Design and evaluation of PV-wind hybrid system with hydroelectric pumped storage on the National Power System of Egypt

1 Introduction

In Egypt, the daily demand for electricity significantly fluctuates and the late evening peak demand is one and half times more than the demand of the off-peak hours.Thus, the development of generation facilities to serve the peak demand is an important requirement in the Egyptian Power System. According to the Long Term Generation Expansion Plan of the ministry of electricity and energy of Egypt, a large amount of renewable energy generating capacities will be connected to the power system in near future. In the early period of last two decades,the ministry of electricity and energy of Egypt and the authority of renewable energy noticed that conventional energy sources would not be able to cover the demands for electric energy in the early future. By 2022, Egypt plans to cover 20 percent of the demand for energy with nonconventional energy sources (12% wind energy, 5.8%hydroelectric power and 2.2% solar energy) [1, 2]. The power generated by Wind Turbine Generators is cubically dependent on wind speed, as shown below in (1), which makes the wind power an oversensitive source of electrical energy [3].

where PWT represents the output power from the wind turbine (W), Cp is the coefficient of performance, ρ is the air density (kg/m3), A is the swept area of the wind turbine rotor blades (m2), and U is the wind speed normal to the hub of the wind turbine (m/s). Similarly, other forms of RES, such as photovoltaic power, are susceptible to fluctuations of solar radiation and environmental temperature. Because of the stochasticity and irregular nature of renewable energy [4, 5], as well as fluctuations of load, energy storage systems must be designed in high penetration systems of these volatile sources. The function of energy storage systems can be simply defined as capturing energy produced at one time and used later. The stored energy can be found in form of gravitational energy in reservoirs, chemical energy in the batteries, pressure energy in the form of compressed air, and thermal energy in form of molten salt.

Battery storage is one of the most widely used forms of energy storage in form of chemical energy. The lifetime of the battery is highly affected by discharge level. Since batteries are constituted by chemicals, their performance,cost and lifetime are highly affected by the way and conditions under which they are used.

Generally, 60-80% storages in grids are batteries.Rather than grid storage, the battery storage is more suitable for power system stability. Their main advantage is their short response time (seconds milliseconds). They enhance the security and reliability of the power system.They are extremely useful to provide support for shortterm active power imbalances (seconds to hours) that cause the system frequency to diverge from its load. The main drawback at present of the battery storage when used in utility grid storage applications is their lower lifecycle as compared to other sources of energy storage. Other drawbacks are reduction of capacity with life, high capital cost, high environmental impact and short backup duration.The above drawbacks make them uncompetitive to pumped storage.

Molten salt can be utilized as a thermal energy storage method to keep thermal energy acquired by a solar tower or solar trough of a concentrated solar power plant. The typical molten salt extended mixture includes sodium nitrate, potassium nitrate and calcium nitrate. It is nonflammable and non-toxic and has already been used in chemical and metallurgical industries as the thermal fluid transport, therefore experience with such energy storage systems exists in non-solar applications. The main drawback of the thermal storage, when used in utility grid storage applications is their overall lower efficiency (only 15%) as compared to other sources of energy storage[8]. Other drawbacks are high capital cost as we have to build a thermal power plant to convert thermal energy into electricity as molten salt is used as fuel source to heat water. The above drawbacks make them uncompetitive to pumped storage.

A pumped storage project has an upper reservoir to store water using surplus energy during off-peak hours and a lower reservoir to which the water is drained back generating electricity during peak hours. The pumped storage schemes (PSS) could serve as a bulk storage application. The efficiency of this system is usually from 70% to 85%, which makes it one of the most effective ways of storing energy [9]. The size of the reservoir depends upon the capacity and time period for which the energy is to be stored. Unlike fossil/nuclear boiler plants, they can quickly ramp up and down, and achieve full load in matter of seconds or minutes (for very high capacity). PSS technology is the ideal tool to improve the use of variable generation for a long time.

When discharging these units during the period of peak demand, the system is better exploited. With the increasing uncertainty in generation, it is important to have the flexibility to ensure supply security. Lower maintenance and operation costs, as well as high reliability, are also important. The lifespan of the mechanical equipment of hydroelectric plants is typically 40 to 50 years, but hydro plants themselves may have a lifespan of a hundred years.

