Computational Techniques in Water Resources ENCE 370 Chapter 3: Method of Characteristics | Elective I | Complete Guide (IOE New Syllabus) | Notes, Numerical Problems & Solutions | MOC, Water Hammer, Open Channel Flow, Unsteady Pipe Flow
Method of Characteristics
ELECTIVE I: COMPUTATIONAL TECHNIQUES IN WATER RESOURCES ENGINEERING (ENCE 370)
Chapter 3: Method of Characteristics
12 Hours  |  16 Marks

Computational Techniques in Water Resources ENCE 370 Chapter 3: Method of Characteristics

Overview of the Method of Characteristics Notes

The Method of Characteristics (MOC) is one of the most authoritative numerical and analytical techniques used in computational techniques in water resources to analyze highly dynamic, transient flow phenomena. Specifically designed to handle hyperbolic partial differential equations, the Method of Characteristics transforms complex non-linear PDEs governing water hammer transients in closed conduits and gradually varied unsteady open channel flows into ordinary differential equations (ODEs) along specific characteristic lines ($C^+$ and $C^-$).

With a major weightage of 16 marks in the IOE engineering syllabus, this chapter provides students with deep insight into hydraulic transient modeling, wave speed calculation, boundary conditions (valves, reservoirs, junctions), rectangular grid algorithms, and numerical stability criteria.

Syllabus: Method of Characteristics (16 Marks)

3 Method of Characteristics (MOC) (12 hours)

3.1 Characteristics; Solution requirements; Advantages; Limitations

3.2 Characteristics equations to open channel and pipe flow

3.3 Finite difference solution of characteristics equations of unsteady pipe flow

3.4 Algorithm for solving water hammer problem using rectangular grid

3.5 Stability of MOC solution

3.6 Boundary conditions for unsteady pipe flow problem

3.7 MOC for 1D gradually varied unsteady open channel flow

3.8 Numerical (FDM) solution of MOC for 1D unsteady open channel flow

Fundamentals of the Method of Characteristics in Water Resources

In hydraulic engineering, fluid transients caused by rapid valve closures, pump trips, or sudden surge operations generate high-frequency pressure waves known as water hammer. Similarly, dam breaks and gate operations initiate unsteady wave propagation in open channels. Direct numerical integration of hyperbolic partial differential equations describing these events can introduce numerical dispersion and instability. The Method of Characteristics solves this challenge by reformulating partial derivatives into total derivatives along pathlines called characteristic curves.

1. Concept, Solution Requirements, Advantages, and Limitations of the Method of Characteristics

The core concept behind the Method of Characteristics involves finding trajectories along the space-time ($x-t$) plane along which information about physical variables (such as piezometric head $H$ and discharge $Q$, or water depth $y$ and velocity $v$) propagates at specific characteristic wave speeds ($a$ or $c$).

A. Solution Requirements

  • The governing partial differential equations must be strictly hyperbolic in nature.
  • Accurate physical wave speed propagation velocities (e.g., speed of sound in water through elastic pipes $a$, or shallow water wave celerity $c = \sqrt{gy}$) must be determined.
  • Proper spatial discretization ($\Delta x$) and temporal discretization ($\Delta t$) satisfying the Courant criterion must be chosen.
  • Well-defined boundary conditions at all physical terminals (reservoirs, valves, junctions, dead ends) must be established.

B. Advantages of the Method of Characteristics

  • High Accuracy: Accurately tracks steep wave fronts and sharp pressure peaks without introducing excessive numerical smoothing or artificial numerical diffusion.
  • Physical Clarity: Preserves the physical concept of wave propagation along directional lines ($C^+$ and $C^-$).
  • Flexibility with Boundary Conditions: Simplifies complex boundary evaluations by isolating upstream and downstream wave contributions.

C. Limitations of the Method of Characteristics

  • Restricted primarily to 1D and simple 2D hyperbolic transient systems.
  • Requires strict adherence to grid spacing constraints; handling variable pipe diameters or complex networks requires spatial interpolation, which may introduce interpolation errors.

2. Governing Equations and Characteristics Formulation for Pipe Flow

Unsteady flow in elastic closed conduits is governed by 1D hyperbolic continuity and momentum equations:

$$\text{Continuity: } \quad \frac{\partial H}{\partial t} + \frac{a^2}{g A} \frac{\partial Q}{\partial x} = 0$$ $$\text{Momentum: } \quad \frac{\partial Q}{\partial t} + g A \frac{\partial H}{\partial x} + \frac{f Q |Q|}{2 D A} = 0$$

Where $H$ is the piezometric head, $Q$ is discharge, $a$ is wave speed, $g$ is acceleration due to gravity, $D$ is pipe diameter, $A$ is cross-sectional area, and $f$ is the Darcy-Weisbach friction factor.

