Page 51 - DCAP601_SIMULATION_AND_MODELING
P. 51

Unit 3: Simulation of Continuous System (II)



            An instantaneous description of  the state of the system is given by the outputs of the  two  Notes
            integrators, i.e., by variables y  and y . We will use, once again, the fourth-order Runge-Kutta
                                    l     2
            method to obtain the values of y  and y  as a function of time. We will choose the step-size
                                       1     2
            t = H = .001 second and simulate the system for 5 seconds. Thus the number of steps will be
            N = 5,000. This is too large a number of outputs to be plotted or examined. We will, therefore,
            print out the values of y  and y  only once every 100 integration steps. This can be implemented
                               1    2
            by keeping a counter K which is decremented by I for each integration step. Every time K equals
            zero, y  and y  are printed and K is reset to 100. The following FORTRAN program performs the
                 1     2
            simulation.
            C     H IS TIME STEP, N IS NO. OF STEPS, Yl, Y2 INITIAL VALUES
                  T = O.
                  YI = 1.
                  Y2 = 0.
                  H = .001
                  N = 5000
                  K= 1
                  DO 120 I – I, N
                  K = K – I
                  IF (K.NE.O) GO TO 110
            C     PRINT ONCE IN 100 STEPS
                  PRINT, T, YI, Y2
                  K = 100
            C     CALCULATE THE RUNGE-KUTTA TERMS
              110 Ull = H*Y2
                  UI2 = H*(.1*(1. – Y1*Y1)*Y2 – Y1)
                  U21 = H*(Y2*+ .5*UI2)
                  U22 = H*(.1 *(l. – (YI +.5*U11)*(Y1 +.5*U11)*(Y2+.5*U12) – (Y1 +.5*U11)
                  U31 = H*(Y2 + .5*U22)
                  U32 = H*(.1 *(1. – (Y1 +.5*U2l)*(Y1 + .5*U21))*(Y2+ .5*U22) – (Y1 + .5*U21)
                  U41 = H*(Y2 + U32)
                  U42 = H*(.I*(1. – (Y1 + U31)*(Y1 + U3l)*(Y2 + U32) – (Yl + U31)
            C     CALCULATE Y1 AND Y2
                  Yl = Y1 + (U11 + 2.*U21 + 2.*U31 + U4l)/6.
                  Y2 = Y2 + (U12 + 2.*U22 + 2.*U32 + U42)/6.
                  T = T + H
            120 CONTINUE
                  PRINT, T, YI, Y2
                  STOP
                  END
            By studying the three examples in this unit so far, you may have acquired the incorrect impression
            that (i) every dynamic continuous system must first be expressed as a set of  simultaneous
            differential equations before being simulated and, that (ii) such a system is always deterministic.
            The next example is meant precisely to dispel this notion. Moreover, in this example we are
            simulating  a system which is too large and expensive to ex·  periment with and  where an
            incorrect design could become very costly.






                                             LOVELY PROFESSIONAL UNIVERSITY                                   45
   46   47   48   49   50   51   52   53   54   55   56