Energy Efficiency of Low Voltage Direct Current Supplies

Including PV Sources

Anis Ammous

, Ammar Assaidi

, Abdulrahman Alahdal

and Kaiçar Ammous

Department of Electrical Engineering, College of Engineering and Islamic Architecture, Umm Al-Qura University, K.S.A.

Department of Electrical Engineering, National School of Engineers of Sfax, University of Sfax, Tunisia

Keywords: Renewable Energy, Power Electronics, LVDC, Energy Efficiency.

Abstract: The low Voltage Direct Current (LVDC) system concept has been growing in the recent times due to its

characteristics and advantages like renewable energy source compatibility, more straightforward integration

with storage utilities through power electronic converters and distributed loads. This paper presents the

energy efficiency performances of a proposed LVDC supply concept and others classical PV chains

architectures. A PV source was considered in the studied nanogrids. The notion of Relative Saved Energy

(RSE) was introduced to compare the studied PV systems energy performances. The obtained results

revealed that the employment of the LVDC chain supply concept is very interesting and the use of DC loads

as an alternative to AC loads, when a PV power is generated locally, is more efficient. The installed PV

power source in the building should be well sized regarding to the consumed power in order to register a

high system RSE.

1 INTRODUCTION

The photovoltaic electricity was primarily

established for standalone applications deprived of

any connection to a power grid. Such was the case of

satellites or isolated habitations. Currently, PV's are

found in many power applications like personal

calculators, watches and other objects of daily use,

they can supply many individual DC loads without

difficulty. The main objective of this paper is to

investigate the energy efficiency performance of a

proposed Low-Voltage Direct Current (LVDC) PV

system regarding to a classical LVDC architecture

and classical PV systems using AC loads. All the

studied PV chains are on-grid ones and are supposed

supplying offices.

In general, electric energy consumption in office

applications and housing is achieved by using the

alternative current plugs even for Grid Tie PV panel

systems. In this case the use of AC's can increase

system losses especially when DC current is used at

the load levels.

LVDC systems have been gaining more interest

during the past few years both in academia and

industry. LVDC systems offer many advantages

covering higher energy efficiency and easier

integration of modern energy resources in

comparison to conventional AC systems.

Direct use of DC power would reduce many

power conversion losses by exploiting self-

consumption of the energy produced on site and

decreasing imports of electricity from the grid. DC

loads used in households and office buildings, also

operate on DC, heating/cooling systems and larger

equipment used in industry such as variable

frequency drives have also adopted DC motors.

Direct current power systems are essentially more

efficient than their AC counterparts; since in DC

systems do not suffer from skin effect or reactive

power (Leonardo et al., 2016; IEC, 2016; D. Kumar

et al., 2017; Gyuyoung Yoon et al., 2019).

A literature research has exposed the study of the

first system analysis explored the use of very low

voltage (<120 V) in small-size systems, particularly

residential dwellings (J Pellis, 1997).

Subsequently, Lasseter R.H proposed the

concept of the DC Microgrid as a low voltage

distribution network. This concept was projected as

the future low voltage distribution systems which

were facing revolutionary variations at the time due

to emanation of distributed generation and market

liberalization. The basic idea behind this concept is

to combine micro sources and loads into one entity

188

Ammous, A., Assaidi, A., Alahdal, A. and Ammous, K.

Energy Efﬁciency of Low Voltage Direct Current Supplies Including PV Sources.

DOI: 10.5220/0010471001880195

In Proceedings of the 10th International Conference on Smart Cities and Green ICT Systems (SMARTGREENS 2021), pages 188-195

ISBN: 978-989-758-512-8

which could be interpreted as a single dispatch-able

load that could respond in short time to meet the

transmission system needs (R. H. Lasseter, 2002).

For many years, The LVDC system has been

developed for specific applications like aerospace,

automotive and marine (A. Ghareeb et al., 2013),

(Ahmed T et al., 2015), (Jifei Du et al., 2019).

