Energy 93 (2015) 235e244

Contents lists available at ScienceDirect

Energy
journal homepage: www.elsevier.com/locate/energy

Optimal planning of combined heat and power systems within

microgrids
Aboelsood Zidan a, c, Hossam A. Gabbar a, b, *, Ahmed Eldessouky a
a

Faculty of Energy Systems and Nuclear Science, University of Ontario, Institute of Technology, 2000 Simcoe Street North, Oshawa, L1H7K4 ON, Canada
Faculty of Engineering and Applied Science, University of Ontario, Institute of Technology, 2000 Simcoe Street North, Oshawa, L1H7K4 ON, Canada
c
Department of Electrical Engineering, Faculty of Engineering, Assiut University, 71515 Assiut, Egypt
b

a r t i c l e i n f o

a b s t r a c t

Article history:
Received 29 April 2015
Received in revised form
5 September 2015
Accepted 10 September 2015
Available online xxx

In this paper, an optimal deployment with respect to capacity sizes and types of DG (distributed generation) for CHP (combined heat and power) systems within microgrids was presented. The objective
was to simultaneously minimize the total net present cost and carbon dioxide emission. A multiobjective GA (genetic algorithm) was applied to solve the planning problem including the optimization of DG type and capacity. The constraints include power and heat demands constraints, and DGs
capacity limits. The candidate technologies involved in this study include CHP generators (with different
characteristics), boilers, thermal storage, renewable generators (wind and photovoltaic), and a main
power grid connection. The surplus/decient electricity can possibly be sold to/bought from the main
grid. Costs of CHP generators are based on their types and the capacity range. The approach was applied
to a typical CHP system within microgrid system as a case study, and the effectiveness of the proposed
method was veried.
2015 Elsevier Ltd. All rights reserved.

Keywords:
Combined heat and power
Renewable
Microgrid system
Gas-power
Multi-objective
Genetic algorithm

1. Introduction
Society's full dependence on energy leads to electricity network
congestion. For centuries, energy systems were based on centralized energy generation with a transmission and distribution system. With the rapid increase in fuel prices, capital cost of central
generation plants, and electricity/head demand growth; there is a
real need to have an alternate generating system with higher efciency of energy use. Microgrid is a relatively small-scale localized
energy network which includes loads, control system and a set of
energy resources such as generators and energy storage devices
[1,2]. Microgrid can operate in a grid-connected mode where energy resources interact with the main electrical grid, or in an
islanding mode where a microgrid feeds its local loads without the
use of the main electrical grid. Microgrid is widely recognized as an
alternative generating system which can compete with traditional
centralized electricity plant [3]. As an alternative to centralized
energy systems, microgrids provide energy locally by utilizing DG

* Corresponding author. Faculty of Energy Systems and Nuclear Science, University of Ontario, Institute of Technology, 2000 Simcoe Street North, Oshawa,
L1H7K4 ON, Canada.
E-mail address: Hossam.gabbar@uoit.ca (H.A. Gabbar).
http://dx.doi.org/10.1016/j.energy.2015.09.039
0360-5442/ 2015 Elsevier Ltd. All rights reserved.

(distributed generation) with minimum energy transmission from/

to remote regions [4].
Diversifying the nature of energy sources improves reliability,
power quality, and economics of a system [5]. Therefore, microgrids comprise several types of DGs such as energy generation,
energy storage and load management options. Currently, natural
gas is the primary fuel for DGs [4]. Renewable energy sources such
as wind power and PV (photovoltaic) are receiving a wide acceptance in the power generation industry as they are inexhaustible
and nonpolluting. Renewable energy sources are characterized as
intermittent power sources due to climate changes in wind speed
and solar irradiance. One of the main applications of DGs is CHP
which generates heat and electricity simultaneously [4]. CHP and
renewable energy sources are key elements in future clean energy
systems [6,7]. Flexible CHP production can help along the integration of uctuating electricity production from renewable energy sources [6,7]. Furthermore, thermal energy storage
represents another fundamental element in a microgrid [5]. It
reserves the thermal energy surplus to be used during higher
demand times, thus effectively shaving the peak of the thermal
energy demand. With the gradual depletion of fossil fuels and the
keen desire for many countries to meet the environmental constraints established by the Kyoto Protocol in order to reduce

236

A. Zidan et al. / Energy 93 (2015) 235e244

greenhouse gas emissions, microgrids with CHP and renewable

energy resources provide an effective organization form of
distributed energy supply units [2].
Microgrids with CHP and renewable energy resources have
many benets such as high reliability, and reduction of: energy loss,
emission, wasted heat, capital and running cost of transmission
networks, and risk of energy supply failure [1e5,8]. Planning CHP
systems is receiving a great deal of attention [1e12] as a reasonable
planning is an important premise and guarantee for stable and
efcient microgrid operation. A lot of works have been carried out
in the pertinent literature, which can be generally divided into
three aspects: the assessment indicators and methods, the operation strategy, and the optimal planning of CHP system within
microgrid system. With regard to assessment indicators and
methods, the performance indices of economy, reliability, emission
and energy utilization efciency are the KPIs (key performance
indicators) to analyze the performance of a microgrid system [2].
For example, ref. [1] proposed a hierarchical framework to realize
an economic generation schedule of microgrids. However, thermal
storage and emissions are not included. Refs. [2,3] focused on
evaluating different optimal output sets of DG-mix on the basis of
multi-objective optimization compromising between fuel cost and
emission. Renewable energy sources and thermal storage are not
included in Ref. [3]. Ref. [4] proposed an optimal design of microgrids with CHP units which integrates the results of a life cycle
assessment of the microgrids. The objective was to minimize the
overall operating cost and emission. However, their work did not
consider the integration of renewable energy resources in microgrids. Ref. [5] presented operational results of a real life residential
microgrid which includes six apartments, photovoltaic plant, a
solar based thermal energy plant, a geothermal heat pump, and a
thermal energy storage. Authors in Ref. [6] illustrated why electricity smart grids should be seen as part of overall smart energy
systems and emphasized the inclusion of exible CHP production in
the electricity balancing and grid stabilization. For example, CHPs
should operate in a way that they produce less energy when the
output of renewable energy sources is high and more when the
output of renewable energy sources is low. Ref. [7] presented solutions to integrate renewable energy sources into electricity systems using small and medium sized CHPs. Furthermore, the
proposed solutions and software tools allow partnerships to offer
services which are currently only offered by big power plants to
electricity markets.
In Ref. [8], an algorithm to nd the optimal types, sizes, and
placement of DGs in a microgrid was proposed. The objective
considered is to minimize the capital and operational cost and they
did not include emissions. Furthermore, renewable energy sources
and thermal storage are not included. Ref [9] aimed to reduce the
fuel consumption rate of DGs in a microgrid. However, emissions
and maintenance costs are not included. An optimization algorithm
for optimal dispatching of DGs and storage systems in an islanded
microgrid was proposed in Ref. [10]. The objective was to minimize
the overall operating cost and emission. However, their work did
not consider the thermal energy demand. Authors in Ref. [11]
presented a comprehensive analysis of seven different technologies to integrate uctuating renewable energy sources into the
electricity supply. The seven technologies are electric boilers, heat
pumps, electrolysers with local CHP, electrolysers with micro CHP,
hydrogen fuel cell vehicles, battery electric vehicles and exible
electricity demand. These different technologies were analyzed and
compared in terms of their ability to integrate renewable energy
sources and their fuel efciency in different scenarios. Ref [12]
proposed a deterministic model to analyze the economic feasibility of using compressed air energy storage plant in the Danish

