• No results found

An Innovative Methods On Voltage Profile Improvement Using Anfis Based System To Improve The Power Quality Of Photovoltaic (Pv) Energy Generation System Using Hysteresis Loss Current Control Of D-Statcom

N/A
N/A
Protected

Academic year: 2020

Share "An Innovative Methods On Voltage Profile Improvement Using Anfis Based System To Improve The Power Quality Of Photovoltaic (Pv) Energy Generation System Using Hysteresis Loss Current Control Of D-Statcom"

Copied!
15
0
0

Loading.... (view fulltext now)

Full text

(1)

170

An innovative methods on Voltage Profile

Improvement using ANFIS based system to

improve the power quality of Photovoltaic

(PV) energy generation system using

Hysteresis Loss Current Control of

D-STATCOM

P

1

P

Adarsh kumar, P 2 P M.F. Qureshi P 1 P

Department of Computer Science, Scholar, Dr. C.V. Raman University, Bilaspur, India

P

2

P

Department of Electrical Engg., Additional Director, DTE Raipur, India

Abstract

In this piece of research, an ANFIS is designed for output voltage and frequency control

of a wind power generation system. Photovoltaic (PV) energy generation system (PVEGS) provide

the opportunity to capture more power than other turbines. Distributed power generation systems

based on renewable energy sources are attracting the market. In this regard, grid-connected

photovoltaic (PV) plants are becoming a common technology to generate energy and its penetration

level is gradually increasing. Microgrid, a recently emerging power technology, is expected to

possess an increasing role in future power systems due to its immense advantages. The micro grid

concept has been proposed as a solution to the problem of integrating several power generations

without disturbing the utility, especially when renewable energy sources are utilized.

The difficulties of a complete stability analysis mainly come from the analytical complexity

of the PV system dynamical model. The PV array current-to-voltage characteristics with the

irradiance and temperature and, PV inverter grid-connection. The fluctuations in Voltage can be

regulated if reactive power is maintained properly. A Neuro Fuzzy controlling technique is

proposed for the fast controlling of Distributed Power flow Controller (DPFC). Also the results are

compared with existing PI controlling technique. An IEEE 4 bus Transmission system with DPFC

at load bus is simulated in MATLAB/SIMULINK.

A D-STATCOM a controlled reactive source, which includes a Voltage Source Converter

(VSC) and a DC link capacitor connected in shunt, capable of generating and/or absorbing reactive

power is proposed for effective control of reactive power integrated with ANFIS. The operating

principles of D-STATCOM are based on the exact equivalence of the conventional rotating

(2)

171 Keywords: Power Quality, ANFIS Controller, Voltage Sag, Voltage Swell. Photovoltaic (PV)

energy generation system, Matlab/Simulink,D-STATCOM

1. Introduction

Photo Voltaic Systems

A photovoltaic (PV) system directly converts sunlight into electricity. The basic device of a

PV system is the PV cell. Cells may be grouped to form panels or arrays. The voltage and current

available at the terminals of a PV device may directly feed small loads such as lighting systems and

dc motors. A photovoltaic cell is basically a semiconductor diode whose pn junction is exposed to

light. Photovoltaic cells are made of several types of semiconductors using different manufacturing

processes. The incidence of light on the cell generates charge carriers that originate an electric

current if the cell is short circuited.

Fig.1.Equivalent Circuit of a PV Device including the series and parallel Resistances

The equivalent circuit of PV cell is shown in Fig.1. In the above diagram the PV cell is

represented by a current source in parallel with diode. Rs and Rp represent series and parallel

resistance respectively. The output current and voltage from PV cell are represented by I and V.

A grid-connected PV system based on a single central inverter is one of the most prevailing

configurations since, under uniform irradiance conditions of the PV panels, it represents a good

trade-off between the extracted energy and the design complexity of the power inverter.

