Nonlinear Programming Methods in Reservoir System Management

摘要

In a recent state-of-the-art review of optimization methods for managing reservoir systems, it was concluded that nonlinear programming methods had less viability than other, more commonly used alternatives. The report considers the status of existing nonlinear programming algorithms, and reaches a more favorable conclusion. Several existing techniques not included in the earlier survey are discussed in detail; these are: (1) the Kuhn-Tucker Theory; (2) reduced gradient methods (with conjugate and superbasic sets); and (3) global selection strategies for superbasic sets. A new approach based on the concept of truncated Newton algorithms is also presented.