I got this equation to solve in VBA:
Energy = y + (0.5) ^ 2 / ( 2 * 9.81 * y ^ 2 )
but I got the Energy
value and want to solve for the y
value.
In R is easy: uniroot
. There is something like this?
I got this equation to solve in VBA:
Energy = y + (0.5) ^ 2 / ( 2 * 9.81 * y ^ 2 )
but I got the Energy
value and want to solve for the y
value.
In R is easy: uniroot
. There is something like this?
The equation is a cubic polynomial with two positive roots and one negative. We can eliminate the negative one, and consider a direct evaluation of the two positive roots with the following VBA code in a Module.
Global Const g As Double = 9.81
Global Const PI As Double = 3.14159265358979
Public Function CalcE(ByVal y As Double) As Double
CalcE = y + (0.5) ^ 2 / (2# * g * y ^ 2)
End Function
Public Function SolveForY(ByVal E As Double, Optional ByVal second As Boolean = False) As Double
' Solve "E = y + (0.5) ^ 2 / ( 2 * g * y ^ 2 )" for "y"
If Not second Then
SolveForY = E / 3# - 2# * E * Sin(Asin((16# * E ^ 3# * g - 27#) / (16# * E ^ 3# * g)) / 3#) / 3#
Else
SolveForY = E / 3# + 2# * E * Sin(Asin((16# * E ^ 3 * g - 27#) / (16# * E ^ 3 * g)) / 3# + PI / 3#) / 3#
End If
End Function
and the supporting Asin()
function defined in terms of Atn()
.
Public Function Asin(ByVal x) As Double
If Abs(x) < 1# Then
Asin = Atn(x / Sqr(1# - x * x))
Else
Asin = 2 * Atn(1) * Sgn(x)
End If
End Function
Some example results are
? SolveForY(1#) 0.12035578724076 ? SolveForY(1#, True) 0.986917853921696
And checking the solutions
? CalcE(0.12035578724076) 0.999999999999997 ? CalcE(0.986917853921696) 1
Consider using algebra:
Energy = y + (0.5)^2 / ( 2 * 9.81 * y^2 )
Energy = y + .25 / ( 19.62 * y^2 )
Energy * ( 19.62 * y^2 ) = y * ( 19.62 * y^2 ) + .25
Energy * 19.62 * y^2 = 19.62 * y^3 + .25
0 = 19.62 * y^3 - Energy * 19.62 * y^2 + .25
and apply standard techniques:
Here's a translation of John Alexiou's answer from VBA to Excel formulas. If energy in in cell B2
then the three solutions are given by:
=B2/3-2*B2*SIN((ASIN((16*B2^3*9.81-27)/(16*B2^3*9.81))+0*PI())/3)/3
=B2/3-2*B2*SIN((ASIN((16*B2^3*9.81-27)/(16*B2^3*9.81))+4*PI())/3)/3
=B2/3-2*B2*SIN((ASIN((16*B2^3*9.81-27)/(16*B2^3*9.81))+2*PI())/3)/3
The third equation gives negative y
.