Photovoltaic Power Generation

Photovoltaic Power Conversion

• Sun light striking to the PV cell frees electrons from their atoms, are directed towards the surface
of the cell. This rush of electrons causes flow of current.

• Typically, a PV cell generates a voltage around 0.5 to 0.8 volts

depending on the semiconductor and the built-up technology.

• This voltage is low enough as it cannot be of use. Therefore, to

get benefit from this technology, tens of PV cells (involving 36 to 72
cells) are connected in series to form a PV module. These modules
can be interconnected in series and/or parallel to form a PV panel
Equivalent Model
• The equivalent circuit model of a PV cell is needed in
order to simulate its real behaviour.
• A cell can be modelled as a DC current source in parallel
with two diodes that represent currents escaping due to
diffusion and charge recombination mechanisms.
• Assuming that the current due to charge recombination is small
enough to be neglected, a simplified PV cell model can be
reached as shown in fig. 2 known as single-diode model.
• Series resistance (Rs) is for representing hindrance in the path
of flow of electrons from n to p junction and parallel resistance
(Rp) is due to the leakage current.
Modelling contd.
The equivalent circuit of a PV cell can be developed with a current source representing the photo current
(IPh) and a diode. Its output current can be written as:
  qV  
I  I ph  I rs  exp  
 1 ; I ph   I sc  K I Tc Tref  S
 kTc A  

(1)
where Irs is the cell reverse saturation current (of p-n junction), Tc is the cell’s working temperature, A is
the ideality factor, S is irradiation, ISC is the cell’s short circuit current at reference temperature Tref and KI
is the short circuit current temperature coefficient of the cell. The net output PV current, power with Np
number of parallel cells, Ns number of series cells, and Vdc being the output voltage are given by (2).

  q Vdc   
I pv  N p I ph  N p I rs  exp   1 
  kTc A N s    (2)
Ppv  Vdc *I pv  f (Vdc , S ,Tc ) 

 IV and PV characteristic
• Open-circuit voltage (Voc):
• Terminal voltage when the terminals of the module
are disconnected.
• Short circuit current (Isc)
• Maximum Power PMPP where the voltage versus current
product is maximum
• VMP≈ 0.8 * Voc
• IMP≈ 0.9 * ISC
• The best conditions, are the "standard test
conditions " happen at Irradiance equal to 1000W/m²,
cells temperature equals to 25°C, and spectral
distribution (Air Mass) AM is equal to 1.5
 Maximum Power Point Tracking
• Basic methods include:
• Fractional Open Circuit Voltage
Vmpp ≈ K * Voc
• Fractional Short Circuit Current
Impp ≈ K * Isc
• Look-up Table Technique
• Perturb and Observe
• Incremental Conductance
• Fuzzy Logic
• Neural network based
P&O MPPT algorithm and working principle
• Here dc voltage reference (Vdcref) is perturbed with a fixed amount (∆V) and the next perturb is
decided based on the status of change in power (∆P).
• For correct execution of the P&O MPPT, fMPPT should not be greater than the inverse of 20ms
(settling time).
• The sensing of Ipv and Vpv done at higher frequency than the frequency of updating (the Vdcref),
fMPPT may result in erroneous results since the decision of perturb is taken based on an unsettled
outputs as it may destabilize the system.
702
Start
701

in V
Sensing/Sampling of

dc
Vdc(n) and Ipv (n) 700

V
Vdc Ref
699
Vdc
P ( n)  V ( n) * I ( n)
pv dc pv 1.95 1.96 1.97 1.98 1.99 2 2.01 2.02 2.03 2.04 2.05
Time in second
dP ( n )  P ( n )  P ( n  1) (b)
pv pv pv
dV ( n )  V ( n )  V ( n  1)
dc dc dc
VMPP2

