Application of mixed integer nonlinear programming (MINLP) optimization through GAMS for component selection in vapor compression refrigeration.

Number: 2352

Author(s) : BRENDEL L. P. M., BRAUN J. E., GROLL E. A.

Summary

Vapor compression system and component modeling tools are essential for feasibility or design studies of HVAC&R solutions. Such tools frequently rely on scaling factors, for example to estimate the needed heat exchanger surface area or swept volume of the compressor. However, when designing systems using existing components, their capacities or dimensions are not variable and selecting components becomes a mixed integer nonlinear programming (MINLP) optimization. In this type of optimization, not the optimal swept volume of the compressor is sought, but rather whether one or two units of model A, B, or C result in the best value for the objective function. The Generic Algebraic Modeling System (GAMS) is an established and powerful modeling environment for MINLP but is rarely used in the field of vapor compression refrigeration. This paper demonstrates the use of GAMS in making optimal selections of refrigerant, compressor, evaporator and condenser for a chiller from a library totaling 6000 possible combinations. The total computational time for the optimization in GAMS was 11 seconds, a task for which the Engineering Equation Solver (EES) needed 621 seconds. The GAMS language also allows a more convenient implementation using integer variables and set representations.

Available documents

Format PDF

Pages: 10 p.

Available

Free

Details

  • Original title: Application of mixed integer nonlinear programming (MINLP) optimization through GAMS for component selection in vapor compression refrigeration.
  • Record ID : 30030688
  • Languages: English
  • Subject: Technology
  • Source: 2022 Purdue Conferences. 19th International Refrigeration and Air-Conditioning Conference at Purdue.
  • Publication date: 2022

Links


See other articles from the proceedings (224)
See the conference proceedings