electricity system in comparison with other exible technologies

including heat pumps and hydrogen storages.
For the operation strategy of CHP within microgrid system,
there are two basic strategies: FTL (following thermal load) and FEL
(following electrical load) [2]. When microgrid works at FTL mode,
cogeneration units put the priority on heat production. If the
microgrid could not meet the electrical demand, the decient
electricity can possibly be purchased from the main grid. When
microgrid works at FEL mode, cogeneration units put the priority
on electricity production. If the microgrid could not meet the heat
demand, the decient heat can possibly be supplied by a boiler.
However, the strategy of following hybrid thermal-electric load
which can switch between the two basic modes according to load
represents the higher economic and environmental performance
[13,14].
Investment in natural gas-power technologies increases
continuously. Power generation using natural gas has received
signicant attention due to the following reasons [15]: 1) short
construction times and low initial investment costs which makes it
attractive in a deregulated market; and 2) burning natural gas emits
less harmful emissions compared with other fossil fuels such as
coal and oil. DGs based on gas engines and micro turbines represent
a reliable option generally for countries that have an access to
natural gas [16]. Gas technologies such as NGFCs (natural gas fuel
cells), H2FCs (hydrogen gas fuel cells) and NGTs (natural gas turbines) can be used to balance the required power/heat demand in
microgrids. Furthermore, power-to-gas (P2G) storage can be
implemented to converts the surplus power from renewable energy to hydrogen fuel to be used where and when it is needed [17]
through one or more of the following ways: 1) Clean fuel source; 2)
To be fed to the natural gas grid as fuel for low carbon heating or/
and fuel for generating controllable electricity; and 3) Industrial
customers.
Modern energy systems are faced with a high degree of
complexity and uncertainty due rising factors such as penetration
of renewable energy resources, new loads (e.g. EVs (electric vehicles)), rise of energy storage devices, variable demand, and demand
response programs. For example, some consumers are producing
electricity/heat themselves. However, if they use PV and/or wind
based generators and are temporality not provided with energy,
they may still need to be fully supplied by the main grid. Thus, the
net load prole will be less predictable [18]. Scenarios help in
capturing the conceptual system as visualized by a user by means of
operational examples. Scenarios simulate events a user would
experience in performing tasks that constitute the operation of a
system. Optimization under uncertainty can be solved using
scenario-based modeling with three phases [19]: 1) gathering and
processing of relevant background data and judgmental knowledge; 2) creating a coherent set of scenarios; and 3) building and
solving a stochastic optimization model. Furthermore, several
techniques have been applied for modeling uncertainties such as
Monte Carlo [20], scenario reduction technique [21], fuzzy
modeling [22], and information gap decision theory [23].
Several challenges may appear in dealing with the future energy
system [18,24]. Examples of these challenges are [18,24]: 1) the
demand at customer side will be surrounded with more uncertainty; 2) emerging new technologies and their timing and penetration levels; 3) the degree in which residential loads are
controllable; 4) use of smart grid information and communication
technologies to intelligently integrate actions of users; and 5) uncertainties regarding economic development, international policies
and trade, and climate policies.
Microgrid performance is mainly dependent on its DGs type and
capacity. The goal of the work presented in this paper is therefore to

A. Zidan et al. / Energy 93 (2015) 235e244

propose an optimal planning for CHP and renewable energy resources within microgrids in a manner that will include:
e Electrical and heat demands with considering the hourly change
during the day
e Uncertainty associated with renewable DG units' output power
e Economic objective (total capital and operational costs) and
environmental objective (CO2 emissions)
e Optimal mix of gas-power (natural gas turbines, natural gas fuel
cells, and hydrogen gas fuel cells), renewable energy resources
(wind and PV), natural gas boiler, electrical heater, thermal
storage, and the main electrical grid
From this perspective, the proposed work may act as a useful
modeling and design tool, to assess the opportunity of employing
alternative energy technologies and strategies. Other DG technologies can be included according to the microgrid under study and
the regional primary energy attribute which reects the potential
of various types of energy resources in different regions.
The remainder of this paper is organized as follows: Section 2
presents the problem description. Section 3 presents the models
used for the system components. The problem formulation is
explained in Section 4. Sections 5 and 6 detail the test results, and
Section 7 presents the conclusions.

237

and environmental (carbon dioxide emission) viewpoints. Fig. 1

shows the energy ow schematic of CHP system within microgrid
system. The system load consists of heat and electrical load, where
the heat load can be used for space heating, cooking, bathing,
showers, washing, and laundry purposes, etc. [2]. NGFC and NGT
consume natural gas fuel and H2FC consumes hydrogen fuel and
they work as CHPs to produce electricity and heat simultaneously.
Electrical load is mainly met by renewable energy resources (wind
and PV) and CHPs while surplus/decient electricity can be sold to/
purchased from the main grid. The heat load demand is met by
CHPs heat together with natural gas-red boiler and thermal
storage, and decient heat energy, if any, will be supplied from an
electrical heater. Furthermore, for the sake of verication and
comparison, Table 1 shows four congurations of a microgrid that
will be implemented in this work.
3. Modeling of DG units
In this paper, the types of DG units considered are NGFCs (natural gas fuel cells), H2FCs (hydrogen gas fuel cells), NGTs (natural
gas turbines), TS (thermal storage), WT (wind turbines), and PV
(photovoltaic). Other DG types can be included and modeled with
similar approaches. DG models describe gas fuel consumption, CO2
emission, power loss, and total net present cost (i.e., capital and
operational costs).