In the specific architecture considered PV panel are series-parallel connected to the input of

a single full-bridge power inverter by means of a DC-link capacitor, as shown in Figure 2. Another

interesting fact about this GCPV inverter circuit is that it can be found as output power stages in

other configurations such as the widely used two stage boost-buck configuration shown in Figure 2

and the multilevel configuration. Due to this common power stage it is possible to extrapolate the

(3)

172

Fig.2 Single-Phase Single-Stage Grid-Connected Photovoltaic Inverter Schematic Diagram

In the case of the GCPV inverter, input-output linearization is applied. Input-output

linearization aims to find a direct and simple relation between the system output and the control

input. Recall the system dynamic equations (1) and (2) rewritten here for convenience,

𝜑 = 𝐼𝑠𝑛𝑝 (1)

𝛼 = 𝑛𝑠

𝜂𝑣𝑇 (2)

In this case, if the switching frequency is sufficiently high, the dynamical behavior of the

GCPV system can be approximated by the following set of differential equations:

𝐶𝑍̇1 = −𝜇𝑧2+ 𝑓𝑝𝑣(𝑧1)

𝐿𝑍̇2 = −𝜇𝑧2− 𝑣𝑔

Where 𝜑 and 𝛼 represent non-negative parameters of the photovoltaic generator.

Where 𝑛𝑠and 𝑛𝑝 are the number of PV panels connected in series and parallel, respectively.

Where 𝜂 is the emission coefficient which depends on semiconductor material of the cell, 𝐼𝑠 is the

reverse saturation current of the pn-junction and 𝑣𝑇 is the thermal voltage.

The photovoltaic generator can be modeled by means of a function 𝑓𝑝𝑣: 𝑈 → 𝐼where the

instantaneous voltage and current of the photovoltaic generator are denoted by𝑣𝑝𝑣 ∈ 𝑈and 𝐼𝑝𝑣 ∈ 𝐼

respectively.

(4)

173

it is possible to obtain an explicit relationship between the output y and the input μ differentiating

the output once

𝑦̇ = 𝑧2̇ =1𝐿 (𝜇𝑧1− 𝑣𝑔)

Defining a “new” control input w to be of the form 𝜔 = 𝜇𝑧1 yields the following linear

relationship between the system’s output, ZR2R, and the defined control input 𝜔

𝑧2̇ =1𝐿 (𝜔 − 𝑣𝑔)

Notice that

• By means of the input-output linearization, the dynamics are decomposed into an external part

(the dynamics of the output current and an internal “unobservable” part, i.e., the dynamics of the

capacitor voltage.

• The external part consists of a linear output relation between ZR2R and w which makes it possible to

use the well known linear control techniques.

• The internal part needs to be analyzed in order to assure the stability of the closed-loop system.

The controllers designed under ideal conditions represent a good start for the development

of more robust control schemes. Hence, a controller that assures the stability (local or global) of the

GCPV inverter.

The performance of electrical networks can be improved with the management of reactive

power. Also reactive power compensation is the best method of maintaining good power quality.

The concept of Flexible AC Transmission System (FACTS) technology is concerned with the

management of reactive power. Proper control of reactive power also results in maintaining the

voltage profile i.e., voltage quality of the transmission system. To reduce the cost of reactive power

control through FACTS devices, D-FACTS technology is proposed in the previous work. To

regulate the controlling action of D-FACTS device, Distributed Power flow Controller (DPFC),

firstly a Conventional PI controller is proposed. To improve the rate of convergence of controlling

action of the DPFC a soft computing technique is proposed in this paper. The structure of DPFC is

given in the below figure 3.

(5)

174 Principle of D-STATCOM

A D-STATCOM is a controlled reactive source, which includes a Voltage Source Converter

(VSC) and a DC link capacitor connected in shunt, capable of generating and/or absorbing reactive

power.

