Control of an offshore wind turbine modeled as discrete system Pedro Guimarães, Suzana Ávila To cite this version: Pedro Guimarães, Suzana Ávila. Control of an offshore wind turbine modeled as discrete system. 2nd ECCOMAS Young Investigators Conference (YIC 2013), Sep 2013, Bordeaux, France. <hal- 00855865> HAL Id: hal-00855865 https://hal.archives-ouvertes.fr/hal-00855865 Submitted on 30 Aug 2013 HAL is a multi-disciplinary open access archive for the deposit and dissemination of scientific research documents, whether they are published or not. The documents may come from teaching and research institutions in France or abroad, or from public or private research centers. L archive ouverte pluridisciplinaire HAL, est destinée au dépôt et à la diffusion de documents scientifiques de niveau recherche, publiés ou non, émanant des établissements d enseignement et de recherche français ou étrangers, des laboratoires publics ou privés.
YIC2013 Second ECCOMAS Young Investigators Conference 2 6 September 2013, Bordeaux, France Control of an offshore wind turbine modeled as discrete system Pedro Varella Barca Guimarães a, pedrobarca@hotmail.com, Suzana Moreira Avila b,avilas@unb.br a Dpto. de Engenharia Mecânica - Faculdade de Tecnologia - Universidade de Brasília Caixa Postal 04386-70910-900 - Brasília, D.F. - Brazil b Faculdade UnB Gama, Área Especial 02 Lote 14 Setor Central Gama - DF Abstract. Offshore wind turbines are installed at the sea and present a series of advantages such as wind tends to blow stronger and consistently with less turbulence. This work studies the dynamic behavior of wind turbines modeled as an inverted pendulum. Besides, as this type of pendulum is originally unstable, it is proposed a proportional control to stabilize its behavior. Keywords: Inverted Pendulum; offshore wind turbines; control. 1 INTRODUCTION Wind energy presents itself, nowadays, as one of energy sources in fast development and implementation all over the world. The design, construction and maintenance of the so called wind farm still present lots of challenges for engineers and researchers. Thus, offshore wind turbine are installed at the ocean next to coast presents advantages like getting benefits of more intense and consistent wind with less turbulence in these regions (Pao and Johnson, 2009). One of the conceptions of wind turbines is the floating device. This kind of structural system can be analyzed like a discrete model of an inverted pendulum. This article presents one modelling of inverted pendulum analyzing its dynamic behavior and stability. A Proportional, Integral and Derivative control system (PID) is based in approximate the value of the variable to the target value applying an acting value, which is transformed in a signal and sent to actuator used in the process to guarantee a stable and accurate control. To the problem discussed in this article, it will be applied the proportional control to minimize the vibration. 2 INVERTED PENDULUM MODEL For modeling the dynamics behavior of an offshore wind turbine, this article studies the behavior of an inverted pendulum s discrete model (Drummond et al, 1999) like is shown at Figure 1, in which there is a concentrated mass at the top of the bar. The system s excitation is done by a harmonic force applied at the concentrated mass, simulating the wind s force applied at blade/nacelle set. The motion equations of this system are given by: (1) Where a is half of the bar s length, c is the damping coefficient, Km is the torsional spring constant, g is the gravity acceleration, M is the concentrated mass, mb is the bar s mass, mc is the car s mass, F is the force applied at concentrated mass, θ is the bar s angle with the vertical and u is the car displacement. The system s Transfer Function is given by:
2 F. Author et al. Young Investigators Conference 2013 (2) Where A = (mb + 2.M).a², (mc - M) + (M + mb + mc).i; B = (I + (mb + 2.M).a²).c; C = (M + mb + mc).