Help needed graphing a 3D linear system of equatio

[rmix22-disabled]

Hope somebody can help me out.
The problem seems simple - I want to solve a system of three linear equations in three variables and then graph the three planes.
Goal was to do so without having to copy or retype part of the initial equations.

I found a solution but its not real elegant and I have the feeling it could be done a better way using a vector of function definitions or the like. Unfortunately I was not able to work out a better way.
Maybe somebody can point me in the right direction

RMix

The attached sheet was made with MC14(M030) and saved as MC11.

[AlvaroDíaz]

Hope this helps.

Regards. Alvaro.

[rmix22-disabled]

Thanks for your reply.

While your way of obtaining the three functions is not really different from the one I posted, your sheet answers another question of mine - how to plot the point of intersection - thanks for that.

Regards
RMix

[ptc-1368288]

On 4/12/2010 6:36:10 AM, rmix22 wrote:
>Thanks for your reply.
>
>While your way of obtaining
>the three functions is not
>really different from the one
>I posted, your sheet answers
>another question of mine - how
>to plot the point of
>intersection - thanks for
>that.
>
>Regards
>RMix
_______________________________

From recollection, you have 3 planes crossing, then not a point of intersection, rather a line in the 3D space. In what respect do you think that line that may be curve will help ?

jmG

[ptc-1368288]

On 4/12/2010 10:13:12 AM, jmG wrote:
>On 4/12/2010 6:36:10 AM, rmix22 wrote:
>>Thanks for your reply.
>>
>>While your way of obtaining
>>the three functions is not
>>really different from the one
>>I posted, your sheet answers
>>another question of mine - how
>>to plot the point of
>>intersection - thanks for
>>that.
>>
>>Regards
>>RMix
>_______________________________
>
>From recollection, you have 3 planes
>crossing, then not a point of
>intersection, rather a line in the 3D
>space. In what respect do you think that
>line that may be curve will help ?
>
>jmG
>________________________________

Off my head, decide of two Y, X planes and solve linearly for the Z in the 3D domain. You can then construct the X, Y grid matrix, let say a 10 by 10 grid with the Z values. At this point you can interpolate for intermediate values. The very same principle as if the matrix would be a real Engineering surface plot.

jmG

[rmix22-disabled]

While I like to see the point of intersection, another questions comes to mind - is there a way to AUTOMATICALLY deal with a system that will not solve, like three parallel planes or the like.

I was able to make the point show up at (0/0/0) in that case which really did not look well. Another idea was to plot a point with very large coordinates to make it vanish but I woul have to turn off autoscaleing for that. Unfortunately the plot options are not programmable or able to deal with variables.

Any good ideas, is there anything valid, P could be set to which will not show up in the diagram?

RMix

[RichardJ]

On 4/12/2010 12:58:10 PM, rmix22 wrote:

>Any good ideas, is there
>anything valid, P could be set
>to which will not show up in
>the diagram?

Unfortunately, 3D plots don't behave the same as 2D plots in that regard. If it's not a valid point, the graph chokes.

The only thing I can thnk of is to take a look at Stuart's sheet on functional graphs:

http://collab.mathsoft.com/read?95744,12e#95744

Richard

[rwells]

I just posted the same question last week and got several good examples returned. I suppose you can find them by clicking my avatar and investigating my posts.

[ValeryOchkov]

May be this help you:

[Werner_E]

Valery Ochkov schrieb:

May be this help you:

I fear it doesn't. The point was to deal with a system of three equations which has no unique solution automatically. That is: the point of intersection is plotted if it exists, otherwise it is suppressed.

Not an easy task considering that 3d graphs are not parameter driven. Not even the axis limit can be variables.

If its still of interest after nearly three year. As you wrote you can make the point show up at (0/0/0) in case of a not unique solution you could put it out of the way completely? Of course this would mean that you have to chose the acis limits manually.