No abstract

As a joint research project sponsored by the French–Norwegian Foundation for Scientific and Industrial Research, the Norwegian Geotechnical Institute, Geodia/Geocean, and École Polytechnique have carried out in the summer of 1990 a series of special in situ pressuremeter and cone penetration tests at a permafrost site in Longyearbyen, Svalbard. The purpose of the testing program was to compare the results of the two different field testing methods and to check the potential of such methods for serving as a basis for prediction of the behaviour of foundations in permafrost. As pressuremeter test results have been presented in a previous publication, this paper shows only the results of cone penetration tests and their interpretation in terms of strength and creep parameters of the frozen soil. Subsequently, the same data are used to predict the behaviour under long-term axial load of two prefabricated and one cast-in-place concrete pile that were tested by the Selmer Furuholmen Anlegg A.S. in cooperation with the Norwegian Geotechnical Institute at the same site for 3 years. Key words : permafrost, cone penetration tests, piles field loading tests, creep settlement.

No abstract

In 1995 an extensive load testing program was conducted on 30' pipe piles in dense silica sands similar to those found in the southern North Sea. A highly instrumented pile was driven at one location, extracted and redriven at a second location. A total of twelve static compression and tension load tests were performed at three penetration depths between 30 and 47 meters. Test objectives included acquiring reliable data on capacity of offshore type piles in sands which hitherto were hardly available and to understand the failure mechanisms in order to improve current foundation design. The paper describes the design of the testing programme and project organisation and funding. Details of the lest pile and instrumentation to measure the loads in the pile and the pile-sand interaction are highlighted with an emphasis on quality of the measurements. The field-work and innovative testing facilities are also described. Introduction Estimates of axial capacity of open-ended pipe piles in sand are an area of great uncertainty. Various authors (e.g. Pelletier et al., 1993) have argued that the world-wide used API recommended practice (RP2A, 1982-1991) does not properly account for fundamental parameters (pile length, plug behaviour, horizontal stresses, driving history, etc.) and underestimates the capacity of driven piles in very dense sands. On the other hand no other widely recognised design method is currently available. The base for a design method should be an understanding of the pile-sand load transfer in parallel with representative full scale pile load tests to confirm this (Hobbs, 1992). Some new promising design methods have been published in recent years (Randolph et nf., 1992, Lehane et al., 1992), but the existing data base on pile tests in sand is limited to piles of relatively small size and/or low capacity. Moreover few tests provide instrumented data of enough quality and resolution to confirm the load transfer in detail. While this unsatisfactory situation has been recognised by most parties involved in foundation design it appeared very difficult to resolve because high capacity pile tests in sand are relatively costly. In 1985 Heerema and Fugro developed a plan to test three 42' piles in very dense sands nearshore Eemshaven in The Netherlands (Fig. 1). On each pile compression and tension tests were foreseen at four penetration depths with a potential ultimate compression load of 90 MegaNewton (MN). The programme costs were estimated at US$ 12 million. Even though a 40% European subsidy was available not enough support from industry could be generated. This contrasts with the fact that about US$ 50 million is spent annually on pile foundations in the southern North Sea and a 30% increase in allowable load could lead to a saving of about US$ 60,000 per pile installed (van Zandwijk, 1986).

SAGITAIRE is an artificial intelligence software specifically developed to pezform in an integrated environment the complete chain of operations related to offshore foundation designs, from the acquisition of site investigation data to the design of pile foundations. The paper outlines the main advantages of the computer-aided design system, presents the architecture of the program and describes how the dataflow is managed and controlled during the elaboration of a geotechnical project. 1. INTRODUCTION Geotechnical engineering calls to a large extend upon the individual experience of skilled engineers and is certainly less computerized than other engineering fields. This is mainly because the soil is a complex, multiphasic, heterogeneous natural material and also because soil mechanics is a rather young science. In the particular case of foundation engineering of offshore jacket platforms the need for developing efficient and rational dimensioning tools has progressively emerged. The context is relatively favorable : an important flow of data has to be managed in a very short time, soil investigation techniques are practically standardized and dimensioning procedures are similar worldwide. The availability of efficient computer-based techniques and expert systems, the improved capabilities of graphic and database management tools and the recent hardware developments have opened the door to a more comprehensive, rational and efficient application of computers in the geotechnical process. An artificial intelligence software, named SAGITAIRE, has been developed in order to assist, in an integrated user-friendly environment, the geotechnical engineer in the complete chain of operations related to offshore foundation designs. This paper focuses on:some specific aspects of the offshore geotechnical practice;the objectives and general architecture of the software;the interest and applications of a computer-aided system in offshore geotechnics. It is particularly demonstrated how human errors and non rational choices can be avoided and quality control can be provided efficiently at any stage of the project;the detailed description of the software: description of the various databases; procedures used to interprete geotechnical data, derive boring logs and design parameters; methods introduced for the design of pile foundations; provision of typical documents and drawings;the possible further developments. 2. OFFSHORE GEOTECHNICAL PRACTICE The engineering of offshore jacket platform foundations is characterized by four successive and complementary steps:site investigation and associated laboratory testing,interpretation of soil data and construction of boring logs,assessment of synthetic soil profiles and design soil parameters,foundation design computations. Site investigation and laboratory testing The techniques and tools used to perform offshore soil investigations are practically standardized: wire-line operated equipments, push or hammer sampling depending to soil conditions, quasi exclusive use of the cone penetrometer (cpt) as in-situ measurement technique. As the number of geotechnical contractors involved in offshore soil investigation is very limited, the procedures applied and the equipments used are quasi normalized and well controlled.