Deformations of the Sayano-Shushenskoe dam ? Measurement results and problems

Abstract

Numerous on-site observations by geodetic methods were carried out during construction on the gravity-arch dam of the Sayano-Shushenskoe hydroelectric station with a maximum height of 245 m and crest length of 1066 m (67 sections). The measurement methods and results in the initial stages were elucidated in the literature [1-7]. The composition and results of measurements at the final stage of construction are examined below. By 1989 it was possible to place marks on the crest, to reliably tie-in the inner (in the dam, adits) and outer (bank surface in the upper and lower pools) geodetic networks, and to stabilize the geometric schemes of these networks. In 1990 the reservoir was filled for the first time to the elevation of the normal pool level (NPL). Composition of Measurements. The schemes of the inner and outer geodetic networks are shown in Figs. 1 and 2. Longitudinal galleries in the dam (near the upstream face, in column 1) and their continuation in the banks, in grouting adits, are used for installing the monitoring and measuring equipment (MME) and taking measurements in the inner network. The inner network includes also (not shown in Fig. 1) stationary hydrostatic leveling systems in longitudinal and transverse galleries at elevation 310 m, where measurements, for reasons independent of the surveyors, were performed irregularly. The composition of the measurements, cyclicity, number of marks (points being determined), and volume of measurements of them during 1990 (by this time the geometric scheme of the geodetic network became constant) are given in Table 1. Approximately the same volume of measurements was taken in 1989. Settlements of the dam, its foundation, and bank surface are determined by differential leveling from clusters (groups) of initial bench marks 1 and 2 at a distance of 3 kin from the dam. In each cluster there are three bench marks in adits at a distance of 10 m from the surface. Leveling from bench marks 1 and 2 is carried out twice a year at the maximum (fall) and minimum (spring) levels of the upper pool. In the remaining cycles the settlements are determined from clusters 3 and 4, also of adit bench marks, the regularities of the change in the elevations of which are known from preceding measurements. The leveling lines run from the reference bench marks to the gallery of the dam at elevation 344 m and to the crest. The measuring elevations are interconnected in the dam by means of permanent devices Invar wires constantly suspended in shafts having on the ends scales and a weight at the bottom for tension (the Moscow Institute of Geodesy, Aerial Photography, and Cartography Engineers (MIIGAiK) which developed these devices called them height elevators). The length of the wires (differences between the zero lines of the upper and lower scales) are periodically compared with the length of a standard tape measure suspended alongside for comparison. Thus, the height network consists of closed traverses including sections of differential and hydrostatic leveling and height elevators. Adjustment of the network is carried out by the least-squares method. The root-mean-square error of determining settlements of the network marks does not exceed 1 mm. The horizontal network inside the dam includes plumb lines and traverse at three elevations (344, 413, and 467 m). The angles in the traverse are determined indirectly by means of stationary systems of extended triangles in which the heights are measured [2]. Owing to construction noise and damage of the wires forming the extended triangles, measurements at the two upper elevations were carried out irregularly, in particular, measurements were not taken at 413 m in 1989. The elevations are tied-in by alignments of reverse and direct plumb lines forming, together with the survey traverse, a geodetic network of closed traverses adjusted by a special program. The inner and outer networks are tied-in through points on the crest secured next to the plumb lines. Settlements of the Dam and Its Foundation. After filling the reservoir to the NPL (540 m) in 1990, the maximum measured settlements of the foundation were 45 ram. They were recorded in the central part of the dam (section 33) on the downstream side in a gallery at elevation 310 m (about 5 m above the foundation). More complete data on the character of the