Literature review reveals that over the last decades,

LVDC systems are growing rapidly for industrial

applications, essentially in the telecommunication

industry, ships and buildings. Adopting Direct

Current in data centers improve efficiency, decrease

capital cost, increase reliability and boost power

quality (AlLee et al., 2012). In data centers, LVDC

architectures have been widely studied. Various

leader projects have been installed in Europe, the

United States, Oceania and Asia. From these

projects it was registered that the profit of DC in

data-centers are about 10% to 30% reduction in

energy consumption, about 15% lower capital costs,

simpler design, potential increase in reliability, less

physical area requirements, a smaller carbon

footprint and less cooling demand (Tomm Aldridge,

2009), (Brian Fortenbery, 2011).

The most significant challenge that DC

distribution systems face today is the lack of

standardization inducing varied architectures and

operations of DC distribution systems (Kiran and

Hassan, 2020), (Paul, Robert and Sean, 2010).

The work presented in (V. Vossos et al., 2014)

was accomplished in many various locations through

the country, with different types of system

topologies. Further, distribution topologies were

carried out for both two cases with and without

energy storage.

Studies conducted in (V. Vossos et al., 2014),

(Paul, Robert and Sean, 2010) aim to accomplish 25-

30% of energy savings. The environment

conditioning loads are very significant part in

Buildings and should be explored in further studies.

The authors in (Patterson, B. T, 2012) reported

that the majority of electricity used in on office

building passes through power converters enclosing

further conversions. Average conversion efficiency

is closed 68%. When using high quality electronics,

only a 10% loss from each stage of conversion is

considered as generous number.

The DC power is directly produced from

residential solar panels and inverter is commonly

added to supply AC loads.

Despite that the multi-stage conversion is basic

to extract power from the solar panel into the server;

losses resulting from these conversions are expected

to be between 10% and 25%.

Through review of the available literature

(Leonardo et al., 2016), (Gyuyoung Yoon et al.,

2019), (Kiran and Hassan, 2020), (Yu Zhang et al.,

2020), local DC grids are a promising option for

buildings to link natural DC power sources such as

photovoltaic power systems with DC loads like

lighting applications and data centers (Tomm

Aldridge, 2009), (Kiran and Hassan, 2020).

(A. AMMOUS and H. Morel, 2014) reported that

DC microgrids are alternatives promising to

conventional AC distribution networks especially for

the integration of renewable energies. They allow,

for example, to reduce consumption energy of 25%

when supplying buildings directly from sectors and

by photovoltaic panels.

The majority works on DC distribution grids

assume that converters are installed at each

household, which connect the local DC or AC

nanogrids (Patwa and Saxena, 2020) ,(X. Yue et

al.,2018). In case of distributed energy resources,

nanogrids in buildings could be functioned

separately from the main grid in islanding mode (X.

Yue et al.,2018) and typical low voltage subsystems

like 48V, 24V, or 12 V can be applied (Rodriguez-

Diaz et al., 2016). They could, for example, be used

for low power LED lighting or for connecting loads

by USB Type-C connector and USB Power

Delivery.

In the first part of the paper, we focused our

study on the state of the art related to the use of the

LVDC supply concept and the proposition of an on-

grid LVDC PV chain. The disadvantage of the use

of classical on-Grid PV systems and of using AC

plugs to supply electric DC loads are shown. The

used average model of power converters is then

presented in the second part. This model allows the

evaluation of the different converters efficiencies in

the studied PV chains. The last part of the paper

treats the energy efficiency performances of the

proposed LVDC system compared to others classical

ones. For this purpose, offices loads are considered

and Jeddah location (in KSA) was chosen in our

study.

2 THE PROPOSED LVDC PV

SYSTEM

Electronic appliances, such as computers, gaming

consoles, printers, economic LED lights, televisions

and so on need DC supplies. Additional AC to DC

converters are needed in such equipment.

Energy Efﬁciency of Low Voltage Direct Current Supplies Including PV Sources

189

Figure1 shows one of classical chain used to

connect PV panels to the low voltage grid with a

transformerless solution. The DC/DC converter,

Converter #1, allows the extraction of the maximum

power from the PV panels when climatic conditions

changes. The Maximum Power Point Tracking

(MPPT) algorithm is used to act on this DC/DC

converter control.

The DC/AC converter, Converter #2, transfers

the PV generated power to the grid and ensures the

regulation of the DC voltage value (400 V) of the

inverter input. This DC value is mainly used value

for single phase PV systems allowing to obtain

easily the standard 230 V AC. The injected current

to the grid has a quasi-sinusoidal waveform.

