strange deSolve results

[Thomas Hu]

I'm solving a seemingly simple ODE for a harmonic oscillator driven by 3 sinusoids of frequencies 10, 20 and 50 rad/s:

deSolve(u''+18260.176014467*u=5.6371639192022*(cos(50.*t)-0.46496268156585*(cos(20.*t)-0.12482457123238*cos(10.*t))) and u'(0.)=0. and u(0.)=0.,t,u)

but the result nspire gets contains (cos(t))^48, etc, (listed below) and a plot looks chaotic.

wonder if there's a bug with desolve and how this can be fixed?

Thanks!

Thomas

u=−2.2894504808711−4*cos(135.13021873167*t)-201358231204.4*(((sin(t))^(2)+11.5)*(cos(t))^(48)+73.437499999999*((sin(t))^(2)+2.6702127659574)*(cos(t))^(44)+692.87109375001*((sin(t))^(2)+0.91333333333333)*(cos(t))^(40)+1958.4106445313*((sin(t))^(2)+0.16777408637874)*(cos(t))^(36)+2144.0448760987*((sin(t))^(2)-0.23983739837398)*(cos(t))^(32)+1010.4900932312*((sin(t))^(2)-0.49300699300699)*(cos(t))^(28)-212.33959794045*(cos(t))^(26)+71.736350655558*(cos(t))^(24)-19.642096012831*(cos(t))^(22)+4.321261123205*(cos(t))^(20)-0.75463199946238*(cos(t))^(18)+0.10290436736248*(cos(t))^(16)-0.010719209287111*(cos(t))^(14)+8.2805739444705−4*(cos(t))^(12)-4.554446410199−5*(cos(t))^(10)+1.6830223689554−6*(cos(t))^(8)-3.8303081437275−8*(cos(t))^(6)+4.6669881641014−10*(cos(t))^(4)-2.3706847991706−12*(cos(t))^(2)+2.5946549377866−15)