2 Sea-water pumped storage hydroelectric power plant

Sea water pumped storage is a modified form of pumped storage technology. These can be built near sea coasts with a high mountainous coastline. Unlike the normal pumped storage it has only one reservoir (upper)instead of two (upper and lower). This reduces its cost appreciably. Further, it has no problem of water shortage as sea water is always available. The water from upper reservoir is carried to turbine in power house through penstocks and discharged to sea to generate electricity during peak hours. During off-peak hours, the turbine will be operated in pumping mode in reverse direction and the sea water is pumped to upper reservoir. This round cycle efficiency of generating electricity during peak hours and pumping the water back to upper reservoir is around 80-85%. In case of sea water pumped storage scheme,as sea itself acts as the lower reservoir, the construction investment of lower reservoir is not required. However, to avoid any damage to the ecosystems due to high volume water discharge into sea, proper energy dissipation system is required to be constructed [10].

A detailed comparison of various storage technologies is presented in Table 1 [11, 12]. As seen from Table 1,except for low energy density due to weak gravimetric field, pumped storage is quite suitable for energy storage.

3 Main components of the proposed hybrid system

A typical structure of the proposed renewable energy system with a sea water pumped storage system including the basic parameters and the direction of power flow is shown in Fig.1. The intergration of RES into the power system makes its management very complicated. That is because the output energy yielded from such systems like PV and WT mainly depends on the solar radiation and wind speed, which can only be predicted to a very small degree of accruacy. Storage system is used to store excess energy during the day and generate electricity at night. The daily functioning of such a system must be compatible with the daily profile shown in Fig.2. A dual-penstock hydraulic unit helps in the voltage regulation and frequency stability with appropriate control strategies. This article uses a storage system with a dual-penstock seawater pumping system instead of a single penstock pumping system [13, 14].

Table 1 Comparison between different types of energy storage [11, 12].

Battery(Advanced Lead Acid)Battery(Na-S)Battery(Li-Ion)Molten Salt(Thermal Energy δT= 250 Deg C)Pumped Storage(100 m head)Energy Density(Practical) kJ/kg 200 kJ/kg 400 kJ/kg 540 kJ/kg 400 kJ/kg 1 kJ/kg Overall Efficiency 80%(charge-discharge)80%(charge-discharge)80%(charge-discharge)15% (Solar to Electricity)80% (from Utility)Life Span (Utility Scale) 10-15 years(4500 cycles- 33-80%DoD)15 years (4500 cycles- 80% DoD)10-15 years(4500 cycles- 85%DoD)30 years 30-50 years Technical Maturity Moderate High Moderate Low Very High Scale of Installation Few Kw – 30 MW Few Kw – 30 MW Few Kw – 30 MW 1 MW – 200 MW Few Kw-350 MW Usual Storage Duration 30 mins-6 hrs 30 mins-6 hrs 30 mins-4 hrs 2 hrs-12 hrs 2 hrs-24 hrs Initial Installation Cost/kw [8,9]$1800- 6000/kw $3000/kw $2000-3500/kw $2500/kw(Excluding cost of Solar plant)$1500-2500/kw Cost per Kwh installed Cap.[8,9]Levelised Cost of Energy(LCOE)/kwh$200-1000/kwh$440-512/kWh$400-500/kWh—$60-150/kWh$0.2-0.4/kwh$0.26/kwh$0.2-0.35/kwh—$0.1-0.2/kwh Environment hazard Yes Yes No No No Response Time Milisec Milisec Milisec Few mins-hour Few Mins Reliability High High High Moderate High

Fig.1 The fanciful design of the proposed hybrid power system with the flow of energy

Fig.2 Potential contribution from different sources in covering the demand for energy

The main objects of this work are:

• To formulate a mathematical model describing the parallel operation of a hybrid energy system consisting of wind turbine, PV and double penstock sea-water pumped storage hydropower plant integrated with electric utility.

• To introduce an optimized model results in minimizing the impact of such a system on the national grid by selecting the optimum parameters of the hydropower plant.

• To investigate and introduce the degree of responsibility of each parameter on the value of the energy exchanged with the grid (surplus and deficit).

3.1 Wind turbine

The wind turbine’s power output mainly depends on the wind speed in the selected site and the characteristics of the selected wind turbine described by the turbine’s power curve. In this study, the power curve of the wind turbine T600-48 is considered. The turbine starts operation when the wind speed normal to the hub exceeds 3 m/sec (Ucutin). When the wind speed exceeds 12.5 m/sec (Urated),the turbine starts to generate its rated power (600 kW) and when the speed reaches 25 m/sec (Ucut-off) the turbine is shut down for safety reasons. The power curve of the proposed turbine for this study has been presented in Fig.3.

Depending on the fundamentals of wind energy discussed above, the energy obtained from the wind turbine can be calculated as follow [15]:

Where: EWT – energy generated from wind turbine (kWh),U – speed of wind (m/s), n – number of turbines, PWT–output power rated of wind turbine (kW), t – time (hour).

