frequency regulation of micro-grid connected hybrid power ......solar pv array ac bus micro grid...

Technol Econ Smart Grids Sustain Energy (2017) 2: 13DOI 10.1007/s40866-017-0028-3

ORIGINAL PAPER

Frequency Regulation of Micro-grid Connected HybridPower System with SMES

Shailendra Singh1 ·Rohit Kumar Verma2 ·Ashish Kumar Shakya3 ·Satyendra Pratap Singh1

Received: 25 October 2015 / Accepted: 26 June 2017 / Published online: 22 July 2017© Springer Nature Singapore Pte Ltd. 2017

Abstract This paper presents the frequency regulationanalysis of a micro-grid connected hybrid power systembased on solar Photovoltaic (PV), Wind and Diesel-EngineGenerator (DEG) with Superconducting Magnetic EnergyStorage system (SMES) unit. Abrupt change in load demandand power fluctuations from PV and wind power sourcecauses frequency variability in the system. In order to miti-gate this issue, the hybrid power system incorporates alongwith the energy storage devices. A case study of the impactof SMES operations on the performance of frequency sta-bility of hybrid system has been carried out. Simulationof a Hybrid Power System (HPS) model has been carriedout in two different scenarios with the two phases. Firstis the steady state investigation of HPS using PI controllerwith step load change response. Second is the dynamicperformance study throughout a typical day with arbitraryvariation in load demands. Controller parameters are tunedwith Genetic Algorithm based optimization approach. Thelimits on the inductor (coil) current due to the possibili-ties of discontinuous conduction in the presence of largedisturbances and limited storage capacity have also beenincorporated. HPS has analyzed in the Matlab Simulinkenvironment. Simulation results indicate that SMES system

� Shailendra [email protected]

1 Electrical Engineering Department, Indian Instituteof Technology (BHU), Varanasi, India

2 Electrical Engineering Department, Indian Instituteof Technology (ISM), Dhanbad, India

3 Electrical Engineering Department, GLA University,Mathura, India

can contribute towards the frequency stability in both casesfor variable power generations from renewable sources andirregular demand loads.

Keywords Renewable power · Photovoltaic · Energystorage · SMES · Micro-grid · Frequency stability · Hybridpower system

Introduction

The adverse effect of the greenhouse gases on atmosphereand decreasing fossil fuel reserves drags attentions towardsclean and pollution free energy. In modern era, the leadingclean energy sources are hydro energy, solar energy, windenergy and biogas mainly [1, 2]. Moreover, generationsfrom the wind and solar have enhanced their contribution inmost of the world’s electricity generation [3].The primaryapplication of solar power system is to provide power to theoff-grid area where the extension of the grid is not possible.However, wind power is more commonly used for grid inte-grations. Solar PV arrays have more installation flexibilityand can be easily mounted on domestic locations [4]. In thecurrent scenario, the wind, and solar PV power plants arewidely deployed for distributed generations and microgrids(MGs). Generated power from PV and wind sources is vari-able in nature; it can be used to compensate the load demandlocally and/or injected into local grids [5]. Renewable tech-nology is also facing a lot of technical, economic and socialbarriers because of their intermittency nature and uncer-tainty issues [6]. Moreover, the interconnection of theserenewables into the grids/MGs creates some technical prob-lems such as frequency and voltage regulations. Operationalissues in distributed system also raised due to high/low pen-etrations of distributed renewable sources [7]. Therefore,

http://crossmark.crossref.org/dialog/?doi=10.1007/s40866-017-0028-3&domain=pdfmailto:[email protected]

13 Page 2 of 13 Technol Econ Smart Grids Sustain Energy (2017) 2: 13

these problems need the attention of researchers and reg-ulatory authorities to provide the best feasible solution. Inthis article, the frequency regulation issue raised during theoperation of MG connected HPS has been discussed. Thefrequency stability of the system is mainly affected by mis-matching of load generation balance and irregular loadings.Therefore, energy storage technology has been introducedto mitigate the frequency stability issue in micro-grid. Cur-rently, storage technologies are capable of handling multi-functional tasks as backup powers, active/reactive powerinjection or absorption from/to the grid [8–10]. The conven-tional energy storage such as Battery energy storage system,Hydro pump storage, etc. is proposed by various researchersin the literature [11–13]. However, these storages have sev-eral problems such as high response time, limited Lifecycle,storage capacity, large unit volume and hazardous environ-mental impact, etc. Therefore, in this paper, SMES basedtechnology has been proposed for the stable operation ofhybrid power system. The potential capabilities such as fastresponse, inherently huge storage efficiency and injecting/absorbing real or reactive power ability endorse the SMESas a viable solution [14, 15]. The effect of SMES during thetransient disturbance condition for a grid-connected powersystem has been analyzed in [15]. Various authors [15–17]have presented the application of SMES with the on gridand off grid mode. Neural network based frequency controlof single and two area system is well documented in [18].Authors in [19] have been presented the power flow con-trol and management scheme with energy storage devices. Aknowledge domain based concept for power flow controllerhas been discussed in [20]. Various efforts had been carriedout by the authors to minimize the frequency fluctuation inhybrid power system with SMES in [4, 5, 21].