2. Problem description
3.1. Natural gas turbines (NGTs)
A CHP (combined heating and power) microgrid with DG units
provides an effective solution to energy-related problems,
including increasing energy demand, higher energy costs, energy
supply security, and environmental concerns. This paper presents a
planning algorithm for the CHP system within microgrid in order to
determine the optimal deployment with respect to capacity sizes
and types of DGs. The performance of a CHP system within
microgrid is evaluated from the economical (total net present cost)

Natural gas is one of the leading energy sources for DGs. Because
of the extensive natural gas supply infrastructure and the environmental benets of using natural gas, it is one of the leading
choices for on-site power generation. NGTs operate by burning
natural gas. Hot gasses produced during the combustion process
turn the turbine and generate electricity. This type of DGs is typically favored for meeting peak loads, as the turbines can quickly

Fig. 1. Energy ow schematic of CHP system within microgrid system.

238

A. Zidan et al. / Energy 93 (2015) 235e244

Table 1
Sample conguration of microgrid system.
Conguration number

1
2
3
4

CHP

Connection with the main grid

NGT

H2FC

NGFC

achieve full generation capability. A DG model can be represented

by the following equations:

ffue;i t

Pi t
ct2T;
ui hi;P

Hi t Pi t

0  Pi t  Pr;i ; i2G

hi;H
ct2T; i2G
hi;P

Ei t Ki ui ffue;i t ct2T; i2G

Ploss;i t


Pi t

1  hi;P  hi;H ct2T; i2G
hi;P

(1)

(2)

Thermal storage

Renewable energy
Wind

PV

sources without CO2 pollutants, also, there are no fuel costs, and
their operating and maintaining costs are considered in the model.
3.4.1. Uncertainty modeling
In this work, the wind speed and solar irradiance for each hour
of a day are modeled by Weibull and Beta PDFs (probability density
functions), respectively, using historical data for the site under
study [25,26]. The uncertainty in WT and PV behaviors is modeled
as follows [25,26]:

(3)

(4)

where, G is the set of DGs; ffue, i is the consumed fuel by DG i at hour

t; Pi is output power from DG i at hour t; T is set of hourly periods
(i.e., 8760 for the entire year); ui is the energy density of the fuel
consumed by DG i in kWh/kg; hi,P power efciency of DG i; Pr,i rated
power for DG i; Hi is the generated heat power from DG i at hour t;
hi,H heat efciency of DG i; Ei is the CO2 emissions from DG i at hour
t; Ki is the carbon footprint for the energy produced by DG i in kg
CO2/kWhr; Ploss,i is power losses for DG i at hour t.
3.2. Natural gas fuel cells (NGFCs)
A fuel cell is a device that converts the energy present in a fuel to
electricity using an electro-chemical process. Thus, unlike conventional power plants, fuel cell extracts electricity from a fuel
without going through a combustion process. This results in a clean
emission signature. Fuel cells require hydrogen as the fuel source.
However, hydrogen can be produced from a readily available fuel
source such as natural gas (NGFCs). Similar to NGTs, NGFC model is
represented by Eqs. (1)e(4) using its own parameters (i.e.,
i NGFC).
3.3. Hydrogen fuel cells (H2FCs)
At presence of hydrogen fuel infrastructure, hydrogen gas fuel
will be supplied to a local storage tank and/or feed directly into a
fuel cell which will produce electrical power. This results in a clean
emission without CO2 pollutants (i.e., EH2FC 0 in Eq. (3)). Similar to
NGTs and NGFCs, H2FC model is represented by Eqs. (1), (2) and (4)
using its own parameters (i.e., i H2FC).

e Based on seasons, the entire year is divided into three seasons

(winter, mid-season, and summer), and each season is being
represented by one day, which is subdivided into 24 h segments.
e The mean and standard deviation for each time segment are
calculated utilizing the historical wind speed and solar irradiance data.
e The Weibull and Beta PDFs (probability density functions) are
generated for each hour using the mean and standard deviation
for each segment.
e In order to integrate the output power of wind turbines and PV
modules as multistate in the formulations, the continuous PDF
of each is divided into a proper number of states. In this paper,
the PDF of the wind speed and solar irradiance for each hour is
divided into 12 and 10 states respectively. Then the probability
of each wind speed and irradiance state is calculated.
e The corresponding output power of the wind turbine and PV
module in each state is calculated using the wind turbine power
performance curve and PV module characteristics.

3.4.2. Combined states model

The combined states model describes all the system states and
their probabilities of occurrence. In this work, the hourly average
load is considered and the variations within the hour are neglected.
CHPs (NGT, H2FC, and NGFC), boiler, electrical heater, and thermal
storage are assumed to be rm without uncertainty. This is because
their status (i.e., output of CHPs, boiler, and electrical heater, and
charging/discharging status for thermal storage) will not change
within the hour. Assuming that wind speed states and solar irradiance states are independent, the probability of any combination
of them is obtained by convolving the two probabilities. Therefore,
the total number of states (Ns) for each 1 h time segment is 120 (12
wind state  10 PV state).
3.5. Thermal storage (TS)

3.4. Renewable energy sources

Renewable energy sources such as WT (wind turbines) and PV
(photovoltaic) become a main element in energy production systems as they are inexhaustible and nonpolluting. Renewable energy
sources can increase reliability, reduce greenhouse gas emissions,
and reduce energy losses for power systems [25]. Renewable energy sources are characterized as uctuating power sources due to
changes in wind speed and solar irradiance. WT and PV are clean

Thermal storage is a technology that stores excess thermal energy for later use for heating/cooling applications. TS systems are
used particularly in buildings and industrial processes. TS integration enables microgrid self-consumption. TS systems help in
balancing between energy demand and supply on daily basis. TS
systems can also reduce peak demand, energy consumption, CO2
emissions and costs, and increase the overall efciency of energy
systems. Furthermore, storing the variable renewable energy in the

A. Zidan et al. / Energy 93 (2015) 235e244

form of thermal energy may help in increasing the share of renewables in energy mix. As shown in Eq. (5), the heat stored cannot
exceed the installed capacity of the thermal storage at any time.
Charge and discharge rates for a thermal storage are the rates at
which heat is added to or reduced from the thermal storage. These
rates depend on the characteristics of the thermal storage equipment and they are limited by constraints (Eqs. (6) and (7)). As
shown in Eq. (8), the energy stored in a thermal storage depends on
the energy stored from the previous time period, the energy
charged/discharged and the turn-around efciency. In order to
guarantee that no heat is accumulated from day to day, at the end of
a day the thermal storage state returns to its initial value at the
beginning of that sample day (Eq. (9)).