The AC terminals of the VSC are connected to the Point of Common Coupling (PCC)

through an inductance, which could be a filter inductance or the leakage inductance of the coupling

transformer, as shown in Fig.4. The DC side of the converter is connected to a DC capacitor, which

carries the input ripple current of the converter and is the main reactive energy storage element.

This capacitor could be charged by a battery source, or could be recharged by the converter itself. If

the output voltage of the VSC is equal to the AC terminal voltage, no reactive power is delivered to

the system. If the output voltage is greater than the AC terminal voltage, the D-STATCOM is in the

capacitive mode of operation and vice versa. The quantity of reactive power flow is proportional to

the difference in the two voltages.

An IEEE standard 4 bus system as shown in figure 5 is used for the simulation. DPFC

is proposed to connect at load bus as the variations in the load are the main cause for the variations

in magnitude of voltages in the entire transmission system. Analysis is done with conventional PI

controller and with ANFIS controller.

Fig.5 IEEE 4 bus system

5. Adaptive neuro-fuzzy inference system

ANFIS is the combination of neural networks and fuzzy system to determine parameters of

the fuzzy system. The main purpose of using the Neuro-Fuzzy approach is to automatically realize

the fuzzy system by using the neural network methods. In ANFIS control system, Fuzzy Sugeno

models are involved in framework of adaptive system to facilitate the learning and adaptation

(6)

175

descent training algorithms from the neural networks area are used to obtain fuzzy system

parameters. For this reason, the approach is generally expressed as Neuro-Fuzzy modeling. To

express the ANFIS structure, two fuzzy if-then rules under Takagi-Sugeno (TS) model are given as

follows:

Rule 1: If (x is AR1R) and (y is BR1R) then fR1R = pR1Rx+qR1Ry+rR1

Rule 2: If (x is AR2R) and (y is BR2R) then fR2R = pR2Rx+qR2Ry+rR2

Here, ri, pi and qi are the design parameters determined during the period of training phase.

Fig.6. ANFIS controller system

The general block diagram of ANFIS controller system performed for two-rule fuzzy system is

given in Figure 6. The ANFIS controller system realizes TS rules in 5 layers by using

multi-iteration learning procedure and hybrid learning algorithm.

The duty of learning or training algorithm for ANFIS is to change all the adjustable

parameters to compare ANFIS output with trained data. aRiR, bRiR and cRiR membership function

parameters define the center of sigma, slope and bell type membership function respectively.

𝑧 =𝑤 𝑤1

1+ 𝑤2𝑧1+

𝑤2

𝑤1+ 𝑤2𝑧2

= 𝑤1(p1x + q1y + r1) + 𝑤2(p2x + q2y + r2)

= (𝑤1𝑥)p1 + (𝑤1𝑦)q1+ (𝑤1)r1+ (𝑤2𝑥)p2+ (𝑤2𝑦)q2+ (𝑤2)r2

ANFIS Controller Design

The design of the controllers presented in this section are primarily based on similar

control schemes used for the full-bridge rectifier. In both cases the inductor current has to be in

phase with the utility grid voltage and the capacitor voltage needs to be regulated to a desired DC

value. In accordance with the previous, the first control scheme presented is based on the intuitive

time-scale separation between the fast dynamics of the inductor current and the slow dynamics of

(7)

176

enables to analyze the inductor current dynamics and the capacitor voltage dynamics separately

under certain assumptions.

The output of Sugeno Fuzzy system z is linear in the result parameters p, q and r. ANFIS output

is a linear combination of adjustable parameters. However, if parameters of membership function

are not fixed and permit changing, the area to be trained becomes wider and convergence of training

algorithm slows down. In such cases, the hybrid learning algorithm with combination of gradient

descent and least- squares gives more effective results.

Fig.7For Photovoltaic (PV) energy generation system (PVEGS), structure of Sugeno type Fuzzy

Inference System consists of three input and one output signals.

