International Journal of Ambient Energy

ISSN: (Print) (Online) Journal homepage: https://www.tandfonline.com/loi/taen20

A distinctive methodology to recover renewable

penetration cost into electric grid

Mahaboob Shareef Syed, Ch. V. Suresh & M. Ravindrababu

To cite this article: Mahaboob Shareef Syed, Ch. V. Suresh & M. Ravindrababu (2022) A
distinctive methodology to recover renewable penetration cost into electric grid, International
Journal of Ambient Energy, 43:1, 8463-8471, DOI: 10.1080/01430750.2022.2091024

To link to this article: https://doi.org/10.1080/01430750.2022.2091024

Published online: 21 Jul 2022.

Submit your article to this journal

Article views: 45

View related articles

View Crossmark data

Full Terms & Conditions of access and use can be found at

https://www.tandfonline.com/action/journalInformation?journalCode=taen20
INTERNATIONAL JOURNAL OF AMBIENT ENERGY
2022, VOL. 43, NO. 1, 8463–8471
https://doi.org/10.1080/01430750.2022.2091024

A distinctive methodology to recover renewable penetration cost into electric grid

Mahaboob Shareef Syed a , Ch. V. Suresh b and M. Ravindrababuc
a EEE Department, VLITS, Vadlamudi, India; b Department of EEE, VVIT, Nambur, India; c EEE Department, UCEN, JNTUK, Narasaraopet, India

ABSTRACT ARTICLE HISTORY

The current work portrays a unique methodology for the optimal penetration and operation of wind and Received 25 November 2020
solar power plants into the real-time Andhra Pradesh (AP) state electricity grid. The objectives of the pre- Accepted 30 May 2022
sented work include identifying the suitable locations for the installation of wind and solar power plants, KEYWORDS
forecasting the wind and solar powers available using the proposed persistent extreme learning machine Forecasting; persistent
algorithm and finally obtaining the payback period of the installation with the optimal operation. It was extreme learning machine
observed that the suggested algorithm forecasts very precisely, which was validated with the help of hard- algorithm; optimal power
ware setup. The payback period was calculated with the incorporation of wind alone, solar alone and both flow; payback period;
wind-solar combination. With the obtained results, it can be derived that the incorporation of wind-solar renewable generation
combination into the AP real-time system yields to less payback period over the trio.

Abbreviations sizing and siting of the same by scheduling the charging cycle
of the battery considering the minimisation of energy losses as
AP Andhra Pradesh
an objective. A hybrid power system comprising wind and solar
CI clearness index
has been installed in a Bahamas standalone isolated system. The
CP Betz limit
economic, sensitivity, risk analysis and system feasibility stud-
CSA Cuckoo search algorithm
ies have been carried out with the help of RETScreen software
DHI diffuse horizontal solar radiation
in Bingham, Agelin-Chaab, and Rosen (2016). The optimal size
ELM extreme learning algorithm
of wind, solar and hydro units in an isolated electric system
GHI global horizontal solar radiation
has been identified for cost minimisation using a Cuckoo search
MEM match evaluation method
algorithm (CSA) in Mallikarjuna, Balachandra, and Manohar Potli
MNRE Ministry of New and Renewable Energy
Venugopal (2016) where the optimal size of the renewable
MOPSO multi objective particle swarm optimisation
hybrid system for a standalone system has been carried out.
NREDCAP New and Renewable Energy Development Corpora-
The wind and solar resource availability have been evaluated,
tion of Andhra Pradesh
and the plant size has been chosen optimally for a residential
NSGA non-dominated sorting genetic algorithm
OPF optimal power flow building (Belmili et al. 2014). The authors also considered the
P-ELM persistent extreme learning algorithm seasonal effects of developing a techno and economic software
PV photo voltaic (Maklad 2014). In Maouedj et al. (2014), a hybrid system has been
RES renewable energy sources formulated constituting photo voltaic (PV) arrays, wind power
SVM support vector machine and battery backup. The authors also implemented the exper-
imental setup for the application of public street light and the
performance has been evaluated. In Engin (2013), a method-
1. Introduction
ology has been proposed for optimal sizing of PV and wind
The electric power demand has been increasing over the years generation system with battery charging control criteria. The
significantly. It is well known that the per capita power consump- solar energy institute lighting was considered as an application,
tion is an index for the development of any country. The deple- and the performance analysis was carried out. The critic of the
tion of fossil fuel resources, environmental concerns and huge above literature is that the hybrid system has been formed with
capital investments has been made the entire world look for renewable power generators for a standalone system or for a
alternate energy sources. The renewable energy sources (RES) distribution system only.
are the only admitted answer to all of the above issues. The out- A novel methodology has been proposed for the integra-
standing benefits of the RES are the core motivation for the exe- tion of RES into the power grid to minimise the life cycle cost
cution of the present work. A lot of research has been proposed (Gonzalez, Riba, and Rius 2015). A new multi-objective formula-
for integrating RES into the electrical power system. tion has been developed to optimise annual renewable energy
The battery coupled photovoltaic generation is proposed in cost and system reliability at the same time (Bilil, Aniba, and
Ali et al. (2017), a methodology has been suggested for optimal Maaroufi 2014). To solve the problem elitist multi-objective

