GATE Civil: Differential Equations (ODEs, Laplace, heat and wave equations), previous-year questions solved, part 2
Part 2 of 2 of the Differential Equations hub: 10 more official GATE CE questions on partial differential equations: Laplace, heat, wave and classification, each solved step by step and checked against the official answer key. Part 1 covers ordinary differential equations.
Questions from the ten official GATE CE papers, sessions CE1 and CE2 of GATE 2022 (IIT Kharagpur), 2023 (IIT Kanpur), 2024 (IISc Bengaluru), 2025 (IIT Roorkee) and 2026 (IIT Guwahati); solutions by Skill-Lync. · Updated 17 Nov 2026 · 10 of 20 questions on this page
Which partial differential equation does z = sin(y + it) + cos(y - it) satisfy?
Consider the following expression:
where , , and are variables, and is a complex number. The partial differential equation derived from the above expression is
- (A)
- (B)
- (C)
- (D)
Answer: (A) Official key: A
- Write and , so .
- With respect to : and .
- With respect to , the chain rule brings a factor from and from : , and differentiating again, .
- Adding the two results: . This is the Laplace equation, option (A). In general any satisfies it.
Trap: (B) is the wave equation, which belongs to real arguments ; the factor squares to and flips the sign. The first-order options each remove only one of the two functions: and , neither of which is zero.
Laplace's equation governs steady seepage through soil. The flow nets drawn under concrete weirs and through earth dams, and seepage software such as PLAXIS or GeoStudio SEEP/W, solve exactly this equation for the total head to get seepage quantity and exit gradients.
If f satisfies the Laplace equation on a unit disc and equals 3 on the boundary, what is f(0, 0)?
The function satisfies the Laplace equation
on a circular domain of radius with its center at point with coordinates , . The value of this function on the circular boundary of this domain is equal to 3.
The numerical value of is:
- (A)0
- (B)2
- (C)3
- (D)1
Answer: (C) 3 Official key: C
- A solution of is a harmonic function, and harmonic functions have the mean value property: the value at the centre of any circle inside the domain equals the average of the values on that circle.
- On the boundary the function is 3 everywhere, so the average is 3 and .
- Directly: satisfies and the boundary condition, and this problem has only one solution (by the maximum principle a harmonic function takes its largest and smallest values on the boundary, both 3 here). So and (C) 3.
Trap: choosing 0 because the Laplacian is zero. is a statement about curvature, not about the value of . A Gauss-Seidel relaxation on a grid over the disc, with the boundary held at 3, converges to 3.000 at the centre.
Steady seepage and steady heat conduction obey Laplace's equation, so the same averaging property holds: the head at an interior point of a flow net under a weir is the average of its surroundings, which is exactly the update that finite-difference seepage solvers repeat until the heads settle.
What is the steady-state pollutant concentration at the centre of a 1-D reservoir with no-flux ends?
The concentration of pollutants in a one-dimensional reservoir at position and time satisfies the diffusion equation
on the domain , where is the diffusion coefficient of the pollutants. The initial condition is defined by the step-function shown in the figure.
The boundary conditions of the problem are given by at the boundary points and at all times. Consider , , and .
The steady-state concentration at the center of the reservoir (in ) is ___________. (in integer)
Answer: 2 µmol/m Official key: 2 to 2
- The condition at both ends means no pollutant crosses either end. The reservoir is sealed, so the total amount never changes.
- Initial amount: .
- At steady state , so and is a straight line; with zero slope at the ends it is a constant .
- Mass balance: , so 2 µmol/m everywhere, including . only sets how fast this is reached (time scale s), not the final value.
Trap: answering 0 because the step does not reach at , or 5 (the value ) by forgetting that the pollutant spreads over the full 10 m. An explicit finite-difference run of the equation settles at 2.000 at the centre.
This is the sealed-tank limit of pollutant mixing: a spill into a closed bay, a sump or a long detention tank eventually spreads to a uniform concentration equal to the total mass divided by the length or volume, and the diffusion time, length squared over D, tells how long that takes.
