Scribd
Upload
55 views

ANN Switched Z-Source Inverter Based PV Generation System: Abstract

1) The document proposes an ANN switched Z-Source inverter system for solar energy generation. A feedforward neural network determines the switching scheme of the Z-Source inverter to improve efficiency. 2) Simulation results show the proposed ANN switched Z-Source inverter reduces total harmonic distortion to 4.81% compared to 6.17% for a conventional PWM inverter. 3) The Z-Source inverter topology can boost or buck voltages using a single circuit, unlike conventional VSIs, allowing it to perform efficiently even under partial solar panel shading conditions.

Uploaded by

sathish
Copyright
© © All Rights Reserved
Available Formats
Download as DOC, PDF, TXT or read online on Scribd
55 views

ANN Switched Z-Source Inverter Based PV Generation System: Abstract

1) The document proposes an ANN switched Z-Source inverter system for solar energy generation. A feedforward neural network determines the switching scheme of the Z-Source inverter to improve efficiency. 2) Simulation results show the proposed ANN switched Z-Source inverter reduces total harmonic distortion to 4.81% compared to 6.17% for a conventional PWM inverter. 3) The Z-Source inverter topology can boost or buck voltages using a single circuit, unlike conventional VSIs, allowing it to perform efficiently even under partial solar panel shading conditions.

Uploaded by

sathish
Copyright
© © All Rights Reserved
Available Formats
Download as DOC, PDF, TXT or read online on Scribd
You are on page 1/ 9

ANN Switched Z-Source Inverter based PV Generation System

Abstract:
Solar Energy Systems (SES) are grooming nowadays due to their own merits. The system
consists of DC-AC converter network. In this paper, Z-Source inverter based SES has
been proposed. In order to improve the efficiency of the system, Artificial Neural
Network (ANN) based switching scheme is suggested to trigger the switches of the
inverter circuit. The proposed scheme has been simulated using MATLAB/Smulink and
the results have been resented. The results depict that the proposed scheme has reduced
THD of 4.81%, compared to its conventional counterpart and can be used eve under
partial shadowing conditions of the SES.

Keywords: Z-Source Inverter, Multi-Level Inverters, THD, ANN, feedforward

1. Introduction

Photo Voltaic (PV) System based electrical power generation; called Solar Energy
Systems (SES) has become popular nowadays, due to the revolution of green and clean
energy. Solar energy is abundantly available. The system is noiseless and needs a little
maintenance. The size of the system depends on the power capacity, which makes the
system to be more robust and compact. Day by day, much advancement has been taking
place in the solar energy conversion technologies. Efforts have been taken by researchers
on the aspects of delivering the quality power generated from the solar energy system to
the consumers. The main problem with the solar energy systems is the ways to increase
the output voltage. Generally, a number of solar panels are being connected in series to
increase the output voltage. The drawbacks of this method include low reliability and
poor efficiency. The output voltage of the PV cell being DC is converted into AC through
inverter and is fed to the consumer. The conversion of DC to AC is effected through
inverters and in particular Multi-Level Inverters (MLI). In these MLI based SES
modules, the output of the PV cell are modulated through DC-DC converters and then fed
to the MLIs. Z-Source inverter is another type of inverter topology that doesn’t require
DC-DC converter. The recent investigations are focusing on the development of new
algorithms for DC-AC conversion, in order to improve the conversion efficiency. The
conventional method is to step-up the AC output voltage through a step-up transformer.
3-Phase MLIs are also employed in the SES[1]. The problems associated with these
conventional methods are electromagnetic interference, size, acoustic noise and high
cost[2]. To overcome, these drawbacks, the Z-Source inverters have been suggested. [3-
5]. These inverters employ less number of switches compared to the conventional
inverters.
The model of the Z-source inverter has been developed and analyzed for its
transient characteristics [6 & 7]. The proper switching of the switches in the z-source
inverter yields better performance and efficiency [8 - 10]. In this paper, the Artificial
Neural Network (ANN) based switching scheme for the Z-source inverter has been
proposed for improving the performance of the SES. The proposed scheme has been
simulated using MATLAB/Simulink. A comparative analysis has been made between the
conventional Pulse Width Modulated (PWM) Inverter and the proposed Z-Source
Inverter scheme.

