Application of Wireless Communication to Small WECS with Induction Generator

Transcription

1 Application of Wireless Communication to Small WECS with Induction Generator Marek Adamowicz, Ryszard Strzelecki Department of Ship Automation Gdynia Maritime University Gdynia, Poland Zbigniew Krzeminski, Janusz Szewczyk, L. Lademan Department of Electrical Drives and Automation Gdansk University of Technology Gdansk, Poland Abstract - Wind energy conversion systems (WECS) seem to be self-evident elements of the future smart grids. Among many generator types the squirrel cage induction generator (SCIG) characterizes robustness and low cost. A full-scale power converter can perform smooth grid connection over wide speed range of multiple-stage geared SCIG. Increasing number of sensors can improve the WECS control and diagnostics but increasing number of thin wires i.e. between a nacelle and the converter housing can reduce reliability. A wireless communication using ADF7020 transceiver for data acquisition and transmission is proposed in the paper. I. INTRODUCTION The future smart grids will incorporate increased share of wind power and renewable energy sources[1], [2], [3]. That presents significant challenges to many aspects of operation, control and communication. Wind energy conversion systems (WECS) seem to be self-evident elements of the future smart grids. It can be installed and interconnected in a relatively short period of time. Among many generator types the squirrel cage induction generator (SCIG) characterizes robustness and low cost for mass production. In order to fulfill the variable speed operation a variable speed multiplestage geared SCIG is equipped with a full-scale converter. Compared with the variable speed concept with a partialscale power converter used for doubly fed induction generators, the full-scale power converter can perform smooth grid connection over the entire speed range. The decreasing cost of power electronics may make variable speed multiple-stage geared concepts more and more attractive [4]. The advances in wireless communications have enabled the implementation of low-cost and multifunctional wireless sensors and communication modules in wind power plants [11] and drive systems [12]. Two main factors that describes wireless sensors are link quality and power consumption [13]. In the case of power converter integrated transceiver, the latter factor can be omitted in comparison with whole power consumption of control card of WECS. The concept of the windmill diagnosis using wireless network has been very recently proposed in [11]. The measurement data from sensors related to windmill condition monitoring may not be time critical because typical faults including contamination of oil in gear box, rise in temperature of generator windings or excessive nacelle vibration due to structural faults develop slowly [11]. Unlike diagnosis problems the use of wireless communication for the control purpose requires significant computational effort of DSP and higher performance of the link quality. In this paper an attention is paid to application of wireless communication to small WECS dedicated for distributed generation and microgrids. It can serve for communication with user interface as well control purposes and diagnostics purposes. Example of wireless protocol is described. Some experimental and simulated waveforms are presented to show the performance of investigated 40 kw WECS with SCIG. II. DESCRIPTION OF INVESTIGATED WECS An overview of smart WECS under investigation is shown in Figure 1. It comprises of: a wind turbine coupled with an induction generator within a moving nacelle, full-scale converter, control board with ADSP-21065L providing sensorless control of the induction generator and power flow control, wireless communication with user interface, wireless wind measurement. The wind turbine contains a 10 meter three-blade rotor coupled with gearbox [6] /10/$ IEEE 944

2 A speed observer-based Maximum Power Point Tracking (MPPT) control strategy is used to capture the maximum power from the wind energy. The optimal speed referred to maximum power is found according to the aerodynamic characteristics of wind turbine blades [1], [7], [8]. A general overview of the signal flow within investigated WECS is shown in Fig. 2. Mechanical power P m of the wind turbine depends on the third power of the wind speed, that is mainly influenced by the installation site and the tower height [1], [8]: 2 3 ( λ β) ρ υ P = 0.5 C, R (1) m A tip speed ratio λ in (1) is given by: p Figure 1. Small WECS with applied wireless communication A nacelle angle against the wind is forced by small assisted drive (M) with the FPGA based controller. Wind sensor on the nacelle provides a signal to the yaw controller to point the turbine. The windmill operates from 3.5 m/s with the nominal wind speed of 14.5m/s. It is turned-off with the wind speed of above 20 m/s. In basic configuration the nacelle is positioned at 90 degree angle in case of strong winds to protect the wind turbine against damages. For the 40kW wind energy conversion system an induction generator is chosen due its robustness, low investment costs and low maintenance requirements. For achieving a high performance power conversion the flux level of the induction generator should be optimized to reduce the core loss and to maximize the output power. The full-scale power converter interfaces the generator with a microgrid or single end user. Among many features described in referred literature [5], [7], [8] the back-to-back converter topology offers possibility of energy storage interconnection to DC-link. A wireless communication shown in Fig. 1 can be used for remote data acquisition, diagnosis, control and monitoring of wind generator. Moreover, WECSs can communicate with each other. In present application a 868 MHz band is utilized to provide a communication between WECS and user interface as well between sensors mounted in the nacelle and converter controller. III. DESCRIPTION OF CONTROL SYSTEM The proposed control system has been implemented on a Sh65 control card with an ADSP-21065L digital signal processor developed in Gdansk University of Technology. Both generator control and the control of grid side converter are realized in single DSP. The program interrupt is generated every 150 us. That gives the switching frequency of power converter equal 3.333kHz. ωr λ = (2) υ where ω is the rotational speed of the turbine blades, R denotes a blades radius, ρ denotes an air density and υ denotes the effective wind velocity. The power coefficient C p in (1) is used in control of wind turbine. The power coefficient indicates how large share of total power from wind can be converted by the wind turbine. Its theoretical maximum value is C p =0.593 however in practice the power coefficients occur in the range of C p = The structure of the WECS control scheme using MPPT algorithm is shown in Figure 3. REF The reference signal of mechanical torque T e is calculated from maximum power P m of wind turbine: Pm where optimal value of K is given by: 3 = K ω (3) 2 Kopt = 0.5 ρ R Cp _ max (4) Therefore: REF Pm Te = (5) ω where C p_max in (4) is taken from the curve shown in Fig. 4. Figure 2. A general overview of investigated system 945