(km (mb + 2M).a.g); D = (Km (mb+2m).a.g).c. 3 NUMERICAL ANALYSIS Figure 1: Inverted pendulum with concentrated mass. For numerical analysis of the system, the following numerical values were considered for the variables. a = 45 [m] - Half of bar s lenght; c = 2400 [N.s/m] Damping coefficient; Km = 2,5667.10 14 [N/rad] Torsional spring constant; I = 9,5769.10 8 [kg.m²] Bar s moment of inertia in relation to the contact point with the torsional spring; g = 9,81 [m/s²] Gravity acceleration; M = 110000 [kg] Concentrated mass; mb = 3,5469.10 5 [kg] Bar s mass; mc = 240000 [kg] Car s mass; F Force applied to the concentrated mass; θ bar s angle with the vertical; Initially, it is performed an analysis varying the value of torsional spring constant Km to verify its influence over a possible system s stability. The initial value of Km adopted was calculated considering the bar rigidly attached to the car, case wherein the system would be stable. Using MATLAB, the transfer function s poles were calculated to verify system s stability and reducing the Km s value until the system becomes unstable. Results are shown in Table 1: Table 1: Relation between Km value and system s poles. Km values Poles Km = 8,2763.10 14-0,0000 + 840,21i -0,0000-840,21i -0,0000 10-6 Km = 8,2763.10 8-0.0014 + 0.6997i -0.0014-0.6997i -0.0034 10-7 Km = 8,2763.10 7 0.3805-0.3832-0.0034
F. Author et al. Young Investigators Conference 2013 3 The stability is verified when the real part of all poles are negative. If any of the poles is null, the system can be stable or unstable. The system is unstable if any of pole s real part is positive (Ogata, 2002). Thereby, the system became unstable between 10-6 Km and 10-7 Km, because the real part of one of the poles became positive. This result will be used for proportional control. 4 PROPORTIONAL CONTROL The proportional control applies an acting value to the variable s value in a proportional way. To implement this control to the offshore wind turbine, an external torsional spring can be used, because its resultant torque is proportional to the angular amplitude of the bar. To apply proportional control, it was used the value 0,3.10-6 Km, which makes the system unstable. The system s unstable state would be the worst case for the problem. The block diagram can be seen in Figure 2. Figure 2: Block diagram. Using previous data and applying the proportional control with constant P = mc + mb + M, the system s simulation was performed via software MATLAB and SIMULINK. The wind s force was considered as harmonic function added to a constant value. Figure 3 shows the graphic produced by simulation. 15 x 10-3 Angular Amplitude X Time 10 5 0-5 0 100 200 300 400 500 Figure 3: Graphic obtained from SIMULINK for angular amplitude versus time. Analyzing the graphic, it s clear that the angular amplitude is small enough to guarantee the linear behavior hypothesis, which should be less than 0,26 rad. For a better analysis, the block diagram for open loop was calculated considering the proportional control. Herewith, a new transfer function is found and the analysis of poles can be made to strengthen the result that proportional control was able to stabilize the system.
4 F. Author et al. Young Investigators Conference 2013 Calculating the new transfer function and its poles: Figure 4: Block diagram for open loop. (3) Where A = (mb + 2.M).a², (mc - M) + (M + mb + mc).i; B = (I + (mb + 2.M).a²).c; C = (M + mb + mc).(km (mb + 2M).a.g) + (mb+mc).(mc + mb + M); D = (Km (mb+2m).a.g).c + 2.c.a.(mc + mb + M). The poles were given by: -0,0000 + 0,1656i; -0,0000-0,1656i; -0,0061. By the result of the poles, it can be verified the proportional control made the system stable as expected, since the stability was already verified by the simulation. 5 CONCLUSIONS With the preliminary study of wind turbine s dynamic behavior by discrete model of inverted pendulum, it was possible to identify a critical situation to apply the proportional control. The inverted pendulum is a system originally unstable and with a non-linear behavior. The proportional control was effective for this problem, because it stabilized the system and kept it within the regime of small displacements to guarantee the hypothesis of linear behavior. For further studies, another control system can be analyzed, as the tuned mass damper, or TMD, which uses a pendulum to stabilize the vibration (Gordon and Matthew, 2011). REFERENCES [1] Ogata, K., Modern Control System, 4ª Edição, Prentice-Hall, New Jersey, 2002 [2] Pao L.Y. and Johnson K.E., A tutorial on the dynamics and control of wind turbines and wind farms Proceedings of the American Control Conference, St Louis, EUA, 2009 [3] A. C. Drummond, K. C. Oliveira, A. Bauchspiess, Estudo do Controle de Pêndulo Inverso sobre Carro utilizando Rede Neural de Base Radial, IV Congresso Brasileiro de Redes Neurais, 1999. [4] Gordon M. Stewart, Matthew A. Lackner, The effect of actuator dynamics on active structural control of offshore wind turbine, 2011