(This is a sample cover image for this issue. The actual cover is not yet available at this time.)

Transcription

1 (This is a sample over image for this issue. The atual over is not yet available at this time.) This artile appeared in a journal published by Elsevier. The attahed opy is furnished to the author for internal non-ommerial researh and eduation use, inluding for instrution at the authors institution and sharing with olleagues. Other uses, inluding reprodution and distribution, or selling or liensing opies, or posting to personal, institutional or third party websites are prohibited. In most ases authors are permitted to post their version of the artile (e.g. in Word or Tex form) to their personal website or institutional repository. Authors requiring further information regarding Elsevier s arhiving and manusript poliies are enouraged to visit:

2 Author's personal opy Energy Conversion and Management 52 (2011) Contents lists available at SiVerse SieneDiret Energy Conversion and Management journal homepage: Assessment of eletriity generation and energy ost of wind energy onversion systems in north-entral Nigeria M.S. Adaramola a,, S.S. Paul b, S.O. Oyedepo a Department of Energy and Proess Engineering, Norwegian University of Siene and Tehnology, Trondheim, Norway b Department of Mehanial and Manufaturing Engineering, University of Manitoba, Winnipeg, Manitoba, Canada Mehanial Engineering Department, Covenant University, Ota, Ogun State, Nigeria artile info abstrat Artile history: Reeived 5 May 2011 Reeived in revised form 18 July 2011 Aepted 18 July 2011 Available online xxxx Keywords: Wind speed Eonomi analysis Energy output Capaity fator Nigeria In this study, the wind energy potential and eonomi analysis in seleted six loations in north entral part of Nigeria were investigated using wind speed data that span between 19 and 37 years measured at m height. The performane of small to medium size ommerial wind turbine models were examined and eonomi evaluation of the wind energy in the seleted sites was made by using the levelised ost method. The results showed that the ost of energy prodution per kwh for the seleted sites vary between 4.02 and It was shown that Minna is most viable site while Bida is found to be least among the sites onsidered. Using three seleted wind turbine models (in Minna) as ase study, an inrease in the esalation rate of operating and maintenane ost from 0% to %, lead to an inrease in the unit energy ost by about 7%. It was further shown that by inreasing the esalation rate of inflation from 0% to 5%, the ost of energy dereases by about 29% while the disount rate (return on investment) dereases from 11.54% to 6.23%. Ó 2011 Elsevier Ltd. All rights reserved. 1. Introdution The inreasing global energy demand and the adverse effets of non-renewable fossil fuels on environment had motivated onsiderable researh attention in a wide range of engineering appliation of renewable resoures suh as: solar, geothermal, and wind. It is widely aepted that wind energy has beome the fastest growing renewable soures of energy in both developed and developing ountries. This is beause wind energy by nature is lean, abundant, affordable, inexhaustible, and environmentally preferable. It has been estimated that the total available wind power surrounding the earth is of the order of 11 GW. This is several times more than the urrent global energy onsumption. Prior studies have shown that the key requirements for proper and benefiial development of wind power at any loation are the wind data analysis and aurate wind energy potential assessment. Several studies on renewable soures of energy in Nigeria have been performed to date. The analysis of available data for seleted ities has onfirmed a high prospet of wind energy resoures in Nigeria. A detailed review and disussion of these studies an be found in (e.g. [1 5]) and are not repeated here. Overall, the results of these studies show that the wind speed varies in general from low regime in the southern part of the ountry to relatively high Corresponding author. Tel.: ; fax: address: muyiwa.adaramola@usask.a (M.S. Adaramola). speed regime in the northern part. The annual mean wind speeds in Nigeria was found to vary between about 2 and 9.5 m/s with an overall annual mean wind speed of about 4.62 m/s. Although the harateristis and pattern of wind speed aross Nigeria have been studied, it is noted that less attention have been given to sites in north entral region until reently. For example, based on the wind data from 1971 to 2007 for five seleted loations in this region, Ohunakin [6] found that the wind resoure in this region an broadly be lassified into lass 1 ategory. It should be remarked that the fous of the study like others are limited to wind speed distributions. The eonomi and performane of wind energy onversion systems has not been thoroughly investigated. The fous of this study therefore, is to evaluate the wind energy potential in six seleted loations (Abuja, Bida, Ilorin, Lokoja, Makurdi and Minna) in the north entral region and to perform the eonomi analysis on seleted small to medium size ommerial wind turbines. This information will be helpful to government and any organization to make an informed deision regarding investment in wind energy resoure in this part of Nigeria. 2. Wind harateristis in north-entral Nigeria The wind data used in this study were obtained from the Nigerian Meteorologial Ageny (NIMET), Oshodi, Lagos. The geographial oordinates of the meteorologial stations where wind speed data were aptured at m height by a up-generator /$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved. doi:.16/j.enonman