3 Figure 3. WECS control scheme using MPPT algorithm Figure 4. Experimental curve of power coefficient C p versus tip speed ratio The algorithm of maximum K coefficient tracking includes four steps: registration of measurements of wind velocity ν, wind turbine mechanical torque T m and rotor speed in certain time period, calculation of C p coefficient and tip speed ratio λ, determination of the actual operating point on the characteristic curve power conversion coefficient versus tip speed ratio, correction of the K coefficient. Application of speed observer [9], [10] eliminates the rotor speed measurement for control system. Application of wireless sensor [11], [12] for wind speed measurement can further improve reliability of the WECS system. IV. INTERCONNECTION OF ADF7020 WITH SH65 The range of wireless communication depends on the transceiver parameters: sensitivity, output power and operating frequency. The huge influence of the range has also an environment in which the modules work. Factors such as air humidity, range of sight, building materials, metal film sun-screened windows etc. limit maximum useful range. Two recently developed chips for industry-oriented wireless communication applications Texas Instruments CC1100 and Analog Devices ADF7020 were compared in [13]. Both transceivers operate in the license-free ISM band at 868MHz. Both also perform frame synchronization, bit tracking, and detection, and both are characterized by high date rate above 100Kbps. However, detailed comparison described in [13] indicates lover noise floor and higher sensitivity of ADF7020. ADF7020 transceiver was used in proposed WECS for wireless communication with user interface. For ADF7020 board matching to Sh65 control board voltage levels a simple interface was designed [14] which enables to supply the transceiver from control board. ADF7020 systems do not possess neither built-in controller nor memory for writing initialization procedure. Therefore ADF7020 registers have to be programmed from control board after turning on. Configuration of first two registers in proposed system is shown below: Register 0 - N REGISTER: FRACTIONAL_N = (transmit mode Tx); INTEGER_N = 78 (transmit mode Tx); FRACTIONAL_N = (receiving mode Rx); INTEGER_N = 78 (receiving mode Rx); TX_RX = 0 (transmit mode Tx, 1 for Rx); PLL_ENABLE = 1 (PLL on); MUXOUT = 3 (not used); Register 1 - OSCILLATOR/FILTER REGISTER: R_COUNTER = 1; XTAKL_DOUBLER = 0 (off); CLOCKOUT_DRIVE = 0 (off); XOSC_ENABLE = 1 (generator on); VCO_BAND = 0 (862 MHz to 956 MHz); CP_CURRENT = 3 (current pump = 2.1 ma); VCO_BIAS = 10 (VCO BIAS current = ma); VCO_ADJUST = 0 (650 MHz to 920 MHz); IF_FITLTER_BW = 2 (band pass 200 khz). On the basis on Register 0 and Register 1 set-up the operational frequencies are: f out = Hz for Tx mode f out = Hz for Rx mode V. SIMULATION AND EXPERIMENTAL RESULTS The wireless data transmission between WECS and user interface using ADF7020 transceivers was tested in laboratory. First transceiver was connected to Sh65 control board within investigated WECS and the second was connected to user PC. Suitable packet formats were developed by authors [14]. A detailed description of packet formats very similar to used here can be found in [15]. The Data Link layer packet consists of 7 bytes of a preamble and 3 bytes of synchronization used for recognition 946

