1

Control of DC-DC Converter and Inverter for Stand-alone Solar Photovoltaic Power

Plant Ramesh P

Senior Engineer CDAC (Power Electronics Group)

Trivandrum, India rameshp@cdac.in

Lekshmi K.R Senior Engineer

CDAC (Power Electronics Group) Trivandrum, India lekshmikr@cdac.in

Aby Joseph Scientist E

CDAC (Power Electronics Group) Trivandrum, India

abypj@cdac.in

Abstract – Solar energy is one of the most promising sources of energy for the future energy scenario. Being clean and renewable energy source, a lot of research is happening across the world in the solar energy sector. Solar photovoltaic is considered to be the best means of tapping solar energy. Power electronics is considered to be the key driving technology for harvesting energy from solar photovoltaic. In this paper, an efficient energy conversion scheme for a stand-alone solar photovoltaic power plant is presented with details on control schemes. Simulation studies and test results on laboratory proto unit are also presented.

Key words—DC-DC Boost converter, Inverter, Digital Signal Processor (DSP), Field Programmable Gate Array (FPGA), Predictive Control, Pulse Width modulation (PWM) strategies, Space Vector Pulse Width modulation (SVPWM), Power Conditioning Unit (PCU)

I. INTRODUCTION A stand alone photovoltaic power plant is usually used in

locations where there is no grid connectivity. Such power plants may be used in a smart grid environment considering the energy economics. The stand-alone solar photovoltaic power plant generally consists of an input DC-DC converter, output inverter, filter, transformer and battery bank for storage connected to the intermediate DC link. The variable DC output from the solar photovoltaic array has to be converted into a fixed DC and this is done by the input DC-DC converter. The DC-DC converter proposed in this paper is realized with the boost topology. Battery banks and battery management become subsystems of such power plants. The inverter is used to generate regulated AC supply at the output so that loads can be connected on the AC side. A transformer is provided at the output to match the voltage and provide isolation. Moreover the transformer acts as output impedance of the power plant which will limit the fault current. The transformer also attenuates DC injection from the inverter, if any. Figure 1 shows the single line diagram of the solar photovoltaic power plant.

The inverter considered in this paper is of 3phase full bridge configuration, whose control strategy has been implemented so as to provide stiff control of voltage and frequency at the output. A detailed analysis is carried out for different PWM strategies used in inverters and variable structure scheme is proposed for higher energy efficiency. Thermal analysis is carried out to study the thermal performance of various PWM schemes and results are presented. All the schemes mentioned above are analyzed and validated using simulation studies and actual experimental results are shown for proto PCU for specific test conditions. The simulation studies are carried out using PSIM software. The proto PCU is rated for 10kW and the control strategy is implemented using a digital controller hardware platform developed with DSP and FPGA.

II. INPUT DC-DC BOOST CONVERTER The input DC-DC converter in boost mode, as shown in

Figure 2 , interfaces the input solar PV sources to the inverter. Battery is connected at the output of the boost converter. The battery will act as energy storage element so that during low solar insolation or night time the battery will supplement the power to the load. In a micro-grid where many power plants are operating in parallel, the battery can provide the inertia to the generating system. Preliminary function of the boost converter is to regulate a constant DC voltage with minimum ripple so that the inverter can be controlled well.

Fig.2. Scheme for Input DC-DC Boost Converter

Solar PV Cin

Cout L1

L2 L3

S1

S2

S3

S4

S5

S6

Fig.1. Scheme of stand-alone photovoltaic system

Inverter

Battery Bank

Converter Solar PV Interface transformer

Output filter

2014 IEEE International Conference on Power Electronics, Drives and Energy Systems (PEDES)

978-1-4799-6373-7/14/$31.00 ©2014 IEEE

Downloaded from http://www.elearnica.ir

2

This paper analyses all the subsystems and brings out a practical design with controllers for the DC-DC converter and the output inverter. The DC-DC boost converter is controlled such that the output inverter is fed with more or less constant DC input voltage at the interconnecting bus. Since the battery is connected to the DC bus, battery management is also incorporated in the control scheme of the DC-DC converter. The DC-DC converter also tracks the Maximum Power Point (MPPT) using Perturb and observe method in which the photovoltaic array has to operate [1].

