Is there an upper bound for the variable in 'addGenConstrPow()' ?

Answered

chi hsuan lin

March 24, 2021 05:32 Edited

• 
• 
• 

Hi,

I am trying to add a general constraint into my model. However, it seems like there's an upper bound for the variable in 'addGenConstrPow()'. 

Here is an example: 

x = m.addVar(ub=5,lb=0,vtype=GRB.CONTINUOUS,name='x')
y = m.addVar(ub=GRB.INFINITY,lb=0,vtype=GRB.CONTINUOUS,name='y')
m.update()
m.setObjective(y, GRB.MAXIMIZE)
m.addGenConstrPow(x, y, 10)
m.optimize()

Optimal solution found (tolerance 1.00e-04)
Best objective 1.000000000000e+06, best bound 1.000000000000e+06, gap 0.0000%
1000000.0000000005

Apparently, the objective value should be y = 9765625 (5^10), but I got 1000000 as the 'objVal'. Besides, I also meet the same problem while implementing  'addGenConstrLog()' .

I am wondering if there is a misunderstanding or how can I deal with the limitation?

Thanks,

Lin

0

• Official comment

  

  Simranjit Kaur

  • Gurobi Staff

  December 06, 2024 00:34

  This post is more than three years old. Some information may not be up to date. For current information, please check the Gurobi Documentation or Knowledge Base. If you need more help, please create a new post in the community forum. Or why not try our AI Gurobot?.
• 

  Alison Cozad

  • Gurobi Staff

  March 29, 2021 12:14 Edited

  You are correct, there is an upper limit placed on \(y\).  According to the General Constraint documentation,

  For some of the supported functions, modest  values can lead to enormous  values (and vice-versa). This can cause numerical issues when solving the resulting piecewise-linear MIP model. To avoid such issues, we limit the range of any x or y that participates in a function constraint to [-1e+6, 1e+6]. The parameter FuncMaxVal allows you to change these limits, but we recommend that you proceed with caution.

  Therefore, you can simply change this value using the FuncMaxVal parameter.  However, as the documentation notes, such large numbers can lead to numerical issues.  You can learn more about numerical instability in the Guidelines for Numerical Issues.

   

  0
• 

  chi hsuan lin

  March 27, 2021 09:24 Edited

  • 
  • 
  • 

  I've changed the parameter successfully! Thanks for your help!

  Best,

  Lin

  0

Post is closed for comments.