CHAPTER ONE INTRODUCTION 1.1 Background of the study Ordinary Differential Equations (ODEs) play a key role in modeling dyna mic systems across diverse fields like physics, engineering, biology, and economics. These equations describe how quantities change over time or other variables, often subject to specific initial or boundary conditions that define the system's state at certain points. Many real-world pheno mena-such as disease spread, planetary motion, and heat transfer—ca n be formulated as initial or boundary value problems (IBVPs) involving ODEs. However, for complex systems, finding exact analytical solutions may not be feasible, leading to the need for numerical approaches. Multi-step methods are a powerful class of numerical techniques for ap proximating solutions to ODEs. Unlike single-step methods, which rely 0 nonly the most recent point in the solution, multi-step methods use mul tiple previous points to progress. This approach improves both efficienc y and accuracy by incorporating more information from past points. Co mmon examples of multi-step methods include the explicit Adams-Bash forth and implicit Adams-Moulton methods, along with Backward Differe Ntiation Formulas (BDF), which are particularly useful for stiff ODEs. Multi-step methods are especially advantageous for initial or boundary v alue problems (IBVPs). These problems involve finding an ODE solution t hat meets specified initial and boundary conditions over a given interval. Multi-step methods provide an efficient and accurate framework for solv ing such problems, allowing for step-size adjustments to balance precisi on with computational load. Additionally, implicit multi-step methods are beneficial for maintaining stability in stiff problems, where rapid change s can pose challenges for other numerical methods. -This study delves into multi-step methods as a robust approach for solvi ng IBVPs in ODEs, highlighting their theoretical foundations, accuracy, st ability, and practical applications. By doing so, it demonstrates how mult i-step methods serve as valuable tools for initial boundary problems, wh ere analytical solutions are impractical or unavailable. 1.2. Statement of the problem. Analytical solutions to ordinary differential equation are generally impos sible. Consequently, numerical methods, including iterative techniques a nd computational approaches, must be utilized to approximate solution s, These methods involve starting with an initial guess and incrementall y refining it through successive tests until convergence. This project foc uses on developing a multi-step approach to numerically solve differenti al equation. 1.3 Aim of the study The aim of this study is to apply multi- steps to solving ordinary differ ential equations Objective of the study Understand the concept of Multi-step methods to solving ordinary dif ferential equation Solve the questions using Multi-step methods like Adams Moulton, A dams Bashforth methods, Compare results from the method with the results obtained from Ana lytical method. 1.4 SOME BASIC DEFINITIONS 1.41 DIFFERENTIAL EQUATION A Differential equation is a relation between a function its derivatives an d the variables upon which they depend. “] lThe general form of an ordinary differential equation is given by Where n represents the highest order derivatives and y and it's derivativ es are function of x. Numerical methods by M.K. Jain, S.R.K. lyengar, R. K. Jain (2009). 1.42 ORDER OF A DIFFERENTIAL EQUATION 1. An order of a differential equation is the highest derivatives prese nt in the equation. An equation containing only first derivatives is c alled a first-order differential equation, an equation containing the second order derivatives is called a second-order differential equat ion, and soon 1.43 DEGREE OF A DIFFERENTIAL EQUATION. The degree of a differential equation is the power of the highest or der derivatives after the equation has been rationalize. Examples are differential equations of degree 1 and 2 respectively.1.45 1.46 ORDINARY DIFFERENTIAL EQUATION (ODE) An ordinary differential equation (ODE) is an equation containing on e independent variable and its derivatives. The term “ordinary” is us ed in contrast with the term partial differential equation which may be with respect to more than one independent variable. The first ord er differential equation is given as: Where x is the independent variable and y is the dependent variable, the types of first order ordinary differential equation include separabl e variable differential equation, homogeneous differential equation, li near differential equation, Bernoulli differential equation. LINEAR AND NON-LINEAR ORDINARY DIFFERENTIAL EQUATIONS A Linear Ordinary Differential Equation: A differential equation is linear if it can be written in the form wy w! 0) Gea Fn OD dy Ftv tay (x) & tay (x) y = g(x) de dx y = y(X) is the unknown function (dependent variable), x is the independent variable, 68), Ay AQ), oy @0(X) Are functions of x (they could be constan tas well) g(x) isa given function or could be zero. Ina linear differential equation:The unknown function y and its derivatives appear to the first power (no exponents, squares, or other nonlinear terms). There are no products or functions of y and its derivatives Example of linear equation: dy dy ee Fe 189 Nonlinear Ordinary Differential Equation: A differential equation is non linear if it involves nonlinear terms in the unknown function y or its der ivatives. This could include: dy,* + Powers of y or its derivatives, such as y2 or ‘ds dy + Products involving y and its derivatives, such as y+ dx + Functions of y other than simple multiples like sin(y), n(y) Example of a nonlinear equation: d’y dy y= =sin(y) ae dx 1.5 Numerical solutions to ordinary differential equation Numerical solutions to differential equations involve using computational met hods to find approximate solutions, rather than exact, closed-form solutions. T his approach is useful when the differential equations are too complex to solv e analytically. 1.51 Concept of Numerical solutions 1. Initial value problem (IVP) and Boundary valued problem (BVP): The O -DE that is specified along with an initial (one) condition is called an initial value problem while that specified along two conditions is called a bo undary value problem. 2. Step size: The interval between discrete points at which the solution i s approximated 3. Error Analysis: This tells us how much the numerical solution deviate s from the true (exact) solution. error Analysis helps us understand a nd control the accuracy of the numerical methods we use. 1,52 Steps for solving Numerical methods Numerical methods can be categorized into Single step method and Multi-ste p methods 1. Single step method In Single step method the solution at any point is obtained using the sol ution at only the previous point. The General Single step method can be written as Yost Yn th * D0 Yin BD where: his the step size, is an approximation function that gives an estimate of the slope, b ased on the function f the current point (x,,y,.), and possibly other poi nts within the interval. The function is chosen depending on the specific single-step method, such as Eule 1's Method or Runge-Kutta. Numerical Methods by . M.K. Jain, S.R.K lye ngar, R.K jain (2009). Advantage: Easy to implement; low computational costDisadvantage: First order accuracy may require small steps size for stab ility. Example: a. Euler's Method: For Euler’s method Yost = Sut An Yo) b. Midpoint Method (Second-Order Runge-Kutta) The Midpoint method, a simple second-order method, improves u pon Euler’s method by taking an intermediate slope: h h wr =Yoth f(x, + Sy £(%qs Yn) c. Fourth-Order Runge-Kutta (RK4): The widely-used RK4 method uses a more complex ® by combinin g multiple evaluations of f = 6 (x). Ys) h h k= f(t 5. Yat ki) ket h hy = Fat govt 5h) ky= f(x,+h, y,-thky) h Yast = 3,4 5 (Rit 2kat ky t+ ky) 2. Multi-step methods.Multi-step methods utilize information from several previous points to com pute the next value. They can be implicit or explicit. Example: a. Adam-Bashforth method (explicit): A family of methods based on previous values. The first order is given by: Yours Nath YOb fx 5 Yu) where: Yn is the next approximate value of the solution, y, is the current va lue, his the step size f(x y) isthe function defined by the differential equa on ak f(x,y) 5; are the coefficients that depends on the number of steps. b. Adams-Moulton method (implicit): This method requires solving an equation at each step since they depend ont he future value. The first order is given by: “7k Yor Yath Sob, fer where: The main difference is that “+ is included, making the method implicit Num erical Analysis by Richard L.Burden , J.Douglas Faires (2011).