Thursday, 10 April 2014

Numerical Methods Root Finding

Methods in Finding the Common Numerical Root:

In the common root of finding the schemes for in the functions in the one independent with are variable are mainly briefly in the discussed in the following section.

Bisection Method:


In the idea of the bisection method is mainly based in the fact in a function will change the sign when there is passes through the zero. In the evaluation of the function in the middle of the interval and in replacing whichever the limit in the same sign, in the bisection method can be halve the main size of the interval in the each iteration and eventually in the root.
https://www.efunda.com/math/num_rootfinding/images/Bisection2.gif
When there is an interval in the roots, in the bisection of the method is the one that will not fail. In the cases it is the slowest. When there contains more than the one root, and then the bisection can be find one.

Secant Method:

For the improve in the slow convergence in the bisection method, in the secant method of a certain function there is approximately linear for the local region in the interest and in the uses of the zero-crossing in the line connecting the limits of the intervals as in the new sort of the reference point. In the next interaction starts from the function at the new point of the reference. And in the interaction starts in from the evaluating in the function of the new reference point and then he forms of another line. And then the process is repeated until the main root is found.

https://www.efunda.com/math/num_rootfinding/images/Secant2.gif
In a mathematical form this secant method is will converges and more rapidly in a near root than the false of the position method. In fact there in the secant method always will not always bracelet the main roots, for not sufficiently smooth functions the algorithm may not cover the functions.

False Position Method:

In a similar way to the secant method, in the false position of the method also mainly uses in the straight line for an appropriate of the function in the local interest of the region. And the main and only difference in the two methods are that the secant methods keeps the most of the recent two estimates, and when there is a false position method then they retains the most of the recent estimations and the next one which was recent which has a main opposite sign in the value of  function.

https://www.efunda.com/math/num_rootfinding/images/FalsePosition2.gif

In the false position method it sometimes makes to keep the older reference point in order to maintain the opposite sign bracket in the around the root, which has a lower and the uncertain convergences of the rate when compared to the secant method and in this the emphasis the bracketing the main root and in sometimes there they will restrict to the false positioning to the method and in the difficult situations and when there is a need to solve the highly nonlinear equations.

No comments:

Post a Comment