A Comparative Analysis Between the Performances of Outdoor Hybrid System Located in Burj Al-Arab and Complete Real System Model of Wind Turbine Power Generation Which Was Built in MATLAB/SIMULINK using Permanent Magnet Synchronous Machine

A Comparative Analysis Between the Performances of Outdoor Hybrid System Located in Burj Al-Arab and Complete Real System Model of Wind Turbine Power Generation Which Was Built in MATLAB/SIMULINK using Permanent Magnet Synchronous Machine

1. M. Aboras and A. A. Hossam El-din Ahmed H. H. Ali

   Department of Electrical Engineering Department of Energy Resource and Enviromental Engineering Alexandria University Egypt-Japan University of Science and Technology

   Alexandria 21544, Egypt. Alexandria 21934, Egypt.

   Abstract – With the current rapid industrial development in the world, energy shortage has become one of the biggest issues that many countries are facing. As a consequence of rising cost of fossil fuel and advanced technology, renewable, clean and green energy is the perfect alternative for non-renewable energy. In this paper, well compare between real values modeling of wind turbine power generation system has been built in MATLAB/Simulink and experimental reading for output dc current and output dc voltage before battery storage line Vs wind speed applied on outdoor Hybrid system located in Burj Al-Arab trying to prove that this complete real system wind turbine system has been built in SIMULINK model is valid for any micro turbine power generation of power with permanent magnet synchronous machine (PMSM) located in any site. The WT model was built using the real values of mechanical torque and power. The fixed controlled pitch wind-driven turbine model has been used for generating power with (PMSM). Each part of the system is built from the dynamics of every part of the generation system with their interconnections. The model is built in the MATLAB/Simulink using SimPower Systems library. The performance of the built model is studied for an isolated load. The SIMULINK model of the whole system with wind turbine has been developed and some relations and curves are deduced by XY graph and scopes from MATLAB/Simulink model.

   Index Terms – Micro-turbine, permanent magnet synchronous machine, wind turbine, hybrid system.

   1. INTRODUCTION

      The global concern about the environmental problems such as pollution and the depletion of natural resources due to using fossil fuels for generating electrical power, researchers are now focusing on obtaining new renewable and environmentally friendly resources of electrical power. The wind energy conversion system is becoming one of the most popular systems all over the world for emission-free, clean and green electrical power generation. Electrical power generation using wind energy is possible in two ways, constant speed operation and variable speed operation using

      power electronic converters. Variable speed power generation is used more because all wind velocities can be used to generate maximum electric power. Many generations used wind energy for thousands of years for primitive purposes,

      1. milling grain, pumping water and sailing [1]. It was not until 1888 when the first electricity generating windmill system was assembled in Cleveland, Ohio but it was very inefficient producing only 12 kilowatts. During the 1930s, two notable primitive large-scale wind generators were established: (The Balaclava generator on the Black Sea and the Smith-Putnam generator in Vermont). The wind energy industry has experienced a growth of almost 30 percent each year through the last decade [2] There are two main types of wind energy conversion systems (WECS)s:-

         1. Fixed-speed WECS and

         2. Variable-speed WECS

      Power maximization techniques and algorithms are used in the variable-speed WECS in order to extract as much power as possible from the wind.

   2. WIND TURBINE CHARACTERISTICS AND MODEL

      The wind turbine operates at constant speed with variable pitch angle control which permits the machine to generate electrical power directly at 60/50 Hz so it can be connected to the grid. A variable-speed wind generation system generates variable voltage and variable frequency power that gets converted to constant voltage and constant frequency before connecting it to the grid. The fixed cost of this system is high, but the energy capture is large enough to make the life-cycle cost lower. The constant-speed system practically neglected due to recent technological advances in wind turbine field, power electronics, and AC drives.

