Comparison Of Field Measurements With Finite Element Results--Capitol Hill Relief Sewer Section II (Washington, D.C.) --

INTRODUCTION The state-of-the art in the area of soil-structure interaction is not sufficiently developed that one can accurately predict the earth load distribution on buried structures. This lack of knowledge has led to the use of high factors of safety in their design and conservative assumptions for the loading diagrams. As a step towards the improvement of the design methodology for buried structures, it is important to be able to apply new analytical techniques and monitor the performance data obtained on full-size buried structures. The performance data should preferably include: 1) the actual stress distribution on the buried structure, 2) stress distribution in the buried structure and surrounding soil media, 3) vertical and lateral movements in the soil adjacent to the buried structure, and 4) deflection of the buried structure. Brown (1967), Davis (1972), Krizek, et al (1974), Krizek and MacQuade (1978), and some few others have compared the performance of full-size culverts with finite element results. Despite all these efforts, a complete understanding of the soil-structure interaction problem has not yet been fully achieved. In an effort to make improvements in the existing design procedures for buried structures constructed in an open-cut, the Department of Environmental Services (DES), Washington, D.C., in conjunction with Howard University, undertook a detailed study of Section II of the Capitol Hill Relief Sewer (CHRS). Inclinometers were used to measure lateral movements of the excavation wall and the total normal interface stresses acting on the concrete sewer were obtained using glotzl total pressure cells. This paper describes the results of the field instrumentation program for acquisition of data, their comparison with the design assumptions, and the results of a series of post-construction finite element analyses of the CHRS project.