Assessment of wind power capacity credit in Morocco: Outlook to 2020

Abstract

The present article focuses on the calculation of the wind capacity credit by integrating the Moroccan project on the wind energy of 1000 MW in 2020. After an introduction to the Moroccan Integrated Wind Energy Project, a wind capacity credit assessment program will be implemented on Matlab software including the whole information about “installed capacity, number of plants, failure rate, types of installed units, peak demand etc.” This program will be used to calculate the safety rate of an electrical system as well as the capacity credit of Morocco’s electricity production network. This section will be built in two phases: the first phase will examine the impact of TAZA wind farm with an installed power of 150 MW, while the second phase will focus on the generalization of this study on all the wind farms that will be injected to the Moroccan grid in 2020. The research provides conclusion according to comments and assessment of the impact of this electric energy integration based on the wind generation.

Keywords

Capacity credit wind energy conventional production loss of load expectation

Introduction

As part of its energy strategy, Morocco undertakes a vast wind energy program, to support the development of renewable energy and energy efficiency in the country. As a result, a very ambitious program for the development of these renewable energies has recently been adopted by the Moroccan government. Also, Morocco has an important wind resource with regions exceeding 10 m/s for the average annual wind speed (Figure 1), in particular:

The average annual speeds in towns of Essaouira, Tangier, and Tetouan are between 9.5 and 11 m/s.

Tarfaya, Taza, and Dakhla are the towns with average annual speeds between 7.5 m/s and 9.5 m/s.

Figure 1.

Map of the average wind speed in Morocco (m/s) (Kousksou et al., 2014).

A first wind map of the country showed that the northern zone (Tangier to Tetouan) and the coastal strip from Tarfaya to Lagouira present exceptional sites with regular winds and average speeds sufficient to develop profitable projects.

Thus, the Moroccan Integrated Wind Energy Project (IEP), spanning a period of 10 years for a total investment estimated at 31.5 billion Moroccan dirhams, will enable the country to increase the installed wind power capacity from 280 MW in 2010 to 2000 MW in 2020. The goal of Moroccan IEP is to install 1600 MW of new wind farms in 2020 as follows (Azeroual et al., 2018; Kousksou et al., 2014):

600 MW under development: Tarfaya (300 MW), Akhfenir (200 MW), Bab El Oued (50 MW), and Haouma (50 MW);

1,000 MW planned on five new sites chosen for their great potential: Tangier 2 (150 MW), Koudia El Baida in Tetouan (300 MW), Taza (150 MW), Tiskrad in Laayoune (300 MW), and Boujdour (100 MW) (Azeroual et al., 2018; Oumounah, 2018).

The objectives of the Moroccan IEP are not only to ensure customer satisfaction but also to protect the environment. They can be listed as follows:

Increase the share of wind energy in the total power capacity to 14% in 2020;

Achieve a production capacity of 2 GW from wind power and an annual production capacity of 6600 GWh, corresponding to 26% of the current electricity production (Azeroual et al., 2018; Oumounah, 2018);

Save 1.5 million tons of oil annually, and avoiding the emission of 5.6 million tons of CO₂ per year. This represents a no less committed to the fight against global warming (Choukri et al., 2017; Kousksou et al., 2015; Oumounah, 2018).

With a large capacity of wind energy that will be injected into the national grid, a particular importance must be attached to the impact of this insertion on the safe operation of the electrical system. In addition and since we are interested in wind power, some constraints are worth mentioning (Bloess et al., 2018):

The wind blows when it wants.

The wind blows as it wants; it is difficult to predict its exact intensity.

The wind blows where it wants and, unfortunately, it does not blow when we need a lot of electricity.

Therefore, the fact that electrical energy cannot be stored, the grid must be balanced to the nearest second (Appino et al., 2018; Bloess et al., 2018). Thus, the large-scale integration of wind energy constitutes a major challenge for the electrical system because it is difficult to stop accurate predictions of energy production. The uncertain nature of wind generation could also have an impact on the safe operation of the system. The operation reliability is considered in this case as the ability of the system to cope with the multiple risks that could disrupt its operation.

