Optimization models in solar power tower plant designdesign of the heliostat field and the tower receivers

  1. Domínguez Bravo, Carmen Ana
Supervised by:
  1. Emilio Carrizosa Priego Director
  2. Manuel Quero García Director
  3. Enrique Fernández Cara Director

Defence university: Universidad de Sevilla

Fecha de defensa: 08 January 2016

Committee:
  1. Juan Antonio Mesa López-Colmenar Chair
  2. Rafael Blanquero Bravo Secretary
  3. Martin Frank Committee member
  4. Peregrina Quintela Estevez Committee member
  5. Alfredo Marín Pérez Committee member

Type: Thesis

Teseo: 396949 DIALNET lock_openIdus editor

Abstract

The design of solar power tower (SPT) plants started to be studied in 1970s and is still being an active field of research nowadays. In this dissertation, the optimal design of an SPT plant under different considerations is addressed, focusing on the tower-receiver and heliostat field design. The first chapter provides the general ideas on SPT plants, with a description of the components involved in the problem and a presentation of the mathematical modelling used in this document. An algorithm to solve the basic problem is proposed in Chapter 2. It is a greedy-based heuristic algorithm, which introduces an innovative approach solving the heliostat location problem without fixing geometrical patterns. This algorithm is extended in Chapter 3 to address the solar field design with heliostat pod systems (triangular structures having several heliostats on it). Chapters 4 and 5 concern heuristic procedures adapted to the solution of innovative problems arising in this technology: multiple receivers system design and multi-size-heliostat field design. Both are challenging problems due to their high complexity. The heuristic algorithm in Chapter 2 has been modified and combined with different optimization procedures in order to furnish a solution which, although not optimal, is competitive and can be considered as a good solution. Along all the chapters, the results have been compared to the state-ofthe- art results, when available. This has allowed to validate the proposed algorithms. The general conclusions and some comments on further work are commented in the last chapter.