Figure 1: Classical configuration (syst1) of the On-Grid

PV system and DC load powered from AC sockets.

A main used LVDC photovoltaic architecture is

shown in figure 2. This system involve direct current

chain form PV to load. The regulated DC bus (400

V) was used for this purpose and a DC/DC converter

(η

) adapts LVDC loads supplies to this DC bus.

The efficiency of each PV conversion chain

depends on different considerations like the type of

used power semiconductor devices and the

magnitude of the transferred power.

Following we will present the proposed new PV

architecture for DC loads supplies and the developed

average model of power converters. The proposed

LVDC PV chain uses the DC bus available directly

after the PV panels. This bus is not regulated but it's

value varies in a given range depending on PV

panels associations (parallel/series) and open circuit

voltage across each panel.

The regulation of the DC voltages supplying

loads are regulated by DC/DC converters (in general

Buck ones). The proposed architecture is the one

shown in figure 3. The DC/DC converter connecting

the PV panels to the inverter become a reversible

DC/DC converter (Buck-Boost).

Figure 2: Possible LVDC classical architecture (syst2).

The last converter allows the transfer of power from

PV panels to the inverter when the generated power

is higher than the consumed power. This reversible

converter works as a Buck converter in the case

when the consumed power is higher than the PV

generated one. Both, system2 and system3 use the

same number of converters, the advantage of the

proposed architecture (syst3) is that the path of the

power from the PV source to the Load is shorter

during a day. during a luck of PV power, sure the

path of the power become higher from the grid to the

load. In order to judge the efficiency of the different

configurations a simulations was performed during a

typical days (24 hours/day) in each season and then

the saved energy will be evaluated during a

complete year (365 days). The effect, of the

generated energy magnitude by the load consumed

energy magnitude, on the systems performances was

studied too.

Figure 3: Proposed LVDC architecture (syst3).

3 POWER CONVERTERS

MODELS

Modeling is required to analize the dynamic

behavior of a power converter in several

applications. Since both accuracy and simulation

SMARTGREENS 2021 - 10th International Conference on Smart Cities and Green ICT Systems

190

rapidity are essential particularly for long time

simulations and for complicated circuits, the

averaging method is the widely used technique for

complex power electronics systems. Based on the

classical averaged model, the converter is considered

to be a linear system using ideal switches, however

the non-linear averaged model is established on

semiconductor device models including static and

dynamic characteristics of the switches. Figure 4(a)

shows the considered inverter leg with two active

switches (IGBTs or MOSFETs) directly controlled

by external control signals and two passive switches

(DIODEs). In Figure 4(b), the adopted leg circuit

based on the used averaged model is presented. In

this developed model, the leg switches are replaced

by a controlled voltage source V

in series with a

controlled current source I

given by (A. AMMOUS

et al., 2003

VU= 

and

1e2

Ii= 

(1)

With

U

and

i

are the time averaged values of

the instantaneous terminal waveforms of U

(t) and

(t) respectively over one cycle Ts (switching

period of the controlled switches).

Figure 5 shows the adopted switching waveforms

of the active switch (U

(t), i

(t)) and the passive

switch (U

(t), i

(t)) during the switching period T

The driving signals e

and e

are the control signals

of T

and T

respectively.

Figure 4: (a) The PWM-switch, (b) The corresponding

averaged model.

The power losses of semiconductors (P

switch

and

diode

) are estimated using this analytical

representation of the switching characteristics which

including both conduction and switching losses, and

considering the various conduction and switching

times.

Different static and dynamic power devices

parameters can be deduced from their data sheets or

by experiments.

(t)

don

IRM

doff

(t)

str

/dt

(t)

(dI

/dt)

off

-(V

-V

)

/dt

.(dI

/dt)

(t)

Figure 5: Adopted switching characteristics for the switch

and the diode components in the PWM-switch cell.

The developed average model allows computing

all the dissipated power in the semiconductor

devices and then deducing the different converters

efficiencies (A. AMMOUS et al., 2003). Time

domain simulations will be more rapid for the whole

PV chain system.