CONTACT Mahaboob Shareef Syed [email protected]

© 2022 Informa UK Limited, trading as Taylor & Francis Group
8464 M. S. SYED AND CH. V. SURESH

genetic algorithm (NSGA-II) has been used. A match evaluation carried out by integrating wind, solar powers, and the combina-
method (MEM) was proposed depending on multi-objective par- tion of both wind and solar powers.
ticle swarm optimisation (MOPSO) and non-dominated sorting This paper contributes for the development of a sophisticated
genetic algorithm (NSGA-II) for optimal sizing of RES. For achiev- and precise forecasting algorithm, proposes a methodology for
ing an optimum capacity of RES a multi-objective function has the optimal operation by incorporating RES and obtaining the
been developed, comprising of correlation, inequality and annu- payback period. The entire work is organised as in the follow-
alised cost coefficients (Yazdanpanah 2014). The efforts also ing sections. The problem statement is described in Section
have been made to replace (Moriarty and Honnery 2020) all the 2, Section 3 projects the methodology for optimal siting of
conventional sources with RES. A study was carried out (Gomaa wind and solar plants based on clearness index (CI) method.
et al. 2019) to assess the environmental impacts of incorpo- The forecasting is carried out in Section 4 by proposing P-ELM
ration of the wind power with the help of the proposed LCA algorithm. Optimal power flow problem is carried out in Section
method and implemented for a real-time system Tafilah of Jor- 5 by incorporating wind and solar powers, the detailed solution
dan. The analysis was held by considering the design from the methodology is demonstrated in Section 6 with justified results
scratch of the work to the ultimate development of the complete in Section 7. The conclusions of the proposed work are presented
system. in Section 8.
The main problem posed by the RES is their intermittent
nature of power generation. Therefore, it is not suggested to
integrate RES directly into the grid. To ensure the most secure
2. Problem statement
operation of the power system, it is indeed necessary to use It is apparent that fossil fuel resources are not only going to
the most sophisticated forecasting methodology. By reviewing deplete in the near future but also causes a polluted environ-
Wang, Guo, and Huanga (2011); Ambach and Vetter (2016); Lei ment. To avoid the effect caused by greenhouse gases, the
et al. (2009); Antonanzas et al. (2016) and Ren, Suganthan, and non-conventional energy sources or RES are the only remedy.
Srikanth (2015), for the present work, a novel hybrid persistent The Andhra Pradesh (AP) state is naturally endowed with wind
extreme learning machine (P-ELM) algorithm has been used to and solar resources. The authors in this work tried to find the
forecast availability of wind speed and solar insolation for long most suitable location of wind and solar plants installation where
term time horizons. A generalise approach is adopted (Sohani there is high wind and solar potential. The optimal ratings of the
et al. 2021) to enhance the payback period, particularly, with the plants are decided to minimise the total fuel cost, thereby finally
incorporation of solar power. With the help of HOMER software obtaining the payback period for each renewable resource.
(Chisale and Mangani 2021), the system is designed, optimised,
analysed and obtained the payback period.
The novelty of the current work is a sophisticated step-by- 3. Siting of wind and solar plants
step methodology is proposed to obtain the payback period. A The present work considers the optimum placement and size
hybrid machine learning approach is proposed to forecast wind of RES in an Indian real-time 14 bus AP state electrical system.
and solar powers in the identified areas. The OPF problem was For this purpose, it cannot be ignored to identify the areas,