A. Converter control There are two control methods widely used for DC-DC

converter, current mode control and voltage mode control. Current mode control is widely employed to achieve high dynamic performance of the system. Current mode control for a DC-DC converter is a two loop system. The outer Voltage controller can be a simple PI controller as shown in Fig 3.Its function is to regulate the Solar PV voltage to reference value by generating current reference. An inner current loop will regulate the currents within prescribed limits. The current loop monitors the inductor current and compares it with its reference value. The reference value for the inductor current is generated by the voltage control loop [2]. This scheme will be suitable for paralleling even dissimilar converters.

Predictive method of current mode control will have the

faster step response [3]. Predictive control will generate all the duty cycle in advance based on the reference current and the sensed inductor current, input voltage and output voltage. This method will be simpler than other control methods of DC-DC converter. In the Predictive Digital Current Program Mode control (PDCPM), the inductor current is sampled only at their peak instants. This is achieved by sampling the inductor current of a particular converter immediately after its switch is turned off. The output voltage obtained from predictive control is shown in figure 3.

For PDCPM technique, the outer loop will be a voltage loop. The output of voltage loop is fed to the predictive controller loop. In the predictive controller loop, the inductor current at the beginning of the next switching cycle is determined by the inductor current at the beginning of present switching cycle, input voltage, output voltage, duty cycle at the present of switching cycle.

L

i sdcoc

1]Td[nVII ++∆+= (1)

From equation (1) duty ratio can be derived as

( )1][nV1][nV

11][nI[n]I1]T[nV

Lddc

pvLc

sdc1][n −

−−+−−

−=+

(2)  

From equation (2) the duty ratio can be found in advance. In short, the predictive mode of current control increases

the speed of the inner current loop and the dynamic performance of the overall system. B. Interleaved operation of boost converter

Interleaved operation of boost converter is obtained by phase shifting the carriers according to the number of parallel converters. Interleaved performance will reduce the ripple in current and voltage thereby reducing the filtering requirements. Carrier waveform of the interleaved operation is shown in figure.4. The effective peak current will get reduced by a factor of 3 thereby improving efficiency. Since the control action itself is getting corrected thrice, for the same

0 0.1 0.2 0.3 0.4 0.5Time (s)

0

100

200

300

400

500

Vdcout

Fig.5. Simulated Output voltage waveform using predictive control

0.4711 0.4712 0.4713 0.4714 0.4715Time (s)

0

0.2

0.4

0.6

0.8

1

Vcarrier1 Vcarrier2 Vcarrier3

Fig.4. Simulated Interleaved carrier waveforms

Fig.7. Converter Limb currents (9 kW Load) -Experiment

Fig.6. Converter Limb currents (9 kW Load) - Experiment

Fig.3. Closed loop control of DC-DC Converter

MPPT Voltage

Controller Current

Controller DC-DC

Converter + -

VPV IPV

Vref e Ic d1 d2 d3

3

Fig.9(a) Battery charging Current = 1.3 A

Fig.9(b) Battery charging Current = 2.5 A

Fig.9(c) Battery ripple current

Switching frequency the dynamic performance of paralleled converters will be better than single converter with the same rating.

Major benefits of carrier interleaving are increased equivalent switching frequency, reduced filtering requirement, high dynamic performances and improvements in reliability. The current sharing between interleaved converters is achieved using PDCPM algorithm by generating appropriate duty cycles even in transient load variation. This helps in paralleling of dissimilar converters with stiff current control. This requirement is crucial in PV power plants as the capacity addition may be a frequent task. The Simulated waveforms for interleaved carriers and the output voltage are shown in Fig 4, Fig 5 respectively. The experimental results for current sharing between individual limbs in DC-DC Converter are shown in Fig 6 &7.