In light of all this, a study on the long-term impact of wind power production is necessary. To which level, in terms of capacity and flexibility, can wind power replace conventional power plants? To answer this question, we will use the concept of wind capacity credit (CC).

The wind CC measures the ability of wind plants to replace conventional production capacity in a given system (Castro and Ferreira, 2001; Clack et al., 2017). This definition may seem incomplete because it doesn’t include the impact of wind generation on the system reliability. For this reason, we take the proposed definition by Karki and Po (2007), which adding the influence of the intermittency of this energy on the system reliability. Thus, it is possible to define CC as the potential of wind farms to replace conventional plants without endangering the system or degrading reliability.

For CC calculation, two approaches are commonly used: chronological methods (Castro and Ferreira, 2001) and probabilistic methods (Castro and Ferreira, 2001; Clack et al., 2017; Karki and Po, 2007; Milligan and Parsons, 1999; Voorspools and D’haeseleer, 2006); the results show that the chronological approach can be very useful for the grid operator, while the probabilistic approach is especially useful for grid planning. The chronological methods provide information on the capacity of wind resources to cover peak demand, while the probabilistic methods provide information on the expected CC, obtained through statistical calculations.

Therefore, the aim of this article is to analyze the impact of the wind farms in the safety of the electrical system in accordance with an injection scenario of high wind energy. First, we will study the reliability of Moroccan electrical grid in presence of TAZA wind farm with an installed power of 150 MW and after, we will generalize this study on all wind farms that will be injected to national grid in 2020.

This article is structured as follows: In the second section, the production system reliability in the presence of wind energy is represented. In the third section, the CC calculation with the probabilistic model proposed is analyzed. In the fourth section, the calculation model of wind CC is applied to TAZA wind park. The fifth section presents a generalization on the analysis of CC for wind capacities installed on the national grid by 2020. Finally, the conclusion is summarized in the sixth section.

The reliability of production systems in the presence of wind

The goal of an electrical system is twofold. First, it consists in satisfying the demand at a reasonable cost and ensures with safety the continuity and the quality of service (Allan et al., 1989). Moreover, the determination of a safety criterion is a regular part of the operational reliability studies, namely the reliability of the electrical grid. Therefore, reliability refers to the ability of a system to perform a required function; in the case of the electrical grid, the primary function is the supply of electricity from the power plants to the final consumer (Allan and Billinton, 2000; Borges and Dias, 2017). Moreover, the ability of a system to meet the conditions established in the production system is quantified by various criteria called “reliability indices.” In the scientific literature, two methods are mainly mentioned (analytic approach and simulation approach). The analytical approach is the most used in the planning of the electric grid because it has fewer computer constraints and it is faster during its running. Thus, analytic approach will be adopted for this study.

The CC analysis

The analytical approach that can be used to evaluate the adequacy of a particular production system is composed of three parts as shown in Figure 2. In addition, the combination of the production and load model through a risk model makes it possible to determine the safety degree of a particular system. Also, the determination of the risk model consists in calculating the reliability index for covering the total demand of the electrical system. Indeed, reliability calculations are based on the calculation of probabilistic indicators at the time of peak demand. These indicators indicate the level of loss of production to cover the demand (Billinton et al., 1996; Schenk et al., 1984; Wangdee and Billinton, 2006). Furthermore, the most failure criterion used for the production planning and future capacities for the electrical system is the LOLE (loss of load expectation) index (Wangdee and Billinton, 2006). It can be described as the expected number of hours of failure during which the peak demand can exceed the available generation capacity. This indicator also gives the number of hours during which the load loss can occur (Singh and Kim, 1988).

Figure 2.

Evaluation model of the adequacy of the electricity production systems.

In our study, in order to determine the wind CC we will model the electrical system in a probabilistic way. The probabilistic model of an electrical system must make it possible to describe the behavior of the uncertain variables of the system and the correlations between these uncertainties (Figure 3). Thus, the input parameters of the probabilistic model of the electrical system as we define it give way to a set of probability distributions, each probability characterizing the variation of a parameter (Mengshoel et al., 2010).