HSTS t  HTS;r

c t2T

239

Ns
XX

Cope t

rs

Cgas;i ffue;i t; s Cm;i Pi t; s

i2G s1


Csu;i SUi t; s Cm;TS HTS;in t; s ct2T; s2Ns
(15)



Helec heater t; s
Pgrid t; s
helect:heater
s1

Cgas;NG
 Csell t Pgrid t; s Hbo t; s
uNG hbo

ct2T; s2Ns
Cm;bo

Ns
X

Cpur t

(5)

rs

Cbuy t

(16)
HTS;in t  CTS
HTS;out t  DTS

c t2T

(6)

c t2T

HSTS t HSTS t  1 HTS;in hTS 

(7)

Egrid t

HTS;out
hTS

c t2T

HSTS 24 HSTS 0

(8)

4. Problem formulation
The objective of the microgrid planning problem is to minimize
the total net present cost and the carbon dioxide emission for all
possible combinations (states) of load and DG output power. The
costs and emissions for each of the combined generation and load
states (Ns) are evaluated and weighted based on their probability of
occurrence. The multi-objective planning problem for a CHP system
within microgrid system can be dened as:

4.1. Objective (tness) function

Minimize OF1 ; OF2

Cope t Cpur t

(10)
(11)

t2T

OF2

ct2T; s2Ns

(17)

Egrid t EDG t Ebo t

Ns
XX

rs Ei t; s

ct2T; s2Ns

(18)

i2G s1

(9)

where, HSTS and HSTS,r are the heat stored and the rated installed
capacity of the thermal storage respectively; HTS,in/HTS,out are the
heat sent/received to/from the thermal storage; CTS/DTS are the
maximum charging/discharging rates for the thermal storage; and
hTS is the turn-around efciency of the thermal storage.

rs Kgrid Pgrid t; s

s1

EDG t

OF1 fcap

Ns
X

Ebo t

Ns
X

rs Kbo Hbo t; s

ct2T; s2Ns

(19)

s1

where, G is the set of DGs {NGT,H2FC,NGFC,WT,PV}; Ns is number of

states per hour; r(s) is probability of state s; OF1(cost) and OF2(emission) are the objectives required to be minimized; fcap is the
capital cost of DGs; Cope(t) is the operational cost at hour t; Cpur(t) is
the energy purchase cost at hour t; Egrid(t) is the CO2 emissions from
the main grid at hour t; EDG(t) is the total CO2 emissions from all
DGs in the microgrid at hour t; Ebo(t) is the CO2 emissions from the
boiler at hour t; Ccap,i is capital cost of installing DG i in $/kW; CRFi is
the capital recovery factor of DG i; r is the interest rate; ni is the
lifetime of DG i in years; CRFTS is the capital recovery factor of
thermal storage; nTS is the lifetime of thermal storage in years; Cgas,i
gas price required for DG number i in $/kg; Cm,i maintenance cost of
DG number i; ffue,i(t,s) is the consumed fuel by DG i at hour t and
state s; Pi(t,s) is output power from DG i at hour t and state s;Cm,TS
maintenance cost of thermal storage; Csu,i start-up cost of DG
number i; SUi (t,s) start-up status of DG number i at hour t and state
s; Cbuy/Csell buying/selling price of electricity from/to the main grid
at hour t; Pgrid(t,s)/Pgrid-(t,s) power bought from/sold to the main
grid at hour t and state s; Hbo(t,s)/Helec heater(t,s) is heat supplied by
the boiler/electrical heater at hour t and state s; hbo/helec heater efciency of boiler/electrical heater; Cm,bo maintenance cost of boiler;
Kgrid is the carbon footprint for the energy purchased from the grid
in kg CO2/kWhr; Kbo is the emission from the boiler in kg CO2/kWhr.

(12)

t2T

fcap

4.2. Constraints

Ccap;i Pr;i CRFi r; ni Ccap;TS HTS;r CRFTS r; nTS

i2G

(13)
r 1 rni
ci2G
CRFi r; ni
r 1 rni  1

(14)

Pi; t; s Pgrid t; s  Pgrid t; s

i2G

Pld t; s Plosses t; s ct2T; s2Ns

(20)

240

A. Zidan et al. / Energy 93 (2015) 235e244

0

Helec heater t; s
Pgrid t; s ; Pgrid t; s
helect:heater

 Pgrid;max ct2T; s2Ns

(21)

Pi;min  Pi t; s  Pi;max ci2G; t2T; s2Ns

X

Hi t; s Hbo t; s Helec

heater t; s

(22)

HTS;out t; s

i2G

 HTS;in t; s  Hld t; s ct2T; s2Ns

(23)

where, Pld(t,s) and Hld(t,s) power and heat demand at hour t and
state s; Pgrid,max upper capacity of the main grid; Pi,min and Pi,max are
lower and upper power generation of DG i respectively; HTS,in (t,s)/
HTS,out(t,s) are the heat sent/received to/from the thermal storage at
hour t and state s.

Step 5: Generate a new population (Pk1) through the application of the following operators to (Pk): selection through
choosing some parents to participate in the next generation
based on their tness function values, elitism in order to guarantee the presence of the best individuals of the current generation in the new one, crossover which combines parts of two
parents to produce children that contain some parts from both
parents, and mutation which applies random changes to a single
chromosome to create a child.
Step 6: Check the constraints of the distribution networks that
correspond to the individuals in (Pk1).
Step 7: Evaluate the tness functions for the individuals in
(Pk1).
Step 8: Check for the termination condition. If the optimal
pattern remains unchanged after a preset number of iterations
or MI has been reached, go to step 9; otherwise go to step 5.
Step 9: Report the results.
5. Case study

4.3. Implementation of the genetic algorithm (GA)

In this work, a GA has been applied in order to solve the CHP
system within microgrid planning problem based on the proposed
formulation. As shown in Fig. 2, the algorithm includes the
following steps [26,27]:
Step 1: Read the following information that has been input to the
algorithm: Annual microgrid power and heat demand proles,
population size, chromosome length, and maximum number of
iterations (MI).
Step 2: Generate an initial population (P0). As shown in Eq. (24),
the chromosome length equals the total number of decision
variables (optimal rating for each DG unit and thermal storage,
and supplied heat from boiler and electrical heater).


X P1 P2 / Pi P G Hbo Helec heater HTS;r c i2G

(24)

Step 3: Check the constraints that correspond to the individuals

in (P0). Infeasible solutions are then removed from the solution
space through the assignment of a large penalty cost.
Step 4: Evaluate the tness functions for individuals in (P0) using
Eqs. (11) and (12). The population is then denoted by iteration
number k (i.e., population Pk).

