联系方式

  • QQ:99515681
  • 邮箱:99515681@qq.com
  • 工作时间:8:00-23:00
  • 微信:codinghelp

您当前位置:首页 >> C/C++编程C/C++编程

日期:2023-11-29 09:22

PX390 2023-24, Assignment 3

1 Intro

The purpose of this exercise is to further expand your knowledge of C and your

ability to understand and test numerical code. You are given a code which

should solve two coupled partial differential equations. The code is, however,

broken and needs to be fixed. Your task is to read the specification, understand

the program, and ‘debug’ it. At the end of this task the program should do

what these specifications say it is meant to do. Some of these problems with

the code are basic C programming errors, other are more subtle mismatches

between the required and actual code behaviour.

You should submit a single C source file with a list of corrected bugs, at the

top of the C source file (in comments), where I have added a placeholder (this

description will not be marked but helps both of us keep track of what you have

changed). You may submit your code via the link in the assignment section of

the moodle page.

The code must compile and generate no warnings when compiled with

gcc -Wall -Werror -std=c99 -o assign3 assign3.c -lm

There are a finite number of bugs, but we will not tell you how many: make

sure that your code actually works by testing it rather than just looking for

obvious errors. Some errors may be repeated in multiple statements, or may

effect multiple lines of code.

1.1 How is it marked

Marks are given for the absence of each bug (in an otherwise correctly working

code) and for the code compiling correctly. You may lose marks in more than one

category for sufficiently severe problems. A correctly working code will contain

a stable and consistent finite difference method for solving the equations below.

1.2 Tips

1. The compiler is your friend. Start with the first warning/error messages

and work through them until there are none left.

2. Some bugs are hard to catch just by inspecting the code. Adding printf

statements is usually the easiest way to check that the code is doing what

you think it is: check the maths at the first timestep, for example.

3. You can often catch bugs in numerical code by looking at the output

graphically: does it do something strange at the boundaries? A variety of

tools (Matlab/matplotlib/Origin) exist which can take numerical output

and plot it for you. This is one of the things you should learn while doing

this assignment as it will be essential later on.

1

4. Learn how to use debuggers (gdb). Memory checking tools like valgrind

can help catch issues to do with reading/writing into an incorrect memory

location.

5. It is expected that you will allocate arrays on the heap and that the

allocated memory is released (freed) after use.

2 Specification

The code must use a simple, convergent, finite difference scheme to solve the

coupled differential equation for real-valued functions u and v with

?u

?t ? K

?

2u

?x2

? v(u + 1) = 0 (1)

?v

?t ? K

?

2v

?x2

+ u(v + 1) = 0 (2)

The equation is to be solved on a 1D x domain with x ∈ [0, length], as an

initial value problem. The domain length, number of grid points, length of time

over which to solve and the coefficients are defined at the start of the main.

The initial condition is u(x, 0) = 1.0 + sin(2πx/[length]), v(x, 0) = 0.0. The

first grid point is at x = 0, and the final point at x =[length]= (nx?1)×δx. The

function and its derivatives are periodic, so periodic boundary conditions should

be enforced, that is, u(x = 0) = u(x =[length]) and v(x = 0) = v(x =[length]).

The provided code may take a slightly different approach to periodic boundaries

than your used to so examine closely. Please note that at no point should the

values of u or v blow up to infinity.

2.1 Output

The code outputs simulation data at a fixed interval in time equal to the chosen

time step dt: it should output the initial values (at t=0), and at dt, 2dt, 3dt . . .

(but not necessarily the final value).

The time t, x coordinate, and u and v are written in that order in a single

output line for each gridpoint. Do not add extra comments/blank lines to

the output, which should contain only numbers! Remember that I will need

to read the numbers, so if it is output in an unreadable format then you’ll lose

marks.

2


相关文章

版权所有:留学生编程辅导网 2020 All Rights Reserved 联系方式:QQ:99515681 微信:codinghelp 电子信箱:99515681@qq.com
免责声明:本站部分内容从网络整理而来,只供参考!如有版权问题可联系本站删除。 站长地图

python代写
微信客服:codinghelp