Fig.3 The shape of the wind turbine T600-48 power curve

3.2 Solar photovoltaic array

The solar cell is the main component of the photovoltaic array, which is connected in series and / or parallel to form PV modules. In general, a typical module will contain 24/72 cells connected in series. The to form a PV array the PV models are connected in series and parallel. As the long-term temperature and humidity for different sites in Egypt agree with the SOC of the mono-Si-HIP-200BA3 PV-module from Sanyo, these PV modules are selected in this study [16]. The output power from the photovoltaic system is affected by the intensity of radiation, operating temperature and the degree of cleanness of the panels.

The energy yield from the PV array can be expressed as[17, 18, 19]:

where EPV is the energy yield (kWh); nPV is the number of photovoltaic panels in the system; Irad(t) is the ambient solar radiation intensity (kWh/m2); ISTC is the intensity of solar radiation under standard operating conditions (kWh/m2); PPV is the installed capacity of the PV panels under standard operating conditions (kW); ηPV is the efficiency of the system, which is related to the operating temperature and cleanness of panel.

3.3 The pumped storage system

Water flow is the volume of water passing through a turbine (m3/s). In general, the energy yield from a hydroelectric power station is calculated using the following formula [20]:

Where, EH – generated energy from water turbine in(kWh), ηH – water turbine and generator efficiency,ρ – sea water density in (kg/m3), g – Gravitational acceleration in (m/s2), Q – water discharge through the turbines in (m3/s), h – effective head (height difference)in (m).

3.4 The operation of the PW/wind/pumped storage hybrid system

The operation of the hybrid power supply is determined by the values of the energy yield from the PV and wind turbine compared with the values of the load demand. Five modes of operation are described as follows:

Mode 1: If the energy generated from the PV and wind turbine is greater than load demand, the extra power has to be raised to the upper reservoir of the hydropower system.

Mode 2: If the energy supply exceeds the demand for energy and the upper reservoir is fully occupied with water,the surplus power will be sent to the national grid.

Mode 3: If the energy supply is equal to the energy demand, there is no need to store energy in the form of water potential or to send energy to the grid utility.

Mode 4: If the energy generated from the PV and wind turbine systems is less than the load demand, some water has to be discharged to cover the load.

Mode 5: If the energy supply is less than the load demand and the stored energy is insufficient to cover the desired load, the deficit energy will be drawn from the national grid.

The operating mode of the HPSP will be determined by a new variable called the balance of energy, which is calculated from the following equation:

where EB is the balance of energy (kWh), and ED is the load demand (kWh).

As mentioned above, the value of EB determines the mode in which the hydroelectric facility will act. The negative value of EB indicates that some water needs to be discharged from the upper reservoir and the system is working in the generating mode. The positive value of EB indicates that the system will be operated in the pumping mode to store the surplus of energy in the upper reservoir.

For the purpose of simplification in programing the upper reservoir is assumed to have a cube shape.The energy storage potential depends on the volume of stored water:

where, a, b, and h2 are the dimensions of the reservoir in (m).

The length of water column is not constant during the operating period and is affected by the volume of water stored or discharged from the reservoir during the period of simulation. In the first period of simulation, the volume of stored energy is assumed to be half of the maximum capacity of the reservoir.

During the simulation period, it is assumed that the amount of stored water must not be less than the minimum allowable value Vmin. The reserved energy in the tank will be sent to the grid in emergency situations related with a fault occurrence or the outage of a power plant. The energy reserved is taken as a percentage of the maximum capacity of the upper reservoir. Amount of water stored in the tank at any time in the year is calculated from the following equations:

where Vmax is the maximum capacity of the upper reservoir(m3), Qdis is the volume of water discharged from the tank and used to cover the load demand during the periods of low wind and solar energy (m3), Qpump is the volume of water raised to the reservoir from the sea water using the excess of energy from solar and wind power plants (m3).

During the period of pumping and discharging, the head of the column of water changed with time due to the additional head, which results from the occupation of the upper reservoir, besides the main head of the hydropower.