Figure 3.

The probabilistic model for wind CC analysis.

The proposed wind CC determination algorithm presented in Figure 3 receives as inputs the characteristics of the electrical system:

The conventional production units and their default rates;

The system consumption data;

The wind profile of the site on which the wind turbines will be installed;

The power curve of the wind turbines that will be installed.

The wind CC will be determined after a comparative study of the reliability of the system without and with the presence of wind energy production. The wind CC analysis is done according to the criterion of replacement of the conventional production; this means that a unit A of capacity X MW will be removed from the system. The objective is to estimate the minimum capacity of wind power to be installed by keeping the same level of reliability.

The explanation in parametric form is as follows: When an X MW capacity is removed from the system, the safety rates $(τ_{o r i g i n a l})$ or the LOLE is affected. The new security rate $τ_{X}$ will meet the requirements of the system and cover the demand. We are therefore looking for the wind capacity $(K_{n})$ which can be installed to reach $τ_{o r i g i n a l} .$

Consequently, the credit of capacity can be calculated by the following expression

C C = X / K_{n} \times 100 %

In order to automate this computation which is expensive, we will implement the algorithm proposed by National Office of Electricity Water (ONEE; 2018) (Figure 4).

Figure 4.

CC determination algorithm.

Conventional production model

The model of the production system is defined as the sets of power generation Units available in a production park. Each unit is characterized by its failure rate. The unavailability rate (forced outage rate (FOR)) is the probability that a unit will be unavailable at a given time.

At the end of 2016, Morocco’s national production plant, with an installed capacity of 11,278 MW, is composed of thermal, hydraulic, interconnection plants, and also the local production of certain dealers. The details of the park and the FOR are given in Table 1.

Table 1.

Power installed on the national power grid (National Office of Electricity Water, 2018).

Power stations	Installed capacity (MW)	FOR
Hydraulic plants	1306	0.01
Pumped-storage	464	0.01
Coal	4146	0.04
Fuel oil	600	0.05
Gas-turbine	1230	0.02
Combined cycle	850	0.04
Diesel thermal	202	0.1
Nuclear	2480	0.12
Total national network	11,278

FOR: forced outage rate.

The installed capacity increased to 12,813 MW at the end of December 2017 and to 12,983 MW at the end of December 2018 against 11,278 MW in 2016, an increase of 15.11% due to the commissioning of the two coal groups of Safi (2 × 660 MW), Jerada coal group (350 MW) followed by the decommissioning of the existing power plant groups (–135 MW) in 2017, and the commissioning of the M’dez and El Menzel hydropower plants (170 MW) in 2018 (Azeroual et al., 2018).

Load model

The load model is acquired with the construction of the monotone consumption for a given period corresponding to 1 year (8760 h or 365 days; Figure 5).

Figure 5.

The monotone of load in 2020.

To construct the monotone of consumption or the cumulative model of load, it is necessary to have the information of the hourly or daily consumption of the system which we will have to classify in descending order, and then we can be able to calculate the weight of each state of consumption.

In the medium term, demand forecasts mainly reflect the needs of industrial customers and private electricity distribution companies in the main cities of the country. These forecasts also cover the assessment of the distribution market in Morocco, which mainly corresponds to rural areas. Thus, the expected growth rate for 2010–2013 is about 6%. Moreover, the evolution of electricity consumption results from the combination of factors from different reasons: economic activity, demography, user behavior, technical progress, development of new uses of electricity, market shares between energy sources, energy control actions, and so on. The basic scenario, relying mainly on economic reforms, focuses on an evolution of 5.82% over the period 2015–2020. This emerging scenario, used as the reference scenario for the development of the Moroccan grid equipment plan, translates a 6% long-term growth (Coopération République du Sénégal—BAD, 2011).