This paper presents the extension work performed pre-vious research work in [16]. However, for the aspect ofnovelty of the following additional work has been incorpo-rated in this paper. (I) Small stability analysis of the solarbased hybrid system with step load response (II) System fre-quency response has been controlled with PI controller. (III)Controller parameters are tuned with Genetic Algorithmbased optimization approach. (IV) An additional hybridpower system based on purely renewable power generationis analyzed with SMES. The effect of SMES during thesteady and dynamic state operation has also been discussedin this study.

Description and Modeling of HPS

In this section, description and modeling of hybrid powersystem have been carried out. Figure 1 shows the schematicdiagrams of HPS connected to the microgrid. HPS is cate-gories in three parts as follows:

Bidirec�onalconverter

MonitoringSystem

Diesel EngineGenerator

SMES

~

LOAD

Solar PV Array

AC BUS

MICRO GRID

Wind power

Fig. 1 Schematic diagram of hybrid Power system with SMES

i Power Generation Sources (Solar PV, Wind, and DEG)ii Storage Systemiii Power Conditioning System (PCS) and load Demand.

Power generation sources and storage facility have con-nected to the local grid. SMES has been consumed andsupplying power from the local grid. Modeling of differentpower generating plants such as PV, wind, DEG, and storagesystem has been carried out below in this section. Modelingof PCS system has not been considered in this study. It isassumed that all required PCS system is working properly inthe system. The system model is also useful for distributedand isolated mode of operation.

Solar Photovoltaic Model

The Solar PV system is composed with PV arrays, connec-tions and protective parts [13]. Arrays are the combinationof solar cells. The equivalent circuit of a PV cell is the com-bination of a current source in parallel with a diode. Thecurrent source output is directly proportional to the lightfalling on the cell. The equation of ideal solar cell [1, 7] thatsymbolizes the solar cell model is [13]:

I = IL − IR[exp

(V

AVi

)− 1

](1)

Fig. 2 Wind speed to power generation model [23]

Technol Econ Smart Grids Sustain Energy (2017) 2: 13 Page 3 of 13 13

Fig. 3 Schematic diagram ofthe SMES unit

ControlUnit

3-phase network

Y YTransformer

Supe

r con

duct

ing

Coi

l

Power ConditioningSystem

Id+

_

Ed

ΔF,ΔV

where: “IL” is photocurrent (A); “IR” is reverse saturationcurrent (A);“V” is diode voltage (V); “Vi” is thermal volt-age, “A” is diode ideality factor. The composition of solarcells in series and parallel combination with their protectiondevice is known as the solar module [13]. Current-voltagecharacteristic equation of equivalent circuit for a PV mod-ule arranged in series Ns and parallel Np can be defined as[7, 22]:

IM = NpIL − NpIR[exp

(VM/Ns + IM/Ns

AVi

)]

−(Np/Ns

)VM + IMRSeRSh

. (2)

“Np” is cells parallel number; “Ns” is cells series number.Detailed modeling of PV system has discussed in [1, 13].

Wind Speed to Power Conversion Model

Wind velocity to power conversion model has been usedto represents the wind power model. The detailed model-ing of wind power has been already discussed in previouslyreported research work [16]. Wind velocity is the combi-nation of following four components: average speed (vwa),

wind speed ramp (vwr), wind gust (vwg) and wind tur-bulence (vwt) in general. Therefore, the resultant windvelocity can be given in [12, 23]:

vw(t) = vwa(t) + vwr(t) + vwg(t) + vwt(t) (3)To approximate the fluctuation effects low pass filter isdeployed in the wind power conversion model as shown inFig. 2. The following equation governs the power generatedby wind turbine:

Pw = cp (λ, β) σ2

π r2v3w (4)

where cp is the power coefficient of wind turbine and λ isthe tip speed ratio, which is defined by:

λ = ωTrvw

(5)

Diesel Engine Generator Model

Diesel engine power generation is governed by followingequation generally.