3 Author's personal opy 3364 M.S. Adaramola et al. / Energy Conversion and Management 52 (2011) anemometer, are given in Table 1. Monthly wind data that span between 19 and 37 years were obtained for Abuja, Bida, Ilorin, Lokoja, Makurdi and Minna. The aquired data were obtained on hourly basis, from whih monthly wind speed and other wind speed parameters were determined. There are many soures of measurement unertainty in up-anemometer measurements. The guidelines and steps neessary to minimize these errors are outlined in Mawell et al. [7]. Following the methodologies proposed and explained in the ISO guide [8] to the expression of unertainty in measurement, the unertainty in the mean veloities at 95% onfidene level was determined to be ±2%. The two-parameter Weibull probability density funtion was used to analyse the wind data. The site wind harateristis and Weibull parameters are presented in Table 2 for eah loation. The annual mean wind speeds are 3.61 m/s, 2.75 m/s, 4.39 m/s, 3.16 m/s, 4.58 m/s and 4.29 m/s for Abuja, Bida, Ilorin, Lokoja, Makurdi and Minna, respetively. The annual most probable wind speed (V F ) varied between 2.77 m/s in Bida and 4.62 m/s in Makurdi, while the annual wind speed arrying maximum energy (V E ) is in range of 3.50 m/s in Bida and 6.80 m/s in Minna. It is noted that the effiieny of a wind turbine is losely related to V E and that should be as lose as possible to the design or rated wind speed of the system [9]. Subsequently, if wind turbines with the same design parameters are installed in all the sites onsidered in this study, the system will generate highest amount of eletriity in Minna and least eletriity in Bida. Furthermore, the annual mean power densities are, respetively, W/m 2, W/m 2, W/m 2, W/m 2, W/m 2 and W/m 2 for Abuja, Bida, Ilorin, Lokoja, Makurdi and Minna. Based on PNL wind power lassifiation sheme [,11], the wind resoure in all the sites onsidered fall into Class 1 wind resoure ategory (P D 6 0). Detailed analysis of wind speed distribution in five of the loations onsidered in this study an be found in Ohunakin [6]. For a modern wind turbine, ut-in wind speed required in order for turbine to start generating eletriity is generally between 3 and 5 m/s and it depends on the size of the turbine, peak power output that an be attained when the wind speed (rated wind speed) is in the range of 15 m/s [5]. For water pumping, wind turbine an be operated at lower wind speed; however, they an funtion effetively when the wind speed is more than 3 m/s. This, however, depends on the required quantity of water. It should also be noted that a site with mean wind speed of about 2.0 m/s an be onsidered for wind-powered pump appliation Table 1 The geographial loation of the seleted stations. Station Latitude (N) Longitude (E) Altitude (m) Measurement period Abuja Bida Ilorin Lokoja Makurdi Minna Table 2 Annual and seasonal of the site wind speed harateristis and Weibull parameters at height m. V m (m/s) P D (W/m 2 ) V E (m/s) V F (m/s) k (m/s) Abuja Bida Ilorin Lokoja Makurdi Minna [12]. Therefore, the wind resoure in Ilorin, Makurdi and Minna an be used for eletriity generation while the wind resoure in Abuja, Bida and Lokoja an be effiiently developed for water pumping and small sale eletriity generation. It an therefore be onluded that based on the end use of the generated power, the above named loations are suitable for utilization of wind energy. 3. Eletrial power output and apaity fator A wind energy onversion system an operate at maximum effiieny only if it is designed for a partiular site. This is beause the rated power, ut-in and ut-off wind speeds must be defined based on the site wind harateristis [13]. It is essential that these parameters are seleted so that energy output from the onversion system is maximized. The performane of a wind turbine installed in a given site an be examined by the amount of mean power output over a period of time (P out ) and the onversion effiieny or apaity fator of the turbine. The apaity fator C f is defined as the ratio of the mean power output to the rated eletrial power (P er ) of the wind turbine [13,14]. The mean power output P out an be alulated using the following expression based on Weibull distribution funtion [13]: 0 e v ð Þ k vr e ð Þ k P out ¼ P k k v v r e v f 1 k A and the apaity fator C f of a wind turbine is given as: C f ¼ P out P er Therefore, v ð Þ k vr C f ¼ e e ð Þ k k k v v r e v k f where v, v r, v f are the ut-in wind speed, rated wind speed, and utoff wind speed, respetively. It is reognized that the apaity fator is a funtion of the site parameters (whih is dependent on the hub height) and the wind turbine design wind speed properties. The ost effetiveness of a wind turbine an be roughly estimated by the apaity fator of the turbine. This fator is a useful parameter for both onsumer and manufaturer of the wind turbine system [15]. For an investment in wind power to be ost effetive, it is suggested that the apaity fator should be greater than 0.25 [9]. In this study, the performane assessments of seleted ommerial wind turbine models were examined by the values of their apaity fators. It is observed that in most ases, the available wind data are measured at height different from the wind turbine hub height. Consequently, the wind speed at the hub height is of interest for wind power appliation and the available wind speeds an be adjusted to the wind turbine hub height using the following Power law expression (e.g. [5,13]): V ¼ h a ð4þ V o h o where V is the wind speed at the required height h, V o is wind speed at the original height h o and a is the surfae roughness oeffiient and is assumed to be (or 1/7) in most ases. The surfae roughness oeffiient an also be determined from the following expression [15]: a ¼ ½0:37 0:088 lnðv o ÞŠ 1 0:088 ln h o ð5þ ð1þ ð2þ ð3þ

