Jorge-Luis Mírez-Tarrillo

Número 16 Vol. 2 (2016)

TECHNICAL-ECONOMIC ANALYSIS OF A AC/DC MICROGRID

FOR PUBLIC HEALTH INSTITUTIONS WITH LOW ELECTRICAL

DEMAND. CASE STUDY: PERÚ

In this paper, we analyze the implementation of a microgrid with a photovoltaic solar plant in health

facilities of the Level I and II according to the categorization of the Ministry of Health of Peru. The

study includes both technical (microgrid control and power management) and economic develop-

ments under a project investment horizon of 15 years. The mathematical modeling and numerical

simulations in Matlab/Simulink are used to demonstrate the feasibility of the project investment.

In addition, details of the PV system design and connectivity scheme with external grid are shown.

Also, its main characteristic is to allow only the entry of missing energy from external power supply

and not have storage systems. The mathematical model has taken into consideration 3-scenarios:

pessimistic, average and optimistic. This paper is a contribution to the implementation of microgrids

in the society and contribution to places that have not yet taken advantage of the solar resource for

electricity generation in health facilities.

KeyWords: Microgrid, Health building, Microgrid control, Power management.

Palabras Clave: Microred, Edicación de salud, Control de microred, Gestión de potencia.

En el presente artículo se analiza la implementación de una microred con una planta solar fotovoltai-

ca en establecimientos de salud del tipo I y II de acuerdo a la categorización del Ministerio de Salud

del Perú. El estudio evalúa tanto el aspecto técnico (control y gestión de potencia de microredes)

y económico bajo un horizonte de proyecto de inversión de 15 años. Modelamiento matemático y

simulación numérica en Matlab/Simulink son usadas para demostrar la factibilidad del proyecto. En

adición, detalles del diseño del sistema fotovoltaico y esquema de conectividad con la red externa

son mostrados. La principal característica de la microred es permitir sólo el ingreso de energía fal-

tante y no tener sistema de almacenamiento. En el modelo matemático tres scenarios han sido consi-

derados: pesimístico, promedio y optimístico. Este artículo es una contribución a la implementación

de microredes en la sociedad y en especial en los lugares en que no se ha aprovechado el sol para la

autogeneración de electricidad en establecimientos de salud.

esumen

bstract

Jorge-Luis Mírez-Tarrillo

Faculty of Sciences

National University of Engineering (UNI), Lima, Perú.

e-mail: jmirez@uni.edu.pe

INTRODUCTION

The Public Health Institutions (PHI) in Peru are classi-

ed according to their level of complexity in three levels:

I, II and III. The Level I and II are of low and medium

complexity and are located at different latitudes and al-

titudes covering the entire national territory in three

regions which are Coast, Highlands and Jungle respec-

tively. Each PHI needs electricity to operate lighting,

biomedical equipment and other loads. For this, the most

common solution is the connection to the external power

supply provided by a utility. However, for this it has to

pay a bill each month.

In Peru, there are places with high va-

lues of wind speed and solar radiation

(1) which can serve for the self-genera-

tion of electricity and/or heat and can be

protable.

In this regard, the Peruvian government

through the Ministry of Health (MOH)

has been implementing the Program

of Support to the Reform of the Heal-

Revista Cientíca

ISSN 1390-5740

ISSN 2477-9105

th Sector II (PARSALUD II), which is

the improvement of the 748 strategic

PHI through the development of Public

Investment Projects (PIP) with a time

horizon of 15 years (2). Each PHI has

roof areas potentially useful for the use

of radiant energy incident and to imple-

ment them with microgrids (MG) with

a good impact on the environment and

reduction of operational costs.

To analyze this case, a mathematical

model was constructed to estimate the

feasibility, return time and volumes of

production/consumption of electricity.

MG are small-scale, supply networks

designed to supply electrical and heat

loads for a small community. From a

grid point of view, the main advantage

of a MG is that it is treated as a contro-

lled entity within the power system. It

can be operated as a single aggregated

load. This ascertains its easy controllabi-

lity and compliance with grid rules and

regulations without hampering the relia-

bility and security of the power utility.

From a customers’ point of view, MGs

are benecial for locally meeting their

electrical/heat requirements. They can

supply uninterruptible power, improve

local reliability, reduce feeder losses

and provide local voltage support. From

a environmental point of view, MGs re-

duce environmental pollution and global

warming through utilization of low-car-

bon technology (3).

MG concept relates to a system which

coordinates locally the demand and su-

pply of energy. MG is essentially an

active distribution network because it

provides a platform for the integration

of various energy sources distributed

through a communications system that

allows control actions at distribution

voltage level (3). There are many pos-

sible congurations which may contain

generation of renewable and non-re-

newable electricity, storage and con-

trollable loads with priority categories

according to the user (4). Therefore, we

want to have: an MG with photovoltaic

(PV) plant; control, monitoring and supervision in auto-

nomous real time in PHI’s electric system and comply

with Peruvian legislation (5).