Figure 1. AP state with availability of wind and solar resources.

 INTERNATIONAL JOURNAL OF AMBIENT ENERGY 8465

where abundant wind and solar resources are available. Accord- Equation (2) in matrix form can be written as:
ing to the data provided by the NREDCAP (New and Renew-
able Energy Development Corporation of Andhra Pradesh), the GY = A⊗ (3)
state is naturally gifted with high wind resources in the areas of
Ananthapur, Kadapa, Chittur and Kurnool. Installing wind farms
The activation function is given as:
in those identified areas is recommended.
At the same time, the following two steps have been adopted ⎡ ⎤
to identify the locations for the feasible installation of solar G(x1 a1 + b1 ) G(x2 a1 + b2 ) . . G(xK a1 + bK )
⎢ . . . . . ⎥
plants. ⎢ ⎥
G=⎢
⎢ . . . . . ⎥
⎥
Step 1: Calculate the CI (Page 1961) value for all 13 districts of ⎣ ⎦
. . . . .
the AP state using Equation (1). The 14th bus is a slack bus, which
G(x1 aN−1 + b1 ) G(x2 aN−1 + b2 ) . . G(xK aN−1 + bK )
supplies the surplus amount of power required in the state. The
CI at the slack bus cannot be calculated. (N−1)XK
  (4)
DHIm
CI = 0.885 ∗ 1 − (1)
GHIm
A⊗ represents target matrix, which is described as: A⊗ =
where DHIm is a monthly mean of the daily diffuse horizontal [a2 , a3 , a4 , . . . , aN ]T .
solar radiation, GHIm is the monthly mean of the daily total global After selecting the values of X and b randomly, the objec-
horizontal solar radiation, which has been obtained from the tive function, which is to be minimised has been developed
Indian Solar Resource Data portal. depending on the back propagation approach:
Step 2: Identify the four areas having the highest CI value.
It has been observed that Ananthapur, Kadapa, Nellore and ⎡ ⎤2

N−1 
K
Kurnool areas have longest sun hours (Figure 1). A⊗
BP =
⎣ Yp G(xp · aq + bp ) − aq+1 ⎦ (5)
q=1 p=1

4. Wind and solar forecasting

Finally, the output weight matrix is determined using SVM
After identifying the wind and solar locations in Section 3, the
and Moore–Penrose approach
next step involves forecasting the available wind speed and solar
insolation for the selected time period. The P-ELM algorithm
has been used to forecast the same for the next 20 years. The Y = G+ A⊗ (6)
methodology of P-ELM has been presented in the succeeding
section. Wind speed forecasting is carried out by considering pres-
sure in mbar, the temperature in °C and speed of wind in m/sec
4.1. P-ELM algorithm for the present instant as the input parameters for the P-ELM
and, the output parameters considered are pressure, temper-
The P-ELM algorithm is developed by extracting the features ature and wind speed for the next instant. Similarly, for solar
of naïve persistent and extreme learning machine (ELM) algo- forecasting, the input parameters chosen are pressure, DHI and
rithms. Persistent is very much effective for forecasting in short GHI measured in Watt/Sq·m and temperature in °C, the out-
time horizons. At the same time, ELM is a machine learning put parameters are pressure, DHI, GHI and temperature in the
algorithm that is effective for longer time period horizons. By succeeding instant. The available wind speed and solar insola-
combining features of these two, P-ELM has been developed to tion have been forecasted in the identified areas using P-ELM
forecast wind and solar power. method for the next 20 years.
Mathematically, P-ELM is modelled as (Syed, Suresh, and Siva
Nagaraju 2017) described below.
Let (ap , ap+1 )N−1
p=1 be the samples for the P-ELM network,
5. Optimal power flow (OPF) incorporating cost of
where ap ∈ RN represents the input samples and ap+1 is the wind and solar generations
respective output samples. The layout of P-ELM is shown in The main objective of OPF problem is to minimise the total gen-
Figure 2. eration cost while satisfying the system equality and inequality
The P-ELM function with K hidden neurons and G (.) activation constraints. The total generation cost includes the generation
function: cost of conventional plants that already existed in system, the
cost of wind and solar generations. The mathematical model can