      The wind turbine model in this thesis is based on the real value system. Conventional wind model was not based on real

      value system. The simulink model of WT is modeled using real value system in order to make the wind turbine model compatible with the real value component. Certain assumptions have been made to the given MATLAB/Simulink wind turbine model. First, in order to make the turbine model compatible with the real value components of the WECS, the wind turbine model is based on real value system. Second, the fixed pitch turbine model is made to represent a controlled fixed pitch turbine. Pitch control is achieved through hydraulic manipulation. Therefore, actually the controlled variable pitch model was used to isolate the effects of electrical control rather than mechanical control. The power coefficient characteristic is a non-linear curve which reflects the aerodynamic behavior of a wind turbine so this curve must be defined. The custom turbine model basis is formed by the characteristics of the wind turbine. The dimensionless, non- linear Cp characteristic is represented as:

      Js = JT + JG (G) (7)

      Where JT is the inertia of the turbine, JG is the inertia of the machine, and G is the gear ratio between the turbine and the generator.

      (8)

      TABLE I

      Number of blades	3
      Blade radius	0.585 meter
      Gear ratio	1\1
      Pitch	0
      Air Density	1.11 kpa

      CUSTOMIZED WIND TURBINE PARAMETERS

      Where:

      (1)

      (2)

      The complete real wind turbine system was built in SIMULINK as shown in Fig.1. Corresponding to real value system of the wind turbine model, wind turbine subsystem can be simulated in MATLAB/SIMULINK as in Fig.2. Aerodynamic characteristics and the energy transfer characteristics within a wind turbine are absent in this model.

      C1 = 0.5176, C2 = 116, C3 = 0.4, C4 = 5, C5 = 21, C6 =

      0.0068

      First, in order to make the turbine model compatible with the real value components of the WECS, the wind turbine model is based on real value system. Second, the fixed pitch turbine model is made to represent a controlled fixed pitch turbine. Pitch control is achieved through hydraulic manipulation. So, actually the controlled variable pitch model was used to isolate the effects of electrical control rather than mechanical control. So, the new power coefficient equation is derived as:

      (3)

      The new power coefficient curve is shown in Fig1. The power and torque characteristics of a wind turbine are represented by equations Eq. (4) and Eq. (5). Using the power coefficient function given by Eq. (3), the mechanical power of the turbine can be represented now as:

      (4)

      The torque is defined as:

      (5)

      The outdoor hybrid system located in Burj Al-Arab specifications are summarized in Table I. The relationship between the turbine output torque and the generator rotor speed is given by Eq. (6).

      (6)

      Where Tm is the mechanical torque, Js is the total inertia of wind turbine, and is the angular velocity of the turbineshaft. Js, is given by:

      The WTG system drive control system can be implemented by building the PWM MPPT controller in MATLAB/ Simulink for variable wind turbine speed [3].

      From wind turbine subsystem MATLAB model we can draw those relations.

      The mode of operation is controlled by the sign of the mechanical torque Tm (positive represents motor mode and negative represents generator mode). After certain interval of time (0.4 sec.) the turbine torque becomes negative and operates as a generator as shown in Fig.8 and Fig.9. Between the turbine shaft and generator rotor shaft, a gear box (G = h

      / l) is installed to increase the turbine low shaft speed to high Generator rotor speed when the gear ratio in this case is 1.

      Fig.1 Customized Model of Wind Turbine

      Fig.2 Custom Wind Turbine structure

      1. Relation between Cp Vs TSR at zero pitch angle

         From XY graph shown in Fig.3 in model we can deduce the relation between The Cp Vs TSR (tip speed ratio) for constant pitch angle equal to zero. Cp max is obtained for certain opm as illustrated by the dotted line in Fig.3.

         Fig.3 Cp Vs TSR at ZERO pitch angle

      2. Relation between Cp Vs TSR at different pitch angles

         By changing the pitch angles for different values, we notice that the curve of Cp Vs TSR will take different shapes as in Fig.4. The maximum value of Cp max can be obtained at zero pitch. Cp max is inverse proportional to pitch angle for certain opm.

         Fig.4 Cp Vs TSR at different pitch angles

      3. Relation between mechanical output power from WT and generator speed

         In the subsystem wind turbine, the relation between the output mechanical power Pmech max and generator speed has been determined keeping the wind velocity 13 m/s, and the simulation time is taken as 50 sec as shown in Fig.5. Here, we notice that at certain opm we get max mechanical power.

