Reliability Analysis And Optimization Method Research For Offshore Wind Turbine Pile Foundation

Over the past decade, continually growing energy demands as well as global warming and other pollution concerns have drawn considerable attention to the need for alternative and renewable sources of energy-offshore wind energy. However, offshore wind turbines are costly and complex, usually working in severe environment on a long term basis, as many source of uncertainty are inherent in structural design such as variability in material properties, structural geometric dimensions and load) , absolute safety cannot be achieved. The reliability and economics of offshore wind turbines especially foundations is one of the currently challenges.Considering comprehensively the characters of offshore wind turbines, the optimum design theory and reliability method for offshore wind turbines and some relevant problems are researched based on those existing in the literature, the major contents in this thesis are summarized as follows:(1) A deterministic optimization method based PSO for offshore wind turbine foundationThe optimization design of offshore wind turbines foundation is a complicated, non-linearconstrained optimization task, traditional gradient-based optimization algorithms are limited to handle the non-linear problem, a particle swarm optimization (PSO) incorporating a penalty function was applied to the conceptual design of the foundation of offshore wind turbine in this paper. PSO is a population-based optimization algorithm, which mimics the social behavior of animals in a flock. It makes use of individual and group memory to update each particle position allowing global and local search optimization. The objective function considered was the total weight of the structure subjected to stresses, displacements and forces constraints. The effects of the parameters were investigated as well and such combination of tuning parameters promote a better global search behavior avoiding premature convergence while rapidly converging to the optimal solution. Results showed the effectiveness of the proposed method by comparing with ANSYS Design Optimization Tool (zero-order method). The PSO with the tuning parameters makes it an ideal method for offshore wind turbines foundations optimization tasks.(2) A reliability analysis method for pile foundation bearing axial loads based PSOThe performance function of pile foundation's axially bearing capacity sometimes is nonlinear and complex, on the basis of geometric meaning of structural reliability index, an optimum model with PSO for structural reliability analysis under arbitrary random variables was established, The PSO algorithm is very efficient to solve global optimization problemsIts use in structural reliability field presents not only the advantage of its facility of implementation, but also the possibility to obtain the design point and the failure probability with a good accuracy. In addition, PSO is a zero order algorithm, for no derivative is necessary for its implementation. Several examples of the literature studied in this paper show that the results yielded are quasi-exact with respect to those yielded by Monte Carlo Simulation (MCS) and response surface methods. Therefore, this algorithm can be considered as an additional efficient algorithm to those existing in the literature based on gradient method for geotechnical engineering.(3) A system reliability analysis method based SVM for offshore wind turbine foundationThe pile-soil structure systems of offshore wind turbines are complicated and theperformance functions of piles foundation's laterally and vertical bearing capacity are implicit, simulation based methods are often used to calculate an accurate value for the system reliability of these complicated structures. One of the major disadvantages however is the large number of simulations required to obtain an accurate estimate of the failure probability. This might result in an unrealistic processing time, making the method unusable for practical purposes. To address this issue, this paper presents a SVM-based system reliability simulation method. The idea of the proposed method is to construct a response surface of the indicator function, instead of constructing a response surface of the limit state function itself. After the response surface of the indicator function is constructed and refined, MCS simulation is implemented. The results have shown the proposed method is compared well with those obtained by the conventional reliability methods such as the direct MCS, IS and is applicable to structural system reliability analysis involving implicit performance functions such as pile foundation.(4) A RBDO method research of offshore wind turbine foundationThe cost of offshore wind turbines especially foundations is large, in the deterministic-based optimization, designs are pushed to the limits of the constraints boundaries leaving no room for tolerances in modeling materials, loads and performance uncertainties and could result in unreliable designs. In modern structural optimization design, the uncertainties must be considered. In this study, a new RBDO method based on SVM has been presented. The method makes use of a two-tier response surface approximation strategy to significantly reduce the computational cost of RBDO for complex structure the first-tier of response surface is analysis response surface (ARS), which is fitted to limit state functions in terms of both design variables and random variables. The second tier is design response surface (DRS) that is fitted to probability of failure as a function of design variables. As illustrated by optimizing the tripod foundation of an offshore wind turbine with subject to multiple reliability constraints, the two-tier response surface approximation strategy exhibits more efficient results than the conventional approximation strategy, while maintaining good accuracy, this strategy is a potential competitive method for RBDO for offshore wind turbines foundations.(5) Selection of the type of foundations for offshore wind farm on fuzzy optimization methodBecause of the difficulties met in conception design in optimum lectotype for the offshore wind turbines foundations, a multi-level and multi-factor fuzzy optimization model has been used. In order to decrease the subjective influences, the multi-level fuzzy optimization has been introduced to cover most of the effect factors, and the prior relation method has been used to determine the membership in comprehensive optimization and weigh ting. Considering various 13 factors of optimization objects and 4 projects, the method has been successfully used to testify by an offshore wind farm, and very ideal results have been obtained, the mufti-level fuzzy optimization model is remarkable.