Probabilistic model (risk model)

Once the production and load models are founded, a convolution of these two models is necessary for the risk model establishment that quantifies the risk of the system. This method leads to a possible determination of the reliability indices, which is the risk of loss of load mentioned above. In this study, we will work with the LOLE index, which is the mathematical expectation of failure hours (the number of hours per year during which the available production resource is not sufficient to cover the entire demand) (Yang et al., 2018).

This index can be represented as follows

L O L E = \sum_{i = 1}^{n} P_{i} (C_{i} - L_{i})

where $C_{i}$ denotes capacity available for day i, $L_{i}$ denotes consumption forecast for day i, and $P_{i}$ is the probability of loss of load for day i.

The LOLE is the most widely used index for determining the production capacities required in the medium- and long-term horizons. The reliability level of the national park is determined after the running the MATLAB program; this helps us to calculate the LOLE reliability index, as presented in this chapter. Table 2 shows the result obtained after program simulation on the Moroccan National Grid (MNG).

Table 2.

MNG reliability level.

System	LOLE (simulated)
MNG	0.0413 (h/year)

MNG: Moroccan National Grid; LOLE: loss of load expectation.

Application to the TAZA wind farm

Wind production model for the TAZA park

First, it is necessary to create the unavailability table of the intermittent production. Then, the unavailability of the intermittent production model must be integrated into the unavailability table of the conventional production model. This subsequently allows the convolution between the production model and the cumulative load model to be made, and also to determine the risk level of the system. The safety studies in Karki et al. (2006) and Catalão (2016) present an approach using a specific time series model called “Auto-regressive moving average” to predict a series of wind speed data in a particular region. This approach is difficult to implement if the access to a complete database of wind speed of a particular site is impossible. In our study, we choose to model the distribution of wind by the normal distribution—getting as parameters the mean μ and the standard deviation σ at the site where the wind turbines are installed and the number Nb of wind samples that it is desired to simulate (Karki and Po, 2007). Also, it is necessary to inspect the power variation of each installed turbine, according to the wind speed (Yang et al., 2018).

In order to establish the model of wind generation, we will combine the data of the wind resource on the TAZA park with the wind turbines installed there.

The distribution of wind at the TAZA site is represented in Figure 6.

Figure 6.

Wind distribution in the TAZA park.

For this probability distribution, we will consider wind speeds ranging from 0 to 10σ to account the extreme wind variables. The distribution is divided into Nb intervals, with each interval having a length of 10σ/Nb.

In our study, the wind resource data is the mean and the standard deviation. The Taza wind distribution requires a mean μ = 9 m/s and standard deviation σ = 2.79 m/s.

Capacity credit calculation

Based on the previous study, we are going to examine the impact of the TAZA park integration to the national electric grid. In addition, the new safety rate will be calculated and interpreted. Also, the obtained results are represented in Table 3.

Table 3.

Reliability of the hybrid park (conventional + wind).

System	LOLE
National grid + TAZA wind park	0.0398 h/year

LOLE: loss of load expectation.

Subsequently, it will proceed to the calculation of the CC granted to the national electric grid following the introduction of 150 MW of wind energy. The addition of 150 MW of wind in the mix of national production has an impact on improving the safety rate of the electrical system; in effect, the safety rate has increased from 0.0413 h/year to 0.0398 h/year.

In order to analyze the CC of the wind park, we will proceed by the same way as presented above; we will remove a certain capacity of thermal energy and replace it by the wind turbines. Furthermore, using the Matlab program already implemented, we will simulate the different scenarios of existing capacities removed from the national park.

The plotted curves in Figure 7 are obtained after having to remove simultaneously 30 MW and 33 MW of conventional production capacities; for each case, the simulation was run to calculate the LOLE in function of the injection of [0, 200] MW wind capacity. According to the curves and after the withdrawal of the two conventional capabilities, the LOLE was degraded. However, once the wind parks were injected, the LOLE marked significant improvement.

Figure 7.

Safety ratio according to the injected wind capacity for 30 MW and 33 MW withdrawn.

