Abstract-- In this paper proposed method of maximum power point tracking using adaptive fuzzy logic control for grid connected photovolatic system. The system composed of boost converter single-phase inverter connected to utility grid. The maximum power point tracking control is based on adaptive fuzzy logic to control MOSFET switch of boost converter and single phase inverter uses predicted current control to control four IGBTs switch for grid-connected control. Adaptive fuzzy logic controllers provide attractive features such as fast response, good performance and it can also change fuzzy parameter for improving control system. The fuzzy logic predicted current control provide current with sinusoidal waveshape and inphase with voltage. This system can provide energy with low harmonics and high power factor.
Index Terms -- Adaptive Fuzzy Logic, Maximum Power Point Tracking, Photovoltaic System.
I. I NTRODUCTION
he photovoltaic (PV) energy is increasing interest in electrical power applications. It is crucial to operate PV
energy conversion systems near maximum power point to increase the output efficiency of PV. However, the nonlinear nature of PV systems is apparent from Fig. 1. i.e. the current and power of PV
array depend on the array terminal operating voltage. In addition, the maximum power operating point is changing with insolation level and temperature. Therefore, the tracking control of maximum power point is the complicated problem. To overcome these problems, many tracking control strategies have been proposed such as perturb and observe [1]-[2], incremental conductance [3], parasitic capacitance [4], constant voltage [5], neural network [6]-[10] and fuzzy logic control [11]-[14]. These strategies have some disadvantage such as costly, difficulty, complexity and non-stability.
The general requirements for maximum power point tracker are simple and low cost, quick tracking under condition
N.patcharaprakiti is with Electrical Engineering Department, Rajamagala Institute of Technology, Chiang Rai, 57120 Thailand (e-mail : pnopporn@hotmail)
S.premrudeepreechacharn is with Department of Electrical Engineering, Faculty of Engineering, Chiang Mai University, Chiang Mai, 50200 Thailand (e-mail : u.ac.th)
Fig.1 PV array characteristics.
change, and small output power fluctuation. A more efficient method to solve this problem becomes crucially important. Hence, this paper proposed the method to track power maximum power point by
using adaptive fuzzy logic control. Fuzzy logic control is appropriate for nonlinear control and it has no complex mathematical. However, the fuzzy logic controller behavior depends on the membership functions, their distribution, and the rules that influence the different fuzzy variables in the system. There is no formal method to determine accurately the parameters of the controller. However, choosing fuzzy parameter to yield optimum operating point and good control system is up to experience from control engineer. For this reason, adaptive fuzzy logic control can solved this problem because it can re-adjust fuzzy parameter to obtain optimum performance.
The grid connected PV system presents in this paper can directly feed energy into the existing AC grid system, where the cost of batteries for energy storage can be reduced. First, we describes the grid-connected PV system. Then, the adaptive fuzzy logic controller is described. The controller for
Maximum Power Point Tracking Using Adaptive Fuzzy Logic Control for
Grid-Connected Photovoltaic System
N. Patcharaprakiti and S. Premrudeepreechacharn, Member, IEEE
T
0-7803-7322-7/02/$17.00 © 2002 IEEE
grid connected PV system, which is predictive current control, also is presented. Finally, the simulation results of MPPT by adaptive fuzzy logic controller compared with conventional fuzzy logic control is discussed.
II. G RID -CONNECTED P HOTOVOLTAIC S YSTEM
In order to show the feasibility of maximum power point tracking using adaptive fuzzy logic control, the photovoltaic power system with boost converter and single phase inverter is constructed as shown in Fig. 2.
Control system Fig.2 Grid-connected photovoltaic system.
A. Boost converter Boost converter use to increase voltage for inverter circuit and also it use to control maximum power point tracking by using adaptive fuzzy logic control and pulse width modulation method to generate pulse for drive MOSFET (SB). Output voltage of boost converter can calculated from (1) )(1 Duty
11
in V o V −=
where V in = input voltage (output voltage of PV array) V o = output voltage
Duty = duty ratio of power switch. B. Single phase Inverter
Inverter circuit converts direct current to alternate current by predicted current control to control current to be sine wave for utility grid-connected. Inverter circuit composed of DC source from boost chopper circuit, four switch IGBTs
(S1-S4) inductance and transformer. The controller for single phase inverter will be described later.
III. A DAPTIVE F UZZY L OGIC C ONTROLLER
Traditional fuzzy logic control requires the expert knowledge of the process operation for fuzzy logic control parameter setting, and the controller can be only as good as the expertise involved in the design. Fuzzy logic control with fixed parameters are inadequate in application where the operating conditions change in a wide range and available
expert knowledge is not reliable. To make the controller less dependent on expert knowledge, the adaptive fuzzy logic control is a solution. Adaptive fuzzy logic shown in Fig. 3 is composed of 2 parts: fuzzy knowledge base controller and learning mechanism described, as described below.
Fig.3 Structure of adaptive fuzzy logic controller.
A. Fuzzy Knowledge Base Controller
The fuzzy knowledge base controller is basic part of fuzzy
logic control which is composed of 3 parts: fuzzification, inference engine and defuzzification as described below.
