Scribd Logo
Upload

Welcome to Scribd!

  • 24 views

    2021 09 24 - Solvers

    Uploaded by

    Koal ferdl
    This document summarizes and compares the Euler explicit and Runge-Kutta 4th order methods for numerically solving ordinary differential equations (ODEs). It presents the methods, applies them to the example ODE dx/dt = sin(t), compares the results to the exact solution, and analyzes how the error decreases as the time step size is reduced. The Runge-Kutta 4th order method is shown to produce significantly smaller errors than Euler explicit, especially for larger time step sizes.

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd

    2021 09 24 - Solvers

    Uploaded by

    Koal ferdl
    24 views10 pages
    This document summarizes and compares the Euler explicit and Runge-Kutta 4th order methods for numerically solving ordinary differential equations (ODEs). It presents the methods, applies them to the example ODE dx/dt = sin(t), compares the results to the exact solution, and analyzes how the error decreases as the time step size is reduced. The Runge-Kutta 4th order method is shown to produce significantly smaller errors than Euler explicit, especially for larger time step sizes.

    Original Title

    2021-09-24_solvers

    Copyright

    © © All Rights Reserved

    Available Formats

    PDF, TXT or read online from Scribd

    Did you find this document useful?

    Is this content inappropriate?

    This document summarizes and compares the Euler explicit and Runge-Kutta 4th order methods for numerically solving ordinary differential equations (ODEs). It presents the methods, applies them to the example ODE dx/dt = sin(t), compares the results to the exact solution, and analyzes how the error decreases as the time step size is reduced. The Runge-Kutta 4th order method is shown to produce significantly smaller errors than Euler explicit, especially for larger time step sizes.

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd
    Download as pdf or txt
    24 views10 pages

    2021 09 24 - Solvers

    Uploaded by

    Koal ferdl
    This document summarizes and compares the Euler explicit and Runge-Kutta 4th order methods for numerically solving ordinary differential equations (ODEs). It presents the methods, applies them to the example ODE dx/dt = sin(t), compares the results to the exact solution, and analyzes how the error decreases as the time step size is reduced. The Runge-Kutta 4th order method is shown to produce significantly smaller errors than Euler explicit, especially for larger time step sizes.

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd
    Download as pdf or txt
    You are on page 1of 10

    Solvers in

    Simulink

    Mads Høi Rasmussen


    Associate Professor
    SDU Mechanical Engineering, Department of Mechanical and Electrical Engineering
    Given an Ordinary Differential Equation (ODE):
    𝑑𝑑𝑑𝑑
    = sin 𝑡𝑡 , 𝑥𝑥 0 = 0
    𝑑𝑑𝑑𝑑

    Solution:
    � 𝑑𝑑𝑑𝑑 = � sin 𝑡𝑡 𝑑𝑑𝑑𝑑
    ⇔ 𝑥𝑥 𝑡𝑡 = − cos 𝑡𝑡 + 𝑐𝑐

    𝑥𝑥 0 = 0

    ⇒ 0 = cos 0 + 𝑐𝑐
    ⇔ 𝑐𝑐 = 1

    ⇒ 𝑥𝑥(𝑡𝑡) = −cos 𝑡𝑡 + 1

    2
    Numeric method for solving ODE’s: Euler Explicit
    ODE:
    𝑥𝑥̇ = 𝑓𝑓(𝑡𝑡), 𝑥𝑥 𝑡𝑡0 = 𝑥𝑥0

    Time step:
    𝑡𝑡1 = 𝑡𝑡0 + ∆𝑡𝑡

    First step:
    𝑥𝑥1 = 𝑥𝑥0 + ∆𝑡𝑡 � 𝑥𝑥(𝑡𝑡
    ̇ 0)

    Second step:
    𝑥𝑥2 = 𝑥𝑥1 + ∆𝑡𝑡 � 𝑥𝑥(𝑡𝑡
    ̇ 1)

    Third step:
    𝑥𝑥3 = 𝑥𝑥2 + ∆𝑡𝑡 � 𝑥𝑥(𝑡𝑡
    ̇ 2)

    Infinitely small time step:


    ∆𝑡𝑡 → 0

    𝑥𝑥𝑖𝑖 → 𝑥𝑥(𝑡𝑡𝑖𝑖 )
    3
    Given ODE:
    𝑥𝑥̇ = 𝑠𝑠𝑠𝑠𝑠𝑠(𝑡𝑡), 𝑥𝑥 0 = 0
    Time step:
    ∆𝑡𝑡=0,2 s.
    Euler Explicit:
    𝑥𝑥0 = 0

    𝑥𝑥1 = 𝑥𝑥0 + ∆𝑡𝑡 � 𝑥𝑥(𝑡𝑡


    ̇ 0)
    ⇒ 𝑥𝑥1 = 0 + 0,2 � 𝑠𝑠𝑠𝑠𝑠𝑠 0 = 0

    𝑥𝑥2 = 𝑥𝑥1 + ∆𝑡𝑡 � 𝑥𝑥(𝑡𝑡


    ̇ 1)
    ⇒ 𝑥𝑥2 = 0 + 0,2 � 𝑠𝑠𝑠𝑠𝑠𝑠 0,2 ≈ 0,0397

    𝑥𝑥3 = 𝑥𝑥2 + ∆𝑡𝑡 � 𝑥𝑥(𝑡𝑡


    ̇ 2)
    ⇒ 𝑥𝑥3 = 0,0397 + 0,2 � 𝑠𝑠𝑠𝑠𝑠𝑠 0,4 ≈ 0,1176

    𝑥𝑥4 = 𝑥𝑥3 + ∆𝑡𝑡 � 𝑥𝑥(𝑡𝑡


    ̇ 3)
    ⇒ 𝑥𝑥4 = 0,1176 + 0,2 � 𝑠𝑠𝑠𝑠𝑠𝑠 0,6 ≈ 0,230

    4
    More steps following Euler Explicit and comparing with the analytical solution
    to the ODE:

    t x x (anal.) Error
    (Euler Expl.) 𝑥𝑥(𝑡𝑡) = −cos 𝑡𝑡 + 1 Anal. – Euler expl.
    0 0 0 0

