1 Copyright © 2013 by ASME Proceedings of the ASME 32nd International Conference on Ocean, Offshore and Arctic Engineering OMAE2013 June 9 - 14, 2013, Nantes, France OMAE2013-10592 FINITE-MEMORY NONLINEAR SYSTEM MODELLING OF OFFSHORE STRUCTURAL RESPONSE ACCOUNTING FOR EXTREME VALUES RESIDUES N.I. Mohd Zaki a , M.K. Abu Husain a , H. Mallahzadeh b and G. Najafian b a UTM Razak School of Engineering and Advanced Technology, Universiti Teknologi Malaysia, Jalan Semarak, 54100 Kuala Lumpur, Malaysia b Department of Engineering, University of Liverpool, L69 3GQ, United Kingdom ABSTRACT Offshore structures are exposed to random wave loading in the ocean environment, and hence the probability distribution of the extreme values of their response to wave loading is of great value in the design of these structures. Due to nonlinearity of the drag component of Morison’s wave loading and also due to intermittency of wave loading on members in the splash zone, the response is often non- Gaussian; therefore, simple techniques for derivation of the probability distribution of extreme responses are not available. Monte Carlo time simulation technique can be used to derive the probabilistic properties of offshore structural response, but the procedure is computationally demanding. Finite-memory nonlinear system (FMNS) modeling of the response of an offshore structure exposed to Morison’s wave loading has been used to reduce the computational effort, but the predictions are not always of high accuracy. In this paper, further development of this technique, which leads to more accurate estimates of the probability distribution of the extreme responses, is reported. Key Words: offshore structures, response, extreme values, Morison's equation, identification technique, finite-memory. 1 INTRODUCTION For an offshore structure, wind, wave and gravitational forces are all important sources of loading. The dominant load, however, is normally due to wind-generated random waves. Probabilistic properties of the loading and the resulting responses are therefore required for risk-based design of these structures. The major obstacle in the probabilistic analysis of the response due to wave and current loading, is the nonlinearity of the drag component of Morison's wave loading [1], which results in non-Gaussian probability distributions for both loading and response [2-5]. The problem is further compounded by current and by intermittent loading on members in the splash zone, which have a significant effect on the statistical properties of response [6, 7]. Probabilistic properties of response can be developed in the time, frequency or probability domains. In each case, sea- states are characterised by an appropriate water surface elevation spectrum, covering a wide range of frequencies. In the conventional time simulation (CTS) technique, the Morison-type nodal loads are calculated from simulated water particle kinematics. The analysis of the structure leads to a time history of (dynamic) response, from which its various statistics are estimated [8]. Due to large sampling variability, very long records are needed for stable statistics and hence computer run-time can be excessive [9]. NewWave theory [10-14] is another time-domain approach for derivation of the probability distribution of extreme responses. NewWave theory has successfully been used to evaluate the probability distribution of the extreme global responses for quasi-static (zero-memory) structures [11,15- 17]. Its extension to the case of dynamic structures is discussed in [18-20]. In the frequency domain, second moment statistical information arises from the standard linearised response analysis of the structure, though non-standard nonlinear spectral analysis procedures could be used to determine the first four statistical moments of response [21-27]; however, in their present form, these techniques are only applicable to simple structures. In the probability domain, the first four statistical moments of response are calculated directly in terms of the variances and cross-covariances of water particle kinematics at different nodes [4,28-30]. However, in its current stage of development, the technique, which has only been tested for quasi-static (zero-memory) responses, cannot account for current, load intermittency in the splash zone and wave directionality.