The efficiencies of the different converters used

in the different chains were evaluated by mean of

refined simulations. The evolutions of these

efficiencies as a function of the transferred power

ɳ(P) are shown in figure 6. The used, controlled,

devices in the converters are the N-channel SiC

power Mosfet's SCT3080KL. Internal anti-parallel

diode of the Mosfet is used for reverse current in the

switch.

We defined the saved energy, W

(i=1,2,3), by a

PV chain, the excess of energy injected in to the grid

after satisfying the load need. It is the energy

balance per month. We note that the chosen

maximum power generated by the PV panel is

1500W for a 25˚C Temperature and 1000W/m

Irradiance conditions.

4 STUDY OF LVDC SOLUTIONS

EFFICIENCIES

Jeddah-KSA location (21° 32′ 34″ N, 39° 10′ 22″ E)

was chosen to perform energy efficiency

performances of the different studied PV chains.

Energy Efﬁciency of Low Voltage Direct Current Supplies Including PV Sources

191

Figure 6: Different converters efficiency.

Jeddah features an arid climate under Koppen's

climate classification, with a tropical temperature

range.

The following graphs in figure 7 shows the solar

radiation during a typical day in each month for

Jeddah city, which was monitored from the station

of University King Abdul-Aziz. In the figure graphs

typical days divided by seasons. A higher value of

solar radiation in July and reaches nearly to more

than 984 W/m



. The generated power by the panels

during a typical day of January in Jeddah is shown

in figure 8. The PV panels generated energy in this

case is 6600 Wh/day.

The annual generated energy by the used PV

panels W

is equal to 3.006 MWh.

First, a load profile of an office was chosen in

order to make comparison of the saved energy by the

three studied PV chains. The profile of the load is

shown in the following figure 9 and the daily

consumed energy by one office is 2919 Wh.

The annual consumed energy by one office is

supposed to be about W

Load

= 1.068 MWh.

The considered office load is composed by,

Desktop computer (180 W), Laptop (50 W), Laser

printer (600W peak, 150 W average during a cycle),

two Led lamps (20W), small TV (20 W) and a fan

(25W). Since the DC/DC converter(η

) is common

for the three studied converters and located just

before the load, it's effect was not taken into account

in the study.

Matlab simulator tool was used to study the

different power PV systems efficiencies. The saved

energy by the three models during each month of the

year in Jeddah city is shown in figure 10. It is

evident that the saved energy decreases with the

increase of consumed power. From figure 10, it's

clear that the proposed new LVDC architecture

(syst3) is the best one and register a higher saved

energy (W

=1.467 MWh/year when only one office

load is considered) compared to the other chains

=1.357 MWh/year and W

=1.319 MWh/year

when only one office load is considered too).

Figure 7: Solar radiations of typical days for each month

and seasons.

From figure 10, it's clear that the proposed new

LVDC architecture (syst3) is the best one and

=1.467 MWh/year

when only one office load is considered) compared

to the other chains (W

=1.357 MWh/year and

=1.319 MWh/year when only one office load is

considered too).

SMARTGREENS 2021 - 10th International Conference on Smart Cities and Green ICT Systems

192

Figure 8: Generated power, by the used panels, during a

typical day of January in Jeddah.

This is due to the localization of the load connection,

in the LVDC chain, close to the PV panels which is

the main introduced modification in the proposed

LVDC system regarding to classical LVDC ones.

We note that when the saved energy W

is negative

this mean that the consumed energy is higher than

generated one and so, this energy is transferred

from the grid to the PV system.

This generated energy was calculated based on

Irradiance, ambient Temperature and wind speed in

Jeddah during each typical day/month in the year.

Figure 9: The assumed daily office load consumption (by

1 office).

We will describe the assess of the efficiency of the

proposed LVDC PV chain solution (syst3) compared

to the LVDC classic one (syst1) and the classical

On-grid PV chain (syst2) using AC loads. We varied

the consumed Energy by the load and we registered

the saved energy by each PV system. It was

remarked a very interesting propriety of the PV

LVDC systems related to the increase of their

efficiency compared to classical PV chain. In fact,

we define the Relative Saved Energy (RSE

%) of

the PV LVDC chain (syst j) to the classical PV chain

(syst1), the ratio of the excess of saved energy of the

given LVDC chain (W

or W

) regarding to the

classical PV system (W

) by the value of the annual

generated PV energy W

Figure 10: Annual saved energy by the three PV chains

models (syst1), (syst2), proposed LVDC (syst3)) for each

month of the year in Jeddah city and for one office.