Square plate with one side at 100 degC and three at 0 degC: centre temperature by superposition?
The steady-state temperature distribution in a square plate ABCD is governed by the 2-dimensional Laplace equation. The side AB is kept at a temperature of 100 C and the other three sides are kept at a temperature of 0 C. Ignoring the effect of discontinuities in the boundary conditions at the corners, the steady-state temperature at the center of the plate is obtained as C. Due to symmetry, the steady-state temperature at the center will be same ( C), when any one side of the square is kept at a temperature of 100 C and the remaining three sides are kept at a temperature of 0 C. Using the principle of superposition, the value of is ________ (rounded off to two decimal places).
Answer: 25.00 Official key: 24.90 to 25.10
- The Laplace equation is linear, so solutions with different boundary values can be added.
- Add the four cases in which each side in turn is at 100 C and the rest at 0 C. The sum has all four sides at 100 C.
- With every side at 100 C the steady solution is everywhere, including the centre.
- Each case contributes at the centre, so and 25.00 C.
- Check: the Fourier series solution at the centre sums to 25.000, and a 21 by 21 finite-difference grid gives 25.000.
Trap: treating it as 1-D conduction between the hot side and the opposite cold side gives the midpoint value 50 C. The two side walls at 0 C also pull the centre down, which the superposition argument captures.
The Laplace equation governs steady seepage under dams and sheet piles, so the flow nets drawn for weirs on permeable foundations follow the same mathematics. Superposition of simple cases is also how engineers combine pressure distributions from several sources in groundwater well-field design.
Heat diffusion in a 5 cm rod with both ends at 0 degC: what is the decay constant beta of the series solution?
A 5 cm long metal rod AB was initially at a uniform temperature of C. Thereafter, temperature at both the ends are maintained at 0 C. Neglecting the heat transfer from the lateral surface of the rod, the heat transfer in the rod is governed by the one-dimensional diffusion equation , where D is the thermal diffusivity of the metal, given as .
The temperature distribution in the rod is obtained as
where x is in cm measured from A to B with at A, t is in s, are constants in C, T is in C, and is in .
The value of (in , rounded off to three decimal places) is ________.
Answer: 0.395 Official key: 0.394 to 0.396
- Substitute one term, with cm, into the diffusion equation.
- Left side: .
- Right side: .
- Equating: , so .
- 0.395 . The same works for every , which is why it can be pulled out of the sum.
Trap: forgetting to square the factor from the second derivative gives , outside the key. Only odd appear because the initial temperature is uniform: the even sine coefficients integrate to zero.
The same diffusion equation and series solution govern Terzaghi's one-dimensional consolidation theory, where excess pore pressure decays with a time factor, and heat of hydration dissipating from mass concrete in dams. The decay constant sets how quickly settlement or temperature rise dies away.
What is the general solution of the second-order PDE d2u/dx2 = 2 for an unknown function u(x, y)?
The second-order differential equation in an unknown function is defined as
Assuming , , and , the general solution of the above differential equation is
- (A)
- (B)
- (C)
- (D)
Answer: (B) Official key: B
- Only -derivatives appear, so behaves like a constant and every constant of integration is an arbitrary function of .
- Integrate once in : .
- Integrate again: , with two arbitrary functions of for a second-order equation: (B).
- Check the others: (A) and (C) give , and (D) gives , which are not 2 for an arbitrary .
Trap: adding a function of as the “constant” of integration. When you integrate with respect to , whatever is constant must not depend on , so it can only be a function of .
Integrating twice and fixing the constants is the double integration method for beam deflection from the bending moment equation. In plate bending, seepage or heat flow fields the constants become functions of the other coordinate, exactly as here, and are fixed by the boundary conditions.
Is x f_xx + y f_yy = (x^2 + y^2)/2 elliptic, parabolic or hyperbolic in different regions of the x-y plane?
For the following partial differential equation,
which of the following option(s) is/are CORRECT?
- (A)elliptic for and
- (B)parabolic for and
- (C)elliptic for and
- (D)hyperbolic for and
Answer: (A, D) Official key: A;D
- Compare with : here , , . The right-hand side does not affect the type.