The proposed method is applied to a general microgrid which

involves different types of buildings. The microgrid under study has
ve local customer types: a school, a hotel, a restaurant, an ofce
building and one hundred residential buildings. The ve customer
types have different heat and electricity patterns based on their
functions. The school consumes energy during day time. The
restaurant has its peak demand hours during lunch time and dinner
time. The residential buildings have their peak demand hours
during the evening when people come back from work. The hotel
and ofce building are commercial buildings which have relatively
at energy consumption during working hours. Based on seasons,
the entire year is divided into three periods (120 winter days, 153
mid-season days and 92 summer days in total) and each season is
being represented by one sample day, which is subdivided into 24 h
segments [28]. Hence, the cost (OF1) and emission (OF2) functions
are multiplied by weighting factors of the sample days, e.g. the
weighting factor of winter sample day is 120. Table 2 provides the
basic statistics for the energy demand proles [28]. Fig. 3 shows the
electricity and heat demand proles for the three sample days per
year [28].
In this work and as shown in Table 1, the four congurations for
microgrids planning problem has been solved for two modes of
operation:
1) Mode 1 FEL (following electrical load) where DG units put the
priority on electricity production to follow the required electrical load demand. If the microgrid could not meet the heat
demand, the decient heat can possibly be supplied by thermal
storage, boiler and/or electrical heater.
2) Mode 2 FTL (following thermal load) where DG units put the
priority on heat production to follow the required thermal load
demand. Furthermore, in both modes of operation, the decient/surplus electricity can possibly be purchased/sold to the
main grid.
For the sake of verication and comparison, within each
conguration and mode of operation the following ve case studies
will be implemented:

Fig. 2. Flow chart for GA-based planning algorithm.

Case 1: The electrical load and heat load are completely supplied
by the main grid and the boiler, respectively (DGs are not
included).
Case 2: The electrical load and heat load are completely supplied
by the main grid and the electrical heater (DGs and natural gas
boiler are not included).

A. Zidan et al. / Energy 93 (2015) 235e244

241

Table 2
Statistics of energy demand prole for the microgrid under study.

Annual heat demand (MWh)

Annual electricity demand (MWh)
Peak heat demand (kW)
Peak electricity demand (kW)

School

Hotel

Restaurant

Ofce

Residential building

The entire microgrid

149.4
49.9
42.1
10.7

184.2
66.03
65.6
11.6

8.4
90.1
2.5
17.7

8.2
23.3
2.8
4.1

111.2
68.03
67.4
18.6

11474.7
7032.9
6839.3
1894.6

Table 3
Parameters for calculating corresponding costs and environmental attributes of the
microgrid.
Parameter

Value

Parameter

Value

uNG

13.1 kW h/kg [29]

37% [30]
33.9% [31]
40%
76% [32]
143 kg CO2/MWh [34]
201.6 kg CO2/MWh [35]
0.3037 ($/kg) [37]
0.47 $ [1]
0.15 $ [1]
0.15 $ [1]
0.003 $/kWh [30]
3189 $/kW [30]
5.121 cent/kWh [40]
4002 $/kW [1]
0.01 $/kWh [1]
0.01 $/kWh [1]

uH2

33.33 kWh/kg [29]

48% [30]
46% [31]
50%
100% [33]
201.6 kg CO2/MWh [35]
366.1 kg CO2/MWh [36]
2 ($/kg) [38]
0.003 $/kWh [30]
0.034 $/kWh [30]
0.034 $/kWh [30]
750 $/kW [34]
5000 $/kW [39]
1882 $/kW [1]
30 $/kW [4]
0.002 $/kWh
90% [41]

hNGT,P
hNGFC,P
hH2FC,P
hbo
Kgrid
Kbo
Cgas,NG
CSU,NGT
CSU,NGFC
CSU,H2FC
Cm, bo
Ccap, NGFC
Cbuy & Csell
Ccap, PV
Cm, WT
Cm, PV

Fig. 3. Typical heat and electricity demand for the microgrid.

Case 3: Planning problem for DGs, boiler and the main grid
(microgrid can buy or sell energy to the main grid) for minimum
cost (OF1 only is included as in Eq. (11)).
a. Running cost only (i.e., operational cost and energy purchase/
selling cost)
b. Total running and capital costs
Case 4: Planning problem for DGs, boiler and the main grid
(microgrid can buy or sell energy to the main grid) for minimum
emissions (OF2 only is included as in Eq. (12)).
Case 5: Planning problem for DGs, boiler and the main grid
(microgrid can buy or sell energy to the main grid) for
compromised solution with minimum possible cost and emissions (both OF1 and OF2 are included).
Table 3 presents the parameters used for calculating various
costs and emissions in the simulation study. As the proposed
method is generic; the number and type of customers, proles of
heat and electricity demands, and values of different parameters
can be easily adjusted according to the microgrid under study and
the local distribution company.

hNGT,H
hNGFC,H
hH2FC,H
helec heater

KNGT
KNGFC
Cgas,H2
Cm, NGT
Cm, NGFC
Cm, H2FC
Ccap, NGT
Ccap, H2FC
Ccap, WT
Ccap, TS
Cm, TS

hTS

The interest rate (r) is 8% [8].

The lifetime period (ni) is considered to be 20 years for all equipment [8].

cost of the microgrid is $7.74  105. Furthermore the total amount

of CO2 emissions is 6.93  106 kg. These results are totally system
dependent and signicant differences can occur for other systems.
As shown in Table 4, case 1 provides the highest CO2 emissions
compared to other cases.

6.1.2. Case 2: with electrical heating

In this case, the electrical load is completely supplied by the
main grid as well as the heat load through electrical heating. The
total annually cost of the microgrid is $9.91  105. Furthermore the
total amount of CO2 emissions is 2.77  106 kg. By comparing these
results with the base case, the total annual cost is increased by
28.04%, and the total CO2 emissions decreased by 60.03%. From
these results it is obvious that electrical heating has a signicant
effect in reducing the CO2 emissions compared to the natural gasred boiler. As shown in Table 4, in case 1 and case 2 the electricity and/or heat generations are balanced with electricity and/or
heat demands. This is because the microgrid purchases electricity
and heat from main grid and heaters according to its electrical and
thermal load.

6. Results and discussions

The outcomes of the planning problem for a CHP system within
microgrid system are represented in the following two subsections.
6.1. Results for mode 1: following electrical load (FEL)
The outcomes of the planning problem for a microgrids works in
a following electrical load mode are shown in Table 4.
6.1.1. Case 1: base case results
In this case, the electrical and heat loads are completely supplied by the main grid and the boiler, respectively. The total annual

6.1.3. Results of conguration 1 (CHPs and the main grid)

As shown in Table 4, case 3b represents the minimum cost solution, the system costs are reduced by 47.93%; on the other hand,
the system emissions are reduced by 53.54%. Case 4 represents the
minimum emissions solution with an 86.49% reduction in emissions, while the system costs are increased by 215.25% (almost two
times higher than the base case). This is due to the high cost of H2FC
DG unit installed in the system. For utilities who interested in
reducing both cost and emissions; case 5 provides the compromised solution, the system costs are reduced by 27.52%; on the
other hand, the system emissions are reduced by 82.97%.