III. BATTERY MANAGEMENT In this section, the battery management strategy along with

input DC-DC boost converter is discussed. Storage batteries are indispensable in all standalone solar PV power systems [10] as the solar PV sources are intermittent in nature. Lead acid battery has been considered for experimental tests as it is most commonly used storage battery for solar PV applications. Since the battery is directly connected to the DC bus, the state of the battery during charging and discharging cycles dictates the DC bus voltage. The selection of charging and discharging operation depends on the availability of solar insolation and load demand as depicted in the operating matrix in Table I. If the Photovoltaic power is excess in amount or load is disconnected, charging will take place as listed in cases I and II in Table I.

TABLE I. Operation matrix of Battery management

If the load demand is higher and lower PV generation or

non-availability of PV generation, discharging will take place as listed in cases III and IV in TABLE I.As the storage capacity of battery is limited, it is desired to ensure disconnecting the battery during over charging or over discharging conditions irrespective of the cases listed in Table I.

In this work, a suitable control strategy which is shown in Figure 8 is developed for two step charging systems where the battery is initially charged with constant current until the battery terminal voltage reaches nominal value. After that, a constant voltage is applied till the charging current tails off to 3 percent of its initial value. During constant current charging,

voltage control loop will be bypassed as shown in Figure 8 and it will come into action when the terminal voltage of the battery reaches nominal value. In case of excess generation of PV power and light load conditions, the difference between the PV reference current (Iref) and DC Equivalent of Load Current (IL) is compared so as to limit the charging current of the battery to be less than 10% of rated current. The battery discharging takes place through inverter to the connected loads so as to operate the system under UPS mode (i.e) when there is no PV generation or during night time.

The experimental results are shown in Figures 9(a)-(c) for constant current charging of battery operation. The converter limb current waveforms are same as discussed in previous section which explained independent operation of Input DC-DC boost converter.

Case Rules/Conditions Battery mode

I PPV > PLoad Charging II PLoad = 0 Charging III (PPV+ PBATT) > PLoad Discharging

IV PPv = 0 Discharging

Vdcref

MPPT VPV

IPV

Iref

(Iref-IL) < 10%

Limit Ic= 10% of Imax

Battery status

IDCref = Iref/3

PDCPM

Voltage Controller

+ -Iref

CV mode

CC mode

Yes

No

Fig.8. Battery charging algorithm Vdcfb

4

IV. OUTPUT INVERTER The solar inverter is a critical component in a solar energy

system. The output voltage need to be regulated at rated voltage and frequency when the load is within limits. The controller has to ensure that the voltage is regulated when DC bus voltage has oscillation within the limits. The control methodology should also ensure that the efficiency of operation is kept maximum. In order to achieve the maximum efficiency, the switching losses need to be minimized with proper selection of modulation strategy.

A. Inverter control Multi loop controller will be used for Inverter control as

shown in Figure 10. The outer voltage controller has a sinusoidal template as its reference. The voltage controller output will be the current to be maintained in the filter capacitor so as to regulate the voltage across it. The load current is added as a feed forward term to the capacitor current reference. This improves the dynamics of the voltage controller during load fluctuations. The control law equations of output inverter are mentioned in (3) and (4).

( ) ⎟⎠⎞

⎜⎝⎛ +−+=

sKK IIVV Ii

piinvα*invαLααinv

( ) pvLα*refαLα*αinv KVVII −+=

( ) ⎟⎠⎞

⎜⎝⎛ +−+=

sKK IIVV Ii

piinvβ*invβLββinv

( ) pvLβ*refβLβ*βinv KVVII −+=

Where, Kpi = L / τi ; Kpv = C / τv

B. Variable PWM strategies of three phase inverter Inverters will transform the regulated dc voltage into ac

form. Inverters are controlled by various PWM strategies i.e.

sine triangle PWM, space vector PWM, and the discontinuous PWM. Simulation Results are shown in Figures (11) – (14).