�Pd = −(1

R+ KI

s

) (1

Tsgs + 1)(

KDEG

TDEG s + 1)

�fe

(6)

Fig. 4 SMES control unit blockdiagram [16]


Fig. 5 Mathematical model of Micro-grid connected HPS

where, R is the speed regulation. DEG automatically startsup with proper control.

SMES Model

Schematic diagram of SMES model has been shown inFig. 3. The SMES model has been divided into three partsmainly. First is as superconducting coil i.e. also known asthe heart of the system and second deals with power con-ditioning system. The third is control unit which has beenshown in Fig. 4 separately. A detailed description of SMESmodeling is well explained in [16, 24]. From Fig. 3, DCvoltage is given by:

Ed = 2Vd cosα − 2Id RC (7)

Where Ed is the DC voltage applied to the inductor in KV,α is the firing angle in degrees, Id is the current flowingthrough the inductor in kA, RC is equivalent commutatingresistance in k and Vdo is the maximum circuit bridgevoltage in kV [16]. Controlling of charging and discharg-ing of the superconducting coil is done with the alteringof commutation angle α. The converter operates in thecharging mode (converter mode), when α is less than 90◦and in discharging mode (inverter mode) when α greater

Table 1 Parameters of GA

Parameter Value

Population size 30

Max no. of generation 50

Mutation probability 0.15

Crossover probability 0.8

Length of chromosomes 8 bits

No. of variables 4

Maximum iteration 30

Start

Generate initial population

Evalute each chromosme by

performance index

Selection, Crossover, Mutation

Evalute new formed

chromosome

If

G0

If∆F=0

Id > IdL Id < IdUStand bymode

SMES

Wind Power /+ PV Power

D

I

S

C

H

A

R

G

E

C

H

A

R

G

E

YES

NO

YES

YES

NO

NO

NO

DEGPower

∆F

Fig. 7 Flow chart of the operation strategy


Table 2 Rating and parameters of hybrid model

Parameter Value (Rating)

System power rating 2 MW (Base)

Capacity of SMES 2 KA

Initial Inductor current(Ido) 100KW

Upper limit of Inductor 150% of Initial

current(IdU) inductor current

Lower limit of Inductor 30% of Initial

current (IdL) inductor current

M 0.12

D 0.1

KDEG 1

TDEG 0.5s

R 0. 5

DEG Power 250KW

Peak PV Power 200 KW

Wind power 160 KW

incremental change in the voltage applied to the inductor isexpressed as:

�Ed = [ KSMES1 + sTdc ]�F (8)

Where, �Ed is the incremental change in converter voltage,Tdc is the converter time delay, KSMES is the control loopgain and �F is the actuating signal to the control unit ofSMES. The inductor current deviation (�Id) is calculatedby following Eq. 9:

�Id = �EdsL

(9)

The overall change in SMES power flow �PSMES due tosystem frequency (�F) change is shown below Eq. 10:

�PSMES = �Ed.Id (10)Id = Ido + �Id (11)

Where Id is net inductor current

0 1 2 3 4 5 6 7 8 9 101112131415161718192021222324-0.15

-0.1

-0.05

0

0.05

Time (Hours)

Cha

nge

in lo

ad D

eman

d (p

.u.)

Fig. 8 Load demand Varations profile

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 2426

28

30

32

34

36

38

40

Time (hour)

Tem

pera

ture

(0 C

)

Fig. 9 Temperature Profile throughout a typical day

Mathematical Modelling of HPS

This subsection described the mathematical model of HPS.The transfer function based model of HPS has been shownin Fig. 5. Assumed that, PCS is properly working at theirdefined positions. Table 1 shows the parameters of HPSused in modeling. The load generation balance must bemaintained for the smooth operation of the grid-connectedsystem. The control strategy is determined by the control-ling error which is the difference between the change in loaddemand (�PD) and change in net power production (�PT).

�PT = �PPV + �Pw ± �PDEG±�PSMES (12)

�PS = �PT − �PD (13)where �PS is net power controlling error.