242

Table 4
Results of the studied cases for FEL mode of operation.
Cong. Number Case

Cost ($)

Total CO2 emission (kg) DGs generated energy

Running

7.74
9.91
2.68
2.62
1.56
3.93
2.39
2.24
1.18
3.36
3.42
2.95
1.32
5.47
3.62
2.57
6.98
3.10




















Total

105
105
105
105
106
105
105
105
106
105
105
105
106
105
105
105
105
105

7.74
9.91
4.13
4.03
2.44
5.61
4.03
3.75
2.13
4.99
1.01
4.80
2.14
1.12
8.44
4.78
1.42
5.30




















105
105
105
105
106
105
105
105
106
105
106
105
106
106
105
105
106
105

6.93
2.77
3.29
3.22
9.36
1.18
2.73
2.64
7.92
9.99
4.75
3.32
9.51
1.50
4.39
2.78
9.31
1.02




















106
106
106
106
105
106
106
106
105
105
106
106
105
106
106
106
105
106

Generated heat (MWh)

Rated size (kW)

Electricity (MWh) Heat (MWh) Purchased (MWh) Sold (MWh) Boiler

Electrical heater TS

NGT

H2FC

NGFC TS

WT

PV

0
0
6966.12
6956.15
6858.20
7032.90
6974.09
6939.74
7032.8
6912.6
8166.89
6873.86
7026.89
7355.54
7121.7
7032.9
7032.9
7032.9

0
11474.70
0
0
5583.70
5156.63
0
0
4187.48
4146.43
0
0
5486.21
7167.73
0
488.35
4417.25
4129.21

0
0
1830.80
1841.97
46.70
1855.29
1813.33
1830.8
716.8
1777.6
1900
1703
391
1900
1274
1900
1900
1900

0
0
0.07
0
1728.24
50.99
9.92
0
1728.2
34.3
0
32
1445
270
0
0
900
0

0
0
18
0.06
0
0
31.07
0
0.7
0
0
1
0
0
125
0
2
0

0
0
0
0
0
0
0
0
0
0
1900
46
0
1371
1137
300
0
0

0
0
0
0
0
0
0
0
0
0
395
73
131
64
300
22
263
159

0
0
8949.45
8932.49
8492.89
9023.27
8961.75
8912.02
8800.32
8870.06
2920.36
8516.35
8527.49
4342.52
4477.89
7903.86
8435.75
8732.2

7032.90
18507.60
66.78
76.75
5758.4
5156.63
58.81
93.16
4187.58
4266.73
0
159.04
5492.22
7168.03
44.04
488.35
4417.25
4129.21

0
0
0
0
0
0
0
0
0
0
1133.99
0
0
322.94
132.84
0
0
0

11474.70
0
5320.32
5331.90
0
105.24
4075.45
4113.2
0
0
8862.88
5540.66
0
457.55
7341.47
4395.24
0
84.27

0
0
0
0
0
0
1236.4
1234.7
1320
1368.6
0
0
0
0
161.85
833.41
1084.9
1160.9

0
0
0
0
0
0
3456.5
3454.5
3362
3452
0
0
0
0
1303.6
2884.7
3510
3354

Table 5
Results of the studied cases for FTL mode of operation.
Cong. Number Case Cost ($)

Total CO2 emission (kg) DGs generated energy

Running

1 &2

1
2
3 a
b
4
5
3 a
b
4
5
3 a
b
4
5

Total
5

7.74  10
9.91  105
0.0715
5.49  104
1.66  106
1.87  104
1098.45
1.72  105
1.32  106
1.61  105
347.5
1.36  104
1.41  106
1.75  105

7.74
9.91
4.81
3.19
4.44
3.46
1.17
3.68
5.18
4.52
1.17
3.49
5.30
4.39
















10
105
105
105
106
105
106
105
106
105
106
105
106
105

6.93
2.77
1.10
1.43
3.05
7.61
2.75
1.55
2.00
5.18
2.76
1.38
1.71
4.88
















10
106
106
106
105
105
106
106
104
105
106
106
104
105

Electricity trade with the main Generated heat (MWh)

grid

Rated size (kW)

Electricity (MWh) Heat (MWh) Purchased (MWh) Sold (MWh) Boiler

Electrical heater TS

NGT

H2FC

NGFC

TS

WT

PV

0
0
8901.33
7957.44
9248.05
8762.17
12884.61
7045.31
17058.17
11692.33
7177.13
9099.21
17385.05
12081.7

0
11474.70
0
0
859.10
198.44
0
1236.65
11.384
21.67
0
0
0
0

0
0
5088.72
3404.17
8.53
4636.94
4714.1
2386.8
57.2
5074.5
4717.5
3297
54
5185

0
0
43.42
0
5457.57
13.73
27.4
0
5372.5
0
0
0
5440
0

0
0
215.59
12.05
0
10.30
1475.7
8.8
0
4.00
1481
11.8
0
0

0
0
0
0
0
0
0
0
0
0
510
150
4052
380

0
0
0
0
0
0
1388
0
1808
957
1400
419
1810
1133

0
0
0
0
0
0
128
26
1900
103
150
0
1850
0

0
0
11474.70
10234.01
9967.5
11276.26
11474.70
9001.4
10988.98
11414.5
11325.83
10091.91
9745.67
11348.60

7032.90
18507.60
2172.40
2170.60
2982.00
2368.84
360.18
3384.14
101.68
600.19
350.96
1477.30
91.28
499.80

0
0
4040.83
3095.14
4338.05
3899.67
6211.89
2159.9
10115.57
5237.95
313.24
3494.11
8330.17
5427.5

11474.70
0
0
1240.69
648.10
0
0
1236.65
474.34
38.53
0
1342.29
0
27.02

0
0
0
0
0
0
0
0
0
0
148.87
40.5
1729.03
99.08

A. Zidan et al. / Energy 93 (2015) 235e244

1
2
3 a
b
4
5
3 a
b
4
5
3 a
b
4
5
3 a
b
4
5

Electricity trade with the main

grid

A. Zidan et al. / Energy 93 (2015) 235e244

243

Table 6
Comparison between results for FEL and FTL modes of operation for conguration 1.
Mode