A MG is not an electrical conguration: (a) without a

load, (b) having only electrical charges, without micro-

sources, (c) without monitoring and control despite ha-

ving microsources, since their operation would not be

quantiable or optimized, (d) a conguration that has all

the elements, but insufcient carbon credits (6).

MATERIALS AND METHODS

About the PHI.

The Ofce of Investment Projects (IPO) of the Gene-

ral Ofce of Planning and Budget of the MOH by RD

No. 010-2012 / EF-68 approved the Minimum Contents

Specic 012 (MCS 012), a technical guidethat authori-

zes the use of renewable energy in PHI [2] but it does

not say how nor mentions modern trends as MG. MOH

by means of PARSALUD makes the evaluation of PHI

called “strategic” to improve their operational capacity

(7), This work is a contribution for the evaluation and

installation of MG in PHI of low electricity demand. The

PHI of Category I and II have usually a roof from 2,000

to 13,000 m

Solar Map of Peru.

Peru is politically constituted by 24 regions. A solar map

of Peru is available at (1). The largest PHI are located in

the capital of each region, therefore the solar radiation

in the capital of each regionis considered as a reference.

From (7) three values are assumed: Maximum (Rmax):

6.08 kWh/m

day; Average (Rave) 5.17 kWh/m

day, and

Minimum radiation (Rmin) 4.42 kWh/m

day.

Wind energy was not considered because installation

of wind turbines depends largely on the location; wind

speed varies during the year, depending on the season,

changes daily and have changes in short term (seconds

to minutes) both speed and direction (8). Currently there

exists a wind map of Peru (9) with only general informa-

tion; one needs to installation of measuring equipment in

PHI and needs real time records for an adequate technical

and economic evaluation of the wind resource (10).

PV Solar Panel Technology

There is a continuous improvement in materials and cons-

truction processes in solar panels (11). For this study, a

Jorge Luis Mírez Tarrillo

Número 16 Vol. 2 (2016)

standard panel has been selected: Model E20-327-COM

(12) from SunPower Company with nominal power of

327 Wp, average panel efciency of 20.4 %, panel area

of 1.63 m2, weighing 18.6 kg and 1046 x 1558 mm of

dimensions. (12). It cost in the Peruvian market is US

$ 1/W including installation. The lifetime of solar panel

selected exceeds the duration of PIP in 10 years.

Proposal connection between external electrical ne-

twork and MG of PHI

The proposed conguration is shown below in Fig.1.

where: “PV” are the PV solar panels; “M” is a electricity

meter that measures the energy coming from the external

power grid; “I” is a microsource controller (3,6); DC/

AC converter of energy from the solar panels, can syn-

chronize the phase, can let ow from the external power

supply the amount of missing energy to complete the

need of electricity of IHP; “PCA” is the Point of Com-

mon Coupling that connects/disconnects the MG of the

utility network (13); “C” is MG central controller (3,6)

that communicates bidirectional with General Electric

Board (GEB). Both “I” and “PCA” management makes

the power of the MG under the rule of admitting only

missing input power from the external power supply, too

“C” has the adequate human-machine interface (HMI)

(3,6); “A” brings together the control, management and

monitoring equipments of MG proposal; it is of accor-

ding at the state of art actual in MG with good reliability,

low unavailability and little interruptions (6).. “A” could

be in the future a single assembled element. The cost of

the electricity that companies sell has also been conside-

red. In this there are two scenarios, buy in low voltage

(LV) and/or medium voltage (MV). PHI Level I usually

feed on 220V (single phase) or 380/220 V (three phase).

The PHI level II, usually purchased in MV with nomi-

nal voltage of 10 kV or 22.9 kV. The price of electricity

in both low voltage “PrBT” and MV “PrMT” are [14]:

PrBT = 0.38115 US$/kWh and PrMT = 0.05705 US$/

kWh. In all cases the cost assumed of “A” is C (C_C)

equal to US $ 3,500.00

Figure 1: Proposal of microgrid in electric network

where: PRG is percentage growth rate,

t2 is time at the end of the study period,

t1 is the initial time of the study phase,

X(t1) is the variable to deduce their be-

havior when baseline and X(t2) is the

variable at the end point of the study

period. Something similar is done with

the prices in LV and MV where the per-

centage of annual growth in electricity

prices (PRG) will be equal to 3.9% (16)

to calculate the evolution of prices du-

ring the PIP.

