Design and modeling of solar sea power plants by geometric programming

Abstract

Geometric programming, a nonlinear optimization technique, is used to design Solar Sea Power Plants (SSPP) which convert the thermal energy stored in the tropical water into electricity. First, the conversion process is described, and the hardware necessary to implement a binary-fluid, closed-Rankine cycle is identified. Next, steady-state analytical models for the major components are derived. These models are then used as the constraints of a geometric program whose objective function is the minimization of a particular function of the design variables of the SSPP. A variety of problems are solved. On one extreme, they include simply the design of a minimum surface heat exchanger for a SSPP, and on another extreme the selection of the various water pipes for a given ocean site, accounting for all the hydraulic losses. The geometric programming technique produces the optimum design and, more importantly, the sensitivity of the objective function at the optimum to variations in cost figures, constraint bounds, and arbitrary constants of the model. It is demonstrated in this dissertation that geometric programming is an economical and effective tool for the analysis and design of complex interacting engineering systems involving many variables and constraints. In particular, it is concluded that geometric programmingmore » is effective in a variety of situations encountered in the design of Solar Sea Power Plants.« less