• Join over 1.2 million students every month
  • Accelerate your learning by 29%
  • Unlimited access from just £6.99 per month
Page
  1. 1
    1
  2. 2
    2
  3. 3
    3
  4. 4
    4
  5. 5
    5
  6. 6
    6
  7. 7
    7
  8. 8
    8
  9. 9
    9
  10. 10
    10
  11. 11
    11
  12. 12
    12
  13. 13
    13

Solving Equations by numerical methods.

Extracts from this document...

Introduction

Solving Equations by numerical methods. The fact that the vast majority of the numbers on the real number line are irrational (i.e. cannot be put in the form p/q, where p and q are integers) has an associated consequence that the vast majority of equations involving powers of x (i.e. polynomials) are insoluble by closed analytical techniques. This coursework's solution implies finding values of x say c1, c2-...c3 where f(c1)=0; f(c2)=0 and so on. Further transcendental equations and non- linear equations require numerical methods for their solution. It is to be noted that in no way are numerical methods inferior to analytical solutions, they are indeed the only practical solutions available, however on the down side no exact solution is possible and error bounds have to be placed on the solution given. In this coursework three main methods of solution from one-dimensional equations are given, each of the methods could be extended to the multi-dimensional case. The aim of this coursework is to show example of the methods in action, where these are successful and where they are not. A comparison between the methods will also be attempted. Method 1: Decimal Search SOLVING AN EQUATION WITH THIS METHOD Fig1. Shows the function f(x) = 12 ln(x) - x3/2. (Where ln(x) represents the natural logarithm of x.) To find the roots, the equation is: 12 ln(x) - x3/2 = 0 This equation can not be solved numerically because x appears inside the function ln(x) ...read more.

Middle

in the form x=g(x). A series of iterations attempts to improve on values of x; geometrically the method is relating on the intersection of the curve y=x with that of y=g(x). An initial value of x is chosen say x0. From this x value the value of the curve g(x) is obtained and a line drawn to the y=x curve, dropping down from this intersection point the value x1 is obtained from where the procedure continues as before. Eventually a point is reached simultaneous with the curve y=x and the curve g(x). In applying this method to the equation f(x) = 2ln(x-1) -x + 7.5 = 0, y = x f(x) = 2ln(x-1) - x + 7.5 x = 2ln(x - 1) + 7.5 = g(x) x = e^(0.5(x - 7.5)) + 1 = g(x) We immediately see the problem of this method. There are two choices for the iteration: 1. x n+1= 2ln(xn-1) + 7.5 2. xn+1= e^(0.5(xn-7.5)) + 1 1. Which of the choices of iteration for x should be used? Without a detailed analysis this question cannot be answered. However one criterion that could and should be used by the user is that for a successful iteration on an increasing function of x approaching the intersection with the curve y=x is that g'(x) should be less than 1 and greater than -1. For a decreasing function in this vicinity things are less clear cut. ...read more.

Conclusion

A simple program (see appendix) can be programmed into the calculator that generates the x-values, or alternatively spreadsheet software such as Excel can be used. Overall Conclusion. Decimal search is largely reliable, if not slow. It required little programming, and would only fail in situations where the use was not careful. However even in taking care it is possible to miss roots in highly oscillatory local situations. Newton Raphson also can present problems; one often has to conduct an analysis to find the initial point, (not always possible with more complicated equations) on the range of values of x one can get away with in the initial trial for convergence to a root. The conclusion here must also be that further research is required into this method. Fixed point estimation iteration presents real problems for the user. By experience he must select from a number of options the most likely iterative method (normally it appears to be safest to choose an iteration with a (1/n)th root in it. Also one cannot be sure which root if any your chosen iterate will converge too, one can however for increasing functions use the criteria of the gradient of the iterative function should be less than 1 in the vicinity of the root. In other examples such as xn+1=K (1-xn) chaotic functions arise. Also in the literature using complex numbers Julia sets with beautiful and chaotic boundaries arise. The world of fractals may also be mentioned here as an area to explore if more time was available. Daniel Antelo Pure2 Coursework St Charles Sixth Form College ?? ?? ?? ?? ...read more.

The above preview is unformatted text

This student written piece of work is one of many that can be found in our AS and A Level Core & Pure Mathematics section.

Found what you're looking for?

  • Start learning 29% faster today
  • 150,000+ documents available
  • Just £6.99 a month

Not the one? Search for your essay title...
  • Join over 1.2 million students every month
  • Accelerate your learning by 29%
  • Unlimited access from just £6.99 per month

See related essaysSee related essays

Related AS and A Level Core & Pure Mathematics essays

  1. Marked by a teacher

    C3 Coursework - different methods of solving equations.

    5 star(s)

    root is between -1 and 0 When I do the iterations starting at the point -1, it turns out like below. The green line is my original equation showing my real root and this clearly shows that the root that is being implied by the y = g(x)

  2. Marked by a teacher

    Estimate a consumption function for the UK economy explaining the economic theory and statistical ...

    3 star(s)

    But it will more precisely as it punish the insignificant value in the equation. The bigger the F stat the better. Compare the F-value in this equation with the first equation, which is 1.557e+004 and 1.404e+004. This means that the second equation works better than the first one, which gives opposite view from the R2.

  1. Numerical solutions of equations

    = 1.899865 x5 = 1.353797 x6 = 1.003365 x7 = 0.847116 x8 = 0.819172 x9 = 0.818390 x10 = 0.818389 x11 = 0.818389 From that, I can quite clearly see that there is a convergence beginning to show towards the positive root.

  2. Numerical solution of equations, Interval bisection---change of sign methods, Fixed point iteration ---the Newton-Raphson ...

    They all lie in the same interval [-1, 0]. Bisectional method has failed to find all the roots when several roots are too close with each other. Fixed point iteration ---the Newton-Raphson method An estimate of the root as a starting point has to be set.

  1. Sequence & Series

    For the A.P. . Similarly, for the G.P. Substituting into, we obtain We ignore the value of p=10 and q=10, which are not valid as p and q must be distinct values. Putting p=-5 into the equation gives q=2.5. Thus, p=-5 and q=2.5. 16. In a G.P. the second term exceeds the first by 20 and the fourth term exceeds the second by 15.

  2. Numerical Solutions of Equations

    I am now going to find the other roots. x1 = 1.07692 x2 = 1.20810 x3 = 1.21482 x4 = 1.21484 x5 = 1.21484 So, for the last root I will try x1 as -3. x1 = -3 x2 = -4.85714 x3 = -4.20901 x4 = -4.06333 x5 = -4.05608 x6 = -4.05606 x7 = -4.05606 f(-4.056065)

  1. Mathematical equations can be solved in many ways; however some equations cannot be solved ...

    I shall now solve it using the Newton-Raphson method to find the root in [0, 1] to the same degree of accuracy. ==> I shall start at 0 as it is a reasonable starting point in order to find the root.

  2. Arctic Research (Maths Coursework)

    G 270 370 0.135 H 315 363.54 0.138 Return Journeys A 180 338.67 0.148 B 225 360.68 0.139 C 270 370 0.135 D 315 317.98 0.157 E 000 338.67 0.148 F 45 317.98 0.157 G 90 310 0.161 H 135 317.98 0.157 Total 2.384 Observing from the table, I can

  • Over 160,000 pieces
    of student written work
  • Annotated by
    experienced teachers
  • Ideas and feedback to
    improve your own work