Following equation finds the change in frequency varia-tion (�F):

�F = KPS1 + TPS .�Ps (14)

Since an inherent time delay exists between system frequen-cies. Transfer function for system frequency variation to perunit power deviation is expressed in below equation:

�F = 1D + sM .�PS (15)

D = 1KPS

,M = TPSKPS

0 1 2 3 4 5 6 7 8 9 1 11 12 13 14 15 16 17 18 19 20 21 22 23 240

0.1

0.2

0.30.4

0.5

0.60.7

0.8

0.91

Time (Hour)

Sola

r Ir

radi

atio

n (k

w/m

2 )

Fig. 10 Solar Irradiation proflie throughout a day


0 1 2 3 4 5 6 7 8 9 1011121314151617181920212223240

0.02

0.04

0.06

0.08

0.1

0.12

Time (Hours)

Cha

nge

in P

V P

ower

(p.

u.)

Fig. 11 PV Power profile for a typical day

Where M = The equivalent inertia constant (p.u) & D =Damping constant (p.u).

Optimal Design of PI Controllers with GeneticAlgorithm (GA)

The actuating signal for PI controller is the summation ofproportional and integral of the error signal. Gain of PIcontroller is shown as:

G(s) = KP + KIs

(16)

Where KP and KI are the Proportional(P) and Integral(I)gain constant respectively.

Objective Function

The tuning of PI controller depends upon two parameters(Kp & KI) to regulate the change in frequency error (�f).The values of PI parameters are calculated through geneticalgorithm optimization technique. In order to achieve, opti-mum system response and to minimize the frequency errorintegral time square error method is used here as shownbelow:

F (X) = IT ES =∫ ∞

O

t ∗ (�f )2 dt (17)

X is the variable in terms of the values of Kp and Ki. Theoptimal values of PI parameter are determined with thegenetic algorithm optimization technique.

Genetic Algorithms

Genetic algorithms (GA) are a biologically inspired algo-rithm. GA search algorithm is based on the mechanics ofnatural selection that copycats biological evolution. GA

0 1 2 3 4 5 6 7 8 9 10 111213 14151617 181920 212223 240

0.02

0.04

0.06

0.08

0.1

Time (Hours)

Cha

nge

in W

ind

Pow

er (

p.u.

)

Fig. 12 Wind Power profile through a day

combines the survival of the fittest among string struc-tures with a structured randomized information exchangeto develop a search algorithm with the innovative flairs ofhuman search. Generally, GA used to find out high-qualityoptimization and solution in search problems. In each iter-ation, a new set of artificial string is formed using bits andpieces of the fittest of the previous [25, 26]. A simple GAthat produces good results in many real-world problems iscomposed of three operations as 1) Selection or Reproduc-tion. 2) Crossover. 3) Mutation. Reproduction or selection isa process in which individual strings are copied according totheir objective function values. After reproduction, simplecrossover carried out in two steps. First, a member of newlyreproduced strings in the mating pool is mated at random.Second, each pair of strings undergoes crossing over as aninteger position. Mutation operator shields beside irrecoverableloss [25]. Figure 6 shows the flowchart of Genetic Algorithm.

Step Procedures

Step procedure used for tuning of PI controllers with GA isshown below

i. Initialize the population for variables(x) with 8 bitslength of chromosomes for each one. Size of thepopulation is in between 15-30 and set iteration (r)=1

ii. Calculate the fitness function based on Eq. 9 for allcandidate solutions.

iii. Sort all the fitness functions in ascending order.[Fr1(x), F

r2(x),]. . . . Fp

r(x),], Best cost = Fr1(x), bestcandidate solution =x, where r is no of iterations.

iv. Set the probability of chromosomes and mutationv. Select all the parent chromosomes and apply

crossover on each candidate solutionvi. Apply mutation to the all chromosomevii. Again calculate fitness function for all candidate

solution and sort them in ascending order [Fr+11 (x),Fr+12 (x),]. . . .Fr+1p (x),],


Fig. 13 Frequency Responsewith increased step load (peakload.)

0 5 10 15 20 25 3048.6

48.8

49

49.2

49.4

49.6

49.8

50

Time (second)Fr

eque

ncy(

Hz)

Without SMESWith SMESWith SMES and GA-PI

Fig. 14 SMES power injectionresponse during the peak load(0.25p.u.)

0 5 10 15 20 25 300

0.01

0.02

0.03

0.04

0.05

Time (seconds)

Cha

nge

in S

MES

Pow

er (p

.u.)

Fig. 15 Frequency Responsewith decreased step load (Offpeak load.)