NO YES
dPpv ( n )
0
dVdc ( n )
k = -1 k=1

VMPP1
Vdcref ( n 1) Vdc ( n )  k V 1 2 N Samples
Time
(a) (c)
 Three step operation in P&O
PB>PC
PB<PC
40 40
B MPP
C B MPP
C
30
VA →VB →VC →VB →VA A
VB →VC →VD →VC →VB D
30
in kW

in kW
20 20
pv

pv
P

P
10 VA VC 10 VB
VD

0 0
0 100 200 300 400 500 600 700 800 900 0 100 200 300 400 500 600 700 800 900
Voltage in V VB VMPP VC
Voltage in V VMPP
Fig. 1 Fig. 2
From fig 1 the dP/dV is greater than zero and results in positive perturb (since PA<PB) when the operation is shifted
either from A to B or from B to A. Similarly, when it is from B to C or from C to B the dP/dV is less than zero and hence
the P&O algorithm initiates a negative perturb (since PB>PC). Hence, it can be concluded that a well executed P&O
algorithm contributes continuously shifting of dc voltage, involving three levels. Similarly for the fig 2 can be
explained.
Incremental conductance
• At the MPPT point of the P~V curve
dp dv di di i
0i v  0 
dv dv dv dv v
Comparison of MPPT algorithms
Impact of parallel leakage resistance on IV characteristics
• In ideal PV cell
• 𝐼𝑃𝑉 = 𝐼𝑠𝑐 − 𝐼𝑑𝑖𝑜𝑑𝑒 ; where 𝐼𝑠𝑐 is the short
circuit current and assume 𝑅𝑠 = 0 and 𝑅𝑃 =
∞
• When parallel leakage resistance is
considered
𝑉𝑑𝑐
• 𝐼𝑃𝑉 = 𝐼𝑠𝑐 − 𝐼𝑑𝑖𝑜𝑑𝑒 − ; keeping 𝑅𝑆 = 0.
𝑅𝑃
• This shows that at any given voltage load
𝑉
current will be decreased by ( 𝑑𝑐 ) from the
𝑅𝑃
load current obtained with ideal PV cell
𝑉
• Due to this reduction of current (equal to 𝑑𝑐 ),
𝑅𝑃
1
the IV characteristics shows a slope of ; as
𝑅𝑃
shown in figure.
• For a cell to have less than 1% loss due to
100𝑉𝑜𝑐
the parallel resistance, 𝑅𝑃 >
𝐼𝑠𝑐
 Impact of series resistance on IV characteristics
• Series resistance includes
• Contact resistance associated with the bond between the cell and its wire leads
• Resistance of the semiconductor
• In ideal PV cell
• 𝐼𝑃𝑉 = 𝐼𝑠𝑐 − 𝐼𝑑𝑖𝑜𝑑𝑒 ; Assume 𝑅𝑠 = 0 and 𝑅𝑃 = ∞
𝑞𝑉
• Diode current is given by 𝐼𝑑𝑖𝑜𝑑𝑒 = 𝐼𝑟𝑠 (exp − 1)
𝑘𝑇𝑐 𝐴

• When series resistance is considered keeping 𝑅𝑃 = ∞

• There is a voltage drop in the series resistance
• So the voltage across the diode is 𝑉𝑑𝑖𝑜𝑑𝑒 = 𝑉 + 𝐼𝑃𝑉 𝑅𝑠 and the diode current
𝑞(𝑉+𝐼𝑃𝑉 𝑅𝑠 )
becomes 𝐼𝑑𝑖𝑜𝑑𝑒 = 𝐼𝑟𝑠 (exp − 1)
𝑘𝑇𝑐 𝐴
• The PV current equation becomes 𝐼𝑃𝑉 = 𝐼𝑠𝑐 − 𝐼𝑑𝑖𝑜𝑑𝑒 = 𝐼𝑠𝑐 −
𝑞(𝑉+𝐼𝑃𝑉 𝑅𝑠 )
𝐼𝑟𝑠 (exp − 1)
𝑘𝑇𝑐 𝐴
 Impact of series resistance on IV characteristics (cont.)
• Thus at any given current the voltage gets shifted by ∆𝑉 = 𝐼𝑃𝑉 𝑅𝑠
0.01𝑉𝑜𝑐
• For a cell to have less than 1% loss due to the series resistance, 𝑅𝑆 <
𝐼𝑠𝑐
 IV characteristics considering impact of both series and parallel
resistance
• When both the series and the parallel resistances are considered both
voltage and current reduction is observed and the overall characteristics is
shown in the figure
 Impact of Shading

• Few or all the cells may be completely or

partially under shading because of cloud
movement, shadow of trees etc.