In case of sine triangle pulse width modulation, the control signal is a sinusoidal signal. The sinusoidal signal is compared with switching frequency carrier waveform in order to generate the switching signals. The frequency of the carrier establishes the inverter switching frequency and is generally kept constant along with its amplitude. The space vector PWM technique is used to approximate the continuously rotating reference voltage vector Vref using the eight stationary converter voltage vectors. For this purpose reference voltage vector Vref is sampled at a sampling frequency ‘fs’, which is much higher than the frequency of the reference signal. Reference vector during each sampling period can be assumed to be constant. It can be approximated by applying suitable converter voltage vectors for specified time durations within that sampling period. Discontinuous PWM (DPWM) methods have found application in high performance current controlled drives due to their low switching loss characteristics and low current ripple characteristics. In the over modulation region, DPWM is the only modulators that maintain high gain. Therefore DPWM is the most beneficial to high power PWM VSI drives. Modern modulation methods were separated into two groups namely, continuous PWM (CPWM) and discontinuous PWM (DPWM). The waveform quality and switching loss comparison indicated near zero modulation index regions CPWM methods are superior to DPWM methods.

(3)

(4)

Losses (A)

Power rating Fig.14. Loss comparison of different PWM

methods

Fig.11. Modulating signal of SPWM

Fig.12. Modulating signal of SVPWM

Fig.13. Modulating signal of DPWM

Vrefβ

IRinv IRload

RYB - β RYB - β

Iinv Iinvβ IL ILβ

RYB - β

VLβ

SVPWM

β- RYB

VL

VR VY VB

Vinv Vinvβ Vref

RYB - β Vrefβ

+ -Vref

VL

Kp + +

IL

- + Kp +Ki/s + +

Icap Iinv Vinv

+ -VLβ

Kp + +

ILβ

- + Kp +Ki/s + +

Icapβ Iinvβ Vinvβ

LOAD IYinv

IBinv

IYload

IBload

VRload

VYload

VBload

Fig.10. Closed loop control of output inverter

5

TABLE II. Comparison Of Losses For Different PWM (Simulated)

Power Rating(kW)

Psw SPWM

(A)

Ptotal SPWM

(A)

Psw SVPWM

(A)

Ptotal SVPWM

(A)

Psw DPWM

(A)

Ptotal DPWM

(A) 1 46.02 53.9 37.97 45 30.942 38 2 55.91 69.44 45.28 57.03 37.56 51 3 65.5 85.43 52.41 69.79 42.9 62.9 4 74.92 101.42 59.49 82.49 48.5 75.1 5 84.24 117.6 66.36 95.27 53,94 87 6 93.56 133.96 73.31 108.23 53.36 100 7 102.85 150 80.1 121.2 64.3 112 8 112.12 167.02 87.07 134.5 69.08 124

9 120.87 183 93.23 147 74.05 137 10 132.12 202.42 100.75 161.16 79.5 150

C. Loss analysis of different PWM with actual IGBT module For calculating losses, actual model of IGBT module

PM150CLA060 is used. Losses are in the form of electric current. Thermal analysis will give the transistor conduction losses, transistor witching losses, diode conduction losses, and diode switching loss. If the frequency of IGBT module is 50Hz, the losses results are the average value for an interval of 20ms. If the frequency is set to be same as the switching frequency, the losses in each switching cycle are obtained. Losses are highest in the case of sine triangle modulated three phase inverter. SVPWM have more losses than DPWM technique but less than that of SPWM technique. Loss comparison of different PWM methods is shown in Table II.

V. EXPERIMENTAL SETUP The experiments were conducted at NaMPET laboratory, CDAC-Trivandrum. The experiments were carried out using 10 kW Power converter module along with its sub components in the existing 25 kWp PCU shown in Fig 15(a). The specification of the proto unit is given in the Table III. 10 kW sub module has a structure as shown in Fig.15 (b). The proto unit consists of a back to back converter module having three DC-DC boost converters in parallel and a three phase inverter with intermediate DC link capacitor. The Embedded Digital controller hardware platform developed with TMS320F2812 DSP (Texas Instruments) and EP2C5 FPGA(Altera) is shown in Fig 16. The experimental results of output inverter under different conditions are shown in Figures 17(a) – (d).

