Solve Castigiliano Method for Continuous Beam Problem

  • Thread starter Thread starter Zouatine
  • Start date Start date
  • Tags Tags
    Method
AI Thread Summary
The discussion revolves around applying the Castigliano method to solve for reactions in a continuous beam subjected to a distributed load. Participants clarify the equations for energy and moment, emphasizing the importance of sign conventions in force and moment balances. A key point is the integration of moment squared over the beam length to derive energy, with some confusion about the integration process and differentiation with respect to forces. The conversation also touches on the chain rule in differentiation, specifically regarding the relationship between moment and force. Overall, the thread highlights the complexities of using the Castigliano method in structural analysis.
Zouatine
Messages
19
Reaction score
0

Homework Statement


Problem:
a continuous beam ABC simply presses, loaded with a distributed load q, the length of the beam is 2L.
EI=constant
840411898.png

find the reactions in the beam ,we use Castigiliano method.

Homework Equations


- U=(1/2EI)∫M^2 dx (U: Energy)
- dU/dP=d (d: displacement)

The Attempt at a Solution


first we have 3 reactions (Ra,Rb,Rc) in the supports.
∑F=0→ Ra+Rb+Rc=ql
∑M/A=0 → Rb+2Rc=3ql/2
for the moment flexing : 0≤x<L
- M(x)= -Ra*x .
L≤x<2L
M(x)=-Ra*x+Rb*(x-L)-(qL/2)*(x-L)^2
my problem is how to find the energy equation??
Thanx.
 

Attachments

  • 840411898.png
    840411898.png
    1.9 KB · Views: 611
Physics news on Phys.org
You seem to have some sign inconsistency.
In your force balance you have Ra and Rb with the same sign, but in the torque balance they have opposite signs.
What is your sign convention?

To get the energy, you have to integrate (M2/(2EI)).dx no?
 
First thanks for your answer Sir ,
I made a mistake ∑F=0→ -Ra+Rb+Rc=ql
- the positive sense for the moment is the clockwise.
yes I know to get energy I have to integrate (M^2/(2EI)).dx , but it's complicated to integrate from 0 to L M^2/(2EI);dx with (M(x)=-Ra*x and from L to 2L M^2/(2EI) with M(x)=-Ra*x+Rb*(x-L)-(qL/2)*(x-L)^2 .
I saw in some document this formula :du/df=(1/EI)*∫M*(dM/df) dx (f: force) but I do not know how they are found that.
thanx
 
Zouatine said:
this formula :du/df=(1/EI)*∫M*(dM/df) dx
That is obtained from the first formula simply by differentiating wrt f. Because the integration bounds are fixed, you can simply differentiate through the integral sign as though it wasn't there. You then use the chain rule to get from d(M2)/df to 2M dM/df.

Can you find dM/df?
 
thanx, so to find dM/df , I choose Ra so,
1- dM/dRa =-x
2- dM/dRa =-x
and after that
du/dRa=(1/EI)*∫(-x*Ra)*(-x) dx+(1/EI)*∫(-x*Ra+Rb*(x-L)-(qL/2)*(x-L)^2)*(-x) dx =0
with (0<x<L) and in second (L<x<2L).
and I find Ra in this equation, just one more question about this du/df=(1/EI)*∫M*(dM/df) dx
Mf (x) it is according to f , when we drift : we find this 2Mf(x) * dMf(x)/df
thanx sir
 
Zouatine said:
Mf (x) it is according to f , when we drift : we find this 2Mf(x) * dMf(x)/df
Sorry, I don't understand your question. What is "Mf(x)"? If you mean literally M multiplied by f(x), I do not see any f that is a function of x. If you mean M as a function if f and x, that would be written M=M(f,x).
 
Sorry. My english is not good my question was how you found this :
d(M^2)/df =2M* dM/df
 
thanx , Sir
 
Back
Top