- Discriminant: . Negative means elliptic, zero parabolic, positive hyperbolic.
- (A) True: for , , , elliptic.
- (B) False: the same region is elliptic, not parabolic.
- (C) False: on the discriminant is 0, so the equation is parabolic there (the term vanishes).
- (D) True: for , , , hyperbolic.
- Correct options: (A) and (D).
Trap: the type of a PDE with variable coefficients can change from region to region; it is decided point by point by the sign of , so do not classify the whole equation once.
The type of a PDE decides how it is solved numerically: elliptic equations such as steady seepage under a dam or Laplace flow nets need boundary values all round, while parabolic consolidation and hyperbolic wave or flood-routing equations are marched forward in time in tools like PLAXIS or HEC-RAS.
Laplace's equation on a 2 m square with three edges at T = 50 and one at T = 100: what is T at the centre?
A partial differential equation
is defined for the two-dimensional field , inside a planar square domain of size 2 m × 2 m. Three boundary edges of the square domain are maintained at value , whereas the fourth boundary edge is maintained at .
The value of at the center of the domain is
- (A)50.0
- (B)62.5
- (C)75.0
- (D)87.5
Answer: (B) 62.5 Official key: B
- Laplace’s equation is linear, so split the problem: , where is 0 on three edges and 50 on the fourth.
- Let be the solution with one edge at 1 and the others at 0. Adding the four rotated copies of gives the solution with all edges at 1, which is 1 everywhere.
- By symmetry the four copies are equal at the centre, so there, and .
- (B) 62.5. A 21 by 21 finite-difference solution gives 62.5000.
Trap: (C) 75 averages the two boundary values as if the hot and cold boundaries were equal in length. The centre value is the average over the whole boundary, of which only one quarter is at 100.
Laplace's equation governs steady seepage through and under dams, steady heat flow through a concrete section and groundwater head in an aquifer. The superposition and averaging used here is how flow-net solutions are checked against finite-difference output from seepage software.
Which of these equations are second-order, linear, homogeneous partial differential equations?
Which of the following equations belong/belongs to the class of second-order, linear, homogeneous partial differential equations:
- (A)
- (B)
- (C)
- (D)
Answer: (B) Official key: B
- Three tests. Order: the highest derivative. Linear: and its derivatives appear only to the first power and are not multiplied together. Homogeneous: every term contains (no forcing term).
- (A) is second order and linear, but contains no : it is a forced (non-homogeneous) wave equation. Not in the class.
- (B) is the three-dimensional Laplace equation: second order, linear, every term contains and the right side is 0. In the class.
- (C) is the advection equation: linear and homogeneous, but only first order.
- (D) squares , so it is nonlinear, even though it is second order and has no forcing term.
- Only (B) belongs to the class.
Trap: (A) passes the order and linearity tests, and the term is easy to dismiss. Any term free of makes the equation non-homogeneous.
Laplace's equation governs steady seepage: flow nets drawn for weirs and barrages on permeable foundations, earth dams and sheet-pile cofferdams are graphical solutions of it, and seepage software such as SEEP/W or PLAXIS solves the same equation by finite elements to get uplift and exit gradients.
Which functions are possible solutions of the partial differential equation u_xx - u_yy = 0?
A partial differential equation is given below.
Possible solution(s) is/are:
- (A)
- (B)
- (C)
- (D)
Answer: (A, C) Official key: A;C
- This is the wave equation . For : and , so it works only for (d'Alembert: ).
- (A) , : . TRUE.
- (B) : but . FALSE.
- (C) : both second derivatives equal . TRUE.
- (D) : but . FALSE.
Trap: (B) and (D) are travelling waves too, but they solve (wave speed ratio 2), not this equation. The answer is (A, C).
This is the one-dimensional wave equation. It governs stress waves travelling down a pile while it is driven, which is the basis of high-strain dynamic pile testing and wave equation analysis used to confirm pile capacity on bridge and metro projects, and of seismic waves travelling through soil layers.