         Fig.5 mechanical power Vs speed generator

      4. Relation between mechanical output power from WT and generator speed

      For different wind speed we can get different power curves. Max power value is proportional to wind speed as illustrated in Fig.6. Each power curve has certain value for called opm where max power value can be obtained at constant wind speed. The value of opm for each power curve increases with increasing in wind speed as shown in Fig.6. These values of opm for each different value of wind speed can be calculated by certain mathematical algorithm. It can be done by power electronic technique this which is called max power point tracking (MPPT) as illustrated in Fig.7.

      Fig.6 Relation between mechanical power and speed generator for different wind speed

      Fig.7 illustration figure for MPPT

   3. MATHEMATICAL MODELING OF PMSM

      3.1 Permanent Magnet Synchronous Machine Model

      There is noticeable symmetry between the mathematical model of a PMSM and the wound rotor synchronous machine. As a result of using modern rare-earth variety with high resistivity for permanent magnets in PMSM, induced currents in the rotor are negligible. Also, there is no difference between the back EMF produced by a permanent magnet and the one produced by an excited coil. So, theres actually symmetry between the mathematical model of a PMSM and the wound rotor synchronous machine (SM). The PMSM block operates in either motor or generator mode. The sign of the mechanical torque indicates the mode of operation (positive represents motor mode and negative represents generator mode) [4]. The machines mechanical and electrical parts are represented by a second order state-space model. The flux established by the permanent magnets in the stator is assumed to be sinusoidal in this model; this implies that the electromotive forces are sinusoidal.

      The block represents the following equations expressed in the rotor reference frame (dq frame).

      The PMSM stator dq-axis voltage equations in the rotor reference frame are given by these equations:

      (9)

      (10)

      The flux linkage in d direction, d = Ld id + m The flux linkage in q direction, q = Lq iq

      A rotor current term is not included due to assuming that no rotor winding is present. The following equations are obtained using the above equations.

      Vd = Rsid + Ld (did/dt) – wrLqiq (11)

      Vq = Rsiq + Lq(diq/dt) rLdid + rm (12)

      Fig.8 d-q-axis equivalent circuit model of the PMSM (a) d-axis (b) q-axis

      This model is standard current dynamics model of a PMSM where Rs is the stator resistance, Vd and Vq are dq-axis voltages, id and iq are dq-axis current components, Ld and Lq are the d-axis and q-axis inductances respectively, r is the rotor speed and m is the flux linkage. The (d, q) axis equivalent circuit is shown below in Fig.8. The d q variables are obtained from (a, b, c) variables using the Parks transformation as defined below. r is the rotor angular position.

      (13)

      The (a, b, c) variables are obtained from the (d, q) variables using the inverse of the Parks transformation as shown below:

      (14)

      The total input power across the air gap is given by:

      (15)

      When the stator phase quantities are transformed using the Parks transformation to the rotor d-q reference frame which rotates at a speed r = dr/dt, the equation becomes:

      (16)

      The mathematical output power Pout can be obtained by replacing Vd and Vq with the associated speed voltages where the zero sequence quantities are neglected:

      (17)

      For a machine that has P number of poles and r = (P/2) rm, where rm is the rotor speed in mechanical rad/sec.

      (18)

      By dividing output power with mechanical speed rm, the equation for electromagnetic torque Te is obtained and given as shown in Eq.(19):

      (19)

      If the number of pole pairs is P then electromagnetic torque becomes:

      Fig.9 Machine side converter Controller in SIMULINK

      (20)

      It is clear from the above equation that the produced torque consists of two parts. The first one is called reluctance torque due to the saliency, while the second is called excitation torque. For non-salient PMSM (Ld = Lq) the electromagnetic torque is:

      (21)

      The relationship between the electromagnetic torque (Te) and the load torque (Tl) is given as:

      (22)

      Where Tl is the load torque, J is the moment of inertia and B is the friction coefficient.

      The equation could be written as first order equations for simulation of the dynamic characteristics of the drive as shown below:

      (23)

      (24)

      Where B is the viscous friction of rotor and load, J is the moment of inertia, r is the rotor angular position, r is the rotor speed, Te is the electromagnetic torque and Tm is shaft mechanical torque. The differential equations controlling the system can be represented as:

      Electrical system:

      (25)

      (26)

      (27)

      3..2 Machine Side Converter Control

      Machine side converter control consists of two loop control structure, i.e. inner loops and outer loops in dq synchronous reference frame.

