Mahdi Teimouri Sichani

Failure Probability Estimation of Wind Turbines by Enhanced Monte Carlo Method

This paper discusses the estimation of the failure probability of wind turbines required by codes of practice for designing them. The Standard Monte Carlo (SMC) simulations may be used for this reason conceptually as an alternative to the popular Peaks-Over-Threshold (POT) method. However, estimation of very low failure probabilities with SMC simulations leads to unacceptably high computational costs. In this study, an Enhanced Monte Carlo (EMC) method is proposed that overcomes this obstacle. The method has advantages over both POT and SMC in terms of its low computational cost and accuracy. The method is applied to a low-order numerical model of a 5 MW wind turbine with a pitch controller exposed to a turbulent inflow. Two cases of the wind turbine model are investigated. In the first case, the rotor is running with a constant rotational speed. In the second case, the variable rotational speed is controlled by the pitch controller. This provides a fair framework for comparison of the behavior and failur...

First passage probability estimation of wind turbines by Markov Chain Monte Carlo

Markov Chain Monte Carlo simulation has received considerable attention within the past decade as reportedly one of the most powerful techniques for the first passage probability estimation of dynamic systems. A very popular method in this direction capable of estimating probability of rare events with low computation cost is the subset simulation (SS). The idea of the method is to break a rare event into a sequence of more probable events which are easy to be estimated based on the conditional simulation techniques. Recently, two algorithms have been proposed in order to increase the efficiency of the method by modifying the conditional sampler. In this paper, applicability of the original SS is compared to the recently introduced modifications of the method on a wind turbine model. The model incorporates a PID pitch controller which aims at keeping the rotational speed of the wind turbine rotor equal to its nominal value. Finally Monte Carlo simulations are performed which allow assessment of the accuracy of the first passage probability estimation by the SS methods.

Model Predictive Control of Buoy Type Wave Energy Converter

Abstract The paper introduces the Wavestar wave energy converter and presents the implementation of model predictive controller that maximizes the power generation. The ocean wave power is extracted using a hydraulic electric generator which is connected to an oscillating buoy. The power generator is an additive device attached to the buoy which may include damping, stiffness or similar terms hence will affect the dynamic motion of the buoy. Therefore such a device can be seen as a closed-loop controller. The objective of the wave energy converter is to harvest as much energy from sea as possible. The straight forward solution to this maximization problem is achieved by maximizing the instantaneous range of motion of the buoy. The buoy as a single degree of freedom oscillator will undergo its maximum movements when it is in resonance with the sea state. Hence the best solution to the problem is achieved by forcing this condition. In the paper the theoretical framework for this principal is shown. The optimal controller requires information of the sea state for infinite horizon which is not applicable. Model Predictive Controllers (MPC) can have finite horizon which crosses out this requirement. This approach is then taken into account and an MPC controller is designed for a model wave energy converter and implemented on a numerical example. Further, the power outtake of this controller is compared to the optimal controller as an indicator of the performance of the designed controller.

Failure Analysis of Wind Turbines by Probability Density Evolution Method

The aim of this study is to present an efficient and accurate method for estimation of the failure probability of wind turbine structures which work under turbulent wind load. The classical method for this is to fit one of the extreme value probability distribution functions to extracted maxima of the response of wind turbine. However this approach may contain high amount of uncertainty due to arbitrariness of the data and the distributions chosen. Therefore less uncertain methods are meaningful in this direction. The most natural approach in this respect is the Monte Carlo (MC) simulation. This however has no practical interest due to its excessive computational load. This problem can alternatively be tackled if the evolution of the probability density function (PDF) of the response process can be realized. The evolutionary PDF can then be integrated on the boundaries of the problem, i.e. the exceedance threshold of the response, which results in the accurate values of the failure probability. For this reason we propose to use the probability density evolution method (PDEM). PDEM can alternatively be used to obtain distribution of the extreme values of the response process by simulation. This approach requires less computational effort than integrating the evolution of the PDF; but may have less accuracy. In this paper we present the results of failure probability estimation by the PDEM. The results will then be compared to the extrapolated values from the extreme value distribution fits to the samples response values.

Non-Linear Aeroelastic Stability of Wind Turbines

As wind turbines increase in magnitude without a proportional increase in stiffness, the risk of dynamic instability is believed to increase. Wind turbines are time dependent systems due to the coupling between degrees of freedom defined in the fixed and moving frames of reference, which may trigger off internal resonances. Further, the rotational speed of the rotor is not constant due to the stochastic turbulence, which may also influence the stability. In this paper, a robust measure of the dynamic stability of wind turbines is suggested, which takes the collective blade pitch control and non-linear aero-elasticity into consideration. The stability of the wind turbine is determined by the maximum Lyapunov exponent of the system, which is operated directly on the non-linear state vector differential equations. Numerical examples show that this approach is robust for stability identification of the wind turbine system.

Nonlinear Motion Analysis of the Wavestar Wave Energy Converter With a Focus on the Structural Responses

The scope of this paper is to connect a nonlinear WEC numerical model with a structural response model. The numerical WEC model takes into account the nonlinear hydrostatic restoring moment of the Wavestar float. A parameterized structural model of the Wavestar arm is developed in ANSYS APDL. Based on the assumption that the structural displacements remain small, linear first order theory is used to calculate the structural response. The section moments and forces are thus superimposed according to the superposition law. The effect of the nonlinear hydrostatic restoring moment on the structural response is investigated. Moreover, an analysis is carried out which shows that reactive control, applied as a closed loop control, increases the section moments and shear forces.Copyright