K
A(aq ) = Yp G(Xp · aq + bp ), q = 1, . . . , N (2) be developed as follows (Hetzer, Yu, and Bhattarai 2008; Shi et al.
p=1 2012):

where X matrix represents weights connecting input layer to hid- Min : FT (.) = FT (PT ) + FT (PW ) + FT (PS ) (7)
den layer, b indicates bias of hidden layer neurons and Y is the
weight matrix that connects from hidden layer to the output where FT (.) is the total fuel cost including thermal, wind and solar
layer. powers.
8466 M. S. SYED AND CH. V. SURESH

Figure 2. Structure of P-ELM.

FT (PT ) is the total fuel cost of thermal plants using NT number The total real and reactive power generations of thermal,
of thermal plants and is given as follows: wind and solar should be equal to the real and reactive demands
(PD , QD ) and the respective losses (PL , QL ).

NT
NT NW NS
FT (PT ) = Fi (Pi ) (8) Pi + PW,j + PS,k = PD + PL
i=1 i=1 j=1 k=1
NT
(14)
Fi (Pi ) is the fuel cost of ith generator to generate Pi MW which Qi = QD + QL
i=1
is given as follows:

Fi (Pi ) = ai ∗ Pi2 + bi ∗ Pi + ci in $ /h (9) (b) Inequality constraints:

FT (PW ) is the total cost to generate PW MW using NW number
Voltages at the different buses should be within the specified
of wind plants and is given as follows:
limits

NW Vimin ≤ Vi ≤ Vimax (15)
FT (PW ) = Fi (PW,i ) (10)
where Vimin and Vimaxare the minimum and maximum voltages
i=1
at the ith node, respectively.
Fi (PW,i ) = KW ∗ PW,i (11) Real power generations: The real powers at the different
buses should be within the limits
FT (PS ) is the total cost to generate PS MW using NS number of
solar plants which is given as follows: (i)Pimin ≤ Pi ≤ Pimax (16)
where Pimin and Pimax are the minimum and maximum power

NS
limits at the ith bus, respectively.
FT (PS ) = Fi (PS,i ) (12)
The constraints imposed on wind and solar powers are spec-
i=1
ified as:
Fi (PS,i ) = KS ∗ PS,i (13)
(ii) Pw = 0; VW < VCut in
ai , bi and ci are the cost coefficients of ith thermal generator. 3
= 0.5 ∗ ρ ∗ A ∗ VW ; VCut in < VW < VCut out (17)
KW and KS in Equations (11) and (13) are the direct cost coeffi-
cients of wind and solar generators, respectively, in $/Mw·h. = Pr ; VW > VCut out
Subjected to the following constraints (Reddy et al. 2014): (iii) PS = 0; GHI < GHImin
(18)
(a) Equality constraint: = PR ∗ η ∗ GHI ∗ A ∗ Sz ; GHI > GHImin
 INTERNATIONAL JOURNAL OF AMBIENT ENERGY 8467

Figure 3. Step-by-step procedure for calculating the payback period.

8468 M. S. SYED AND CH. V. SURESH

Figure 4. Experimental setup for wind speed recording.