For those different tests, the TAZA wind farm with its installed capacity of 150 MW may be used to replace only 30 MW units, with the wind power required for the replacement is 135 MW that is below 150 MW of the park. As regards the to replacement of 33 MW units or more, the number of failure hours in the national grid will increase from 0.0413 to 0.0416 (Figure 7). Thus, the TAZA park does not achieve the required level of reliability for the national grid.

The results are summarized in Table 4.

Table 4.

The wind CC of the park of TAZA.

Capacity removed (MW)	Installed wind capacity (MW)	Capacity credit (%)
30	135	22.22
33	160	20.62

CC: capacity credit.

Generalization on the analysis of CC for wind capacity installed on the MNG by 2020

In the previous section, we have studied in detail the integration of TAZA wind farm impact on the safety rate of the national grid, and the conventional production capacity which can be removed without impacting the safety parameters of the grid. In addition, the Moroccan Government envisage the injection of a significant amount of wind parks by the year 2020. Moreover, considering the continuous increase of the annual load consumed at the level of the country, we will go on to consider the impact of all these parks on the national grid in terms of overall operational safety and in terms of the total capacity that can be removed safely.

The Moroccan IEP which consists of injecting a total power of 2000 MW in wind energy before 2020 will contribute certainly in

Development of Morocco’s potential in renewable energy;

Optimization of fossil source production;

Postpone the realization of the new conventional power plants of production;

Improve the safety of the electrical network;

Improved power quality of electric power served to customers;

Face the perpetual increase of the annual consumption;

Winning in the emission of greenhouse gases and contribute to saving our planet against climate change due to greenhouse gases.

The Moroccan IEP is distributed as follows:

280 MW already achieved and in service;

720 MW in development by the private sector;

1000 MW to be built on five new sites, and which will be commissioned between 2018 and 2020 according to the program in Table 5.

Table 5.

The wind capacities planned to be installed in Morocco before 2020.

Wind farm	Installed power (MW)	Estimated date of commercial exploitation
MIDELT	153.6	30 June 2018
TISKRAD	297.6	30 December 2018
TANGER	99.20	30 June 2018
JBEL LAHDID	201.30	30 June 2019
BOUJDOUR	99.00	30 June 2020

The previous approach to studying the impact of the integration of the TAZA wind farm will be used to determine the operational reliability rate of the power grid, and to calculate the CC. Furthermore, the approach will be used to calculate the CC of wind farms planned to be installed on the national grid before 2020.

In a concern to obtain reliable and significant results, the already implemented program is made annually by certain elements, in order to make the assumptions of its functioning closer to reality, particularly:

The growth rate of demand for energy consumed is about 5%.

Complete the Matlab program on existing production facilities by means of conventional production facilities which are scheduled to be commissioned before 2020 as shown in Table 6.

As regard the year of study, the wind farms already in operation are well declared in the Matlab program on existing means of production.

Table 6.

The conventional capacity installed in Morocco by 2020.

The power stations	Installed power (MW)	Estimated date of commercial exploitation
Coal (Safi)	2 × 660	31 December 2018
Coal (Jerada)	350 – 135	31 December 2018
Hydraulic (M’dez and El Menzel)	170	31 December 2019

The scenarios decided for the simulation work are as follows:

The Scenario to add wind capacities of 100, 150, 200, 300, and 1000 MW each time;

The Scenario to remove from the park of conventional production the respective capacities of 33, 39, 42, 75, 95, 100, 111, 143, 153, and 170 MVA.

The two scenarios were made in order to simulate the wind farms needed to be added before 2020 and to verify the traditional production capacities which can be withdrawn each time following the new integration of the parks.

The objective of this work is to

Calculate the LOLE for each scenario;

Calculate the wind capacities required to substitute the conventional capacities removed;

Calculate the CC for different scenarios.

The various simulations realized for the calculation of the safety rate (LOLE) of the electrical system according to the connection of the new wind farms planned before 2020 are represented in Table 7, showing that as long as the part of the wind energies increase in the national energy mix, the LOLE improves advantage, since its variation is almost linear, as represented in Figure 8.

Table 7.