0 5 10 15 20 25 3049.8

50

50.2

50.4

50.6

50.8

51

51.2

Time(Seconds)

Freq

uenc

y (H

z)

With SMES and GA-PI

Without SMES

With SMES


Fig. 16 SMES power consumption response during the off peak load

viii. If, Fr+11 (x) < Best Cost, Update Best cost = Fr+11 (x)& best candidate solution = x, Else, best cost = bestcost

ix. Put r = r + 1, if no of iteration(r) < max iter, go tostep 2

Else, stop and store best cost and best candidate solution.Parameters of GA used to tune the PI controllers are

given in Table 1.

Operating Strategy

In this section, operation strategy for considered HPS isdiscussed and flow chart has been shown in Fig. 7. The sys-tem/grid frequency is very sensitive to load demand change.Therefore, whenever variations in load demand occur, sys-tem frequency changes according to it. Grid frequencymust lie within the acceptable range for stable and smoothoperation of HPS. In order to neutralize, the frequency fluc-tuations, the power supply should be controlled according toload variations. When the frequency of the grid is reportedbelow 50 Hz or change in frequency is negative, SMESstarts injecting the power into the grid. However, SMES

0 1 2 3 4 5 6 7 8 9 1011121314151617181920212223 24-0.05

0

0.05

Time (Hours)

Cha

nge

in S

ME

S Po

wer

(p.

u.)

Fig. 17 Variation of SMES Power in both charging & dischargingmode

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24

-0.1

-0.05

0

0.05

0.1

Time (Hours)

Cha

nge

in D

EG

Pow

er (

p.u.

)

Fig. 18 Change in DEG. Power output

unit must check the status of superconducting coil current(Id) and verify its capacity to lower limit of the inductorcurrent (IdL). If Id >IdL, SMES will turn on and start dis-charging; otherwise, it will remain in off mode. In anothercase, when the frequency of the grid is reported above 50Hz or change in frequency is positive, SMES starts absorb-ing (charging the SMES coil) some excess power from thegrid. Before consuming the power, Id has to check and com-pared to its upper limit of the inductor current (IdU). IfId


Fig. 20 Frequency Responsewith increased step load (peakload.)

0 5 10 15 20 25 3049.1

49.2

49.3

49.4

49.5

49.6

49.7

49.8

49.9

50

50.1

Time(Second)

Freq

uenc

y(H

z)

Without SMESWith SMES and GA-PIWith SMES

response of systemmodel. The second is the dynamic stabil-ity analysis with 24 hours simulations with random variationin load demand. Phase I is further divided into two cases asstep load analysis during the peak load and off-peak loadtime. Variations of a typical day load demand have beenshown in Fig. 8. The temperature and solar irradiations vari-ations for a typical day have been shown in Figs. 9 and 10respectively. The power produced with PV and wind planthas been shown in Figs. 11 and 12 respectively.

Scenario 1- HPS with Solar PV+DEG+SMES

Scenario 1, consists the study of HPS containing the solarPV, DEG, and SMES storage facilities. The penetration of

solar power is hundred percentage of generated PV power.The study of this scenario is carried out in two phases withtwo cases as discussed below:

Phase I: Steady state Stability AnalysisConsidered HPS is analyzed through step load change

response during peak and off-peak load demand with thepresence or absence of SMES. The system is simulatedfor the duration of has carried 30 seconds.

(i) Case I: During peak loadStep load change of 0.25 p.u is considered for

the analysis of this cases. Due to a sudden rise inload demand, the frequency will go down to belowa defined limit. Therefore, to maintain the system

Fig. 21 SMES power injectionresponse during the peak load(0.25p.u.)

0 5 10 15 20 25 300

0.005

0.01

0.015

0.02

0.025

0.03

0.035

0.04

0.045

0.05

Time(seconds)

Cha

nge

in S

MES

Pow

er(p

.u.)


Fig. 22 Frequency Responsewith decreased step load (Offpeak load.)

0 5 10 15 20 25 3049.8

50

50.2

50.4

50.6

50.8

51

51.2

Time(Seconds)

Freq

uenc

y (H

z)

With SMES and GA-PI

Without SMES

With SMES

frequency, SMES will start supplying the power intothe microgrid. Because of this act, SMES is oper-ating in the discharging mode. Frequency responsewith the sudden increase in step load demand hasbeen shown in Fig. 13. The effect in frequencyresponse with, without SMES and with GA-PI con-troller is shown the Fig. 13. From Fig. 13 it isobserved that frequency deviation is minimized max-imum during the operation of HPS with the com-bination of optimized controller and SMES. Powerinjected by the SMES has also shown in Fig. 14.

