Conversation
|  | ||
|
|
||
| Define distance *a* as 3 meters | ||
| Define distance *a* as 3 meters. The ball joint at A can apply reaction forces, but no reation torques. |
There was a problem hiding this comment.
| Define distance *a* as 3 meters. The ball joint at A can apply reaction forces, but no reation torques. | |
| Define distance *a* as 3 meters. The ball joint at A can apply reaction forces, but no reaction torques. |
|
|
||
|
|
||
| As before, start by defining the location of each relevant point as an array. | ||
| As before, start by defining the location of each relevant point as an array. For this problem vertical arrays are more convenient. |
There was a problem hiding this comment.
This is not obvious - in fact, I'd argue that carrying around the extra dimension in the shape makes it more difficult to interpret the arrays. AFAICT, the main reason for doing so is so that the hstack-ing works when constructing the system matrix. Instead of introducing the extra dimension at the beginning of the processing, I'd vote for simply adding a "dummy" dimension for the hstack, i.e. something like:
unknown_forces = np.hstack((Unit_BD[:, np.newaxis], Unit_BE[:, np.newaxis], np.eye(3)))There was a problem hiding this comment.
Just a quick ping on this one @h2o-DS - WDYT? If you don't have the bandwidth to make these updates but are not opposed to them, I'd be happy to push them up!
|
This one has been stale for a while so I'm going to go ahead and close it. The underlying idea is good, so if folks are interested in picking it, please feel free! The place to start would be addressing the comments from the previous review. |
5 participants
Fixed some formatting and extended the static equilibrium tutorial to include np.linalg.solve()