Sizing a (photovoltaic/wind) Hybrid System

H. Belmili, N. Matidji, O. Badaoui, S. Attoui, N.Hanini, O. Nedjmi* Group of research: Hybrid system, Unit of Development of Solar Equipments (UDSE) National road N: 11 Bou-Ismal LP 365, Tipaza 42415, Algiers, belmilih@yahoo.fr * Electronic Department, University Saad Dehleb Blida, Algiers. Abstract - Combining Solar Photovoltaic systems with batteries can guarantee high supply reliability, but in cloudy weather (weak irradiation) this strategy requires large storage capacity and is expensive. It is cheaper to supply peaks of demand and the demand during this period, with either an additional wind generator. Under this objective different types of (photovoltaic/wind) hybrid systems for supplying electricity have been demonstrated in a large number of pilot and demonstrated projects. In this context, the design of hybrid systems is a relevant issue. An ideal system has to supply, at any given time in the year, an instantaneous energy that equals the consumed energy by all system loads. A sub-sized system obviously doesnt satisfy the demand on electric power and on the other hand, an over-sized system can be completely prohibitive due to economical and financial indicators. This work presents the different methods of sizing a small PV-Wind hybrid system and the chose of the method that gives the optimal technico-economic configuration. Key words: Sizing, PV-Wind, Hybrid system. Rsum- La combinaison des gnrateurs photovoltaques solaires avec les batteries peut garantir la fiabilit d'approvisionnement lev, mais pour des temps nuageux (insolation solaire faible) cette stratgie exige la grande capacit de stockage qui rend le systme chre. Pour cela il est mieux de penser un systme hybride et dajouter au gnrateur photovoltaque un gnrateur olien pour minimiser le cot. Sous cet objectif diffrents systmes (photovoltaque/olien) ont t dmontrs dans un grand nombre de projets. Dans ce contexte, la conception des systmes hybrides est une question approprie. Un systme idal doit fournir, n'importe quelle heure donne par anne, une nergie instantane qui gale l'nergie consomme par toutes les charges de systme. Un systme sous dimensionner videmment ne satisfait pas la demande lectrique d'une part et un systme surdimensionn peut tre d compltement prohibitif aux indicateurs conomiques et financiers. Ce papier prsente les diffrentes mthodes de dimensionner dun systme hybride PV-olien et de choisir la mthode qui donne la configuration technico-conomique optimale. Mots cls : Dimensionnement, PV-olien, systme hybride. covered by the hybrid wind-solar combination. Usually, most of hybrid systems are designed to supply electric power for lighting fixtures, radio/TV, domestic appliances, submersible water pumps etc. This is typical in isolated areas for rural households as well as of some public buildings such as schools, cultural establishments, etc. For an average rural household it is necessary to provide the following services: indoor/outdoor lighting (5/6 points x 5 hours/day) and radio/TV (4 hours/day). Based on these loads, the system should deliver both DC power supply (12/24/48 V DC) and AC power supply (220 V AC). It is necessary also ensure the minimum level of power for the functioning of the systems even during least unfavourable periods(at night, on cloudy and windless days) Electric energy for these periods, is provided from energy storage batteries (their autonomy and capacity being determined on a caseto-case basis). For the above-mentioned requirements, the peak power level of the PV-wind electric generator can usually range from 500W to 1000W.

INTRODUCTION Before building a system with several intermittent energy sources and variable consumption, guidance on the dimensioning of the individual components should be obtained by simulating the system operation under the local conditions, including as appropriate the weather, insulation, wind etc. In general a key objective of such systems is to use the maximum proportion of renewable energy, but other factors including the financial investment, social aspects, local infrastructure, durability etc. must also be considered. The components and subsystems of a stand-alone power supply system based on renewable sources are interconnected to optimise the whole system. The design of a hybrid system will depend on the requirements of the user (isolated or not isolated location, rural or urban, DC or AC. power supply), and on the power supply system proper. Off-grid hybrid power systems can also incorporate energy storage in batteries to increase duration of energy autonomy. If some of the loads connected to a hybrid PV-wind system require permanent electric power supply, a backup diesel generator can be connected to the system to provide electric energy for peak loads which cannot be

1.