(ii) Case II: During off peak loadStep load change of 0.045 p.u. has been consid-

ered for the analysis of this cases. Due to a suddendecrease in load demand, there is a rise in systemfrequency is reported. However, to compensate thisfrequency fluctuations, power generation should beminimized or/and store the energy in SMES. In order

to maintain the system frequency, SMES will startconsuming the power from the microgrid. This stateof SMES is known as the charging mode. The effectin frequency response with, without SMES and withGA-PI controller is shown the Fig. 15. From Fig. 15it is observed that frequency deviation is minimizedmaximum in the operation of HPS the combinationof optimized controller and SMES. Power absorbedby the SMES with negative sign has also shown inFig. 16.

Phase II: Dynamic Simulations for 24 HoursThe considered HPS model has simulated for 24 hours

in this phase of the study. The dynamic stability study iscarried out with varying load demand and power genera-tions throughout a day. The study is further divided intotwo cases, i.e. during the increase in load and decrease inload demand.

Fig. 23 SMES powerconsumption response duringthe off peak load

0 5 10 15 20 25 30-0.05

-0.04

-0.03

-0.02

-0.01

0

Time(second)

Cha

ge in

SM

ES p

ower

(p.u

.)


Fig. 24 Frequency response of the system throughout the day

(i) Case I: Increase in load demandDue to increase in load demand, system frequency

decreases. Therefore, power generation should beincreased to maintain the frequency. PV power isregularly supplying the power to the system. Due todip in system frequency, DEG power can be enlargedto supply the power to the system, up to its ratedcapacity. To compensate the load demand, SMESsupplying power to the system as shown in Fig. 17.

(ii) Case II: Decrease in load DemandChange in frequency increases, as a change in

load demand, decreases. So, in order to maintainfrequency, the total power generation should bedecrease and/or increase the load consumptions.Due to variation in frequency, the net change inDEG power shown in Fig. 18. Power supplied bySMES should be decreasing response reached zeroas shown in Fig. 17. However, the surplus power isstill more than load demand so for balancing powerSMES starts consuming power from the system.Now SMES is in charging mode.Figure 19 shows the frequency response of

the considered HPS including both increase anddecrease of the load demand with and without SMESthroughout a day. From Fig. 19, it can be observedthat frequency fluctuations can be mitigated with theaddition of SMES into the system.

Scenario 2- HPS with Solar PV+Wind+SMES

Scenario 2 consists the study of HPS containing the solarPV, Wind and SMES storage facilities. HPS system is alsoknown as 100% power generations from renewable. Thepenetration of solar power is fifty percentage of generatedPV power. The study of this scenario is carried out in twophases with two cases as discussed below:

Phase I: Steady state Stability AnalysisConsidered HPS is analyzed through step load change

response during peak and off-peak load demand with the

Fig. 25 SMES Power variation in both charging & discharging mode

presence or absence of SMES. The system is simulatedfor the duration of has carried 30 seconds.

(i) CASE I: During peak loadStep load change of 0.25 p.u is considered for the

analysis of this cases. The operation of the system issame as discussed in Case I of Phase I of Scenario1(HPS with Solar PV+DEG+SMES). The effect infrequency response with, without SMES and withGA-PI controller is shown the Fig. 20. From Fig. 20it is observed that frequency deviation is minimizedmaximum during the operation of HPS with the com-bination of optimized controller and SMES. Powerinjected by the SMES has also shown in Fig. 21.

(ii) CASE II: During off peak loadStep load change of 0.045 p.u. has been consid-

ered for the analysis of this subcase. The operation ofthe system is same as discussed in Case II, Phase I ofScenario 1(HPS with Solar PV+DEG+SMES). Theeffect in frequency response with, without SMESand with GA-PI controllers are shown in the Fig. 22.From Fig. 22, it is observed that frequency deviationis minimized maximum, in the operation of HPS withthe combination of optimized controller and SMES.Power absorbed by the SMES has also shown in Fig.23.

Phase II: Dynamic Simulations for 24 HoursThe considered HPS model has simulated for 24 hours

in this phase of the study. The dynamic stability studyis carried out with varying load demand and power gen-erations throughout a day. The operating scheme in thisphase is similar to previous section phase II of Scenario1- (HPS with Solar PV+DEG+SMES). However, inthis study, DEG is not incorporated. Frequency responsethroughout the day has shown in Fig. 24. SMES powerprofile during both the charging and discharging modehas shown in Fig. 25. From the above figures, it is


demonstrated that fluctuations in system frequency canbe smoothened or mitigate perfectly with the integrationof SMES units.