• This can cause problems like drop in output

power, heating of cell etc.
 Impact of Shading
Scenario of one of the series connected cells getting completely shaded

• Let n be the number of cells connected

in series
• Under normal operation, i.e. when all
the cells are in the sun, all cells
produce same voltage and short circuit
current
• When any 1 of the cell is completely
shaded, the Isc of that particular cell
drops to 0. The current which is
flowing through the remaining (n-1)
cells flows through the parallel
resistance (𝑅𝑃 ) of the shaded cell
 Impact of Shading
Scenario of one of the series connected cells getting completely shaded (cont.)

• There is a voltage drop across the resistance

• The diode is reverse biased i.e. 𝐼𝑑𝑖𝑜𝑑𝑒 = 0
• The whole current is flowing through 𝑅𝑃 and 𝑅𝑆 of the shaded cell
• Thus the shaded cell withdraws power instead of supplying power
• Due to the power loss in the resistances, the temperature of the shaded cell
increases
• This is called hot spot formation in a PV string
 Impact of Shading
Change in IV characteristics
Scenario of one of the series connected cells getting completely shaded

• When all the cells are under sun

• Let the DC output voltage is 𝑉
• As all cells are producing equal voltage, the output voltage is shared equally by all cells
𝑛−1
• So the voltage developed by the (n-1) cells is 𝑉𝑛−1 = 𝑉
𝑛

• When one of the cells is under shading

• The DC output voltage will be the voltage drop across resistances of the shaded cell
subtracted from voltage developed by the (n-1) cells i.e. 𝑉𝑆𝐻 = 𝑉𝑛−1 − 𝐼 𝑅𝑃 + 𝑅𝑆
𝑛−1 𝑛−1
• On substituting 𝑉𝑛−1 = 𝑉 we obtain 𝑉𝑆𝐻 = 𝑉 − 𝐼 𝑅𝑃 + 𝑅𝑆
𝑛 𝑛
 Impact of Shading
Change in IV characteristics
Scenario of one of the series connected cells getting completely shaded (cont.)

• During shading there is a reduction in

output voltage obtained by
𝑛−1
• ∆𝑉 = 𝑉 − 𝑉𝑆𝐻 = 𝑉 − 𝑉+
𝑛
𝑉
𝐼 𝑅𝑃 + 𝑅𝑆 = + 𝐼 𝑅𝑃 + 𝑅𝑆
𝑛
• Since 𝑅𝑃 ≫ 𝑅𝑆 ; 𝑅𝑆 in the above equation
can be neglected
• So the voltage reduction can be written
𝑉
as ∆𝑉 = + 𝐼𝑅𝑃
𝑛
• The modified IV curve due to this voltage
drop is shown in the figure
 Impact of Shading
Scenario of one of the series connected cells getting partially shaded

• Let n be the number of cells connected in series and one of the cells is
partially shaded
• The characteristics will depend on the short circuit current of the partially
shaded cell
• For example if one of the n cells in 50% shaded, the short circuit current of
𝐼𝑆𝐶
that cell reduces to
2
• The IV characteristics is obtained by considering two operating
circumstances
𝐼𝑆𝐶
• First, when current supplied by remaining n-1 cells is less than
2
𝐼𝑆𝐶
• Second, when current supplied by remaining n-1 cells is greater than
2
 Impact of Shading
Scenario of one of the series connected cells getting partially shaded (cont.)
𝐼𝑆𝐶
1. When I < (Figure (a))
2
• Diode of the 1st cell is forward biased
𝐼 𝑉
• Current 𝐼𝑑 = ( 𝑆𝐶 − 𝐼 − 𝑑 ) will flow through
2 𝑅𝑃
the diode
• The partially shaded cell contributes positive
voltage
𝐼𝑆𝐶
2. When I > (Figure (b))
2
𝐼
• A current equal to (I − 𝑆𝐶 ) will flow through
2
the parallel resistance
• There is a voltage drop across the resistance.
• The diode becomes reverse biased and carries
no current
• Similar to a complete shading scenario, there
𝑉
is a resultant voltage drop given by ∆𝑉 = +
𝐼𝑆𝐶 𝑛
(𝐼 − )𝑅𝑃
2
 Impact of Shading
Change in IV characteristics for various partial shading scenarios

