Abstract
Previous attempts to optimize self-supporting transmission towers have failed to consider many important aspects of the problem. The present work was aimed at including such important aspects in the formulation and as a consequence it was found that a new approach to the optimization process was required. The new approach was based on a combination of existing optimization techniques and their adaptation to the particular features of tower problems. The inclusion of too many variables in the structural optimization problem, would have detrimental effects on the computational efficiency of any of the existing optimization techniques. However, this difficulty was overcome by introducing the concept of dual design spaces together with a multistage design criteria, in which the optimization problem is decomposed into stages, each stage containing a limited number of design parameters. The stage design parameters are further decomposed into compatible subsets. Each subset was improved rationally using a different optimization concept. Following a general introduction in Chapter 1, Chapter 2 sets out a review of some aspects of the general features of transmission lines which are relevant to the present work. Chapter 3 describes the formulation of a technique of linear analysis applicable, to transmission towers. Modifications to the standard stiffness approach were introduced in order to make full use of the bilateral symmetrical features of the towers, and to cope with the multistage nature of the optimization procedure. The fundamentals of the optimization approach are presented in Chapter 4. The multilevel features of the design procedure emphasize the advantages of the formulation in handling efficiently the many characterising features included. Chapter 5 describes the implementation of the theorems of structural variations in the investigation of the internal topological configurations of the panels. The established theorems, reformulated to deal with groups of structural elements j, were used to study the effects of variations on the structural behaviour and the criterion function of a panel. Finally, an approximate method for predicting the effects of varying the properties of one or more groups of members in a structure based on a least squares error minimization technique is presented. A suitable case study provided by a 220 KV transmission line designed by the Finnish company TECALEX and erected in the ARAB REPUBLIC OF EGYPT, is presented in Chapter 6. A comparison of the TECALEX designs and those predicted by the optimization procedure is also presented. Finally, concluding remarks and suggested improvements and extensions are contained in Chapter 7.