The LOLE (h/year), CC, and the IWC for each conventional capacity removed.

The conventional capacities removed (MW)	Injected wind capacities (MW)					CC (%)	IWC (MW)
The conventional capacities removed (MW)	150	300	700	900	1000	CC (%)	IWC (MW)
33	0.0417	0.0393	0.0329	0.0297	0.0281	20.62	160
39	0.0430	0.0406	0.0342	0.031	0.0294	19.02	205
42	0.0434	0.0411	0.0346	0.0314	0.0301	20	210
75	0.0454	0.0430	0.0366	0.0334	0.0318	23.07	325
95	0.0470	0.0447	0.0382	0.035	0.0335	23.17	410
100	0.0485	0.0455	0.039	0.0358	0.0343	20.61	485
111	0.0499	0.0471	0.0399	0.0367	0.0351	19.64	565
143	0.0538	0.0508	0.0437	0.0407	0.0383	19.32	740
153	0.0558	0.0530	0.0457	0.0421	0.0403	17.58	870
170	0.0587	0.0555	0.0479	0.0449	0.0418	16.58	1025

LOLE: loss of load expectation; CC: capacity credit; IWC: injected wind capacity.

Figure 8.

Safety rate (LOLE) according to the injected wind capacity.

The injection of renewable energies based on the wind turbine makes it possible to increase the safety rate of the electrical network in a normal situation and a safe environment.

Figure 9 shows the simulation results for various scenarios realized in Table 7.

Figure 9.

The capacity credit and the wind capacity injected according to the removed conventional capacity.

From the results obtained in Figure 9, it can be concluded that

The advantageous injection of wind energies increases the reliability of the electrical network in an almost linear way.

In 2020, a scenario is envisaged to substitute a conventional power of about 160 MVA without impacting the safety level of the electrical network. This power is equivalent to two groups of 80 MVA about “2 × 80 MVA”.

The CC increased until the end of 2017 and subsequently decreased to reach a value in 2020 of 16.58% against a value of 23.17% in 2017. This means that if the wind energy represents a significant part of the mixing energy, we can say that the CC of the wind energy is reduced.

In the term of this study the Moroccan IEP which aims to achieve a part of wind power of 2000 MW in the national mix energy presents several advantages:

Allow benefiting from an important level of operating safety of the electrical network;

Avoid the power supply interruptions in the event of an incident in the grid;

Allow maintenance operations on conventional production plants, which generally take a long time;

Benefit from the electrical energy generated from renewable sources with an interesting cost price;

Adequacy in primary, secondary, and tertiary reserves, this refers to the capacity of the electrical system to cope with any imbalance, for example, in case of an unforeseen change in demand or in case of a loss of production.

Conclusion

The reliability study conducted in this article has demonstrated that wind energy can contribute to system safety, and replace a part of installed conventional capacity measured by the CC. The CC is estimated in this article basing on the probability of the national grid reliability, which becomes more complex when the system dimension is important. For this reason, we relied on the existing literature about the reliability and the operation safety, using existing models to deepen them with different parametric criteria. It has also been observed that the CC is reduced linearly according to the injection rate of the wind turbine and is effectively dependent on the conventional power stations considered. Consequently, in a perspective of massive integration of wind power, conventional production will have to be put in place to mitigate the deterioration of CC and preserve the reliability of the system.

Footnotes

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Yassir El Karkri

References

Allan

Billinton

(2000) Probabilistic assessment of power systems. Proceedings of the IEEE 88(2): 140–162.

Allan

Billinton

Sjarief

, et al. (1989) A reliability test system for educational purposes-basic data. IEEE Transactions on Power Systems 4(3): 1238–1244.

Appino

Ordiano

JÁG

Mikut

, et al. (2018) On the use of probabilistic forecasts in scheduling of renewable energy sources coupled to storages. Applied Energy 210: 1207–1218.

Azeroual

El Makrini

El Moussaoui

(2018) Renewable energy potential and available capacity for wind and solar power in Morocco towards 2030. Journal of Engineering Science and Technology Review 11(1): 189–198.