The advantages of the hybrid system The main advantages of a hybrid system The possibility to combine two or more renewable energy sources, based on the natural local potential of the users Environmental protection, especially in terms of CO2 emissions reduction Low cost wind energy, and also solar energy can be competitive with nuclear, coal and gas especially considering possible future cost trends for fossil and nuclear energy. Diversity and security of supply Rapid deployment - modular and quick to install Fuel is abundant, free and inexhaustible Costs are predictable and not influenced by fuel price fluctuations, although fluctuations in the price of batteries will be an influence where these are incorporated.

that these other technologies can provide lower cost per Kilowatt-hour if they are scaled up to a certain level. The hybrid PV-Wind systems offer the most adequate solutions for the electrification of small rural settlements, the combination and the ratio of the two types of energy depending greatly on the resources locally available in each geographical area. These resources can be accurately evaluated only after a period of typically one year of monitoring the basic parameters (wind speed, solar radiation), necessary for sizing and implementing such systems in the respective areas.

2. Small hybrid system requirements The hybrid solar-wind power supply system should meet the following specific requirements: The electric input parameters should be compatible with the electric output parameters, especially taking into account: - The wide range of variation of the electric output parameters of the solar or wind generator, due to the variation of solar radiation intensity or of wind speed - The reduced range of variation of the accumulator batteries load - unload electric parameters, this mode having to be controlled by means of a charge controller Compatibility in various operating modes of the load by interconnecting adequate interfaces (both for the DC part and for the AC part). 3. Basic components of a small hybrid system A typical small hybrid power system can contain the following components: Solar PV Generator: containing a number of series/parallel interconnected solar modules (depending on the necessary voltage), including connection and protection elements (bypass diodes and/or anti-return). This element delivers part of the electric energy supply through solar energy conversion. Wind generator: providing part of the necessary electric energy by converting the mechanical energy from the wind. Storage unit (accumulator battery set): Usually Pb batteries are used, but also Ni-Cd or Ni-Fe, dedicated to applications in the area of renewable energy sources. Unit for power conditioning: This can be a DC/DC converter (for DC loads) and/or inverters (for AC loads). 4. The solution based on hybrid systems If the amount of energy consumption increases, it makes sense to combine PV with wind. The reason is

Figure1. Block diagram of PV-Wind hybrid system

5.

PV-Wind Hybrid system Function procedure During day time, DC power generated by the solar PV array is stored in the battery bank through a hybrid controller, which maximises charging current and prevents excessive discharge/overcharge. Wind turbine generator started generating power when wind speed exceeds cut-in speed of the mini wind turbine (above 2.7m/s). Out from the wind battery charger is also stored in the battery bank through hybrid controller. The wind turbine is self-regulated type with protection for Overspeed. Energy stored in the battery is draw by electrical loads through the inverter, witch convert DC into AC power. The inverter has in-built protection for short-circuit, reverse polarity, low battery voltage and over load. The batter bank is sizing to feed loads up to two days, during non sun/wind days. 6. Methods of sizing PV-Wind hybrid system

6.1 The yearly monthly overage sizing method In this method, photovoltaic and wind generators size are measured from the average annual monthly values of energies statement E pv ,

E w and the load E L . This calculates is basing on

the average annual monthly data of sunning and the

wind. The size of the components photovoltaic and wind is given by the following equations [1]:

A pv ,T = P.

EL E pv ,u EL E w ,u

(1)

Aw , T = (1 P)

(2) (3)
3

E pv ,u = Apv .H .R pv
E w,u = 1 / 2. .C P . Aw .V

(4)

stored in the battery at any moment T, expressed in [Wh]. An elaborated simulation program permits to determine the optimum size of battery bank and PV array for an autonomous PV-wind hybrid energy system for a given load and a desired loss of power supply probability basing on the minimum cost of the system [2]. This method presents the advantage that the introducing of the wind generator permits to minimize the cost of the photovoltaic stand-alone system, by the minimizing the size of the photovoltaic generator and the storage capacities (number of battery). Cost of the system For a hybrid system, the final and must important question is the cost per delivered kWh. For a hybrid system sized for a 1.5 kW wind turbine and 1.0 kW solar photovoltaic, the economic analysis for the optimum system solution shows an average running cost of 0.77 Euro per kWh. The lifetime of the plant is assumed to be 20 years and the battery bank should be replaced every 5 years. Using a battery of half the size would reduce the kWh cost to 0.69 Euro and the load coverage from 85.9% to 80.7% if the lifetime of the battery is still considered to be 5 years. The cost of energy also depends on investment cost, operation and maintenance costs, depreciation period and energy produced in one year, plus external trends such as the cost of batteries (subject to legislation affecting the cost of new materials and the cost of disposal), the potential downward trend of equipment costs with rising volumes etc. The competitiveness of a hybrid system also depends on the relative cost of fossil fuels and the demand for renewable energy from the market. A simple cost calculation relation is: 7.

