23-Chem-A6 Process Dynamics and Control · May 2016
Question 6 of 8: Draining Tank — Step and Impulse Response
Nivaar worked solution (AI-drafted; not reviewed by a licensed engineer)
Notes on this paper
Reference texts: D. E. Seborg, T. F. Edgar, D. A. Mellichamp & F. J. Doyle III, Process Dynamics and Control (4th ed., Wiley) — Laplace-domain modelling, linearisation, transfer functions, feedback stability, frequency response and IMC design; G. Stephanopoulos, Chemical Process Control: An Introduction to Theory and Practice (Prentice Hall) — Routh test, Nyquist criterion and dead-time systems; D. R. Coughanowr & S. E. LeBlanc, Process Systems Analysis and Control (3rd ed., McGraw-Hill) — first-order thermal/level dynamics, step/ramp/pulse response and block-diagram algebra. Standard control conventions (deviation variables; unity valve/sensor gains unless stated) are used throughout. This sitting is an open-book paper of eight problems of which any five constitute a complete exam; all eight are solved here.
Problem 6: Draining Tank — Step and Impulse Response (20%)
Given. A gravity-drained tank with a linear resistance:
Quantity
Symbol
Value
Tank cross-section
$A$
$1\ \mathrm{m^2}$
Initial level
$h_0$
$1\ \mathrm{m}$
Outflow law
$F_1$
$R_1 h,\ R_1=1\ \mathrm{m^2/min}$
Inlet pipe length / area
$L,\ A_p$
$1.0\ \mathrm{m},\ 0.01\ \mathrm{m^2}$
Find. The deviation level $\delta h(t)=h(t)-h_0$ for (a) a unit step and (b) a unit impulse in $F_0$.
Problem 6: single gravity-drained tank, imposed inflow $F_0$ through the inlet pipe, level $h$ ($A=1\ \mathrm{m^2}$, $h_0=1\ \mathrm{m}$), linear outflow $F_1=R_1 h$ through valve $R_1=1\ \mathrm{m^2/min}$.
Check — role of the inlet-pipe dimensions
The forcing is specified as a change in the inlet flow $F_0$ itself, so $F_0$ is imposed directly on the tank and the level dynamics are first order; the pipe length and area do not enter this response. They would matter only if the forcing were an upstream pressure step, in which case the liquid slug in the pipe adds a fluid inertance $I=\rho L/A_p=1000(1.0)/0.01=1\times10^{5}$ (SI), giving a second-order inertial mode. We solve the problem as posed (flow-forced, first order) and note the inertance for completeness.
Approach. Write the linear tank balance, reduce it to standard first-order form to read off $\tau$ and gain $K$, then apply the standard step and impulse solutions.
Model and standard form. $A\,\dfrac{dh}{dt}=F_0-R_1 h$. In deviation variables about the initial steady state ($F_{0s}=R_1 h_0=1\times1=1\ \mathrm{m^3/min}$): $\dfrac{dh'}{dt}=F_0'-h'$. This is standard first order with $$\tau=\dfrac{A}{R_1}=1\ \mathrm{min},\qquad K=\dfrac{1}{R_1}=1,$$ so $\dfrac{H'(s)}{F_0'(s)}=\dfrac{K}{\tau s+1}=\dfrac{1}{s+1}.$
(a) Unit-step response. For $F_0'=1$, the first-order step response is $$\delta h(t)=K\!\left(1-e^{-t/\tau}\right)=\boxed{1-e^{-t}}\ \mathrm{m}.$$ The level rises from $1\ \mathrm{m}$ toward $h(\infty)=2\ \mathrm{m}$ with a $1\ \mathrm{min}$ time constant (~99% settled in $5\tau=5\ \mathrm{min}$).
(b) Unit-impulse response. A Dirac impulse has $F_0'(s)=1$, so $\delta H(s)=\dfrac{1}{s+1}$ and $$\delta h(t)=\dfrac{K}{\tau}e^{-t/\tau}=\boxed{e^{-t}}\ \mathrm{m}.$$ The unit-area flow impulse injects a volume $\int F_0'\,dt=1\ \mathrm{m^3}$ instantaneously, so the level jumps by $\Delta h=V/A=1\ \mathrm{m}$ to $h=2\ \mathrm{m}$ at $t=0^+$, then decays back to $1\ \mathrm{m}$.
Consistency check. The impulse response is the time-derivative of the step response: $\dfrac{d}{dt}\!\left[1-e^{-t}\right]=e^{-t}$ — as it must be for a linear system. Both share the $\tau=1\ \mathrm{min}$ decay.
Problem 6(a): unit-step response. The level climbs from $1\ \mathrm{m}$ to the new steady state $2\ \mathrm{m}$ as $1+(1-e^{-t})$, effectively settled after $\approx5\ \mathrm{min}$.
Problem 6(b): unit-impulse response. The injected $1\ \mathrm{m^3}$ raises the level instantly to $2\ \mathrm{m}$, which then decays as $1+e^{-t}$ back to $1\ \mathrm{m}$.
Quantity
Value
Time constant $\tau=A/R_1$
$1\ \mathrm{min}$
Gain $K=1/R_1$
$1$
(a) Step response
$\delta h(t)=1-e^{-t}$; $h(\infty)=2\ \mathrm{m}$
(b) Impulse response
$\delta h(t)=e^{-t}$; peak $h=2\ \mathrm{m}$ at $t=0^+$