- Asked 3/12
- 2079 Shrawan · 8 marks
- 2075 Bhadra · 5+3 marks
- 2074 Bhadra · 6 marks
Steady 2D model: seepage under a dam
Short answer
Laplace equation governs steady seepage. Interior node: φ_{i,j}=¼(φ_{i+1,j}+φ_{i-1,j}+φ_{i,j+1}+φ_{i,j-1}). Fixed heads on upstream/downstream beds; no-flow boundaries use image points (φ_{i,j-1}=φ_{i,j+1}). Gauss–Seidel iteration solves potentials. Seepage per unit width: Q=K∑(φ_{i,j}−φ_{i+1,j}) across a vertical section.
Develop a steady state 2D model (suitable finite difference expression for homogeneous and isotropic aquifer) for the simulation of seepage under a dam. Also describe the iterative procedure for computing potential at each grid and seepage rate.
Physical situation
h1 (upstream) dam h2 (downstream)
~~~~~~~~~~~~~~~~ [=========] ~~~~~~~~~~~~~~
phi=h1 . . . . . no flow . . . . phi=h2
. . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . .
///////// impervious layer (no flow) /////
Water seeps under the dam through the foundation soil from the upstream bed (head ) to the downstream bed (head ). The flow is steady and two-dimensional in a vertical section.
Governing equation
Continuity for steady flow, , with Darcy's law , ( = potential or hydraulic head). For a homogeneous, isotropic aquifer ():
Finite difference expression
Cover the section by a square grid (). Central differences from Taylor's series:
Adding and setting :
The potential at each grid point is the average of its four neighbours.
Boundary conditions
- Upstream bed: ; downstream bed: (specified head).
- Base of the dam and the impervious layer: no flow, . Use an image (ghost) point: for a point on a horizontal no-flow boundary, so
- Far left and right ends: assumed no-flow (or fixed heads equal to , ), placed far enough from the dam.
Iterative procedure (Gauss-Seidel / Liebmann)
- Set up the grid and number the nodes; give the boundary nodes their values.
- Guess the potential at all interior nodes (for example, a linear variation from to ).
- Sweep the grid row by row. At each interior node compute with the equation above, using the latest available neighbour values. Optionally over-relax: , .
- Find the largest change over the grid.
- If it is greater than the tolerance (e.g. 0.001 m), repeat from step 3; otherwise stop.
- Seepage rate. Between two adjacent nodes the flow through one cell face (per unit length of dam) is
Choose a vertical section between two columns and below the dam; the total seepage is the sum over all rows:
(check with a different section; the totals must agree if the solution has converged, and also equal the flow entering at the upstream bed). 7. Print the potentials (to draw equipotential lines and the flow net) and the seepage in m/s per metre length of dam.
✓ Answer checked by Claude Opus 5.5 (high effort)
Related questions
- Continuity equation; seepage under a dam algorithm · Computational Techniques in Civil Engineering, chapter 7
- 2D groundwater FD scheme, boundary conditions · Computational Techniques in Civil Engineering, chapter 7
- Constitutive relation for 2D isotropic material · Computational Techniques in Civil Engineering, chapter 3
- Darcy fluxes around a central grid · Computational Techniques in Civil Engineering, chapter 7
- Finite difference method; explicit and implicit schemes · Computational Techniques in Civil Engineering, chapter 5 · asked 2×
- Courant condition; 1-D implicit coefficient matrix · Computational Techniques in Civil Engineering, chapter 7
- Seepage under embankment dam on pervious foundation · Hydropower Engineering, chapter 4
- Finite difference nodal temperatures in a square plate · Heat Transfer, chapter 2
- Cloud and grid computing concepts · Internet and Intranet, chapter 7 · asked 2×
- Impervious dam seepage discharge · Soil Mechanics, chapter 7
All Computational Techniques in Civil Engineering chapter 7 questions →