6.2 The most unfavourable month method For this method the size of the PV and Wind generators is being calculated in the most unfavorable month. Generally the month most unfavorable in wind is favorable in irradiation. So we are obliged to dimension the system in two most unfavourable months (unfavourable irradiation month and unfavourable wind month). When the system functioned in this month its automatically functioned in the author month [3].

E A pv ,T = P. max L E pv ,u
E Aw , T = (1 P ). max L E w ,u

(5)

(6)

6.3 Loss of power supply probability (LPSP) method This sizing method consists in determining the optimal number of the batteries and the photovoltaic modules according to the optimization principle knowing: the reliability, which is based on the concept of the probability of loss of energy (Loss of Power Supply Probability _ `LPSP) [2], and on the cost of the system. ` LPSP' is defined as being the fraction of the deficiency energy and that required by the load. It explains the rate of not satisfaction of the load, in term of state of batteries charging. ` LPSP' is defined by:

P = C R/E + c e
Where: P = energy cost, C = investment cost, where

(8)

C = C pv + C el + C bat
R = annual discount rate for capital and associated costs, (Taux d'escompte annuel pour des cots capitaux et associs) E = produced energy, ce = annual operation and maintenance costs.

LPSP = Pr {E B (t) E B min for T T}

(7)

i.e. the probability that the load statement, at any moment T, is lower or equal to the minimal edge of the supplied energy in battery EBmin. EB (t) is the energy

8.

Comparison study: The yearly monthly overage sizing method example [4] P= 1 100% PV_system P= 0 100% Wind_system P Number of PV modules 5 5 4 4 3 3 2 2 1 1 0 PV cost installation [euro] 293,9 293,9 293,9 235,12 176,34 176,34 117,56 117,56 58,78 58,78 0 Number of Wind Generators 0 1 1 1 1 1 1 1 1 1 1 Wind Installation cost[euro] 0 327 327 327 327 327 327 327 327 327 327 Total cost [euro] 293,9 620,9 562,12 562,12 503,34 503,34 444,56 444,56 385,78 385,78 327

1 0,9 0,8 0,7 0,6 0,5 0,4 0,3 0,2 0,1 0

Cub = 90euro[HT] , Cupv(amorph)= 58.78 euro[HT], Cuwg = 327euro [HT]

The most unfavourable month method example [4] P= 1 100% PV_system P= 0 100% Wind_system Number PV cost Number of Wind Total of PV installation Wind Installation cost [euro] cost[euro] [euro] modules Generators 6 5 5 4 4 3 3 2 2 1 0 352,68 293,9 293,9 235,12 235,12 176,34 176,34 117,56 117,56 58,78 0 0 1 1 1 1 1 1 1 1 1 2 0 327 327 327 327 327 327 327 327 327 654 352,68 620,9 620,9 562,12 562,12 503,34 503,34 444,56 444,56 385,78 654

1 0,9 0,8 0,7 0,6 0,5 0,4 0,3 0,2 0,1 0

Cub = 90euro[HT] , Cupv(amorph)= 58.78 euro[HT], Cuwg = 327euro [HT]

Loss of power supply probability (LPSP) method example[5] LPSP Number of PV modules 10-1 10-2 10-3 39 51 52 PV System Number Cost of the system[euro] of Battery 27 33 39 12180 15720 16510 Number of PV modules 28 38 38 PV/Wind System Number Number of of Wind Battery Generator 10 16 26 1 1 1 Cost of the system[euro] 7800 10940 11840

Cub = 90euro[HT] , Cupv= 250euro[HT], Cuwg = 100euro [HT]

Advantages The yearly monthly overage sizing method Minimized the total system cost Lacking of the weather station in isolated location. Minimized the total cost of the system

The most unfavourable month method Loss of power supply probability (LPSP) method

Disadvantages This method must require the monthly and annual weather data site of the located installation High cost, because the oversizing of the system This method uses long term data of wind speed and irradiation taken every hour for several years.

9. Requirements for installing PV/Wind system For installing (PV-wind hybrid systems) as sources of electric energy, the following steps should be taken into account: o Acquire data on the local natural potential of available wind energy and solar energy, from the local meteorological stations; o Determine the annual energy consumption for your application (electric loads considered for connection to the hybrid energy system); o Size the system (in this case, composed of two installations, one for solar energy and the other for wind energy) on the bases of simplified methods. For sizing the installation which will provide electricity from a photovoltaic generator, use the simplified method presented in the solar energy module (step by step guide photovoltaic section). For sizing the installation for wind energy, use the indications presented in the wind energy module (step by step guide section). o For sizing this system, the following will be considered: the PV installation will be sized for summer time conditions (good availability of solar energy) and the wind installation will be sized for winter time conditions (good availability of wind energy), for the necessary consumption; o Determine the most suitable location for installation; 10. Case study installation For example we display of the load profile (figure.2).