1) Fuzzification Membership function’s value are assigned to the linguistic
variables, using seven fuzzy subsets : NB (Negative Big),
NM(Negative Medium), NS (Negative small), ZE (Zero), PS
(Positive small), PM (Positive Medium), and PB (Positive
Big). The partition of fuzzy subsets and the shape of membership function which can adapt shape up to appropriate system are shown in Fig. 4. The value of error (e) and change of error (de) are normalized by input scaling factor βe and βde. In this system input scaling has designed between –1 to 1.
Fig.4 Fuzzy logic membership function.
The triangular shape of membership function of this arrangement presumes that for any particular input there is only one dominant fuzzy subset. Input error (e) for fuzzy logic controller can calculated from maximum power point as follows: (2) I
P V P V I V I E(k)∆=∆=+∆=
where I = Output current from PV array
∆I = I(k)-I(k-1)
V = Output voltage from PV array
∆V = V(k)-V(k-1).
2) Inference Method The composition operation by which a control output can be generated. Several composition methods such as MAX-MIN and MAX-DOT have been proposed in the literature. The commonly used method is MAX-MIN as used in this paper. The output membership function of each rule is given by the MIN (minimum) operator, MAX(maximum) operator. Table 1 shows the rule table f
or fuzzy logic controller. T ABLE 1 R ULE BASE OF F UZZY L OGIC C ONTROLLER Change of Error (de) Erro r (e) NB NM NS ZE PS PM PB NB
NB NB NB NB NM NS ZE NM
NB NB NB NM NS ZE PS NS
NB NB NM NS ZE PS PM ZE NB NM NS ZE PS PM PB PS
NM NS ZE PS PM PB PB PM
NS ZE PS PM PB PB PB PB ZE PS PM PB PB PB PB
3) Defuzzification
As the plant usually required a nonfuzzy value of control, a defuzzification stage is needed. Defuzzificaion for this system is the height method. The height method is both very simple
and very fast method. The height defuzzification method in a system of m rules by formally given by )3( n
1
k k W k W *m 1
k C(k)du ∑∑===
where du = change of control output C(k) = peak value of each output
W k = height of rule k.
Output of fuzzy logic control uses to control through PWM which generated pulse to control MOSFET switch (SB).
B. Learning Mechanism The purpose of learning mechanism is to learn the environmental parameters and to modify the fuzzy logic controller accordingly so that the response of the overall system is close to optimum operation point. The learning mechanism is composed of inverse fuzzy model and knowledge base modifier, 1) Inverse fuzzy model
In this part, we use error (e) or change of error (de) of system and control the knowledge base modifier to modify fuzzy parameter to optimize the operation of system. The fuzzy parameter can be adapted by use this condition
If error < ε (limit value) then knowledge base modifier will be done. 2) Knowledge base modifier In this part fuzzy parameter will be modifier as follow [15]: a) Scaling factor Quite simple schemes for altering the scaling factor to meet various performance criteria can be devised. The range of the error, change of error and also output of fuzzy can set like balance between proportional and integral control.
b) Fuzzy set membership function In this part, tuning peak values, such as error in Fig.4, can improve both responsiveness and stability. The large error can improve responsiveness and small error can improve stability. The modification is performed by shifting the membership functions of both input and output.
c) Tuning rule base
Modifying rule base can effect the control system such as overshoot, setting time, stability, responsiveness. Rule base and fuzzy set membership function has relationship each
other up to quantity of error and change of error. To control system optimization rule base are also effect to system. IV. P REDICTED CURRENT CONTROL From predicted current control as in [16], line current can
defined as (4)
(4) /L
s )]T n (t inv V -)n (t s [V )n I(t -)s T n I(t I =+=∆ where I = inverter line current
V s = utility voltage
V inv = output voltage of inverter. Then V inv can calculated from equation: )(5 )]n I(t )s T n )[I(t s (L/T -)n (t s V )
n inv(t V −+= Current of single phase inverter can be controlled by switch
S1-S4. The switch S1 and S2 use to shape the waveform to follow the reference current. While the switches S3 and S4 use to correct the polarity of the waveform. Hence, the V inv can be described as follows:
)(6 dc V k d inv V •= Where d k is the duty ratio for switches S1 and S2 over one switch period and V dc is the DC bus voltage from boost
converter. The change in line current over one period can defined
as:
)(7 )
s t -n I(t -)n I(t )n I(t -)s T n I(t I =+=∆ From equation (5)-(7) can define the duty ratio for single phase inverter as a function of source voltage (V s ) and the change in line current (∆I) as follow: (8) )s I/T L -s [(V dc
V
1I),s f(V k d ∆=∆=
We will use (8) to control the duty ratio of switch S1 and S2 for single phase inverter.
V. S IMULATION R ESULTS
This section discusses the simulation of grid-connected photovolatic system. The maximum power point tracking was controlled by using adaptive fuzzy logic control and inverter current control using predicted current control to control current of inverter to be sinusoidal waveshape.