Figure 5. Comparison of wind speed with P-ELM and experimental setup.

where Tap setting of transformer:

VW = velocity of wind,
V Cutin = cut-in speed of a wind turbine, T min ≤ T ≤ T max (19)
V Cutout = cut-out speed of a wind turbine,
ρ = air density, where Timin and Timax are the minimum and maximum tap setting
A = area of cross section of wind turbine blade, positions, respectively.
Pr = rated power, Limit on reactive power generation: The reactive power gen-
GHI = global horizontal irradiance in kw/m2 , eration at the different buses should be within the specified
GHImin = minimum GHI, limits
PR = performance ratio of solar panel,
η = panel efficiency, Qmin
i ≤ Qi ≤ Qmax
i (20)
A = area of panel and
Sz = size of solar plant. where Qmin
i and Qmax
i are the minimum and maximum reactive
power limits at the ith bus, respectively.
 INTERNATIONAL JOURNAL OF AMBIENT ENERGY 8469

Thermal limitation of transmission line: The MVA flowing Table 1. Comparison of existing and proposed methods of forecasting.
through any line should be less than its maximum limit. Methods MAE RMSE
Wind forecasting
Sl ≤ Smax
l (21) Existing ANFIS (Jinliang et al. 2017) 0.15 0.17
SARIMA (Jinliang et al. 2017) 0.14 0.18
Smax
l indicates the maximum thermal limitation of lth line. Jinliang et al. (2017) 0.05 0.06
Persistent 0.0499 0.0581
ELM 0.0497 0.0502
Proposed P-ELM method 0.0481 0.0492
6. Solution methodology Solar forecasting
Existing SOF (Jabar, Kazem, and John 2017) 0.361 –
The methodology used in this paper has been demonstrated MML (Jabar, Kazem, and John 2017) 3.369
in the flowchart shown in Figure 3 to solve the current prob- PSVM (Jabar, Kazem, and John 2017) 4.537
lem. The entire work is subdivided into six numbers of tasks. The SVM (Huang et al. 2016) – 5.6
FL-RNN (A. Yona et al. 2013) 0.22 –
first task includes identifying the windiest locations, forecasting Persistent 0.2147 0.347
available wind speed for the next 20 years and calculating the ELM 0.218 0.322
possible wind power that can be generated using Equation (17). Proposed P-ELM method 0.1991 0.319
Similar way, the areas having the highest solar irradiance has
been identified with the help of CI method and has been fore- Table 2. Average operating cost/h for 20 years for with and without renewables.
casted for the next 20 years and the solar power available has
Average cost Average cost with
been calculated in Task 2 using Equation (18). After identifying without renewable Average cost Average cost wind and solar
the available solar and wind power, OPF has been performed S. no ($/h) with wind ($/h) with solar ($/h) combinely ($/h)
in Task 3 to identify the optimal generations of thermal, wind 1 1353.8020 1352.5345 1326.5530 1316.0810
and solar to minimise the total operating cost. The capital invest-
ment, revenue generated by selling the power of wind and solar
has been determined and payback period is calculated in Tasks been considered as 4.5% increment for every year. A 20 MW wind
4, 5 and 6, respectively. farm and 10 MW solar farm have been installed in the identified
areas.
The optimal power flow is performed and the average total
7. Results and discussions operating cost per hour for 20 years has been tabulated for with
For accomplishing the objective of the present work, AP state and without renewable sources and is represented in Table 1.
system which was shown in Figure 1 has been considered. The From Table 2, It has been evident that the average running cost
fitness of the proposed method is tested with the experimental per hour is 1316.0810 $/h, which is less for wind and solar com-
setup as shown in Figure 4. binations compared to the individual case. At the same time, the
The hardware setup includes wind turbine, anemometer and average cost relieved per hour for 20 years has been calculated
data acquisition system (DAS). The DAS logs the wind speed for the three cases and a comparison has been made, as shown
available in that particular area. The P-ELM network was trained in Figure 6.
by considering the hourly data from 1st January 2021 to 31st It has been observed from Figure 6. that the cost relieved
January 2021, total 744 h data. In which 520 h data was con- is more for wind and solar combined than that of the individ-
sidered for training and the remaining 224 h data was used for ual one. The payback period is calculated using the formula
testing. The parameters in each data set contain temperature described by Equation (22):
in °C, Pressure in mbar and wind speed in m/s as inputs for the
Investment required
present instant and temperature, pressure and wind speed for Pay back period = (22)
Net cash inflow
the next instant as output parameters. Figure 5 demonstrates
the accuracy of the proposed P-ELM method. The investment in wind or solar farm includes capital cost of
From Figure 5, it is clear that the P-ELM method exactly traces plant, annual maintenance cost, insurance and cost expended
the real-time values. And also, the obtained results are validated because of unexpected failures. The net cash inflow is nothing
with the existing methods proposed in the literature, as shown but the income gained by selling the energy generated. The cost
in Table 1. of wind energy has been fixed at 0.054$/Kwh and the cost of
From Table 1, it is identified as forecasting of the wind speed solar energy is 0.277$/Kwh by the Ministry of New and Renew-
and solar insolation with the proposed method reduces the MAE able Energy (MNRE), India. Table 3 describes the income gener-
and RMSE very significantly compared to the methods which are ated and capital invested per year for wind plant. From Table 3,
already reported in the literature. it can be observed that, the payback period for wind instalment
Depending on the data obtained from NREDCAP, Anantha- is of 10 years.
pur, Kurnool, Kadapa and Chittoor have been identified as the Tables 4 and 5 describe the revenue generated and invest-
wind-rich locations and the wind speed has been forecasted for ment year-wise for solar only and wind and solar combining,
the next 20 years in those areas. Using the method of CI, Anan- respectively.
thapur, Kurnool, Kadapa and Nellore areas are identified as the Table 4 represents the payback period for solar installation.
areas of having the more solar hours. With the help of P-ELM the It is observed that, the payback period for solar integration into
solar irradiance has been forecasted for the next 20 years there grid is of 10 years. Table 5 represents the combination of wind
by calculated the solar power generation. The load growth has and solar. It is observed that the payback period for wind and
8470 M. S. SYED AND CH. V. SURESH