      The inner loop is used to control the electrical dynamics while the outer loop is used to control the mechanical dynamics in dq synchronous reference frame. the electrical dynamics are faster than the mechanical dynamics in the machine drive.

      The mechanical constant is usually many times greater than the electrical time constant.

      The inner loop current controller can force the motor current to follow their dominant values. Therefore, the balanced currents are actually equal to the reference values under the control action.

      The outer loop function is to regulate the speed of the machine at their maximum output by sending current commands to inner current loop. On the other hand, the outer speed loop ensures that the actual speed is always equal to the commanded speed. Therefore, any transient will be an overcome within the system dynamics without exceeding the motor and inverter abilities. The inner currentloop can also assure fast current response within the drive system. The drive is fed in a way that makes the q-axis current provides the desired torque [5-7].

      Fig.9 shows the implemented model of the machine side converter controller in Matlab/Simulink SimPower Systems library [8]. It is the high-efficiency drive control system for the WTG system. The commanded speed ref is pre- calculated according to the turbine output power and set to the optimum speed [8].

      Based on the speed error, the commanded q-axis reference current iqref is determined through the speed controller. The following PI controller is employed as the speed controller in this system.

      Where: all quantities in the rotor reference frame are referred to the stator frame.

      Mechanical system:

      (28)

      (30)

      Where e is the error between the reference speed and the measured speed, and Kp and KI are the proportional and integral gains of the speed controller respectively.

      Based on the current errors, the d-q axis reference voltages are determined by the PI controllers as shown below:

      Where:

      F: Viscous friction of rotor and load. iq, id : q and d axis currents.

      J: Inertia of rotor and load. Lq, Ld: q and d axis inductances. P: Number of pairs of pole.

      R: The stator windings resistance. Te: Electromagnetic torque.

      TM: Mechanical torque of the shaft. vq, vd : q and d axis voltages.

      : Rotor angular position.

      : Flux induced by the PM in the stator windings. r: The rotor angular velocity.

      (29)

      (31)

      (32)

      Where ed = idref id is the d-axis current error and eq = iqref – iq is the q-axis current error and Kpi and KIi are the proportional and integral gains of the controller respectively. [9]. The dq – axis voltages (Vd, Vq) are transformed into a, b, c quantities (Va, Vb,Vc) and given to PWM generator to generate the gate pulse for machine side converter.

      3. 3 Line Side Converter Control

      The function of the Line side converter control is to keep the DC link voltage constant regardless of the magnitude and direction of the rotor power. With the reference frame oriented along the Line voltage vector position, a vector

      control approach is used here to enable independent control of the voltage and frequency between the load and Line side converter control. The PWM converter is current regulated where The DC link voltage is regulated using the direct axis current component. Quadrature axis current component is used for regulating the reactive power [9-10].

      The implemented model of the Line side converter controller in MATLAB/Simulink SimPower Systems library is shown in Fig.10 [8]. The Line side converter controller operates as a controlled power source. The standard PI controller is used to regulate the DC voltage in the outer loops and the line currents in the d-q synchronous frame in the inner control loops. PI controller regulates the DC bus voltage by assuming an id current component which represents the active component of the injected current into the Line and iq is the reactive component of the injected current into the Line. The iq current reference is set to zero In order to obtain only a transfer of active power. The decoupling terms are used to achieve independent control of id and iq. Therefore a PLL is used to synchronize the converter with the Line as the difference between Line phase angle and the inverter phase angle can be reduced to zero using PI controller.

      1. Simulation Model of WTG System

         The complete model of the MTG system which is implemented in the Matlab/Simulink SimPower Systems library is shown in Fig.11. The input of The Wind Turbine Generation (WTG) system is wind speed and angular speed of PMSM. The input mechanical torque (Tm) for the PMSM is the torque output of the wind turbine. The direction of torque Tm is positive during motoring mode and made negative during generating mode of PMSM [4].