The triangle type membership functions of inputs of this fuzzy inference system are given in

below together with their value gaps. As output of Sugeno type ANFIS consists of real numbers,

marking such as input signal membership function is not made. As the output variable consists of

64 real numbers, the value gap is 42.79 and 3.5・109.

(8)

177

Membership functions, linguistic variables and value gaps of input variables of ANFIS are given in

Figure 09 and Figure 10. The rule base of obtained fuzzy inference system consists of 64 rules. The

linguistic variables of error signal [e(t)] , one of the input variables of ANFIS, are: very large

negative (VLN), large negative (LN), medium negative (MN) and negative (N). The linguistic

variables of change in signal [de(t)/dt], the second input variable, are: large negative (LN), medium

negative (MN), medium positive (MP) and large positive (LP). The linguistic variables of the third

input variable, measured power change ( P measured ) are: positive (P), medium positive (MP),

large positive (LP) and very large positive (VLP).

Fig.9Membership functions and value gaps of error [e(t)] and change in error [de(t)/dt] of the input

variables of ANFIS

Fig. 10The membership function and value gap of the third input variable of ANFIS, the measured

power change [(P measured (p.u)].

64 rules in the rule base of ANFIS are obtained by using a logical operator between the input and

output signals of system with (if, then) words. The input variables of ANFIS are combined with

and” conjunction. 64 rules obtained through anfisedit software of Matlab/Simulink program,

Fuzzy Logic Toolbox are given in Table 1.

(9)

178

Results

The output voltages of load bus in the IEEE 4 bus system is observed without DPFC, with

PI controlled DPFC and ANFIS based DPFC. To show that voltage quality in turn power quality is

improved, different situations like voltage sag, voltage swell are created. Then Controllers are

introduced. The output waveforms for both sag and swell mitigation are given below.

(10)

179

Fig. 12 Mitigation of voltage sag

It is observed that the results are better with PI controller and the convergence made fast

with the introduction of ANFIS technology. The waveforms of active power supplement for both

sag and swell are also given here.

(11)

180

Fig.14 Power Flow control during Sag

The improvement of power flow under normal working conditions is also observed, as given

in Figure 15, and it is noted that with the ANFIS technology it is improved and maintained at the

required value.

Fig. 15: Power response with DPFC with PI and with ANFIS controls

D. MATLAB Model for ANFIS controller

The simulation diagram of the proposed D-STATCOM with Fuzzy Logic Controller is

(12)

181

Fig.16. Block Diagram for D-STATCOM with ANFIS Controller

The ANFIS controller is a controller used to calculate the error in Fig 16. The input that is

given to this block is voltage reference, amplitude, time, sine wave function and reference sine

wave angle. The output measured from this block is error. The ANFIS requires that each control

variables which define the control surface be expressed in fuzzy set notations using linguistic labels.

Simulation Results

A. Photovoltaic (PV) energy generation system (PVEGS), Voltage Output

The output voltage waveform of wind energy system is shown in Fig.17.

Fig.17 Voltage Waveform for Photovoltaic (PV) energy generation system (PVEGS)

The proposed D-STATCOM responds to the harmonic reduction, the voltage is as shown in

Fig. 17. The output voltage waveform of wind energy system is carried out by using D-STATCOM

with Fuzzy logic controller with Hysteresis loss current control algorithm. The voltage range of

wind energy system is 420 volts.

D. Photovoltaic (PV) energy generation system (PVEGS), Voltage and Current Output

(13)

182

Fig.18 Voltage and Current Waveform for PVEGS

The proposed D-STATCOM responds to the harmonic reduction, the voltage and current is

as shown in Fig. 18. The simulation is carried out by using D-STATCOM of fuzzy logic controller

with Hysteresis loss current control algorithm. After the harmonic reduction the voltage and current

flowing in the three phase supply is 440 volts and 12 amps.

