bmatrix}2 & 0\end{bmatrix}\frac{1}{s^{2}+4s+3}\begin{bmatrix}1\\ s\end{bmatrix}=\frac{2}{s^{2}+4s+3}\]
Setting \(s=0\):
Equation
\[G(0)=\frac{2}{3}=0.667\]
The poles are at \(s=-1\) and \(s=-3\), both stable, so this DC gain is also the final value of the unit-step response.
✓
Final Answer
Correct answer: 0.667.
[2 marks] Consider the state model \(\dot{x}=Ax+Bu\) with
Equation
\[A=\begin{bmatrix}0 & 1\\ -2 & -3\end{bmatrix},\qquad B=\begin{bmatrix}1\\ \alpha\end{bmatrix}\]
The system fails to be completely state controllable for
- \(\alpha=0\) only
- \(\alpha=1\) or \(\alpha=2\)
- \(\alpha=-1\) or \(\alpha=-2\)
- every real \(\alpha\)
Solution
For a second-order single-input system the controllability matrix is \(Q_c=[B\ \ AB]\) (Chapter 25).
Equation
\[AB=\begin{bmatrix}0 & 1\\ -2 & -3\end{bmatrix}\begin{bmatrix}1\\ \alpha\end{bmatrix}=\begin{bmatrix}\alpha\\ -2-3\alpha\end{bmatrix}\]
Equation
\[Q_c=\begin{bmatrix}1 & \alpha\\ \alpha & -2-3\alpha\end{bmatrix}\]
Equation
\[\det Q_c = 1(-2-3\alpha)-\alpha(\alpha) = -\alpha^{2}-3\alpha-2 = -(\alpha^{2}+3\alpha+2)\]
Controllability is lost when the determinant vanishes:
Equation
\[\alpha^{2}+3\alpha+2=0 \Rightarrow (\alpha+1)(\alpha+2)=0 \Rightarrow \alpha=-1 \ \text{or}\ \alpha=-2\]
These are precisely the two eigenvector directions of \(A\), whose eigenvalues are \(-1\) and
\(-2\). When \(B\) is aligned with one eigenvector, the other mode receives no input and cannot
be steered.
C
Final Answer
Correct answer: (C) \(\alpha=-1\) or \(\alpha=-2\).
[1 mark] The open-loop transfer function of a system is
Equation
\[G(s)H(s)=\frac{100}{s^{2}(s+10)}\]
The slope of the asymptotic Bode magnitude plot at frequencies well below 10 rad/s is
- 0 dB/decade
- \(-20\) dB/decade
- \(-40\) dB/decade
- \(-60\) dB/decade
Solution
Write the loop gain in time-constant form (Chapter 14):
Equation
\[G(s)H(s)=\frac{100}{s^{2}\cdot 10(1+0.1s)}=\frac{10}{s^{2}(1+0.1s)}\]
For \(\omega\) well below the corner frequency \(1/0.1=10\) rad/s, the factor \((1+j0.1\omega)\) is
approximately 1 and
Equation
\[|G(j\omega)H(j\omega)|\approx\frac{10}{\omega^{2}} \Rightarrow 20\log_{10}|GH| \approx 20\log_{10}10-40\log_{10}\omega\]
A decade increase in \(\omega\) changes \(40\log_{10}\omega\) by 40, so the magnitude falls by 40 dB per
decade. Each pole at the origin contributes \(-20\) dB/decade, and this system has two, so its
type number is 2.
C
Final Answer
Correct answer: (C) \(-40\) dB/decade.