By applying linear combination multipliers ($\lambda$) using the Method of Characteristics, these equations transform into two distinct sets of ordinary differential equations valid along the positive ($C^+$) and negative ($C^-$) characteristic lines:

$$\text{Along } C^+ \left(\frac{dx}{dt} = +a\right): \quad \frac{g A}{a} \frac{dH}{dt} + \frac{dQ}{dt} + \frac{f Q |Q|}{2 D A} = 0$$ $$\text{Along } C^- \left(\frac{dx}{dt} = -a\right): \quad -\frac{g A}{a} \frac{dH}{dt} + \frac{dQ}{dt} + \frac{f Q |Q|}{2 D A} = 0$$

3. Finite Difference Solution and Rectangular Grid Algorithm for Water Hammer

To implement the Method of Characteristics on a digital computer, the $x-t$ continuum is discretized into a rectangular grid. Let node $P$ be an unknown interior grid point at time $t + \Delta t$, calculated from adjacent known grid points $A$ (upstream) and $B$ (downstream) at time $t$.

Integrating the differential equations along $C^+$ ($AP$) and $C^-$ ($BP$) using first-order finite differences yields algebraic compatibility equations:

$$C^+ \text{ Equation: } \quad H_P = C_P – B_P \cdot Q_P$$ $$C^- \text{ Equation: } \quad H_P = C_M + B_P \cdot Q_P$$

Where the constants $C_P$, $C_M$, and impedance factor $B_P$ are defined as:

$$B_P = \frac{a}{g A}, \quad C_P = H_A + B_P Q_A – \frac{f \Delta x}{2 g D A^2} Q_A |Q_A|, \quad C_M = H_B – B_P Q_B + \frac{f \Delta x}{2 g D A^2} Q_B |Q_B|$$

Solving simultaneously for an interior node $P$ gives the explicit solutions for head ($H_P$) and discharge ($Q_P$):

$$H_P = \frac{C_P + C_M}{2}, \qquad Q_P = \frac{C_P – C_M}{2 B_P}$$

4. Stability Analysis and Courant Condition in the Method of Characteristics

The stability of the numerical solution generated by the Method of Characteristics requires that the numerical domain of dependence completely encloses the physical domain of dependence. This requires satisfying the **Courant-Friedrichs-Lewy (CFL)** condition:

$$C_r = \frac{a \cdot \Delta t}{\Delta x} = 1$$

When $C_r = 1$, characteristic lines pass exactly through grid intersections ($A$ and $B$). If $C_r < 1$, spatial grid interpolations are necessary, which can introduce numerical damping. If $C_r > 1$, the numerical solution becomes unstable and experiences catastrophic error growth.

5. Boundary Conditions for Unsteady Pipe Flow Problems

At boundaries (e.g., reservoirs, downstream control valves, series pipe junctions), only one characteristic line enters from the internal grid domain. Thus, an auxiliary boundary relation is required:

  • Constant Head Reservoir Boundary (Upstream): $H_P = H_{res}$. The unknown discharge $Q_P$ is computed directly using the incoming $C^-$ equation: $Q_P = \frac{H_{res} – C_M}{B_P}$.
  • Downstream Valve Boundary: Governed by the orifice equation $Q_P = \tau \cdot Q_0 \sqrt{\frac{H_P}{H_0}}$, where $\tau$ is the non-dimensional valve opening ratio. Combining this with the incoming $C^+$ equation yields a quadratic equation in $Q_P$.
  • Dead End / Closed Valve: $Q_P = 0$, giving $H_P = C_P$.

6. Method of Characteristics for 1D Unsteady Open Channel Flow

For open channel flow governed by the full 1D Saint-Venant equations, the characteristic directions are defined by the flow velocity $v$ and shallow water celerity $c = \sqrt{gy}$:

$$\text{Characteristic Slopes: } \quad \frac{dx}{dt} = v \pm c = v \pm \sqrt{gy}$$

Integrating the compatibility equations along these curved $C^+$ and $C^-$ lines on an $x-t$ grid yields the time-marching solution for water depth $y_P$ and velocity $v_P$, making the Method of Characteristics a versatile tool for routing dam-break waves and hydrographs in rivers.

Lecture Notes: Method of Characteristics by Asst. Prof. Dr. Ram Krishna Regmi

Advertisement

*Disclaimer: This resource is intended strictly for educational purposes. All rights reserved by Pulchowk Campus faculty.

×

General Disclaimer

The academic resources provided on this page are compiled strictly as supplementary study materials for students studying Computational Techniques in Water Resources Engineering (ENCE 370). These lecture summaries, model outlines, and PDF documents are structured to enhance learning and preparation for IOE engineering assessments.

Every reasonable effort is made to maintain factual accuracy and academic clarity. However, students are strongly advised to refer to primary academic textbooks and consult faculty members at Pulchowk Campus for official syllabus guidelines and assessments.

We strictly adhere to copyright and intellectual property protections. If you are a copyright holder or university instructor and wish to request attribution adjustments or content removal, please reach out directly via our contact portal.

Scroll to Top