[Ipopt] Initial guess near the solution

[Allan Leal]

I need to performa sequence of optimization calculationsin which the 
initial guess of the i-th calculationis the result of the i-1-th 
calculation.

The objective function remains the same on these calculations---only the 
equality constraints suffer some slight modifications.Therefore, the 
previous result is a excellent initial guess for the next calculation.

I would like to know how I can provide not only the primal variables 
/x/, but also the Lagrangemultipliers /y/ and /z/? I have observed that 
providing only a good /x/ is not enough, since several iterations are 
required to determine /y/ and /z/.

I believe there is a catch also as to the initial value of the barrier 
parameter /mu/. I would be appreciate for any advice on this too.

One important remark: on these calculations some variables will be very 
near to the boundary. This is what I expect for those variables that 
should in fact be on the boundary, and the small inaccuracy is fine for 
me. Therefore, the calculation should be able to start from an initial 
guess very close to the boundary.

Thanks,
Allan
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://list.coin-or.org/pipermail/ipopt/attachments/20130422/a4146224/attachment.html>