2. SES Configuration

Fig.1. Conventional VSI based SES


Fig. 2. Z-Source Inverter based SES

Figures 1 and 2 depicts the system configuration of conventional and proposed Z-Source
inverter based SES. The advantage of Z-Source topology is that it can either buck or
boost the voltage/ current levels with a single circuit. This cannot be achieved through the
PWM based Voltage Source Inverters (VSI). In this proposed system, the Z-source
inverter acts as a boost converter and supplies sinusoidal voltage to the load.
It is generally known that the output voltage of PV array varies widely under
different irradiance and environment temperature, the typical ratio of the maximum
output voltage and the minimum is 2:1, and even bigger. Thus in order to get steady ac
voltage, grid-connected PV system should have the ability to buck/boost voltage. What’s
more, to enhance the efficiency of PV array, grid-connected PV system also should have
the ability to make PV array output maximum power. A model-based control algorithm
can be used to control a switching matrix that connects a solar adaptive bank to a fixed
part of the PV array. Similarly, dynamic electrical array reconfiguration can be used to
improve the PV energy production during partial shadowing conditions. A controllable
switching matrix can also be used between the PV generator and the central inverter to
allow electrical reconnection of the available PV modules so that the maximum energy
efficiency can be achieved.

3. Mathematical Model of Z-Source Inverter


(a) Non-Shoot through state

(b) Shoot through state


Fig. 3. Equivalent circuit model of Z-Source Inverter

Fig. 3 depicts the equivalent circuit model of Z-Source inverter under non-shoot through
and shoot through state conditions. Assumptions made in the design are as follows:
L1 = L2 and C1 = C2. This results in a symmetrical Z-Source network, where the voltage
across inductors and capacitors are as given in Equations 1 & 2.
Vc1 = Vc2 = Vc (1)
Vl1 = VL2 = VL (2)
During the shoot - through zero state for an interval (T0) during a switching cycle (TZ),
VL=VC, Vd = 2VC, V0=0 (3)
Now consider that the inverter bridge is in one of the eight non-shoot-through states for
an interval of T1, during the switching cycle,
TZ VL=V0 – VC, V0= VC – VL = 2VC –Vpv (4)
Where Vpv is the output DC voltage of PV panel and TZ = T0+T1. The average voltage
of the inductors over one switching period T should be zero in steady state.
From Equations 3 & 4, the voltages across inductors and capacitors are obtained as
VL= [T0 VC + T1 (Vpv – VC)] / TZ =0 (5)
VC / Vpv = T1/ (T1- T0) (6)

The average DC link voltage is


V0 = [T1/ (T1- T0)] Vpv (7)

The peak DC-link voltage across the inverter bridge is expressed in Equation 4 can be
rewritten as:
V0 = B.Vpv (8)
Where, B = TZ/ (T1- T0) = 1/ (1-2T0 /TZ) ≥1
The DC-link voltage is the boost factor resulting from the shoot-through zero state. The
output voltage from the inverter is represented by
Vac= M Vo/2 (9)
Here, M is the modulation index.
From Equations 8 & 9, the output of the inverter is expressed as
Vac= M* B* Vpv /2 (10)
Equation 10, states that the output can be efficiently controlled by controlling the buck-
boost factor M*B. Many efficient algorithms have been proposed to control the buck
boost factor, which in turn determines the switching sequences of the switches in the Z-
Source network.

4. ANN based Switching Scheme


Feedforward Neural Network (FNN) is a the simplest type of ANN where the information
moves in only one direction, forward, form the input nodes, through the hidden nodes and
to the output nodes. Loops or cycles are not present in this type of networks. Fig. 4
depicts the schematic of feedforward network.

Fig. 4 Feedforward NN
In this proposed Z-Source inverter the ANN method has been used to determine the buck-
boost factor for switching the inverter switches. The output voltage of the system is
measured and is compared with the reference voltage. The error is processed through the
feedfroward NN. The network in turn produces the necessary M*B function which forms
the gate pulses needed to trigger the switches.