• Based on the analysis shown above different IV curves for different

scenarios is shown in the figure
 Shade mitigation
• The problems arising due to shading can be taken care by the use of
• Bypass diode
• Blocking diode
 Shade mitigation using bypass diode
• The problem of voltage drop and hot spot formation is solved using
a bypass diode
• A bypass diode is connected in antiparallel to the diode of the cell
• As shown in (a)
• The cell is under sun and gives voltage rise
• The bypass diode is reverse biased and no current flows through it
• This is operation of a normal cell where bypass diode is not even present
• As shown in (b)
• The cell is under shading and does not produce any current
• The current has to flow through the parallel resistance leading to voltage
drop across the cell
• This drop makes the bypass diode forward biased and all the current flows
through the diode and no current flows through the cell (resistance)
• This leads to a small voltage drop of the range 0.2 V to 0.6 V depending on
the type of diode used rather than the large voltage drop (∆𝑉 = 𝑉 𝑛 + 𝐼𝑅𝑃 )
that may occur without it.
 Shade mitigation using bypass diode (cont.)
• Generally bypass diode is not provided across each cell in a module
• Few diodes are used such that each covers a number of cells within the module (Figure (a))
• With the use of these bypass diodes more power is obtained as compared to a module without bypass
diode (Figure (b))
• Similar to a bypass diode covering cells in a module; bypass diodes are used across modules in a string in
an array (Figure (c))

(a) Three bypass diodes, each covering (b) IV characteristics and maximum power (c) Use of bypass diode in a string of modules
one – third of the cells in a module of a module with 3 bypass diodes and one
cell shaded
 Shade mitigation using blocking diode
• When strings are connected in parallel
• Instead of supplying current, the shaded string can withdraw current from rest of the parallel
connected strings (Figure (a))
• This problem is solved by using blocking diode; also known as isolation diode
• The blocking diode is placed at top of each string
• The diode blocks the reverse current withdrawn by the shaded string (Figure (b)).

(a) Without blocking diode (b) With blocking diode

 Multiple MPPs for shaded cell with bypass diode

• Due to the complex IV characteristics depending on whether the bypass

diode is conducting or not, there may be multiple local maximum power
points
 Assignment

• Make electrical network in MATLAB / SIMULINK to obtain:

a) IV and PV curves for a PV module at various insolations
b) IV and PV curves for a PV module under various shading scenarios as shown in the
figure

c) IV and PV curves for a PV module under various shading scenarios as in (b) and
each cell having bypass diode connected across it.
Grid connected PV module
• Controlling action
• Active and reactive power control
• DC link voltage control
• MPPT tracking for variable insolation

• Controller configuration
• Single stage inverter
• Inverter handles MPPT, P&Q control
• Double stage inverter
• Boost converter handles MPPT, Inverter handles P&Q control
Decoupled controller philosophy
• The active and reactive power controlled by a Decoupled controller
• For a three phase system total power in dq frame
3 3
𝑆 = 𝑉𝑑 + 𝑖𝑉𝑞 𝐼𝑑 − 𝑖𝐼𝑞 = 𝑉𝑑 𝐼𝑑 + 𝑉𝑞 𝐼𝑞 + 𝑖 𝑉𝑞 𝐼𝑑 − 𝑉𝑑 𝐼𝑞
2 2
• When it is a balanced system, 𝑉𝑞 = 0;
S  P  iQ 
3
2
 (Vd I d  iVd I q ) 

• Hence by controlling Id and Iq the P and Q can be controlled so the

name decoupled control.

• With park’s transformation on the three phase Ipcc and Vpcc its
corresponding Id+iIq and Vd+iVq is found, which needs the information of
the angle and frequency of the supply voltage. This can be achieved
using a PLL(phase locked loop) circuit.
Park’s Transformation

2𝜋 2𝜋
cos 𝜃 cos 𝜃 − cos 𝜃 +
3 3
2 2𝜋 2𝜋
− sin 𝜃 − sin 𝜃 − − sin 𝜃 +
3 3 3
1 1 1
2 2 2
• In this system, the dc-link voltage controller and
reactive-power controller determine reference
components Idref and Iqref respectively.