The defined PV system RSE

regarding to the

classical PV chain is given by the following

equation.

RSE

(%) =(W

-W

)/W

x 100% with j=2,3 (2)

RSE gives an evaluation of the relative saved energy

if we use system2 or system3 instead of system1.

Figure 11: Relative Saved Energy for the two systems, 2

and 3 in each month (number of offices =1).

If the defined efficiency is negative this mean that

classical On-Grid PV chain (syst1) using AC sockets

for load supplies is better (in term of energy saving)

than the PV LVDC supplies concept.

Figure11 shows the relative saved energy for the

two systems 2 and 3 in January (office=1). The

figure 11 illustrates that the relative saved energy for

the systems3 is more important than the system 2.

The waveforms giving the evolution of the

LVDC chains RSE as function of load energy by PV

energy rate (









) is shown in figure 12. Two main

Energy Efﬁciency of Low Voltage Direct Current Supplies Including PV Sources

193

observations can be highlighted when interpreting

these waveforms:

First, it is clear that the RSE of the proposed

LVDC chain (syst3) is higher than the classical

LVDC one (syst2) till a high load consumed energy

(2.5 times the PV generated energy in our case).

Second, an optimum of these waveforms are

registered when the consumed energy is around the

generated energy by the PV panels. The maximum

yearly RSE of the new LVDC architecture (syst3) is

about 12% while the one registered by the classical

LVDC chain (syst2) is about 2.3%.

From figure 12, we can remark that when the

load power consumption increases, the RES

becomes negative, this mean that the classical PV

chain, using AC supply to feed loads, is better

solution than the LVDC one in this case. We can

remark also, that for a high load consumption, the

LVDC chain (syst2) becomes more interesting than

the proposed LVDC chain (syst3).

Figure 12: Yearly Relative Saved Energy of classical

LVDC system2 and proposed LVDC system3.

In addition, the use of system3 allows the increase of

the load energy consumption range (more than two

times the generated PV energy) where the LVDC

supply concept efficiency is higher than other

systems.

All the obtained results show that the use of the

LVDC chain supply concept is very interesting and

the use of DC loads instead of AC loads when a PV

power is generated locally increases the PV system

efficiency.

The installed PV power in the building should be

well sized regarding to the consumed power in order

to register a high system RSE. In this case an LVDC

system RSE can be higher than 10% compared to the

classical AC supply in PV systems.

5 CONCLUSION

In this paper we proposed a new architecture of a

Low Voltage Direct Current (LVDC) supply

concept. The proposed on grid PV chain system,

involving DC loads, can replace the classical on-

Grid PV systems using AC plugs to supply electric

AC loads. To evaluate the efficiency of some

different PV chains, non-ideal averaged models of

the different converters, have been used. These

models are accurate and suitable to complex systems

study. The Energy efficiency of the different PV

chains were estimated by mean of simulations. The

evaluation of the efficiency of the proposed new

LVDC architecture compared to the classical one

was performed in the case of building offices in

Jeddah. The superiority of the proposed LVDC PV

chain was shown, it depends on the consumed load

energy to generated PV Energy ratio.

ACKNOWLEDGEMENTS

This paper contains the results and funding of a

research projects funded by King Abdulaziz City for

Science and Technology (KACST) Grant no. 14-

ENE2677-10.

REFERENCES

A. AMMOUS and H. Morel, “LVDC: An Efficient

Energy Solution for on-Grid Photovoltaic

Applications” Smart Grid and Renewable Energy,

2014, Vol. 5, No. 4, pp. 63-76.

A. AMMOUS et al « Developing a PWM-switch model

including Semiconductor Device Non-linéarities »

IEEE Transactions on Power Electronics, Vol.18,

No.5 ,September 2003.

A. Ghareeb, A. Mohamed, and O. Mohammed, “DC

microgrids and distribution systems: An overview,” in

Power and Energy Society General Meeting (PES),

2013 IEEE, Jul 2013, pp. 1–5.

Ahmed T. Elsayeda, Ahmed A. Mohamedb, Osama A.

Mohammeda,” DC microgrids and distribution

systems: An overview”, Electric Power Systems

Research 119 (2015) 407–417.