The simulation results of system following parameters
Solar array 60 W, V oc=21.2 volt and I sc=3.54 A, R s=0.39, R sh=176, V t=1.0337, I o= 4.3871e-9, T=298K, Boost converter L con = 1 mH, C con= 4700 µF
Single phase inverter L inv = 1 mH
AC source = 220 V, 50 Hz.
This system can be simulated by use MATLAB program. In this simulation insolation level (G) will change from 1000 W/m2 to 1200 W/m2 at 0.08sec and then change from 1200 W/m2 to 800 W/m2 at 0.15sec.
Fig. 5 shows the PV array characteristic by using MPPT control with conventional fuzzy logic control. The operating point of solar cell doesn’t really work on maximum power point. The error and change of error of fuzzy logic controller is shown in Fig. 6. Thus, the track operating point is improved MPPT controller by using adaptive fuzzy logic
control.
Fig.5 PV array characteristic using conventional fuzzy logic control.
Fig.6 Error (E) and change of error (dE) of fuzzy logic control.
The adaptation of rule base as described in above, the new rule base for the controller is shown in Table 2 and fuzzy membership function after adaptation is shown in Fig. 7. As seen from Fig. 7, the membership function of error has changed, while the change of error doesn’t need to modify the membership function.
T ABLE
2 R ULE BASE OF FUZZY LOGIC AFTER CHANGE THE RULE BASE.
Change of Error (de)
Erro
r
(e)
NB NM NS ZE PS PM PB NB NB NB NM ZE ZE ZE ZE NM NB NM NM ZE NM PS PS NS NB NB NB NB PM PS PM ZE NB NB NS ZE PS PM PB PS NM NS ZE PS PM PB PB PM NS PB PB PB PB PB PB PB PB PB PB PB PB PB PB
Fig.7 Fuzzy parameter using adaptive fuzzy logic.
The adaptive fuzzy logic controller can improve MPPT controller as seen from Fig. 8-9. Fig. 8 shows PV array characteristic after fuzzy membership adaptation. When we compared with PV array characteristic in Fig. 5, the operating point of this system operates closer maximum power point track
ing than conventional fuzzy logic control system before parameter adaptation. Fig. 9 shows the array voltage, array current and array power plot with time. Fig. 10 shows output of converter and output of inverter current. This system can provide energy to utility with low harmonics and
high power factor.
Fig.8 PV array characteristic using adaptive fuzzy logic.
Fig.9 PV array voltage, current and power vs time
Fig.10 Voltage output of converter and inverter current
VI. C ONCLUSION
This paper has presented the adaptive fuzzy logic control for control maximum power point tracking of grid-connected photovoltaic system. The proposed algorithm in adapting fuzzy logic control has simulated. The simulation results show that the advantage of this system are adaptation of fuzzy parameter for fast response, good transient performance, insensitive to variations in external disturbances. In addition, the result of simulation shows that MPPT controllers by using adaptive fuzzy logic has provided more power than simple fuzzy logic control. This system also can provide energy to utility with low harmonics and high power factor.
VII. R EFERENCE
[1] C. Hua and C. Shen, “Comparative Study of Peak Power Tracking
Techniques for Solar Storage System”, 1988 IEEE Applied Power Electronics Conference and Exposition Proceedings , Vol.2 pp.679-683. [2] K.H.Hussein, I. Muta, T. Hoshino and M. Osakada , “Maximum
Photovoltaic Power Tracking: An Algorithm for Rapidly Changing Atmospheric Conditions,” IEE Proceedings on Generation, Transmission and Distribution , Vol.142, No.1, pp.59-64, January 1995.
[3] A. Brambilla, “New Approach to Photovoltaic Arrays Maximum Power
Point Tracking”, Proceeding of 30th IEEE Power Electronics Specialists Conference, Vol. 2, 1998, pp. 632-637.
[4] D.P. Hohm and M.E. Ropp, “Comparative Study of Maximum Power
Point tracking Algorithm Using an Experimental, Programmable, Maximum Power Point Tracking Test Bed ”, Proceeding of 28th IEEE Photovoltaic Specialists Conference , 2000, pp.1699-1702.
[5] W. Swiegers and J. Enslin, “An Integrated Maximum Power Point Tracker
for Photovoltaic Panels”, Proceedings of IEEE International Symposium on Industrial Electronic1998, Vol. 1, 1998, pp. 40-44.
[6] T. Hiyama and K. Kitabayashi, “Neural Network Based Estimation of
adaptiveMaximum Power Generation from PV Module Using Environment Information”, IEEE Transactions on Energy Conversion , Vol.12, No. 3, pp.241-247, September 1997.
[7] T. Hiyama, S. Kouzuma, T. Imakubo, and T.H. Ortmeyer, “Evaluation Of
Neural Network Based Real Time Maximum Power Tracking Controller For PV System”, IEEE Transaction On Energy Conversion , Vol. 10, No. 3, pp. 543 –548, Sept. 1995.
版权声明：本站内容均来自互联网，仅供演示用，请勿用于商业和其他非法用途。如果侵犯了您的权益请与我们联系QQ:729038198，我们将在24小时内删除。
发表评论