• The DC link voltage controller counteracts the

voltage variation by specifying an adequate value
of the d-axis inverter current to balance the
power flow of the dc link.

• DC link voltage reference Vref is found from the

mppt controller as discussed previously for single
stage PV.

• The reactive power reference value (Qref) is set to

zero in order to model a unity power factor DG
operation
• From the figure

• Applying parks transformation

 Inner control loop design using pole placement technique
• From the equations obtained above
𝑑𝑖𝑑 𝑑𝑖𝑞
• 𝐿 = −𝑅𝑖𝑑 + 𝑢𝑑 and 𝐿 = −𝑅𝑖𝑞 + 𝑢𝑞
𝑑𝑡 𝑑𝑡
• A transfer function can be obtained using s-transform
𝑖𝑑 1
• = ; similar equations can be written for 𝑖𝑞 and 𝑢𝑞
𝑢𝑑 𝐿𝑠+𝑅
• 𝑢𝑑 is obtained by the inner loop PI controller
𝑢𝑑 𝐾𝐼
• = 𝐾𝑃 +
𝑖𝑑𝑟𝑒𝑓 −𝑖𝑑 𝑠
• The resultant transfer function can be drawn as
 Inner control loop design using pole placement technique (cont.)
• From the equations obtained above
𝐾𝐼 1
• The resultant open loop transfer function is 𝐾𝑃 + ×
𝑠 𝐿𝑠+𝑅
• The transfer function can be simplified as
𝐾𝐼 1 𝐾𝑃 𝐾𝐼 1 1 𝐾𝑃 𝐾𝐼 1
• 𝐾𝑃 + × = × 𝑠+ × 𝑅 × = × 𝑠 + × 𝑅
𝑠 𝐿𝑠+𝑅 𝑠 𝐾𝑃 𝑠+ 𝐿 𝐿 𝐿𝑠 𝐾𝑃 𝑠+ 𝐿
• Take a time constant 𝜏 of around 2-4 ms
𝐿 𝑅
• Set the PI controller gains to 𝐾𝑃 = and 𝐾𝐼 =
𝜏 𝜏
• The resultant open loop transfer function is
𝐿 𝑅 1 𝐿 𝑅 1 1
𝜏 𝜏
× 𝑠+ × = × 𝑠+ × =
𝐿𝑠 𝐿 𝑅 𝜏𝑠 𝐿 𝑅 𝜏𝑠
𝜏 𝑠+ 𝑠+
𝐿 𝐿
1
• The resultant closed loop transfer is
𝜏𝑠+1
 Outer control loop design
• From the equations obtained above
• The outer control loop provides the reference current to the inner loop
• Current reference is usually obtained by applying a limiter to controller output to
limit the current
• Outer control loop time constant is set higher than the inner control loop
• When the inner loop settles 𝑖𝑑𝑟𝑒𝑓 = 𝑖𝑑 ; the inner loop can be just treated as a gain
(= 1)