Figure 6. The operating cost relieved with wind, solar and wind + solar.

Table 3. Revenue vs. investment for wind. 8. Conclusion

Year Revenue obtained (million $) Investment (million $) The present work proposes a unique methodology for the instal-
1 2.0728 61.3643 lation of wind and solar farms and obtaining the payback period.
2 14.5094 66.1026
3 26.946 70.8409
At first, the areas are identified which are having highest wind
4 39.3826 75.5793 and solar resources potential. Later, the wind speed and solar
5 51.8192 80.3176 insolation have been forecasted with the help of the proposed P-
6 64.2558 85.0559
7 76.6924 89.7942
ELM algorithm for the next 20 years. Finally, the payback period
8 89.129 94.5325 is calculated with the installation of wind and solar separately
9 99.4929 98.4811 and also the combination of both. With the obtained simulation
results, it can be derived that the wind and solar combination
achieves less average operating cost per hour and more relieving
Table 4. Revenue vs. investment for solar. cost compared to the remaining two cases. While the payback
Year Revenue obtained (million $) Investment (million $) period for the wind-only case is 9 years, for solar-only case, it is
1 8.1819 67.2464
10 years, and it is 6 years for the combination of both wind and
2 16.3638 68.3769 solar. The main limitation of the proposed method is as the fore-
3 24.5458 69.5075 casting time horizon extends, the forecasting accuracy will be a
4 32.7278 70.638
big question mark. In future, the work can be further extended
5 40.9097 71.7685
6 49.0917 72.899 by incorporating FACTS devices.
7 57.2736 74.0295
8 65.4556 75.16
9 73.6375 76.2905 Disclosure statement
10 81.8194 77.421 No potential conflict of interest was reported by the author(s).

