Lipschitzian global optimization (LGO) program system: Implementation and some recent applications
Consider the optimization problem min f (x) s.t. x belongs to D: x is a real n-vector; D is bounded and robust; f is continuous or Lipschitz-continuous, and (possibly) multiextremal on D. Along the lines of investigating the general convergence mechanism of deterministic algorithms in global optimization, a comprehensive framework was provided to analyses establish and propose globally convergent adaptive partition methods. On the basis of these theoretical results, the LGO program system has been elaborated and implemented, for both PC and workstation environments. LGO includes several global and local optimization procedure options which can be used in interactive or automatic run modes. Among the numerous potential applications of LGO, two recent studies will be reviewed here: (1) the design and operation of {open_quotes}black box{close_quotes} engineering systems, and (2) the calibration of complex system models.
- OSTI ID:
- 36392
- Report Number(s):
- CONF-9408161--
- Country of Publication:
- United States
- Language:
- English
Similar Records
Global optimization in modelling and managing environmental systems: Some applications
The global convergence of partitioned BFGS on problems with convex decompositions and Lipschitzian gradients