• The resultant open loop transfer function can be drawn as

 Outer control loop design (cont.)
• The DC voltage, vdc, shown in figure (a) is
calculated by the following equations
• 𝑖𝐶 = 𝑖𝑃𝑉 − 𝑖𝑑𝑐 (1)
• The capacitor current is given by
𝑑𝑉𝑑𝑐
• 𝑖𝐶 = 𝐶 (2)
𝑑𝑡
• From (1) and (2)
𝑑𝑉𝑑𝑐 (a) PV current and voltages
• 𝐶 = 𝑖𝑃𝑉 − 𝑖𝑑𝑐 (3)
𝑑𝑡
• The transfer function to obtain 𝑉𝐷𝐶 is
𝐼𝑃𝑉 𝑠 −𝐼𝑑𝑐 𝑠
• 𝑉𝑑𝑐 𝑠 = (4)
𝐶𝑠
• The DC current 𝑖𝑑𝑐 is related to direct axis
current 𝑖𝑑
• Power, 𝑃 = 𝑖𝑑𝑐 × 𝑉𝑑𝑐 = 𝑉𝑃𝐶𝐶 × 𝑖𝑑
𝑉
• So 𝑖𝑑𝑐 = 𝑃𝐶𝐶 × 𝑖𝑑
𝑉𝑑𝑐
• Assuming constant 𝑉𝑃𝐶𝐶 and 𝑉𝑑𝑐 , 𝑖𝑑𝑐 and 𝑖𝑑 can be
related by a gain; 𝑖𝑑𝑐 = 𝐾 × 𝑖𝑑
• The DC current obtained can be used to obtain
𝑉𝑑𝑐 as per the figure (b) (b) Transfer function to obtain DC voltage
• The overall loop is a second order system
 Decoupled control details:
Decoupled control involving DC voltage
From the PV grid connected model

Modulation index to be fed to inverter

&
 Sliding Mode Control
• Consider the grid connected PV system
• Current equation in dq reference frame is written as
𝑑𝑖 𝑉
𝐿𝑓 𝑑𝑡 = −𝑅𝑓 𝑖𝑑𝑡 + 𝜔𝐿𝑓 𝑖𝑞𝑡 + 𝑣𝑑𝑡 − 𝑣𝑑𝑝𝑐𝑐 ; 𝑣𝑑𝑡 = 𝑚𝑑 𝑑𝑐
𝑑𝑡 2
𝑑𝑖𝑞𝑡 𝑉𝑑𝑐
𝐿𝑓 = −𝑅𝑓 𝑖𝑞𝑡 − 𝜔𝐿𝑓 𝑖𝑑𝑡 + 𝑣𝑞𝑡 − 𝑣𝑞𝑝𝑐𝑐 ; 𝑣𝑞𝑡 = 𝑚𝑞
𝑑𝑡 2
𝑑𝑉𝑑𝑐 1
• By KCL at DC link capacitor = 𝑖𝑃𝑉 − 𝑖𝑑𝑐
𝑑𝑡 𝐶
• Writing these 3 equations in matrix form for reference values
𝑑𝑖𝑑𝑡 𝑅𝑓 𝑖𝑑𝑡 𝑣𝑑𝑝𝑐𝑐 𝑉𝑑𝑐
− + 𝜔𝑖𝑞𝑡 − 0 0
𝑑𝑡 𝐿𝑓 𝐿𝑓 2𝐿𝑓
𝑚𝑑
𝑑𝑖𝑞𝑡 𝑅 𝑖 𝑣 𝑉𝑑𝑐 𝑚𝑞
= − 𝑓 𝑞𝑡 − 𝜔𝑖𝑑𝑡 − 𝑞𝑝𝑐𝑐 + 0 0
𝑑𝑡 𝐿𝑓 𝐿𝑓 2𝐿𝑓 𝑖𝑑𝑐𝑟𝑒𝑓
𝑑𝑉𝑑𝑐 𝑖𝑃𝑉 1
𝑑𝑡 0 0 −
𝑑𝑥
𝐶 𝐶
• or = 𝑓 𝑥 + 𝑔 𝑥 𝑢
𝑑𝑡
 Sliding Mode Control (cont.)
𝑑𝑥
• The control input matrix 𝑢 is obtained by 𝑢 = 𝑔 𝑥 −1 − 𝑔 𝑥 −1 𝑓 𝑥
𝑑𝑡
𝑖𝑑𝑡
• The output matrix considered is 𝑦 = 𝑖𝑞𝑡
𝑉𝑑𝑐
• In case of SMC we design a sliding surface so that the control input forces the
instantaneous states to the sliding surface which is achieved by adding a switching
control to the equivalent control i.e. 𝑢 𝑡 = 𝑢𝑒𝑞 𝑡 + 𝑢𝑠𝑤 𝑡 = 𝑢𝑒𝑞 𝑡 −
𝜌 𝑠𝑖𝑔𝑛𝑢𝑚(𝜎); where 𝑢𝑒𝑞 𝑡 = −𝑔 𝑥 −1 𝑓 𝑥 and 𝜌 is a positive constant.
𝜎1 𝑖𝑑𝑡 − 𝑖𝑑𝑡𝑟𝑒𝑓
• The sliding surface followed by the output is 𝜎 = 𝜎2 = 𝑖𝑞𝑡 − 𝑖𝑞𝑡𝑟𝑒𝑓
𝜎3 𝑉𝑑𝑐 − 𝑉𝑑𝑐𝑟𝑒𝑓
• The stability of the controller can be analysed by Lyapunov’s method.
• Another function that is used to avoid hard switching of signum function is hyperbolic
tangent function i.e. 𝑢 𝑡 = 𝑢𝑒𝑞 𝑡 + 𝑢𝑠𝑤 𝑡 = 𝑢𝑒𝑞 𝑡 − 𝜌 𝑡𝑎𝑛ℎ(𝜎)
Results with signum and hyperbolic tangent function
0.8