Table 5. Revenue vs. investment with the combination of wind and solar.
Funding
The authors gratefully acknowledge DST-SERB, Government of India, for
Year Revenue obtained (million $) Investment (million $) financial support under the Grant No: ECR/2016/000877.
1 6.2183 62.9437
2 22.8004 69.2615
3 41.4554 76.369
ORCID
4 58.0375 82.6867 Mahaboob Shareef Syed http://orcid.org/0000-0003-0526-8316
5 74.6196 89.0045
Ch. V. Suresh http://orcid.org/0000-0002-9155-9664
6 95.3473 96.9017

References
solar combination is 6 years. From the obtained results, it can Ali, Abid, Nursyarizal Mohd Nor, Taib Ibrahim, and Mohd Fakhizan Romlie.
2017. “Sizing and Placement of Battery-Coupled Distributed Photovoltaic
be concluded that the combination of solar and wind yields Generations.” Journal of Renewable and Sustainable Energy 9 (5): 1–8.
less payback period combined with wind and solar installation Ambach, Daniel, and Patrick Vetter. 2016. “Wind Speed and Power Forecast-
individually. ing a Review and Incorporating Asymmetric Loss.” Paper presented at
 INTERNATIONAL JOURNAL OF AMBIENT ENERGY 8471

Second International Symposium on Stochastic Models in Reliability Engi- Maklad, Yasser. 2014. “Sizing and Costing Optimisation of a Typical Wind/PV
neering, Life Science and Operations Management, Beer-Sheva, Israel, Hybrid Electricity Generation System for a Typical Residential Building in
February 15–18. Urban Armidale NSW, Australia.” International Journal of Energy Economics
Antonanzas, J., N. Osorio, R. Escobar, R. Urraca, F. J. Martinez-de-Pison, and and Policy 4 (2): 163–168.
F. Antonanzas-Torres. 2016. “Review of Photovoltaic Power Forecasting.” Mallikarjuna, A., J. C. Balachandra, and N. Manohar Potli Venugopal. 2016.
Solar Energy 136: 78–111. “Optimal Sizing of Wind/Solar/Hydro in an Isolated Power System Using
Belmili, Hocine, Mourad Haddadi, Seddik Bacha, and Boualem Bendib. SMFFT based Cuckoo Search Algorithm.” Paper presented at International
2014. “Sizing Stand-Alone Photovoltaic–Wind Hybrid System: Techno- Conference on Power Systems (ICPS), New Delhi, India, March 4–6.
Economic Analysis and Optimization.” Renewable and Sustainable Energy Maouedj, R., A. Mammeri, M. D. Draou, and B. Benyoucef. 2014. “Performance
Reviews 30: 821–832. Evaluation of Hybrid Photovoltaic-Wind Power Systems.” Energy Procedia
Bilil, Hasnae, Ghassane Aniba, and Mohamed Maaroufi. 2014. “Multiobjective 50: 797–807.
Optimization of Renewable Energy Penetration Rate in Power Systems.” Moriarty, Patrick, and Damon Honnery. 2020. “Feasibility of a 100% Global
Energy Procedia 50: 368–375. Renewable Energy System.” Energies 13: 1–16.
Bingham, Raymond D., Martin Agelin-Chaab, and Marc A. Rosen. 2016. “Fea- Page, J. K. 1961. “The Estimation of Monthly Mean Values of Daily Short
sibility of a Hybrid Solar and Wind Power System for an Island Community Wave Radiation on Vertical and Inclined Surface from Sunshine Records
in the Bahamas.” International Journal of Renewable Energy Research 6 (3): of Latitude 40 Degree N-40 Degree S.” In Paper Presented at International
951–963. Proceeding of UN-Conference on New Sources of Energy, Vol. 4, 378–390.