Mathematical Model

The radiant energy incident on the area

where the solar panels will be arranged

is calculated based on average values

recorded incident of solar radiation, ac-

cording to:

where: maximum energy radiated for a

day is “Emax”, “Eave” is average value,

and “Emin” is minimum daily radiation

energy. It is considered that “A” has

three scenarios:

Given that Peru is located close to the

equator, the amount of daily hours of

sun is approximately 12 hours, the daily

energy available can supply an electric

charge to daily power average electric

“Pp”, for which, a given amount of solar

panels “Upv” is needed, therefore:

where: “fa” is an adjustment factor that

indicates the amount of energy consu-

med during daylight hours and we con-

sidered fa = 0.70; “Emes” is the mon-

thly electric energy consumed by the

General Considerations

The growth rate of the maximum power

demand to future isassumed as the same

value as the percentage growth rate

of the population (15) of the region to

study. The equations for calculating the

growth in electricity demand will be ac-

cording with (5):

XXe

PRG

() ()

()

−

Emax,ave,min [kWh/day] =

maxave min

Pp [kW] =

fa Emes×

Pec_MTmax,ave,min(t) = PrMT(t) Days(t) Ppmax,ave,min(t)

Revista Cientíca

ISSN 1390-5740

ISSN 2477-9105

The amount of solar panels required is

calculated by the following equation:

The initial cost [US$] of the MG

“C_in” is:

Fifteen years of PIP involves 5475

days. This represents a payment in

purchased energy “Pec” therefore

low voltage “Pec_BT” and MV

“Pec_MT” of:

where “Ppmax”, “Ppave” y“Ppmin” are the

average powers: maximum, average and

minimum deducted from the scenarios:

“Emesmin”, “Emesave” y “Emesmax”.

During PIP’s lifespan, the price of MG

is amortized progressively considering

the following calculation of real value

(VR) (5) for both tariff LV (VR_BT) and

tariff MV (VR_MT):

Similarly, the incident radiant ener-

gy converted into electrical energy for

one day, quantied as “Emax”, “Eave”,

“Emin”; is primarily affected by the ef-

ciency of the solar panel “η”, then the

electrical energy from solar panels for a

day “EEsol” is:

The electricity generated, for this case,

will make the price of energy at low vol-

tage “PrBT” because it resembles the

voltage level and the form of equipment

that is supplied through the distribution

network. The price at the time is calcula-

ted using equation 4. Therefore, during

PIP, sales price of energy produced from

solar panels “CEEsol” is valued accor-

ding:

In short, installation of MG in PHI consists of an initial

cost which is amortized by the cost of energy to stop bu-

ying in LV or MV, likewise, the energy produced repre-

sents a value that gradually increases in the time. In any

time, both values of MG and production are equalized

to represent return time of investment, the MG pays for

itself. Fig. 2 shows the trend where “TIR” is the return

time investment, and; “VR” is real value of MG. Three

curves of scenarios is shown: optimistic (sufx “max” in

variables), average (sufx “ave”) and minimum (sufx

“min”).

Figure 2: Evolution of the actual value of the facility and the value of the energy

produced.

RESULTS AND DISCUSSION

The rst scenario simulated is the power estimated using

20 % of the available area of PHI. For this we considered

the efciency of solar panels, a solar radiation assumed

in Eq. 8 divided by 10 hours of sunshine. Fig. 3 shows

the results, which allow us to verify that the PV plant will

occupy a portion of the available area of PHI.

In the case of Rmin and Amin has 36 PV panels which oc-

cupy 60 m2; in the case of Rmax and Amax are needed 108

Figure 3: Estimated production considering the three scenarios of

“A” and “R”.

Jorge Luis Mírez Tarrillo

PHI in kWh and is considered under the

following three scenarios:

Upv = (Pp

1000)/(Pnom)

C_inmax,ave,min = Upvmax,ave,min

327 + C_C

Pec_BTmax,ave,min(t) = PrBT(t)

Days(t)

Ppmax,ave,min(t)

Pec_MTmax,ave,min(t) = PrMT(t) Days(t) Ppmax,ave,min(t)

VR_BTmax,ave,min(t) = C_inmax,ave,min – Pec_BTmax,ave,min(t)

Pec_MTmax,ave,min(t) = PrMT(t) Days(t) Ppmax,ave,min(t)

EEsolmax,ave,min = Emax,ave,min η

CEEsolmax,ave,min(t) = EEsolmax,ave,min Days PrBT(t)

Número 16 Vol. 2 (2016)

Figure 5: Evolution of the VR and Pec in BT and MT for calculate of TIR

Figure 6: Evolution of VAN into BT and MT for calculate of TIR.

Figure 5 indicates that the return time of the investment

is shorter if there is a higher energy demand. Fig. 6 indi-

cates that it is more benecial for the PHI to buy at LV

instead of buying at MV, but in both cases the return time

is within PIP. It has to be studied at what energy demand

in order to change from LV to MV.