From equation (5), when the value of energy balance EB<0, the system will be operated in the generating mode and the electric energy generated from the generatorturbine set is calculated from the following formula:

where QT is the discharge water speed of the turbine (m3/s), h3 is the mean head of the hydropower system, ηWP is the pipeline conveyance efficiency, and ηT is the generator efficiency. As a result, from the previous equation, the volume of water discharged during that period of time is given in the following equation:

The hybrid system naturally will not be able to satisfy the demand for energy all time. In some cases, we need a support from the grid to cover the demand of energy.The energy deficit in this case is calculated from the next equation:

Similar to equation (9), when the value of the energy balance EB>0, in this case the share of energy surplus is used to raise water from the sea to the upper water tank. In this scenario, the electric energy drawn from the PV and wind turbine, and consumed by the pump system at any period of time is calculated from the following formula:

where QP is the pumping speed (m3/s), and ηP is the efficiency of the pumping unit. Based on the previous equation the amount of water raised to the upper reservoir during a certain period is given in the following formula:

If the hybrid system covers the demand for energy and there is no probability to store additional water in the upper tank, the excess energy given to the grid is calculated based on the following equation:

4 The impact of the hybrid system on the national grid

In this work, the impact of varying the capacities of the pump used in the case of excess energy, the capacities of the turbine-generator set used in case of shortage of energy supply, and the potential of the energy storage represented by the volume of the tank, collaboratively defines the scenarios of energy exchange with the national grid.Generally, 150 scenarios are examined in this study, which result from the permutations of the previously mentioned parameters. The considered parameters are:

• The capacity of the energy storage reservoir of 75,150, 225, 300 and 375 MWh which will cover five hours of average energy demand.

• The turbine maximum discharge rate of 20, 40, 60, 80 and 100 m3/s.

• The pump maximum pumping capacities of 20, 40,60, 80 and 100 m3/s.

The goal of the optimization process is to find the merging between the maximum volume of the reservoir,the maximum turbine discharge rate and the rated capacity of the pump, which will lead to the minimum exchange with the national grid represented by the lowest sum of the energy surpluses and deficits. The objective function is given in equation (15).

5 Results and discussion

For this study, the input data was obtained on energy demand [1, 2], wind speed and solar irradiation [2, 21].Fig.4 summarizes the statistical parameters and annual variability of the data input. Demand for energy and meteorological parameters of climate are characteristics of Ataka-Suez city, which lies on the shore of the Red Sea(latitude 30.0, longitude 32.5).

The balance of energy calculated from the application of (5) over the period of study is shown in Fig.5. The energy balance in winter and summer days are shown in Fig.6 and 7, respectively. The balance of energy is the start point in this study, which determines the mode of operation of the hybrid system.

Fig.4 Annual variation of the input data used in simulation study; (a) Win speed, (b) Solar irradiation,(c) Energy demand

Fig.7 The balance of energy over 24hours of a certain day in summer

Fig.6 The balance of energy over 24hours of a certain day in winter

Fig.5 The balance of energy over the period of study

The pumped storage hydropower plant in general includes two water tanks, but in this case, the lower reservoir is not taken into account as the sea water is considered as the lower reservoir. When excess energy is required, the water used to generate electricity flows through the penstock and drives water turbines. The water is raised back from the sea to the upper tank, when the pumps are run to store excess energy from wind and/or solar power generation. Elevation difference in between the upper reservoir and the seawater level suggested in this work is h3=100 meters and the additional head resulted from the change of amount of water stored in the upper tank is h2=10 meters. Upper reservoir dimensions are proposed as follows: length = 500 meters,width = 200 meters. Then the total volume Vmax of 1000 000 cubic meters. In this study the efficiency of the pump is 90% and for the generator-turbine set is 80%.The maximum capacity of the upper reservoir is 375 MWh, which equals to 5 hours of average load demand.The hourly power generation of such a water turbine is measured by its maximum discharge capacity and efficiency curve, turbine and pump capacities impact on the system configuration will be studied in the following sections.

The demand for energy is time variable so that the ability to predict changes in energy demand with high degree of accuracy is important for power system operator. Over the past decades, many scientific papers have been devoted to the development of new methods of forecasting the demand for energy in the short, medium,and long term [22]. As a result of the application of variable sources of renewable energy, part of the supply began to oscillate remarkably. Similarly, the concept of predicting energy production, for example wind turbines or photovoltaic power plants, has been studied to address the problem of the random nature of renewable energy sources. When a group of consumers using renewable sources without any form of energy storage, their energy needs typically demonstrate a large variability in energy exchange with the grid. As a result, it would be more difficult from the perspective of the grid operators to perfectly plan the work of traditional plants. In the case of the energy surplus is close to the volume of energy deficits, the decisive function of energy storage system becomes obvious. In this case, the energy store could save the available surplus energy and release it when less favorable conditions of solar radiation and wind speed occur. The variability of the energy exchange is visualized in Fig.8. In the following analysis of the hybrid power source configuration shown in Fig.8, it will be used as the reference scenario (S1). The amount of energy exchange with the national grid in a summer and winter day is shown in Fig.9 and 10, respectively.