    0,2 0 0,01993 0,01993

    0,4 0,0397 0,0789 0,0392

    0,6 0,1176 0,1747 0,0571

    0,8 0,230 0,303 0,0730

    1,0 0,374 0,460 0,0860

    5
    Change the time step and see how it influences on the result:

    𝑥𝑥̇ = 𝑠𝑠𝑠𝑠𝑠𝑠(𝑡𝑡), 𝑥𝑥 0 = 0

    Duration of simulation:
    𝑡𝑡 = 1 𝑠𝑠.

    Time step, size x 𝑥𝑥 ∆𝑡𝑡𝑛𝑛+1 − 𝑥𝑥 ∆𝑡𝑡𝑛𝑛


    (Euler Expl.)
    ∆𝑡𝑡1 =0,2 s. 0,374017 N/A

    ∆𝑡𝑡2 =0,1 s. 0,417241 0,043224

    ∆𝑡𝑡3 =0,01 s. 0,455487 0,038246

    ∆𝑡𝑡4 =0,005 s. 0,457593 0,002106

    ∆𝑡𝑡5 =0,001 s. 0,459277 0,001684

    ⇒ 𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶!

    6
    Error check on Euler expl.:

    ∆𝒕𝒕 Error
    0,2 s. 0,0857

    0,1 s. 0,0425

    0,01 s. 0,0042

    0,001 s. 0,00042

    7
    Numeric method for solving ODE’s: Runge-Kutta 4th order
    ODE:
    𝑥𝑥̇ = 𝑓𝑓(𝑡𝑡), 𝑥𝑥 𝑡𝑡0 = 𝑥𝑥0

    Since x depends only on t, Runge-Kutta 4th model reduces to Simpson’s Rule


    of Integration:
    1
    𝑥𝑥𝑛𝑛+1 = 𝑥𝑥𝑛𝑛 + (𝑘𝑘1 +2𝑘𝑘2 + 3𝑘𝑘3 + 𝑘𝑘4 )
    6

    Where:
    𝑘𝑘1 = ∆𝑡𝑡𝑥𝑥(𝑡𝑡
    ̇ 𝑛𝑛 )

    1
    𝑘𝑘2 = ∆𝑡𝑡𝑥𝑥(𝑡𝑡
    ̇ 𝑛𝑛 + ∆𝑡𝑡)
    2

    1
    𝑘𝑘3 = ∆𝑡𝑡𝑥𝑥(𝑡𝑡
    ̇ 𝑛𝑛 + ∆𝑡𝑡)
    2

    𝑘𝑘4 = ∆𝑡𝑡𝑥𝑥(𝑡𝑡
    ̇ 𝑛𝑛 + ∆𝑡𝑡)

    Note: 𝑘𝑘2 and 𝑘𝑘3 are the same since x depends only on t.

    8
    Change the time step and see how it influences on the result:

    𝑥𝑥̇ = 𝑠𝑠𝑠𝑠𝑠𝑠(𝑡𝑡), 𝑥𝑥 0 = 0

    Duration of simulation:
    𝑡𝑡 = 1 𝑠𝑠.

    Time step, size x 𝑥𝑥 ∆𝑡𝑡𝑛𝑛+1 − 𝑥𝑥 ∆𝑡𝑡𝑛𝑛


    (R-K 4th)
    ∆𝑡𝑡1 =0,2 s. 0,45969795 N/A

    ∆𝑡𝑡2 =0,1 s. 0,45969771 2,4 � 10−7

    ∆𝑡𝑡3 =0,01 s. 0,45969769 2 � 10−8

    ∆𝑡𝑡4 =0,005 s. 0,45969769 ≈0

    ∆𝑡𝑡5 =0,001 s. 0,45969769 ≈0

    ⇒ 𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶 𝑖𝑖𝑖𝑖 𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜 𝑚𝑚𝑚𝑚𝑚𝑚𝑚 𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓𝑓 𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐 𝑡𝑡𝑡𝑡 𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸 𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒. !

    9
    Error check on R-K 4th order

    ∆𝒕𝒕 Error
    0,2 s. 2,5 � 10−7

    0,1 s. 1,6 � 10−8

    0,01 s. 1,6 � 10−12

    0,005 s. 1 � 10−13

    0,001 s. 2,8 � 10−16

    10

    You might also like

    pFad - Phonifier reborn

    Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.

    Note: This service is not intended for secure transactions such as banking, social media, email, or purchasing. Use at your own risk. We assume no liability whatsoever for broken pages.


    Alternative Proxies:

    Alternative Proxy

    pFad Proxy

    pFad v3 Proxy

    pFad v4 Proxy