         Fig.10 Line side converter Controller in SIMULINK

         Fig.11 Model of WTG System connected to isolated load in SIMULINK

         The input of the machine side converter controller is the rotor speed and three phase stator current signals of the PMSM. The machine side and line side converters use the sinusoidal pulse width modulation (SPWM) with triangular carrier signal. Both machine side and line side converters are IGBT- based VSC, that is available in universal bridge block in the Simulink of the MATLAB. The function of the three phase- active LCL filter circuit is to reduce the high-order harmonics distortion and supplies the reactive power to the system and maintain the Load end-side voltage and current wave forms as sinusoidal.

         Conversion of ac generator voltages in wind systems to DC link voltages has been dominated by phase controlled or diode rectifiers. Recently, this thesis has changed this diode rectifier to IGBT converter gated by MPPT PWM converter which takes line current and generator speed as input. This change was made because the non-ideal character of the input current drawn by these rectifiers approach creates a number of problems for the power distribution network and for other electrical systems in the vicinity of the rectifier e.g.:

         1. Phase displacement of the current and voltage fundamentals which requires that the source and distribution equipment handle reactive power increasing their volt-ampere ratings.

         2. Low input power factor and high input current harmonics.

         3. Lower rectifier efficiency because RMS values of the input current are large.

         4. Input AC mains voltage distortion due to the associated higher peak currents.

         5. Large reactive components size.

      2. Relation between output electrical power and wind speed

      After getting the previous relations from the subsystem wind turbine, we will draw the relation between the output electrical power after machine side converter and wind speed in micro-turbine generation system as shown in Fig.12 from XY graph in MATLAB/Simulink.

      Fig.12 Relation between output electrical

      5.2.6 The wind speed profile input

      Wind profile for this relation is shown in Fig.13, obtained from scope 8 in MATLAB/Simulink as shown in Fig.11.

      Fig.13 wind speed profile

   4. EXPERIMENTAL WORK

      The module used is a small wind turbine. Zephyr is seeking type certification for the Z-501 wind turbine. Used in E-JUST university small turbine to be certified in Japan, this wind turbine is a part of a hybrid system consisting of a solar panel with a controller and an inverter Fig.14. The maximum output power of the hybrid system is 800W. The experimental work is done directly on the wind turbine. The specifications are shown in Table II .

      TABLE II

      MODULE SPECIFICATIONS IN DATA SHEET

      General Configuration
      Propeller plane diameter	1170 mm
      Rotation operating range diameter	1240 mm
      Weight	6 kg
      Main body length	675 mm
      Start of power generation (cut-in) wind speed	2.5 m / s
      Cut-in rotational speed	500 rpm
      Rated output speed	1700 rpm
      Upper limit voltage adjustment range	DC13.0 ~ 17.0V
      Rated output (rated wind speed 12.5 m/s at the time)	400 W
      Maximum output	450 W
      Rated output voltage	DC12V
      Battery bank voltage (Vdc)	12V
      Generator efficiency	>0.8
      Wind energy utilizing ratio (Cp)	0.4
      Generator type	Permanent Magnet Alternator
      enerator weight (kg)	5.5
      Speed regulation method	Yawing+Electrom agnetism braking
      Blade material/quantity	GRP/3
      Shutting down method	Manual+Automati c

      The experimental work was done on the wind turbine during the first half of Septamber-2013 for 14 days, and max wind speed during summer season is 9 m/s .Temperature, humidity and wind speed are measured using a portable metrological weather station. The data was instantaneously recorded and the average value was reported each 15 minutes on an LCD attached to this metrological weather station. In order to determine the characteristics of the wind turbine, as circuit diagram that is illustrated in Fig.15. Ammeter and voltameter used with ranges of 100 A and 20V respectievley.

      Fig.14 Hybrid system located in Borg Al Arab

      Fig.15 connection diagram

   5. EXPERIMENTAL RESULTS

The results was registered for 5 times at each point and the average value was reported for all current and wind speed. The results are tabulated and drawn relation curves between them.

1. Experimental relation readings between Electrical current in ampere, voltage in volt of wind turbine and Wind speed in

   m/s.