V. Conclusion

The dynamic problem of maintenance of voltage at the nominal per unit value became a

great task in the operation and performance of the power transmission system. A new D-Facts

device, DPFC is proposed in the previous work, but the convergence is low. The proposed Soft

Computing technique, adaptive neuro fuzzy inference system, improved the convergence rate of the

DPFC in improving the voltage profile of the transmission system. It also maintained good power

flow control during all periods including voltage sag, swell and normal working conditions very

effectively.

An ANFIS is designed for output voltage and frequency control of a wind power

generation system. Photovoltaic (PV) energy generation systems (PVEGS) provide the opportunity

to capture more power than other turbines. Distributed power generation systems based on

renewable energy sources are attracting the market. In this regard, grid-connected photovoltaic (PV)

plants are becoming a common technology to generate energy and its penetration level is gradually

increasing.

In this work, a fast and cost effective D-STATCOM is proposed for reducing the problem

of harmonics in industrial distribution systems. Hysteresis loss current control algorithm utilizes the

(14)

183

to trigger the switches of an inverter using a Pulse Width Modulation (PWM) scheme. The

D-STATCOM handled the situation without any difficulties and injected the appropriate voltage

component to correct rapidly any changes in the supply voltage there by keeping the load voltage

balanced and constant at the nominal value. In this study, the D-STATCOM has shown the ability

to reduce the harmonics problem. The effectiveness of ANFIS with hysteresis loss current control

of D-STATCOM is established for RLC load. Through analysis and simulation it is shown that the

proposed scheme is superior compared to the other conventional controller technique in terms of

energy saving and dynamic performance. The Fuzzy logic control with hysteresis loss current

control of D-STATCOM has the ability for good compensation characteristics. By using this

compensation strategy the THD (Total Harmonics Distortion) is reduced up to 3.21%.

References

1. Zhihui Yuan, Sjoerd W. H. de Haan, Jan Braham Ferreira and Dalibor Cvoric,, “A FACTS

Device: Distributed Power–Flow Controller(DPFC),” IEEE transactions on power

electronics, Vol. 25, No. 10, October 2010.

2. Mr. Sandeep R. Gaigowal and Dr. M. M. Renge, “ Some studies of Distributed Series FACTS Controller to control active power flow through Transmission Line,‟‟ 2013, International Conference on Power, Energy and Control (ICPEC), pp 124-128.

3. Anurag S.D. Rai, Dr. C.S. Rajeshwari and Dr. Anjali Potnis,‟‟Modeling of Distributed Power Flow Controller (DPFC)in MATLAB/SIMULINK,‟‟ International Journal of Interdisciplinary Research, Vol. 01, Issue 06, September-2014.

4. Augus Jamal and Ramadoni Syalputra, “ UPFC based on Adaptive Neuro – Fuzzy for power

flow control of multimachine systems”, International Journal of Engineering Science

Invention, Volume 2, Issue 10, October 2013.

5. C. Kocatepe, A. ˙Inan, O. Arıkan, R. Yumurtacı, B. Kekezo˘glu, M. Baysal, A.Bozkurt, Y.

Akkaya, “Power quality assessment of grid-connected wind farms considering regulations in

Turkey”, Renewable and Sustainable Energy, Vol.13, pp. 2553–2561, 2009.

6. A. Gaillard, P. Poure, S. Saadate, M. Machmoum, “Variable speed DFIG wind energy

system for power generation and harmonic current mitigation”, Renewable Energy, Vol. 34,

pp.1545–1553, 2009.

7. IEC 61000-4-30. Electromagnetic compatibility (EMC)—Part 4: testing and measurement

techniques. Power quality measurement methods.

8. Ajay Kumar G.V.N, Jarupula Somlal,“Load Compensation Of A Diesel Generator Set For

(15)

184

9. Cuk Supriyadi A.N, Takuhei Hashiguchi, Tadahiro Goda and Tumiran,“Control Scheme of

Hybrid Wind-Diesel Power Generation System”, 2011.

