GenOpt
A generic multiparameter optimization program for system optimization. It automatically determines the values of userselected design parameters that lead to the best operation of a given system. Optimizes a userselected objective function, such as a building's calculated annual energy use. It also offers an interface for easily implementing your own optimization algorithms into its library.
GenOpt has an open interface on both the simulation program side and the optimization algorithm side. By modifying a configuration file, it allows users to easily couple any external program (like DOE2, SPARK, BLAST, EnergyPlus, TRACE, TRNSYS, etc., or any userwritten program). GenOpt is written entirely in Java so that it is platform independent. An interface for coupling external simulation programs and adding custom optimization algorithms is available. If the simulation input files are available, the time to set up an optimization problem is typically less than one hour. Screen Shots
Keywords
system optimization, parameter identification, nonlinear programming, optimization methods, HVAC systems
Validation/Testing
N/A
Expertise Required
Computer literacy required. No programming skills required for using the standard package. Basics of Optimization Theory for nonexperienced users explained in program documentation. Knowledge of Optimization Theory and basic skills in Java required for adding your own optimization algorithms to the GenOpt library.
Users
100
Audience
Engineers, researchers.
Input
The following input is read from ASCII files: (a) Specification of parameters to be optimized and their minimum and maximum values, (b) settings for the optimization algorithm, (c) configuration file of simulation program (i.e., how to start simulation program, where to read simulation output), (d) input template file (used for generating simulation program input file).
Output
Online plotting of optimization progress (objective function and free parameters). Result of optimization in text file.
Computer Platform
Platform independent (written in Java).
Programming Language
Java
Strengths
Utility classes such as for linear algebra, optimality check, or doing linesearch can be used for user's own implementation of optimization algorithms.
Weaknesses
GenOpt is not aimed at solving optimization problems when the properties of the objective function (e.g., gradient, Hessian) are analytically available. It will work on those problems, but more efficient programs exist for that purpose.
Contact
Company: 
Lawrence Berkeley National Laboratory 
Address: 
MS 903147 
Telephone: 
(510) 4867538 
Facsimile: 
(510) 4864089 
Email: 

Website: 
Availability
GenOpt 1.1 is available free of charge. To download, visit the GenOpt web page.