Conclusion

The mitigation of frequency fluctuations in micro grid con-nected PV, wind based hybrid power systems has beenanalyzed with SMES technology. The performance of theconsidered system has been assessed with the variable loaddemand and renewable power sources. The small signal sta-bility analysis is carried out with and without PI controllers.The parameters of PI controllers is optimized with geneticalgorithm. Simulation results demostrates that SMES basedsystem gives better frequency response in comparison tothe only diesel generator and without SMES. However, itseems that SMES technologies are not so much economi-cally. However, this issue may compensate with its fruitfulapplications such as fast response, high efficiency, capa-bility of control of real power and reactive power flows.With the emergence of smart grids, this technology isreceiving more attention in the field of power and energysystems. Therefore, it is hoped that its potential advantagesand environmental benefits will make SMES units a viablealternative for energy storage source.

References

1. Singh S, Singh AK, Chanana S (2012) Operation and control ofa hybrid photovoltaic-diesel-fuel cell system connected to micro-grid. IEEE Power India Conf

2. Jae Woong S et al (2013) Synergistic control of SMES and batteryenergy storage for enabling dispatchability of renewable energysources. IEEE Trans Appl Supercond 23(2):5701205–5701205

3. Kumar D, Sarkar S (2016) A review on the technology, per-formance, design optimization, reliability, techno-economics andenvironmental impacts of hydrokinetic energy conversion sys-tems. Renew Sust Energ Rev 58:796–813

4. Walid O (2010) Performance analysis of grid-Connected photo-voltaic systems. Dissertation, University of Waterloo

5. Qi C, Ming Z (2012) Photovoltaic module simulink modelfor a stand-alone pv system. International conference onapplied physics and industrial engineering. Phys Procedia 29:94–100

6. Byung-Kwan K et al (2013) Effect of a SMES in Power distri-bution network with PV system and PBEVs. IEEE Trans ApplSupercond 23(2):5700104–5700104

7. Kim AR et al (2011) Performance analysis of a toroid-type HTSSMES adopted for frequency stabilization. IEEE Trans ApplSupercond 21(2):1367–1370

8. Ali MH et al (2010) An overview of SMES applications in powerand energy systems. IEEE Trans Sustain Energy 1(1):38–47

9. Chen L, Liu Y et al (2006) Detailed modeling of superconductingmagnetic energy storage (SMES) system. IEEE Trans Power Deliv21(2):699–710

10. Nakayama T et al (2009) Micro power grid systemwith SMES andsuperconducting cable modules cooled by liquid hydrogen. IEEETrans Appl Supercond 19(2):2062–2065

11. Chanana S, Kumar A (2011) Operation and control of BESSusing frequency-linked pricing in real-time market with high windpenetration. Int J Energy Sector Manag 5(4):585–602

12. Singh S, Singh M, Chanana S, Raveendhra D (2014) Operationand control of a hybrid wind-diesel-battery energy system con-nected to micro-grid. In: International conference on control andautomation, robotics and embedded system

13. Singh S, Singh SK, Chanana S, Singh YP (2014) Frequency reg-ulation of an isolated hybrid power system with battery energystorage system. Power and energy systems: towards sustainableenergy

14. Ali MH et al (2009) Improvement of wind-generator stability byfuzzy-logic-controlled SMES. IEEE Trans Ind Appl 45(2):1045–1051

15. Banerjee S, Chatterjee JK, Tripathy SC (1990) Application ofmagnetic energy storage unit as load-frequency stabilizer. IEEETrans Energy Convers 5(1)

16. Singh S, Joshi H, Chanana S, Verma RK (2014) Impact of super-conducting magnetic energy storage on frequency stability of anisolated hybrid power system. Computing for sustainable globaldevelopment (INDIACom)

17. Tripathy SC, Kalantar M, Balasubramenian R (1991) Dynamicand stability of wind and diesel turbine generators with supercon-ducting magnetic energy storage unit on an isolated power system.IEEE Trans Energy Convers 6(4)

18. Das S, Devesh S, Singh SP (2015) Single and two area LFC usingRBF. In: IEEE International conference on recent developments incontrol automation and power engineering (RDCAPE), pp 355–360

19. Ansari TMM, Velusami S (2010) Dual mode linguistic hedgefuzzy logic controller for an isolated wind–diesel hybrid powersystem with superconducting magnetic energy storage unit.Energy Convers Manag 51(1):169–181

20. Pandey RK, Gupta DK (2016) Knowledge domain states mappingconcept for controller tuning in an interconnected power network.Int J Electr Power Energy Syst 80:160–170