AlLee, G., and Tschudi, W., “Edison redux: 380 vdc

brings reliability and efficiency to sustainable data

centers,” IEEE Power Energy Mag., Vol. 10, No. 6,

pp. 50–59, November2012.

Brian Fortenbery. “DC Power Standards”, Technical Report

March, Electric Power Research Institute, 2011.

D. Kumar et al,” DC Microgrid Technology: System

Architectures, AC Grid Interfaces, Grounding

SMARTGREENS 2021 - 10th International Conference on Smart Cities and Green ICT Systems

194

Schemes, Power Quality, Communication Networks,

Applications, and Standardizations Aspects,” IEEE

Access.,vol, 5,pp. 12230-12256, 2017.

Gyuyoung Yoon, Kyoko Sugiyama, Saya Yoshioka and

Shinji Sakai,” Energy Efficiency and Cost

Performance of Direct-Current Power Supply Systems

in Residential Buildings by 2030s and 2050s”, E3S

Web of Conferences 111 CLIMA 2019.

International Electrotechnical Commission (IEC) Report

on LVDC Electricity for the 21st Century, 2017.

J Pellis. The DC low-voltage house. Graduation project,

Eindhoven University of Technology,1997.

Jifei Du, Trillion Q. Zheng, Yian Yan , Hongyan Zhao,

Yangbin Zeng and Hong Li,” Insulation Monitoring

Method for DC Systems with Ground Capacitance in

Electric Vehicles », Appl. Sci. 2019, 9, 2607, pp.1-15.

Kiran Siraj, Hassan Abbas Khan,''DC distribution for

residential power networks—A framework to analyze

the impact of voltage levels on energy efficiency'',

Energy Reports, Volume 6, 2020, Pages 944-951.

Leonardo Trigueiro dos Santos, Manuela Sechilariu and

Fabrice Locment,” Optimized Load Shedding

Approach for Grid-Connected DC Microgrid Systems

under Realistic Constraints,” Buildings, pp1-15, 2016.

Patterson, B. T,” Dc, come home”. IEEE power & energy

magazine (2012, November/December), pp.60-69.

Patwa, K.., & Saxena, D. R. ''Review on Power System

Performance in High/Low Voltage Distribution System''

smart moves journal ijosthe, Vol.7, No.6, (2020).

Paul Savage, Robert R. Nordhaus, and Sean P. Jamieson.

DCMicrogrids: “Benefits and Barriers”.Technical

report, Yale School of Forestry and Environmental

Studies, 2010.

R. H. Lasseter, “Microgrids,” in Power Engineering

Society Winter Meeting, 2002. IEEE, 2002, vol. 1, pp.

305–308.

Rodriguez-Diaz, E., Chen, F., Vasquez, J. C., Guerrero, J.

M., Burgos, R., and Boroyevich, D., “Voltage-Level

Selection of Future Two-Level LVdc Distribution

Grids: A Compromise Between Grid Compatibiliy,

Safety, and Efficiency,” IEEE Electrif. Mag., Vol. 4,

No. 2, pp. 20–28, June 2016.

Tomm Aldridge.” Direct 400Vdc for Energy Efficient

Data Centers Direct 400Vdc Facility Vision”.

Technical Report April, Intel Corporate Technology

Group, 2009.

V. Vossos et al.” Energy savings from direct-DC in U.S.

residential buildings”, Energy and Buildings 68

(2014), pp.223–23.1

X. Yue, D. Boroyevich, F. C. Lee, F. Chen, R. Burgos

and F. Zhuo, "Beat frequency oscillation analysis for

power electronic converters in dc nanogrid based on

crossed frequency output Impedance matrix model," in

IEEE Transactions on Power Electronics, vol. 33, no.

4, pp. 3052–3064, Apr. 2018..

Yu Zhang, Shuhao Wei, Jin Wang and Lieping Zhang,”

Bus Voltage Stabilization Control of Photovoltaic DC

MicrogridBased on Fuzzy-PI Dual-Mode Controller”,

Journal of Electrical and Computer Engineering,

Volume 2020, Article ID 2683052, 10 pages.

Energy Efﬁciency of Low Voltage Direct Current Supplies Including PV Sources

195