Figure.3 Average wind speed and irradiation in the most unfavourable months 10.2 Choose of the installation equipments The PV generators preferred is the monocrystalline modules. The Wind generators preferred are characterised by : o Maximal power 1kW (small wind turbine). o Starting wind speed 3 m/s o Operating range 2.5-20 m/s o Power curve (see figure.4/5)
3,0

2,5

2,0

Power (KW)

1,5

1,0

0,5

0,0 0 5 10 15 20

Wind speed (m/s)

1 ,0

0 ,8

Demand (KW)

0 ,6

0 ,4

Figure.4: Power curve of the wind turbine used for sizing simulation DC load.
0,5
5 10 15 20

0 ,2

0 ,0

H our
0,4

Power (KW)

Figure.2: Load profile per day 10.1 Choose of sizing method In this study we have only the data of the most unfavorable months, the figure.3 represent the wind speed and irradiation in these months. So we are obliged to use the most unfavourable method of sizing.

0,3

0,2

0,1

0,0 0 1 2 3 4 5 6 7 8 9 10 11 12

Wind speed (m/s)

Figure.5: Power curve of the wind turbine used for sizing simulation AC load.

10.3 Result of sizing :

Load demand per day (kWh/d) 18.1(DC) 18.1 (AC) Size of PV Generator (kW) 3.2 4 Size of Wind Generator (kW) 1 Turbine 1 Turbine Nb of battery 13 13 DC/AC Converter / 2 kW Cost of the system[] 16420 18570
Power(KW)
1,4 1,2 1,0 0,8 0,6 0,4 0,2

PV average electric production Wind avrage elctric production

1,2

DC Load Battery state of charge Wind Power PV Power

0,0 1 2 3 4 5 6 7 8 9 10 11 12

Months
1,2

1,0

1,0

0,8

0,8

Figure.9: Monthly average electric production for AC Load. CONCLUSION This paper presents a comparative study of different sizing method of PV/Wind hybrid system. For objective to choose the method that presents a good reliability and optimal cost. This first step for installing a PV/Wind hybrid system must be studied in carefully to avoid any sub-sizing or over-sizing of the system. In the other hand and independently of the method used, the optimization of the sizing of a photovoltaic/wind hybrid system requires a detailed analysis of the load, data of solar illumination and wind speed with its distribution function. In addition to the economic aspects, the factor of autonomy is one of the significant criteria which must be taken into account in all the study of optimization of dimensioning.

Power (W)

0,6

0,6

0,4

0,4

0,2

0,2

0,0 0 50 100 150 200 250 300 350

0,0 400

Months

Figure.6: Monthly average electric production, DC load and battery state of charge results of simulation.

2,4 2,2 2,0 1,8 1,6

Battery state of charge Wind Power AC Load PV Power

2,4 2,2 2,0 1,8 1,6 1,4 1,2 1,0 0,8 0,6 0,4 0,2 0,0

Power(W)

1,4 1,2 1,0 0,8 0,6 0,4 0,2 0,0

Months

Figure.7: Monthly average electric production, AC load and battery state of charge results of simulation.
PV average electric production Wind average elctric production
1,0

0,8

0,6

0,4

0,2

0,0 1 2 3 4 5 6 7 8 9 10 11 12

Months

Figure.8: Monthly average electric production for DC Load

References [1. W. Kellogg et al Optimal unit sizing for a hybrid wind/photovoltaic generating system Elsevier Science, pp35-38, (1996) [2] A. D. Bagul and al Sizing of a stand-alone hybrid wind-photovoltaic system using a threeevent probability approximation solar energy vol. 56 pp 323-355 (1996) [3] A. El Khadimi et al, Dimensionnement et Optimisation Technico-conomique dun systme dEnergie Hybride photovoltaque Eolien avec Systme de stockage, Energies Renouvelable, Vol 7, pp. 73 83, 2004. [4] S. Diaf et al, Analyse technico conomique dun systme hybride (photovoltaque/ olien) autonome pour le site dAdrar, Energies Renouvelable, Vol 9, pp. 127 134, 2006. [5]. A. Kaabeche et al Optimisation dun systme hybride (olien photovoltaque) totalement autonome Energies Renouvelable, Vol 9, pp. 199-209, 2004.

Power (KW)