Page 37 - ASME DSCC 2015 Program

Technical Program

Model Predictive Control for Energy Maximization of Small Vertical
ConTRIBuTED SESSIon
1-25-1 WP4 Advances in Wind Energy Systems Axis Wind Turbines
Geroge Bellows f 4:00pm–6:00pm Contributed regular paper. DSCC2015-9891
Aykut onol, ugur Sancar, Ahmet onat, Serhat Yesilyurt, Sabanci
Session Chair: Hosam K. fathy, The Pennsylvania State University University, Istanbul, Turkey
Session Co-Chair: Soumik Sarkar, Iowa State University In this paper, a model predictive control (MPC) approach is presented to
maximize the energy generated by a small vertical axis wind turbine (VAWT)
understanding Wind Turbine Interactions using Spatiotemporal
subject to current and voltage constraints of electrical and power electronic
Pattern network
components. Our method manipulates a load coefficient and optimizes the
Contributed regular paper. DSCC2015-9784
control trajectory over a prediction horizon such that a cost function that
Zhanhong Jiang, Soumik Sarkar, Iowa State University, Ames, IA, United measures the deviation from the maximum available energy and the viola-
States
tion of current and voltage constraints is minimized. Simplified models for
This paper presents a data-driven modeling framework to understand spa- the VAWT and a permanent magnet generator have been used. A number
tiotemporal interactions among wind turbines in a large scale wind energy of simulations have been carried out to demonstrate the performance of
farm. A recently developed probabilistic graphical modeling scheme, namely the proposed method at step and oscillatory wind conditions. Furthermore,
the spatiotemporal pattern network (STPN) is used to capture individual impacts of the constraints on energy generation have been investigated.
turbine characteristics as well as pair-wise causal dependencies. The causal Moreover, the performance of the MPC has been compared with a typical
dependency is quantified by a mutual information based metric and it has maximum power point tracking algorithm in order to show that maximizing
been shown that it efficiently and correctly captures both temporal and the instantaneous power does not mean maximizing the energy; and simula-
spatial characteristics of wind turbines. The causal interaction models are tion results have shown that the MPC outperforms the maximum power point
also used for predicting wind power production by one wind turbine using tracking algorithm in terms of generated energy by allowing deviations from
observations from another turbine. The proposed tools are validated using the maximum power instantaneously for future gains in energy generation.
the Western Wind Integration data set from the National Renewable Energy Two-Stage Design of linear feedback Controllers for a Proton
Laboratory (NREL). Exchange Membrane fuel Cell
Multi-objective optimal Control of Wind Turbines for Speed Regulation Invited session paper. DSCC2015-9973
and load Reduction Verica Gajic, Patrick Rose, Garrett Clayton, Villanova University,
Contributed regular paper. DSCC2015-9787 Villanova, PA, United States
jackson g. Njiri, dirk Söffker, University of Duisburg-Essen, Duisburg, The paper considers the eighth-order proton exchange membrane (PEM)
Germany fuel-cell mathematical model and shows that it has a multi-time scale proper-
This paper presents a multi-objective control method for regulating speed ty, indicating that the dynamics of three model state space variables operate
and reducing structural loads on large wind turbines. Structural loads in the slow time scale and the dynamics of five state variables operate in the
become more pronounced as the turbine size increases resulting in the fast time scale. This multi-scale nature allows independent controllers to be
need to develop control algorithms to minimize the load in addition to the designed in slow and fast time scales using only corresponding reduced-
mainstream objectives of regulating speed and maximizing power extraction order slow (of dimension three) and fast (of dimension five) sub-models. The
in the turbines. To realize the two competing objectives of controlling speed/ presented design facilitates the design of hybrid controllers, for example,
power and structural load minimization, two control loops are employed. the linear-quadratic optimal controller for the slow subsystem and the eigen-
The first loop, collective pitch controller (CPC), is designed to regulate value assignment controller for the fast subsystem. The design efficiency
rotor rotation speed by reducing aerodynamic power coefficient in high and its high accuracy are demonstrated via simulation on the considered
speed regime. Then a multi-input multi-output (MIMO) controller based on PEM fuel cell model.
Stochastic Proportional-Integral-Observer is used in the second control loop
to fulfill the objective of load reduction of rotor blades. The performance of
the proposed control algorithm is evaluated against Proportional-Integral
controller (PI-Controller). The results demonstrate that the proposed control
method can achieve the objective of reducing structural load without much
influence on the speed regulation objective.

37