d-axis current
0.6

0.4

1 2 3 4 5 6 7 8
Time in second

(a) Change in reference current

1.5

d-axis current
1

0.5

0
1 2 3 4 5 6 7 8
Time in second

(b) System response with signum switching function

1.5
d-axis current

0.5

0
1 2 3 4 5 6 7 8
Time in second
(c) System response with hyperbolic tangent switching function
 Sliding Mode Control (cont.)
• The reference signals are obtained in the following manner
• 𝑉𝑑𝑐𝑟𝑒𝑓 is obtained from MPPT algorithm
• 𝑖𝑑𝑐𝑟𝑒𝑓 is obtained from the corresponding differential equation
• The 𝑖𝑑𝑡𝑟𝑒𝑓 and 𝑖𝑞𝑡𝑟𝑒𝑓 current references are obtained as per the phasor diagram
• From the phasor diagram
𝑉𝑑𝑐 𝑖𝑑𝑐𝑟𝑒𝑓
𝑖𝑑𝑡𝑟𝑒𝑓 cos 𝜃 sin 𝜃 𝑣𝑝𝑐𝑐
=
𝑖𝑞𝑡𝑟𝑒𝑓 sin 𝜃 − cos 𝜃 𝑄𝑟𝑒𝑓
𝑣𝑝𝑐𝑐

• 𝑣𝑝𝑐𝑐 in above equation is given by 𝑣𝑝𝑐𝑐 = 𝑣𝑑𝑝𝑐𝑐 2 + 𝑣𝑞𝑝𝑐𝑐 2

−1 𝑣𝑞𝑝𝑐𝑐
and 𝜃 is given by 𝜃 = tan
𝑣𝑑𝑝𝑐𝑐
 Simulation – Grid Connected System with Battery
System Configuration

3 phase,
DC
480 V, 60
Hz DC
Circuit
Breaker Multiple PV
loads Array Battery
GATE and its
Grid Side Converter controller

• In grid connected mode, the grid acts as a DC link voltage controller.

• The grid voltage source converter maintains the power balance between AC and DC
side.
• Control of VSC is based on the dq-synchronous frame as control is easier in equivalent
two-phase model.
 Role of battery

• In grid connected mode, the battery always maintains the State-of-

Charge(SOC) with the incoming current from the main grid
• Generally the SOC is estimated using the Coulomb’s counting method.
There are various other methods to estimate SOC since SOC is
estimated using terminal voltage of the battery. However, there will
always be an error involved in the measured and actual SOC.
𝐼0
𝑆𝑂𝐶 = 𝑆𝑂𝐶𝑜 + ∆𝑡
𝑄𝑛
• In islanded mode, the battery maintains the DC link voltage
The governing equations are:

3
𝑃= 𝑣 𝑖 + 𝑣𝑞 𝑖𝑞
2 𝑑𝑑
𝑑𝑖𝑑
𝑅𝑔 𝑖𝑑 + 𝐿𝑔 − 𝜔𝐿𝑔 𝑖𝑞 + 𝑣𝑑 = 𝑣𝑑𝑔
𝑑𝑡
𝑑𝑖𝑞
𝑅𝑔 𝑖𝑞 + 𝐿𝑔 + 𝜔𝐿𝑔 𝑖𝑑 + 𝑣𝑞 = 𝑣𝑞𝑔
𝑑𝑡
𝑑𝑉𝑑𝑐
𝐼𝑑𝑐 =𝐶 + 𝐼𝐿 + 𝐼𝑖𝑛𝑣
𝑑𝑡