CONCLUSIONS

It has been shown that a PIP horizon of

15 years is viable, reaching a payback (a)

in worst case scenario (Amin, Rmin and

electric demand) within approximately

4 years for users who buy at low voltage

and 14 years for users who buy at me-

dium voltage; and (b) in best scenario

(Amax, Rmax and electric demand) is

about 2.5 years for users who buy atlow

voltage and 6.5 years for users who pur-

chase atmedium voltage.

At the end of PIP (15 years) are still

10 years more of optimal performance

inPV solar panels, which entails that if

is necessary, PHI may increase the capa-

city of PV solar power.

Moreover, there is area available for ins-

tallation of solar thermal equipment for

production of hot water. Both cases lead

to an improvement in PHI economy,

quality of care and working environment

and increase the efciency contributing

to a green image.

The calculation described considers the

PV solar power as a source of electri-

cal power of the base type, that is to

provide important part of consumption

and the remaining (which may be conti-

nuous, intermittent or changing) will be

supplied from the public grid.

Not considered is electrical storage, be-

cause it has a high cost in operation and

maintenance, which will increase the

cost and extend the return time of invest-

ment. The electricity produced by a PV

solar plant in a MG pays itself.

ACKNOWLEDGMENTS

The author thanks to the National Coun-

cil of Science and Technology of Peru,

CONCYTEC for funding his doctoral

research at UNI, also at the Electric

Power System Research Group in San-

dia National Laboratories (USA) during

a internship (Jan – Apr 2016).

Figure 4: Projection of MD during PIP time.

PV panels which occupy 180 m

A second simulation scenario is to calculate the maxi-

mum demand (MD) during the years of PIP considering

the growth rate of the population, given that as the po-

pulation increases, so does the health care and therefore

the amount of energy that is required for PHI. For this,

two scenarios have been considered: with Amin, Rmin (see

Fig.4a) and Amax, Rmax (see Fig. 4b). In this aspect the

environmental conditions have been assumed constant

over the years.

Revista Cientíca

ISSN 13905740

ISSN 2477-9105

eferencias

1. Energy and Mines Ministry. Peru Solar Map [Website].http://deltavolt.pe/atlas/atlassolar

2.Ministry of Health of Perú. Strategic Health Establishments: R.M.N° 632-2012/MINSA [Website].

http://www.minsa.gob.pe/Portalweb/EsSa/index.asp

3. Chowdhury S, Chowdhury SP, Crossley P. Microgrids and Active Distribution Networks”.Lon-

don.The Institution of Engineering and Technology. 2009.

4.ArtoNiinistö. Simulation of the Management of a Micro Grid with Wind, Solar and Gas Generators

[Master Thesis]. Aalto University, Finland. 2009.

5. Website Peruvian State. Electrical Legislation: Law of Electrical Concessions and Regulations.

[Website].http://www.peru.gob.pe/docs/PLANES/13049/PLAN_13049_2014_Normatividad_

LCE_y_RLCE.pdf

6. C. Schwaegerl. DG3&DG4 Report on the technical, social, economic, andenvironmental benets

provided by Microgrids on power system operation. Siemens. 2009

7.Ministry of Health of Perú.Program to Support of Health Sector Reform [Website].http://www.

parsalud.gob.pe/

8. Francisco GoncalvesGoianaMesquita. Design Optimization of Stand-Alone Hybrid Energy Sytems

[Master Thesis]. Universidade do Porto, Portugal. 2010.

9.Energy and Mines Ministry.Wind Map of Perú [Website].http://deltavolt.pe/atlas/eolico

10.Masters GM. Renewable and Efcient Electric Power Systems. New Jersey. Jhon Wiley & Sons,

Inc. 2004.

11. Phys.org. NREL demonstrates 45.7% efciency for concentrator solar cell [Website]. United

States of America http://phys.org/news/2014-12-nrel-efciency-solar-cell.html

12.SunPower. Solar Panels for Commercial Use Serie E20-327-COM [Website].http://us.sunpower.

com/sites/sunpower/les/media-library/data-sheets/ds-e20-series-327-residential-solar-panels.pdf

13. Paolo Piagi. Autonomous Control of Microgrids.IEEE PES Meeting 2006.

14. Luz del Sur. Prices for the Sale of Electricity (including tax) Tariff September 2015 [Website].

https://www.luzdelsur.com.pe/media/pdf/tarifas/TARIFAS.pdf. 2015.

15. National Institute of Statistics and Information. Growth Rate of Population for Region [Website].

http://www.inei.gob.pe/media/MenuRecursivo/publicaciones_digitales/Est/Lib0015/cap-52.htm

16. PCR Pacic Credit Rating.Peru Electric Sector: Report at September 2014 [Website].http://www.

ratingspcr.com/uploads/2/5/8/5/25856651/sector_electrico_peruano_201409-n.pdf