21. Molina MG, Mercado PE (2011) Power flow stabilization andcontrol of microgrid with wind generation by superconductingmagnetic energy storage. IEEE Trans Power Electron 26(2):910–922

22. Rawat S, Singh S, Gaur K (2014) Load frequency control of ahybrid renewable power system with fuel cell system. In: PowerIndia international conference (PIICON)

23. Singh S, Singh M, Chanana S, Semwal S (2015) Frequency regu-lation of isolated hybrid Wind/Diesel, power generation with fuelcell system. Lecture notes in electrical engineering, Springer, 326pp. 853–862

24. Ali MH, Tamura J, Wu B (2008) SMES strategy to minimize fre-quency fluctuations of wind generator system. In: 34th Annualconference of IEEE industrial electronics IECON

25. Gupta DK, Naresh R, Singh MK (2013) Genetic aided tuning ofPI controller based AGC for two area interconnected system. In:Proceedings of IEEE 2nd student conference on engineering andsystems

26. Goldberg DE (1989) Genetic algorithm in search, optimizationand machine learning reading. Addison-Wesley Publishing Com-pany, Inc


Shailendra Singh was born inAuraiya, India, in 1986. Hereceived the B.Tech. degreein electrical and electronicsengineering from the UttarPradesh Technical University(UPTU) Lucknow, India, in2010 and M.Tech degree withspecialization in Power sys-tem from National Instituteof Technology, Kurukshetra,India in 2012. After that Heworked as a assistant Profes-sor in Graphic Era hill Univer-sity, Bhimtal and Graphic.

Era University DehradunIndia from 2012 to 2014. He has numerous research papers in reputepeer reviewed national, international conferences and journals. He isalso student member of IEEE(USA), India Smart Grid Forum(ISGF)and various IEEE societies such as power and energy (PES), Indus-trial and Applications (IAS) and Electrification and Transportation.Currently he is perusing Ph.D degree in department of electricalengineering of Indian Institute of Technology (BHU) Varanasi, India.

His research interests include Operation and control of Power sys-tem, Smart distribution system ,energy Storage System, Integration ofRenewable Power Sources and MicroGrids.

Rohit Kumar Verma receivedhis graduation in electricalengineering from Govind Bal-labh Pant Engineering Col-lege, Pauri, India and Master’sdegree from Institute of Tech-nology Roorkee, India. Hehas two years teaching expe-rience. He is currently pur-suing his PhD degree fromIndian Institute of Technol-ogy (Indian School of Mines),Dhanbad, India. His currentresearch interests Mini Grids,Renewable Energy and powersystems.

Ashish Kumar Shakya wasborn in Kanpur, India, in1989. He received the B.Tech.degree in Electrical & Elec-tronics engineering fromUPTU, Lucknow, India, in2010, and the M.Tech. degreein electrical engineering fromthe National Institute of Tech-nology (NIT) Kurukshetra,Haryana, India, in 2012.Currently he is Assistant Pro-fessor in the Department ofElectrical Engineering, GLAUniversity, Mathura, India.His current research interests

Micro Grids Control, Nonlinear Control, Power Electronics andElectrical Machines.

Satyendra Pratap Singh wasborn inVaranasi, India, in 1985.He received the B.Tech. degreein electrical engineering fromthe Uttar Pradesh TechnicalUniversity (UPTU) Lucknow,India, in 2007, and the M.E.degree in electrical engineer-ing with specialization in powersystem and electric drives fromthe Thapar University, Patiala,India, in 2009.

He is currently pursuingthe Ph.D. degree in electri-cal engineering with special-ization in power system from

the Indian Institute of Technology (BHU) Varanasi, India. Hisresearchinterests include wide area measurement in power system, powersystem operation, applications of AI in power system optimization.

Frequency Regulation of Micro-grid Connected Hybrid Power System with SMESAbstractIntroductionDescription and Modeling of HPSSolar Photovoltaic ModelWind Speed to Power Conversion ModelDiesel Engine Generator ModelSMES ModelMathematical Modelling of HPS

Optimal Design of PI Controllers with Genetic Algorithm (GA)Objective FunctionGenetic AlgorithmsStep Procedures

Operating StrategySimulation Results and AnalysisScenario 1- HPS with Solar PV+DEG+SMESScenario 2- HPS with Solar PV+Wind+SMES*-.5pt

ConclusionReferences

frequency regulation of micro-grid connected hybrid power ......solar pv array ac bus micro grid...

Documents