𝑉𝑑𝑐 𝐼𝑑𝑐 = 𝑃 + 𝑉𝑑𝑐 𝐼𝐿

Current and Voltage Controller of VSC

I dref Ki 1 1 Id
+ Kp  R g  L g .s
-
s 1  T a .s
PI Controller PWM Block System transfer
Function

Inner Current Control loop

Vdcref Kiv I d+ref 1 Id 3Vd Idc V dc

K pv 
1
+ +
- C.s  1 R
-
s 1   in s 2Vdc
Inner Current Controller
Loop

Outer Voltage Control loop

Results
Autonomous mode of operation

(b)
(a)

(c)
Nominal behaviour of battery in autonomous mode
(a) Solar Irradiation (b) DC-Link Voltage
(c) Battery SOC
Results
Grid connected mode of operation: SOC > SOCref

(b) (c)
(a)

(d)
Power Balance at(e) higher SOC (f)

(a) Irradiation, (b) DC-Link Voltage, (c) PV Current,

(d) Battery Current, (e) Grid Current, (f) Battery SOC
Results
Grid connected mode of operation: SOC < SOCref

(a) (b)

(c) Power Balance at lower SOC (d)

(a) Irradiation, (b) DC-Link Voltage, (c) Battery SOC,
(d) Battery Current
 Assignment
• Make a grid connected PV system (having a DC/DC converter, battery
with bidirectional converter, inverter and filter).
Inverter mode of operation
• Grid connected mode
• Operates as CURRENT CONTROLLED SOURCE,
• Need to Control net P,Q injected into the grid,
• Terminal Voltage and frequency are fixed as maintained by the
grid.(except in few cases like radial distribution system)

• Off grid mode/isolated mode

• Operates as VOLTAGE CONTROLLED SOURCE,
• Need to control the P and Q through varying the Terminal voltage V and
Frequency.
Inverter Control Techniques
• Vector control/ Decoupled control/ DQ control,
• Nonlinear control,
• Fuzzy based/ Neural based control,
• Sinusoidal component / ICOSϕ based,
Single stage PV module Double stage PV module
Single phase Grid connected PV
• Challenges
• Oscillating active power at twice the supply frequency,
• Tuning of PI controller for sinusoidal signals,
PV with PR controller
• From power balance
Ppv=Pac=Vacrms*Iacrms => Iref_peak=Ppv*sqrt(2)/Vacrms
• A PR controller is , where Wo is the supply
frequency.
 Microinverters
• Instead of using multiple series and parallel connected PV modules (or
strings) followed by a single inverter, an alternative method is to have
microinverter for each PV string and the AC output of all the inverters is
connected at an AC bus
• Advantages
• Reduced down time – If any one of the inverter is malfunctioning the others may
continue to feed power
• Each module (or string) is independent of other strings and so strings of different
ratings can operate together; while in parallel connected modules, each string should
have similar rating
• Less chances of electric shock
• Disadvantage
• Cost is higher as the number of inverters may be large
 Net Metering
• A customer having a PV system (e.g. a roof top PV system) may supply power
to the grid if the power needed by the local loads (e.g. power consumed by
appliances in the home) is less than the maximum power the PV can
generate.

• The customer may withdraw power from the grid if the local loads are more
than the PV power generation.

• Net metering is a concept in which the bill paid by a customer is based on

the net difference between the power and withdrawn from the grid and
power supplied to the grid.
 Assignment
• Write a report on Net metering applications and challenges.
 Derating of PV
• Derating of PV – Operation at reduced power i.e. power less than the MPP
• Required under load following mode or required to achieve desired control
performance
• A derated PV operates like a battery
• PV under derating can enhance the frequency regulation of an
interconnected AC system as shown in figure
Frequency regulation of an AC system with PV Frequency regulation of an AC system without
f(p.u.)

f(p.u.)
participating in frequency control PV participation