FEL
FTL

Minimum cost solution

Minimum emission solution

Total cost ($)

Total cost ($)

4.03  10
3.19  105

Total CO2 emissions (kg)

6

3.22  10
1.43  106

2.44  10
4.44  106

6.1.4. Results of conguration 2 (CHPs, thermal storage, and the

main grid)
As shown in Table 4, in conguration 1 (cases 3, 4 and 5) the
electricity generation is always balanced with electricity demand.
However, the heat generation is higher than the head demand in all
cases. This is because the microgrid works based on FEL mode.
Therefore, to follow the required electrical load demand, unrequired heat energy may be generated. To provide an efcient
microgrid operation, a thermal storage device is added in conguration 2 to store the surplus heat to be used when it is needed. As
shown in Table 4, adding thermal storage (conguration 2) provides
signicant improvement in the microgrid performance. In all cases
(3, 4, and 5) both cost and emissions have been decreased
compared to similar cases in conguration 1.
6.1.5. Results of conguration 3 (CHPs, renewable energy, and the
main grid)
The results in Table 4 show that CHPs (congurations 1 and 2)
are superior to the renewable based DG units (conguration 3) in
reducing the cost objective. This is because of their dispatchable
natures that can provide higher output energy than the renewable
ones. In addition, NGTs are cheaper. Renewable based DG units have
clean and environmental friendly nature. As shown in Table 4, the
generated heat by the natural gas-red boiler for conguration 3 is
higher compared to its generated values for congurations 1 and 2.
This is to satisfy the heat demand requirements, however, it limits
the contribution of renewable based DG units for reducing CO2
emissions.
6.1.6. Results of conguration 4 (CHPs, thermal storage, renewable
energy, and the main grid)
The results in Table 4 (conguration 4) show that adding thermal storage to the microgrid can help for the integration of uctuating electricity production by renewable energy sources. In all
cases (3, 4, and 5) both cost and emissions have been decreased
compared to similar cases in conguration 3.
The results in Table 4 show that both WT and PV units have been
installed in the microgrid system. However, the results show
dominating WT penetration. This is because WT has higher capacity factor and it is cheaper. The results in Table 4 show that NGTs
(natural gas turbines) are superior to fuel cells (NGFCs and H2FCs)
in reducing the cost objective. This is because natural gas turbines
are cheaper than fuel cells in terms of their capital, operation, and
maintenance costs. As a result and as shown in Table 4, the cases
which include the cost objective (case 3 and case 5) have dominating NGTs penetration. From another side, H2FC based DGs are
superior to NGTs (natural gas turbines) and NGFCs (natural gas fuel
cells) in reducing the gas emissions. This is because of the clean and
environmental friendly nature of the H2FC DGs. As a result and as
shown in Table 4, case 4 which includes the emission objective only
has dominating H2FCs penetration.
6.2. Results for mode 2: following thermal load (FTL)
The outcomes of the planning problem for a microgrids works in
a following thermal load mode are shown in Table 5.

Compromised solution

Total CO2 emissions (kg)

5

9.36  10
3.05  105

Total cost ($)

5

5.61  10
3.46  105

Total CO2 emissions (kg)

1.18  106
7.61  105

As shown in Table 5, in case 3, 4 and 5 the heat generation is

always balanced with heat demand. This is because the microgrid
works based on FTL mode. Therefore, to follow the required heat
load demand, unrequired electricity may be generated. However,
this surplus electricity can possibly be sold to the main grid as a
source of revenue for the microgrid. For example, in conguration 1
(case 3a and case 5), the revenue from selling surplus electricity to
the main grid exceeds the required operational cost for gas fuel and
DGs maintenance costs. This explains the negative values of the
running cost (sum of the operational cost and energy purchase/
selling cost) for these two cases.
The results in Table 5 shows that FTL mode provides better results regarding the cost and/or emission objectives compared to FEL
mode for all congurations. Furthermore, because FTL mode follows the required heat load demand, it requires lower capacities of
thermal storage compared to FEL mode.
Table 6 shows a comparison between the FEL and FTL modes of
operation in terms of minimum cost, minimum emissions, and
compromised solution for conguration 1. As shown in Table 6, FTL
mode is more effective than FEL mode for the three selected solutions. For example, the total cost during the FTL mode is lower than
the total cost during the FEL mode for both of the minimum cost
and compromised solutions. Also, the total CO2 emissions during
the FTL mode is lower than the total CO2 emissions during the FEL
mode for the three selected solutions.

7. Conclusion
In this paper, a multi-objective optimization approach based on
GA for CHP system within microgrid system optimal planning is
proposed. The two objectives of the problem are to minimize 1) the
total cost (capital, O&M costs), and 2) the total gas emissions from
the main grid, boiler and DG units. Several solutions have been
presented. This set of satisfactory feasible solutions allows the
utility to identify the best solutions for lower cost or lower gas
emissions to choose based on its priorities. The results reveal the
effectiveness of the thermal load following mode compared to
electrical load following mode. The results showed that natural gas
turbines are superior to fuel cells in reducing the cost objective.
Also, hydrogen based fuel cells are superior to natural gas turbines
and natural gas based fuel cells in reducing the gas emissions. The
results showed that adding thermal storage to the microgrid has
signicant enhancement in the system performance and can help
for the integration of uctuating electricity production by renewable energy sources. Moreover, the proposed algorithm can be
applied to any type of DG units.

References
[1] Wu X, Wang X, Qu C. A hierarchical framework for generation scheduling of
Microgrids. IEEE Trans Power Deliv 2014;29:2448e57.
[2] Guo L, Liu W, Cai J, Hong B, Wang C. A two-stage optimal planning and design
method for combined cooling, heat and power microgrid system. Energy
Convers Manag 2013;74:433e45.
[3] Basu AK, Bhattacharya A, Chowdhury S, Chowdhury SP. Planned scheduling
for economic power sharing in a CHP-based micro-grid. IEEE Trans Power Syst
2012;27:30e8.