Billinton

Chen

Ghajar

(1996) A sequential simulation technique for adequacy evaluation of generating systems including wind energy. IEEE Transactions on Energy Conversion 11(4): 728–734.

Bloess

Schill

Zerrahn

(2018) Power-to-heat for renewable energy integration: A review of technologies, modeling approaches and flexibility potentials. Applied Energy 212: 1611–1626.

Borges

CLT

Dias

(2017) A model to represent correlated time series in reliability evaluation by non-sequential Monte Carlo simulation. IEEE Transactions on Power Systems 32(2): 1511–1519.

Castro

RMG

Ferreira

LAFM

(2001) A comparison between chronological and probabilistic methods to estimate wind power capacity credit. IEEE Transactions on Power Systems 16(4): 904–909.

Catalão

(2016) Electric Power Systems: Advanced Forecasting Techniques and Optimal Generation Scheduling. Boca Raton, FL: CRC Press.

10.

Choukri

Naddami

Hayani

(2017) Renewable energy in emergent countries: Lessons from energy transition in Morocco. Energy, Sustainability and Society 7: 25.

11.

Clack

Qvist

Apt

, et al. (2017) Evaluation of a proposal for reliable low-cost grid power with 100% wind, water and solar. Proceedings of the National Academy of Sciences of the United States of America 114: 6722–6727.

12.

Coopération République du Sénégal-BAD, (2011) “Étude d’interconnexion des réseaux électriques Sénégal – Mauritanie – Maroc - Espagne,” Banque africaine de développement, Département par pays region ouest B (ORWB), Bureau regional de la Banque au Sénégal (SNFO).

13.

Karki

(2007) Impact of wind power growth on capacity credit. In: 2007 Canadian Conference on Electrical and Computer Engineering (CCECE 2007), Vancouver, BC, Canada, 22–26 April. New York: IEEE.

14.

Karki

Billinton

(2006) A simplified wind power generation model for reliability evaluation. IEEE Transactions on Energy Conversion 21(2): 533–540.

15.

Kousksou

Allouhi

Belattar

, et al. (2015) Renewable energy potential and national policy directions for sustainable development in Morocco. Renewable and Sustainable Energy Reviews 47: 46–57.

16.

Kousksou

Bruel

Jamil

, et al. (2014) Energy storage: Applications and challenges. Solar Energy Materials and Solar Cells 120: 59–80.

17.

Mengshoel

Chavira

Cascio

, et al. (2010) Probabilistic model-based diagnosis: An electrical power system case study. IEEE Transactions on Systems, Man and Cybernetics-Part A: Systems and Humans 405: 874–885.

18.

Milligan

Parsons

(1999) A comparison and case study of capacity credit algorithms for wind power plants. Wind Engineering 23: 159–166.

19.

National Office of Electricity Water (ONEE) (2018) Available at: http://www.one.org.ma/

20.

Oumounah

(2018) Intégration des énergies renouvelables dans les systèmes électriques nationaux. ONEE-COP22. Available at: http://www.cop22.org/fr

21.

Schenk

Misra

Vassos

, et al. (1984) A new method for the evaluation of expected energy generation and loss of load probability. IEEE Transactions on Power Apparatus and Systems PAS-103(2): 294–303.

22.

Singh

Kim

(1988) An efficient technique for reliability analysis of power systems including time-dependent sources. IEEE Transactions on Power Systems 33: 1090–1096.

23.

Voorspools

D’haeseleer

(2006) An analytical formula for the capacity credit of wind power. Renewable Energy 31(1): 45–54.

24.

Wangdee

Billinton

(2006) Considering load-carrying capability and wind speed correlation of WECS in generation adequacy assessment. IEEE Transactions on Energy Conversion 21(3): 734–741.

25.

Yang

Wang

Sangwongwanich

, et al. (2018) Design for reliability of power electronic systems. In: Rashid

(ed.) Power Electronics Handbook (4th edn). Oxford: Butterworth-Heinemann, pp. 1423–1440.