To find the optimal configuration of the upper reservoir volume, the capacities of the pump, and turbine-generator set, an additional 150 scenarios beside the base benchmark scenario are mentioned above. Table 2 summarizes the configurations of 8 scenarios, which will be considered in the following sections.

Table 2 Configuration parameters of the selected scenarios

Sc. PV[MW]Vmax[m3] Qpump QT Wind[MW]Z[GWh]S1 40 20 0 0 0 315.5 S2 40 20 200000 20 60 222.8 S3 40 20 400000 20 100 179.6 S4 40 20 400000 100 60 141.9 S5 40 20 600000 100 60 86.83 S6 40 20 800000 60 40 53.14 S7 40 20 1000000 40 60 41.76 S8 40 20 1000000 80 60 24.10

Fig.8 Hourly exchange of energy with the grid over the year

Fig.9 Hourly exchange of energy with the grid over a certain winter day

Fig.10 Hourly exchange of energy with the grid over a certain summer day

It is obviously seen in Fig.11, that in the reference scenario (S1) energy surpluses exceed 10% of the load demand for energy, while over 60% of demand is obtained from renewable sources of energy. In other scenarios, the percentage of excess energy is much lower and does not exceed 5.18% of the load demand, with a minimum value of just over 0.2%. In all scenarios, the annual sum of the demand for energy was 650.34 GWh. As shown in Fig.11,with the application of the pumped storage system for the S2-S8 scenarios, the deficit was mainly replaced by power generated from the hydroelectric power plant. It is noted that, volume of load demand covered by the renewable sources in all scenarios remains constant and the share of the hydropower plant increases gradually. As in scenario S8,more than 35% of the demand is obtained from the storage system. And this leads to the decrease of the unexpected volume of energy exchange with the national power system as shown from the cumulated column below. The results below show that the power system operator has to regulate the operation of the conventional power plants for different energy exchange values between the hybrid power system and the national grid.

Fig.11 The distribution of energy demand between individual energy sources

Fig.12 shows that the energy exchange values were divided into several categories related to their values and directions (positive values main surpluses & negative values for deficits). This graph not only illustrates the volatility of the energy flow but also the most likely demand for reserve power plants, which will adjust their capacity in accordance with changes in demand. The most probable volumes of energy exchange that must be provided in all scenarios are from - 2 to 2 MWh, which have the largest capacity of the upper reservoir as a rule,have lower average energy exchange volumes with the national power system. In general, the total amount of excess energy is relatively small and is concentrated between and 4 MWh. By comparing the appearance of surplus energy in scenarios S1- S8 the function of the upper tank as an energy storage becomes evident.

The results, which are presented in Fig.11 and 12 are also provided in Fig.13 with a visualization of the orders of energy exchange with the grid (the red area indicates the energy deficits, the blue area indicates the energy surpluses and the x-axis represents the percentage of the values of energy within the all period of study). Equations (11) and(14) are used to calculate the values of energy surpluses and deficits for all the studied period. The energy exchange values have been arranged from smallest to largest and presented in a graphical diagram. With the application of the energy storage system (scenarios S2-S8) the total sum of energy delivered and fed to the grid starts to diminish gradually.

Fig.12 Histogram of the observed values of energy exchange for the considered scenarios

6 Conclusion

This article presents a proposed design of hybrid PV-wind hybrid system with hydroelectric pumped storage.Moreover, a complete mathematical model of the system that can be used to simulate and optimize its operation has been introduced. The main purpose of the designed system is to significantly reduce energy exchange between the hybrid system and the power grid without exceeding the nominal capacity of solar and wind power plants.Therefore, in this work, we focus on analyzing the energy exchange statistics between the proposed hybrid system and the national network.

Fig.13 Ordered energy exchange with the grid

Based on the simulation results and analysis, it can be argued that the use of such a hybrid energy source can facilitate the integration of renewable energy sources into the energy system. Moreover, the presented model is universal and can be used elsewhere to test its suitability for developing such a project. Histograms illustrating the amounts of energy exchange with the network show that different approaches can be used to eliminate them, for example, by applying demand management, or at least predict them. The results obtained from the considered scenarios indicate the technical feasibility of a pumped storage power plant using seawater to increase the ability of national networks to ensure high penetration of renewable sources of energy.

It is important to emphasize that so far a large-scale hybrid energy source has been investigated, and its influence on the national energy system of Egypt has been planned for future work.