244

A. Zidan et al. / Energy 93 (2015) 235e244

[4] Zhang D, Evangelisti S, Lettieri P, Papageorgiou LG. Optimal design of CHPbased microgrids: multiobjective optimisation and life cycle assessment. Energy 2015;85:181e93.
[5] Comodi G, Giantomassi A, Severini M, Squartini S, Ferracuti F, Fonti A, et al.
Multi-apartment residential microgrid with electrical and thermal storage
devices: experimental analysis and simulation of energy management strategies. Appl Energy 2015;137:854e66.
[6] Lund H, Andersen AN, stergaard PA, Mathiesen BV, Connolly D. From electricity smart grids to smart energy systems e a market operation based
approach and understanding. Energy 2012;42:96e102.
[7] Andersen AN, Lund H. New CHP partnerships offering balancing of uctuating
renewable electricity productions. J Clean Prod 2007;15:288e93.
[8] Basu AK, Chowdhury S, Chowdhury SP. Impact of strategic deployment of
CHP-based DERs on microgrid reliability. IEEE Trans Power Deliv 2010;25:
1697e705.
[9] Hernandez-Aramburo CA, Green TC, Mugniot N. Fuel consumption minimization of a microgrid. IEEE Trans Ind Appl 2005;41:673e81.
[10] Conti S, Nicolosi R, Rizzo SA, Zeineldin HH. Optimal dispatching of distributed
generators and storage systems for MV islanded microgrids. IEEE Trans Power
Deliv 2012;27:1243e51.
[11] Mathiesen BV, Lund H. Comparative analyses of seven technologies to facilitate the integration of uctuating renewable energy sources. IET Renew Power Generation 2009;3:190e204.
[12] Lund H, Salgi G. The role of compressed air energy storage (CAES) in future
sustainable energy systems. Energy Convers Manag 2009;50:1172e9.
[13] Mago PJ, Chamra LM. Analysis and optimization of CCHP systems based on
energy, economical, and environmental considerations. Energy Build 2009;41:
1099e106.
[14] Mago PJ, Chamra LM, Ramsay J. Micro-combined cooling, heating and power
systems hybrid electric-thermal load following operation. Appl Therm Eng
2010;30:800e6.
[15] Urbina M, Li LZ. A combined model for analyzing the interdependency of
electrical and gas systems. In: 39th North American Power Symposium; 2007.
p. 468e72.
[16] Jouneghani A, Parvizi R, Amidpour M, Chaibakhsh A. Gas based distributed
generation systems, a key to Iran buildings growing energy demand. In: IEEE
2nd Intern. Conf. on Power and Energy; 2008. p. 1592e6.
[17] Hydrogenics. 2015 [Online]. Available: http://www.hydrogenics.com/
hydrogen-products-solutions/energy-storage-fueling-solutions/power-to-gas.
[18] Veldman E, Gibescu M, Slootweg H, Kling W. Scenario-based modelling of
future residential electricity demands and assessing their impact on distribution grids. Energy Policy 2013;56:233e47.
[19] Kaki A, Salo A, Talluri S. Scenario-based modeling of interdependent demand
and supply uncertainties. IEEE Trans Eng Manag 2014;61:101e13.
[20] Soroudi A, Caire R, Hadjsaid N, Ehsan M. Probabilistic dynamic multi-objective
model for renewable and non-renewable distributed generation planning. IET
Gener Transm Distrib 2011;5:1173e82.
[21] Soroudi A, Afrasiab M. Binary PSO-based dynamic multi-objective model for
distributed generation planning under uncertainty. IET Renew Power Gener
2012;6:67e78.
[22] Soroudi A. Possibilistic-scenario model for DG impact assessment on distribution networks in an uncertain environment. IEEE Trans Power Syst
2012;27:1283e93.

[23] Soroudi A, Ehsan M. IGDT based robust decision making tool for DNOS in load
procurement under severe uncertainty. IEEE Trans Smart Grid 2013;4:
886e95.
[24] Lund H, Werner S, Wiltshire R, Svendsen S, Thorsen JE, Hvelplund F, et al. 4th
generation district heating (4GDH): integrating smart thermal grids into
future sustainable energy systems. Energy 2014;68:1e11.
[25] Atwa YM, El-Saadany EF, Salama MMA, Seethapathy R. Optimal renewable
resources mix for distribution system energy loss minimization. IEEE Trans
Power Syst 2010;25:360e70.
[26] Zidan A, El-Saadany EF. Distribution system reconguration for energy loss
reduction considering the variability of load and local renewable generation.
Energy 2013;59:698e707.
[27] Zidan A, El-Saadany EF. Incorporating load variation and variable wind generation in service restoration plans for distribution systems. Energy 2013;57:
682e91.
[28] Zhang. Optimal design and planning of energy microgrids [PhD thesis].
Department of Chemical Eng., University College London; 2013 [Online].
Available:, http://discovery.ucl.ac.uk/1418705/1/Thesis_Di%20Zhang.pdf.
[29] Hydrogen for transport in Europe. 2015 [Online]. Available: http://www.
hylights.org/publications/reports/hyLights_nal_results/H2data_german.pdf.
[30] SENTECH Inc. Commercial and industrial CHP technology cost and performance data analysis for EIA. June 2010.
[31] Karady G, Sirisooriya P, Farmer RG. Investigation of fuel cell system performance and operation: a fuel cell as a practical distributed generator.
Project report. Arizona State University, PSERC Publication; May 2002.
p. 02e26.
[32] Clegg S, Mancarella P. Integrated electrical and gas network modelling for
assessment of different power-and-heat options. In: Power systems computation conference; 2014. p. 1e7.
[33] Natural Resources Canada. Heating with electricity. Ofce of Energy Efciency; 2015 [Online]. Available: https://www.nrcan.gc.ca/sites/www.nrcan.
gc.ca/les/oee/les/pdf/publications/Heating_with_Electricity.pdf.
[34] Zidan A, Shaaban MF, El-Saadany EF. Long-term multi-objective distribution
network planning by DG allocation and feeders' reconguration. Electr Power
Syst Res 2013;105:95e104.
[35] Natural resources Canada, Appendix B e CO2 emission factors. 2015 [Online].
Available: http://www.nrcan.gc.ca/energy/efciency/industry/technical-info/
benchmarking/canadian-steel-industry/5193.
[36] Zangeneh A, Jadid S, Rahimi-Kian A. A hierarchical decision making model for
the prioritization of distributed generation technologies: a case study for Iran.
Energy Policy 2009;37:5752e63.
[37] Ontario Energy Board. Natural gas rate updates. 2015 [Online]. Available:
http://www.ontarioenergyboard.ca/OEB/Consumers/NaturalGas/
NaturalGasRates.
[38] National renewable energy laboratory. 2015. [Online]. Available: http://www.
nrel.gov/hydrogen/production_cost_analysis.html.
[39] Florida solar energy center. 2015. [Online]. Available: http://www.fsec.ucf.
edu/en/consumer/hydrogen/basics/fuelcells.htm.
[40] Independent Electricity System Operator (IESO). 2015. [Online]. Available:
http://www.ieso.ca/Pages/Power-Data/default.aspx#price.
[41] Hawkes AD, Leach MA. Modelling high level system design and unit
commitment for a microgrid. Appl Energy 2009;86:1253e65.