Chisale, Sylvester William, and Patrick Mangani. 2021. “Energy Audit and Fea- Reddy, M. B., Y. P. Obulesh, S. S. Raju, and V. Suresh. 2014. “Optimal
sibility of Solar PV Energy System: Case of a Commercial Building.” Journal Power Flow in the Presence of Generalized Interline Power Flow Con-
of Energy 2021: 1–9. troller.” International Journal of Recent Technology and Engineering 3 (2):
Engin, Mustafa. 2013. “Sizing and Simulation of PV-Wind Hybrid Power Sys- 2277–3878.
tem”. International Journal of Photo Energy 2013: 1–10. Ren, Ye, P. N. Suganthan, and N. Srikanth. 2015. “Ensemble Methods for Wind
Gomaa, Mohamed R., Hegazy Rezk, Ramadan J. Mustafa, and Mujahed Al- and Solar Power Forecasting—A State-of-the-Art Review.” Renewable and
Dhaifallah. 2019. “Evaluating the Environmental Impacts and Energy Per- Sustainable Energy Reviews 50: 82–91.
formance of a Wind Farm System Utilizing the Life-Cycle Assessment Shi, Libao, Chen Wang, Liang zhong Yao, Yixin Ni, and Masoud Bazaegan.
Method: A Practical Case Study.” Energies 12: 1–25. 2012. “Optimal Power Flow Solution Incorporating Wind Power.” IEEE
Gonzalez, Arnau, Jordi-Roger Riba, and Antoni Rius. 2015. “Optimal Sizing Systems Journal 6 (2): 233–241.
of a Hybrid Grid-Connected Photovoltaic–Wind–Biomass Power System.” Sohani, Ali, Mohammad Hassan Shahverdian, Hoseyn Sayyaadi, Siamak
Sustainability 7: 12787–12806. Hoseinzadeh, and Saim Memon. 2021. “Enhancing the Renewable Energy
Hetzer, John, David C. Yu, and Kalu Bhattarai. 2008. “An Economic Dispatch Payback Period of a Photovoltaic Power Generation System by Water
Model Incorporating Wind Power.” IEEE Transactions on Energy Conversion Flow Cooling.” International Journal of Solar Thermal Vacuum Engineering
23 (2): 603–611. 3: 73–85.
Huang, C., A. Bensoussan, M. Edesess, and K. L. Tsui. 2016. “Improvement Syed, M. S., Ch V Suresh, and S. Siva Nagaraju. 2017. “Short Term Solar
in Artificial Neural Network-Based Estimation of Grid Connected Photo- Insolation Prediction: P-ELM Approach.” International Journal of Parallel,
voltaic Power Output.” Renewable Energy 97: 838–848. Emergent and Distributed Systems 32 (6): 1–12.
Jabar, H. Y., Hussein A. Kazem, and B. John. 2017. “Predictive Models for Pho- Wang, Xiaochen, Peng Guo, and Xiaobin Huanga. 2011. “A Review of Wind
tovoltaic Electricity Production in Hot Weather Conditions.” Energies 10 Power Forecasting Models.” Energy Procedia 12: 770–778.
(7): 971. Yazdanpanah, M. A. 2014. “Modeling and Sizing Optimization of Hybrid
Jinliang, Z., W. YiMing, T. Zhong-fu, K. Wang, and Wei Tian. 2017. “A Hybrid Photovoltaic/Wind Power Generation System.” Journal of Industrial Engi-
Method for Short-Term Wind Speed Forecasting.” Sustainability 9 (4): neering International 10: 49. doi:10.1007/s40092-014-0049-7.
1–10. Yona, A., T. Senjyu, T. Funabashi, and C. H. Kim. 2013. “Determination method
Lei, Ma, Luan Shiyan, Jiang Chuanwen, Liu Hongling, and Zhang Yan. 2009. of insolation Prediction with Fuzzy and Applying Neural Network for Long
“A Review on the Forecasting of Wind Speed and Generated Power.” term ahead PV power output correction.” IEEE Transactions on Sustainable
Renewable and Sustainable Energy Reviews 13 (4): 915–920. Energy 4: 527–533.