Neuron-Adaptive PID Based Speed Control of SCSG Wind Turbine System

Hindawi Publishing Corporation Abstract and Applied Analysis Volume 2014, Article ID 376259, 10 pages http://dx.doi.org/10.1155/2014/376259 Research Article Neuron-Adaptive PID Based Speed Control of SCSG Wind Turbine System Shan Zuo,1 Yongduan Song,1,2 Lei Wang,1,2 and Zheng Zhou3 1 Institute of Intelligent System and Renewable Energy Technology, University of Electronic Science and Technology of China, Chengdu 611731, China 2 Intelligent Systems and New Energy Technology Research Institute, Chongqing University, Chongqing 400044, China 3 Web Science Center, University of Electronic Science and Technology of China, Chengdu 611731, China Correspondence should be addressed to Lei Wang; Received 11 March 2014; Accepted 14 April 2014; Published 19 May 2014 Academic Editor: Shen Yin Copyright © 2014 Shan Zuo et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. In searching for methods to increase the power capacity of wind power generation system, superconducting synchronous generator (SCSG) has appeared to be an attractive candidate to develop large-scale wind turbine due to its high energy density and unprecedented advantages in weight and size. In this paper, a high-temperature superconducting technology based large-scale wind turbine is considered and its physical structure and characteristics are analyzed. A simple yet effective single neuron-adaptive PID control scheme with Delta learning mechanism is proposed for the speed control of SCSG based wind power system, in which the RBF neural network (NN) is employed to estimate the uncertain but continuous functions. Compared with the conventional PID control method, the simulation results of the proposed approach show a better performance in tracking the wind speed and maintaining a stable tip-speed ratio, therefore, achieving the maximum wind energy utilization. 1. Introduction With the fast development of wind power generation systems, the generating capacity of wind turbines is expected to reach up to 10 MW [1]. Consequently, the wind turbine weight and size have to be increased simultaneously, with the bald diameter reaching up to 10 meters [2], as shown in Figure 1, which imposes technical difficulty in designing, transporting, and installing such large turbine blades. To address this challenge, novel concept of wind turbine generators with high energy density is urgently needed. High-temperature superconducting (HTS) technology is an expected solution. The research of wind turbines with SCSG has gained worldwide attention during the past decade [3–11]. Among various issues related to SCSG wind power generation systems, speed control represents one of the most crucial ones. Because of the inherent nonlinear and uncertain characteristics of the system, traditional PID control, although simple in structure and used widely in industry, is difficult to achieve reliable variable speed control performance in the blow-rated speed region. To address this issue, several advanced control approaches have been studied, such as single neuron-adaptive PID control approach, BP neural network PID control approach, fuzzy RBF neural network PID control approach, genetic algorithm PID control approach, and adaptive fuzzy PID control approach. However, previous studies show that the response time of single neuron-adaptive PID is comparatively long, and most of the existing algorithms are computationally expensive, and some of them even lead to larger overshoot than traditional PID. The neural network control approach with self-learning and strong self-adaptive characteristics can effectively reduce the negative impact arising from the system parametric uncertainties and stochastic disturbances. Motivated by this fact, in this paper, a single neuronadaptive PID controller based on Delta learning regulation is introduced, in which the RBF neural network is employed to estimate the uncertain but continuous function. Analysis and 2 Abstract and Applied Analysis Main stream geared 5 MW 13–15 m Mass of nacelle +Hub mTop ∼310 to 430 t +Blades Extrapolated for 10 MV mTop ∼750–850 t Possible to go as large as 20 MW with HTS 10 MW 10 m 4.5 MW 13 m 5m Generator Gearbox Hub Optimized HTS direct drive (AMSC) Conventional direct drive 6m 12 m mTop ∼500 t mTop ∼800 t–900 t mTop < 500 t Blade Nacelle Tower Figure 1: Comparison of nacelle sizes for technology options for large systems (image from American Superconductor). simulation results show that the proposed control approach has better performance in terms of robustness, stability, and computational cost as compared with other modified PID methods, thus being mode suitable for the speed control of SCSG wind turbine systems. 2. Dynamic System Modeling 2.1. Configuration of the Superconducting Generator. The SCSG for wind turbine system has a multiple synchronous high-temperature superconducting (HTS) field winding for direct drive train and has been widely studied worldwide. Figure 2 shows the configuration of the 10 MW SCSG wind power generation system, including the wind turbine, the generator, and the convertor [12]. Physical properties and electrical properties of the designed SCSG are given in Tables 1 and 2, respectively [1]. 2.2. Modeling of the Superconducting Synchronous Generation System. It is well known that the expression for power produced by a wind turbine is simply given by 1 𝑃𝑆 = 𝐶𝑝 (𝜆, 𝛽) 𝜌𝜋𝑅2 V3 , 2 Table 1: Physical properties of the designed SCSG. Items Value Items Value Rated power Rated line to line voltage Rated armature current 10 MW Number of poles 24 13.8 kV Rated frequency 2 Hz 418 A Number of phases 3 Rated field current 100 A Rated rotating speed 10 RPM Length of HTS wire Operating temperature 919 km 20 K Table 2: Electrical properties of the designed SCSG. Items Turns of stator coil Number of slots Number of slots per pole per phase Current density of stator wire Space factor of stator wire Turns of field coil Value 28 144 2 5 A/mm2 0.4 1500 (1) where 𝜌 is air density, 𝑅 is the radius of rotor, and V is wind speed passing the rotor. 𝐶𝑝 denotes power coefficient of wind turbine, which is a function of the tip-speed ratio 𝜆 and the pitch angle 𝛽 [13]. Note that the tip-speed ratio is defined by 𝜆= VTip V = 𝑅𝜔 , V where VTip is the tip-speed and 𝜔 is the rotor speed. (2) Abstract and Applied Analysis 3 Rotor dewar Superconducting coils Excitation power supply 3 phase, 13.8 KV, 10 MW, 2 Hz To power grid 3 phase, 13.8 KV, 10 MW, 2 Hz Generator AC-DC-AC-convertor Wind turbine Cooling system Figure 2: Configuration of the 10 MW SCSG system. In the lower-rated wind speed region, the maximum power point tracking (MPPT) control approach is adopted. The maximum power of the wind turbine is expressed as [14] 5 1 𝜌𝜋𝑅 𝐶𝑝 max 3 𝜔. 𝑃max = 2 𝜆3opt (3 (...truncated)