?Optimization of earth-dam structures? and the potential of numerical methods

Abstract

From the Editorial Office:The attention of the paper's readers is called to analysis of the reliability of the results of computer-aided calculations of earth dams from mathematical models. Undisputably, mathematical modeling can be a powerful and reliable means of assessing the serviceability of a structure and ways to improve it; it cannot replace design, however, and as is demonstrated in this paper, the weak sides of mathematical models must also be soberly considered. The authors rightly focus attention on the need for reliable determination of the characteristics and properties of soils placed in the body of a dam, without which a valid result could not be obtained when using even the most complete body of mathematics. Also deserving mention is the analysis of other factors affecting the degree of reliability of computational results. The paper "Optimization of earth-dam structures" [i] is a further development of already several years of evolving ideas and methods of "optimization" [2,3], which assume a selection of "variable factors" (structural -the width and angle of incline of the core, the contour intervals of the slopes, and, technologica I -the density of soil placement, etc.) and their computed values ("levels"), the conditional formalization of factors and the selection of a group of computed variants of the dam profile ("computation plans," or "design matrix"). The cost and "critera of structure reliability," as well as their functional dependencies on variable factors are determined for these variants. The "optimal" variant is then sought proceeding from simultaneous fulfillment of conditions of the maximum reliability function and minimality of the cost function.Without touching upon the methods used to determine the cost indicators of the variants, note that the "reliability criteria," i.e., safety factors for stability, crack resistance, etc., are determined from the results of computatioDs of the stress-strain state [3, p. 32] using the software DAM-2M (the problem of plane deformation in accordance with the method of local variations (MLV), which is based on the "energy" soil model [2,4]. Rasskazov and Sepeda [i] fail to mention these "optimization" fundamentals, however, and the solution of the optimization problem is reduced to some formalized operations with values of the reliability criteria and cost indicators. It is therefore timely and expedient to return to these fundamentals and discuss them on the pages of this journal.i. Optimization of earth-dam structures during their design and construction, which is based on the use of numerical methods of calculating the stress-strain state (SSS), is an extremely urgent problem and promises economic benefits, raising the competitiveness of earth dams in connection with the pronounced cost increase for the construction of concrete dams formed from low-cement concrete.Firstly, however, the problem of optimization in dam-construction practice actually reduces to the search for the possibility of maximum utilization of soils found in the imme...