10.Champa Nandi, A. K. Chakraborty, “Wind Power Plants with VSC Based STATCOM in

PSCAD/EMTDC Environment’’2012.

11.Dhekekar R.S, Srikanth N.V, “ANN Controlled VSC STATCOM with Harmonic Reduction

for VAR Compensation”, IEEE Trans, Vol.2, No. 1, pp. 76-84, March. 2012.

12.Koteswara I.V, Rao,Radha Krishna Reddy.S, Sujesh.K, Subramanyam.JVB,“A Novel

Approach on Load Compensation for Diesel Generator-Based Isolated Generation System

Employing D-STATCOM”, IEEE Trans, Vol. 11, No. 1, May 2012.

13.Linga Retti.P, Siva Kotti Retti.CH, “A D-STATCOM-Control Scheme for Power Quality

Improvement of Grid Connected Wind Energy System for Balanced and Unbalanced Non

linear Loads”, IEEE Trans, Vol. 2, No. 3, June 2012.

14.Srikant Prasad,PPManoj Jha, M. F. Qureshi “Steady State Stability Analysis of a CSI Fed

Synchronous Motor Drive System with Damper Windings Included using ANFIS”

Advances in Modelling, Series C. Automatic Control (theory and applications)ISSN:

1240-4535, Vol. 69, nº1, PP. 39-57, (2014) AMSE France.

15.Mithlesh Singh,PPM.F. Qureshi, Adarsh Kumar, Manoj Jha “Fuzzy based simulation of D-

STATCOM and DVR in Power Systems” Modelling, Measurement and Control, Series A.

General Physics and Electrical Applications, ISSN: 1259-5985, Vol. 87, nº2, pp. 01-18,

(2014), AMSE.

16.Manoj Jha,P PAnita Singh, M.F. Qureshi, “Evolutionary Interval Type-2 Fuzzy Logic

Controller in Power System Transient Stability Analysis” Advances in Modelling, Series C.

Automatic Control (theory and applications) ISSN: 1240-4535, Vol. 68, nº3, PP. 01-18,

(2013) AMSE France.

17.Manoj Jha, Anita Singh, M.F. Qureshi “Design of Genetically Tuned Interval Type-2 Fuzzy

PID controller for Load Frequency Control (LFC) in the Un-regulated Power System”

Advances in Modelling, Series C. Automatic Control (theory and applications)ISSN:

1240-4535, Vol. 69, nº1, PP. 85-105, (2014) AMSE France.

Figure

Table 1 The rule table of ANFIS designed for input/output error control of PVEGS.
Fig. 11 Mitigation of voltage swell
Fig. 12 Mitigation of voltage sag
Fig. 15: Power response with DPFC with PI and with ANFIS controls
+2

References

Related documents

Also, by using the definition of graph folding, we prove that the graph of equivalence classes and the graph folding of commutative KU-algebra are the same,

loaded stratum corneum lipid liposomes (SCLLs), characterization of the SCLLs, and determina- tion of antimycotic efficacy of the formulation against Candida albicans and

the paper smart antenna technology can be used for numerable applications in wireless communication like in. MIMO systems, mobile

Australian Education Index; Canadian Business and Current Affairs Education; Contents Pages in Education; Current Contents; ERIC; EBSCOhost; Education Research Index; Sage;

Herbert Stoddard (1962), an early advocate of prescribed fire (who, by the way, never fin- ished high school but who founded Tall Tim- bers Research Station and, along with Aldo

The objectives of the current study were to examine: (1) the absolute value of the IAA amounts in the cambial region tissues at the lower and upper trunk; (2) the effects of

We had no experience with laparo- scopic bariatric surgery prior to May 2007, and this pro- spective study reviews our first patients focusing on procedure complications,

Step #3 - Use the minimum number of cations and anions needed to make the sum of all charges in the formula equal zero. Why? Answer - One positive two charge is counterbalanced by