4 of the start of the information section. The first part of the information section is the frame length (in bytes) [15]. The second part of the information section is a data packet that contains WECS address and sampled data from WECS. The sync bytes used in experiment are 0x73, 0x55, 0x73 and the number of bytes in frame is 20. Except WECS address only 5 bytes are used for sampled data that are: wind speed, turbine speed, magnitude of machine flux and status. An example of data transmitted between WECS and user interface is shown in Table 1. TABLE I. TABLE I. SAMPLED DATA USED IN EXPERIMENT Data Decimal format Binary format WECS address [-] Wind speed [m/s] Turbine speed [rpm] Output power [kw] Machine flux p.u Status Experimental results of wireless communication with sampled data from Table 1 are shown in Figure 5. The frequency of clock signal was khz. The registered transmission delay between signal transmitted from WECS and received in user interface was μs. Figure 6. Experimental and simulated waveforms of investigated WECS without MPPT control VI. CONCLUSIONS An application of a wireless application to a WECS system with induction generator was proposed in the paper. Application of ADF7020 transceivers guarantees flexible data acquisition and transmission for communication with user interface as well WECS control and condition monitoring. Applied ADF7020 transceivers enable fast data transmission equal kbps at the distance of 100 meters in investigated configuration. The distance can be increased by controlling the power of the transmitter. Proposed wireless data transmission can be extended to wind speed measurement for the control purpose. The transmission speed can be further increased using ADF7025 transceivers. Figure 5. Experimental waveforms of wireless communication Application of MPPT control enables the processing of 95% of energy available from wind turbine which is important particularly at low speeds. Proposed control algorithm of maximum K coefficient tracking ensures an optimization of total energy amount processed in the windmill which can be seen from comparison between Figure 5 and Figure 6. Figure 7 shows picture of investigated WECS [6]. Figure 7. Simulated waveforms of investigated WECS with MPPT control and K coefficient searching algorithm 947

5 a) b) Figure 8. Pictures of investigated windmill of type Enwia 40 [6] with 37 kw induction generator (a) and AC-DC-AC converter (b) ACKNOWLEDGMENT This scientific work is financed from resources for science in years as a LIDER research project of the National Centre for Research and Development (NCBiR). REFERENCES [1] R. Strzelecki, G. Benysek, Eds: Power Electronics in Smart Electrical Energy Networks, Power Systems, Springer London, 2008, 414 pages. [2] Wang J., Huang A.Q., Sung W., Liu Y., Baliga B. J: Smart Grid Technologies. Development of 15-kV SiC IGBTs and Their Impact on Utility Applications, in IEEE Ind. Electron. Mag., June 2009, pp [3] Divan D.: Smart Distributed Control of Power Systems, IEEE PES General Meeting - Conversion and Delivery of Electrical Energy in the 21st Century, Pittsburgh, PA, 2008, pp [4] Li H., Chen Z., Overview of different wind generator systems and their comparisons, IET Renew. Power Gener., 2008, Vol. 2, No. 2, pp [5] Iov F., Blaabjerg F.: Power Electronics and Control for Wind Power Systems, IEEE Conference on Power Electronics and Machines in Wind Applications, PEMWA 2009, CD [6] Enwia 40, [7] Abo-Khali A.G., Kim H.-G., Lee D.-C., Seok J.-K.: Maximum Output Power Control of Wind Generation System Considering Loss Minimization of Machines, in Proc. of the IEEE Industrial Electronics Society, November 2-6,2004, Busan, Korea, pp [8] Pena R., Cardenas, R., Blasco R., Asher G., Clare J.: A Cage Induction Eenerator Using Back-to-back PWM Converters for Variable Speed Grid Connected Wind Energy System, IEEE IECON Rec., pp , [9] Adamowicz M.: Control of the Induction Motor with the Weakened Flux in the Air Gap, PhD Thesis, Gdansk University of Technology, 2008 (in Polish). [10] Adamowicz M., Krzemiński Z.: Novel Adaptive Flux Observer for Wide Speed Range Sensorless Control of Induction Motor, in Proc. European Conference on Power Electronics and Applications EPE, Aaalborg, Denmark, 2007, CD-ROM. [11] Khan Z. H., Thiriet J. M., Genon-Catalot D.: Wireless Network Architecture for Diagnosis and Monitoring Applications, IEEE Conf. on Consumer Communications and Networking Conference, CCNC 2009, CD-ROM. [12] Lu B., Gungor V.C.: Online and Remote Motor Energy Monitoring and Fault Diagnostics Using Wireless Sensor Networks, IEEE Transactions On Industrial Electronics, Vol. 56, No. 11, November 2009, pp [13] He Y., Fikkema P.G.: System-Level Characterization of Single-Chip Radios for Wireless Sensor Network Applications, in Proc. of 10th IEEE Annual Wireless and Microwave Technology Conference WAMICON '09, 2009, CD-ROM. [14] Lademan L., Szewczyk J.: Application of ADF7020 transceivers for wireless data transmissions between SH65L control boards. Research Report. Gdynia Maritime University, July [15] Marinkovic S., Spagnol C., Popovici E.: Energy-Efficient TDMAbased MAC Protocol for Wireless Body Area Networks, in Proc. Of Third International Conference on Sensor Technologies and Applications SENSORCOMM 2009, IEEE Computer Society, pp