   Fig.16 Experimental readings between current in A and wind speed in m/s

2. Experimental relation readings between Electrical Output power of wind turbine in watt and wind speed in m/s

   Fig.17 Experimental readings between power in watt and wind speed in m/s

   VI . MANUFACTURERS LEAFLET DATA FROM ZEPHYR Z-501 WIND TURBINE

   Zephyr company provides me with those relation for the wind turbine module.

   Fig.18 Relation between turbine speed in rpm and wind speed in m/s from manufacturers leaflet

   1. MATLAB MODEL VALIDATION

      A comparison is made between the results obtained from the experimental measurements and the MATLAB simulation, the result showed the error obtained in the different parameters of the characteristics. The maximum error was found in 6.021m/sec wind speed with value of 6% which is considered acceptable value in order to rely on the simulation as a tool of prediction of the performance of the wind turbine in different operation cases. Fig.19 shows the comparison between the experimental readings and manufacturers leaflet in output power and wind speed, Fig.21 shows the comparison between the experimental readings and MATLAB modeling in output power and wind speed, Fig.22 shows the comparison between manufacturers leaflet and MATLAB modeling in output power and wind speed and Fig.23 shows the comparison between manufacturers leaflet, MATLAB modeling and

      experimental readings in output power and wind speed. Table V represents the results of the comparison made between the measured and simulated results.

      TABLE V

      COMPARISON BETWEEN MEASUREMENTS SIMULATION AND MANUFACTURERS LEAFLET

      	P (Watt)	The maximum error was found in 6.021m/sec wind speed value
      Measured Values	55.38 W
      Simulated Values	40.32 W
      Deviation	3.41 %

      Fig.19 Comparison between experimental readings and manufacturers leaflet in output power and wind speed

      Fig.20 Comparison between experimental readings and manufacturers leaflet in output power and wind speed zooming in

      Fig.21 Comparison between experimental readings and MATLAB modeling in output power and wind speed

      Fig.22 Comparison between manufacturers leaflet and MATLAB modeling in output power and wind speed

      Fig.23 Comparison between manufacturers leaflet, MATLAB modeling and experimental readings in output power and wind speed

   2. CONCLUSIONS

   Comparing results from Complete real system model of wind turbin power generation which was built in Simulink using permanent magnet synchronous machine and experimental reading for output dc current and output dc voltage before battery storage line Vs wind speed applied on outdoor Hybrid system located in Burj Al-Arab, the comparison showed a big coincidence between both of the curves with some small systematic errors.

   REFERENCES

   1. T. I. of Electrical and E. E. Inc, IEEE canadian review: Green power, Mathematical and Computational Applications, pp. 10-17,December 2007.

   2. Global wind report, 2008.

   3. 20. R. Datta and V.-T. Ranganathan, A method of tracking the peak power points for a variable speed wind energy conversion system, IEEE Transactions on Energy Conversion, Vol. 18, pp. 163-168, March 2003.

   4. M. Chinchilla, S. Arnaltes, and J.-C. Burgos, Control of permanentmagnet generators applied to variable-speed wind-energy systems connected to the grid, IEEE Transactions on Energy Conversion, Vol. 21, pp. 130-135, March 2006.

   5. Morimoto, M. Sanada and Y. Takeda, Wide speed operation of interior permanent magnet synchronous motors with high performance current regulator, IEEE Trans. Industrial Applications, Vol. 30, pp. 920-926, July/Aug. 1994.

   6. M. Liserre, F, Blaabjerg and S. Hansen, "Design and control of an LCL filter based three-phase active rectifier," IEEE Trans. On Industry Applications, Vol. 4, no. 5, pp.1281-1291, September/ October 2005.

   7. Chee-Mun Ong, Dynamic Simulation of Electric Machinery Using Matlab/Simulink, Prentice Hall 1998.

   8. Matlab/Simulink SimPower Systems Documentation, http://www.mathworks.com

   9. E. Koutroulis and K. Kalaitzakis, Design of a maximum power tracking system for wind-energy-conversion applications, IEEE Transactions on Industrial Electronics, Vol. 53, April 2006.

   10. A.-B. Raju, K. Chatterjee, and B.-G. Fernandes, A simple maximum power point tracker for grid connected variable speed wind energy conversion system with reduced switch count power converters, IEEE Power Electronics Specialist Conference, PESC 03, pp. 748- 753, June 2003.