5. Simulation Results and Discussions

Fig. 5 Simulation Model of proposed Z-Source Inverter

Fig. 5 portrays the MATLAB/Simulink model of the proposed Z-Source Inverter model
with the ANN switching scheme. The output of the inverter and its corresponding FFT
analysis has been depicted in Figs. 6 & 7 respectively.
Fig. 6 Simulated output of Z-Source Inverter

Fig 7 FFT analysis of Z-Source Inverter

Fig 8 FFT Analysis of PWM VSI


The comparative analysis between the PWM VSI an proposed Z-Source inverter
can be made from the figures 7 & 8. Form the FFT analysis made it is evident that the
proposed Z-Source inverter scheme results in reduced THD of 4.81% compared to its
conventional counterpart.
6. Conclusion
The Z-Source inverter topology with ANN based switching scheme has been modeled
analyzed and simulated for the solar energy systems. The results and the corresponding
FFT analysis show that the proposed ANN switched Z-Source Inverter based scheme has
resulted in reduced THD. The output of the system has been observed to be more
sinusoidal. The THD of the proposed scheme is 4.81%, whereas the conventional PWM
scheme resulted in 6.17%. In this way, it is evident that the proposed inverter topology
has an edge over the conventional scheme. Hence, it can be effectively used even when
the SES is subjected to partially shadowing conditions.
References:
1. Eftichios Koutroulis, Kostas Kalaitzaki and Nicholas C.Voulgaris, Development
of a Microcontroller-Based, Photovoltaic Maximum Power Point Tracking
Control System, IEEE Trans. On Power Electrons Vol.16, No.1, Jan. 2001, pp 46
– 54.
2. H.Rostami and D.A.Khaburi, Neural Networks Controlling for Both the DC Boost
and AC Output Voltage of Z-Source Inverter, IEEE Conference, 2010, pp 135 –
140.
3. Rathika, S.; Kavitha, J.; Paranjothi, S.R., “Embedded control Z-source inverter
fednduction motor” : IEEE Conference, INCACEC 2009. Pp 1-7.
4. F. Z. Peng, "Z-source inverter," IEEE Tran. Ind. Appl., vol. 39 no2, pp. 504-510,
Mar./Apr. 2003
5. Tang, Y., Xie, S.J., Zhang, C.H., Xu, Z.G.: ‘Improved Zsource inverter with
reduced Z-source capacitor voltage stress and soft-start capability’, IEEE Trans.
Power Electron., 2009, 24, (2), pp. 409–415
6. F.Z.Peng,M.Shen,A.Joseph,L.M.Tolbert,D.J.Adams, ”Maximum Constant Boost
Control of the Z-Source Inverter”, In. proc.IEEE IAS’04,2004.
7. Zare,Farus and Adabi,Jafar,”Hysteresis Band Current Control for a Single
PhaseZ-Source Inverter with Symmetrical and Asymmetrical Z-network”, In
proceedings Power Conversion Conference-PCC’07,pp.143-148,Nagoya,Japan
2007.
8. M.Shen,F.Z.Peng,”Operating Modes and Characteristics of the Z-Source inverter
with the Small Inductance”, in proc.IEEE IAS,2005
9. U.Shajith Ali,V.Kamaraj,’Sine Carrier for fundamental Fortification in Three
Phase Z-Source PWM Inverters’, Modern Applied
Science,Vol.4,No.1,pp73,January 2010.
10. Shen M S, Wang J, Joseph A, Peng F Z,Tolbert L M, and Adams D J, "Constant
boost control of the Z-source inverter to minimize current ripple and voltage
stress," May-Jun 2006, IEEE Transactions on Industry Applications, vol42, pp.
770-778.
11. N.Vidhyarubini , G.Rohini ,”Z-source inverter for photovoltaic power generation
system “proceedings of ICETECT 2011
12. B.Farhangi,Sh. Farhangi,”Application of Z-Source Converter in Photovoltaic
Grid-Connected Transformerless Inverter”, PELINCEC 2005 Conf.15-20
Oct.2005,Warsaw,Poland.

You might also like