TABLE III. Specifications of proto unit Sl

No Parameter Value

1. Open circuit voltage for the solar PV

array(VOC) 385 V

2. Switching device IPM

3. Maximum Power Point

Voltage(Vmpp) 290 V

4. DC bus voltage(output of the DC-DC

converter) 400V

5. Maximum allowable ripple at the

output < 2% of the

rated

6. Input current ripple <2.5 % of the rated

7. Input voltage ripple <1% 8. Switching frequency of the converter 10 kHz 9. Input inductor 5mH,50A 10. Output capacitor and Input capacitor 440µF 11. Lead acid battery (12V,135Ah)

Fig 17(c). Load Current IR,IY,IB

Fig.17(a). Output Voltages VRN(star) & VRY(delta)

Fig.17 (d). Inverter current IR and Battery current

Battery current

Fig.17(b). Output Voltages VRN VYN VBN (star)

Fig.15(b) Photograph of 10 kW Power

Converter module

Fig.15(a) Photograph of PCU

Fig.16. Photograph of Embedded controller for PCU

6

VI. CONCLUSION This paper presents simulation studies on DC-DC converter with carrier interleaving, Predictive Digital Current Program Mode (PDCPM) control for realizing the multi loop control strategy in DC-DC converter and various modulation strategies for inverters. All the above are studied in connection with realizing stand alone photovoltaic power plants. The modulation strategies of inverters are compared according to their losses. Experiments are carried out on a 10kW PCU and the results are shown for individual operation of the subsystems.

REFERENCES [1]C. Hua and C. Shen, “Study of Maximum Power Tracking

Techniques and Control of DC/DC Converters for Photovoltaic Power System,” 29th Annual IEEE PESC, IEEE Computer Soc. Press, New York, USA, 1998, pp. 86-93.

[2]Souvik Chattopadhyay and Somshubhra Das “A Digital Current-Mode Control Technique for DC–DC Converters” IEEE Trans on power electronics, vol. 21, no. 6, pp.1718 -1726 nov.2006

[3]Jingquan Chen, Member, IEEE, Aleksandar Prodic´, Student Member, IEEE, Robert W. Erickson, Fellow, IEEE, and Dragan Maksimovic´, Member, IEEE “ Predictive Digital Current Programmed Control” IEEE Trans on power electronics, vol. 18, no. 1, pp 411-418 january 2003

[4]H.Z.Azazi, E. E. EL-Kholy, S.A.Mahmoud and S.S.Shokralla “Digital Control of Boost PFC AC-DC Converters with Predictive Control” Proceedings of the 14th International Middle East Power Systems Conference (MEPCON’10), Cairo University, Egypt, pp 721-727 December 19-21, 2010.

[5]Yu Gu, Student Member, IEEE, and Donglai Zhang, Member, IEEE “Interleaved Boost Converter with Ripple Cancellation Network” IEEE Trans. on power electronics, vol. 28, no. 8, pp 3860-3869 august 2013

[6] Keliang Zhou and Danwei Wang, Member, IEEE “Relationship between Space-Vector Modulation and Three-Phase Carrier-Based PWM: A Comprehensive Analysis” IEEE Trans. on industrial electronics, vol. 49, no. 1,pp 186-196 february 2002

[7] Power electronics: Converters, Application and design by Ned Mohan, Undeland, Robbins

[8] Mitulkumar R. Dave1, K.C.Dave “Analysis of Boost Converter Using PI Control Algorithms” International Journal of Engineering Trends and Technology- Volume3 Issue2-pp 71-73 2012

[9] Michael Bierhoff, Henrik Brandenburg, Friedrich W. Fuchs “An Analysis on Switching Loss Optimized PWM Strategies for Three Phase PWM Voltage Source Converters” The 33rd Annual Conference of the IEEE Industrial Electronics Society (IECON) Nov. 5-8, pp.1512-1517 2007, Taipei, Taiwan

[10] Ola Al-Qasem, Jafar Jallad “Experimental Characterization of Lead – Acid Storage Batteries used in PV Power Systems” International Journal of Advanced Research in Electrical, Electronics and Instrumentation Engineering, Vol. 3, Issue 4, April 2014