4 Author's personal opy M.S. Adaramola et al. / Energy Conversion and Management 52 (2011) Alternatively, the Weibull probability density funtion an be used to obtain the extrapolated values of wind speed at different heights. The Weibull parameters at measurement height are related to the parameters at the wind turbine hub height by the following expressions [16]: n h ðhþ ¼ o ð6þ h o kðhþ ¼k o 1 0:088 ln h o 1 0:088 ln h where o and k o are the sale fator and shape parameter, respetively at the measurement height h o and h is the hub height. The exponent n is defined as: n ¼ ½0:37 0:088 lnð o ÞŠ 1 0:088 ln h ð8þ 4. Cost analysis Prior studies have shown that the ost of eletriity generated by a wind turbine depends on several fators whih inlude: the site speifi fators (e.g. wind speed and quantity of eletriity generated, ost of land, and installation ost); ost of wind turbine, and its eonomi life span; operating and maintenane osts; eletriity tariff and inentives and exemptions [9,17]. Apart from the ost of the wind turbine whih is set by the manufaturers, osts of other ativities are loation dependent. The ost of the wind turbine aounted for about 70% of the all total initial investment ost for wind energy development [9,17,18]. As shown in Table 3, the speifi ost of a wind turbine is dependent on the rated power but varies widely from one manufaturer to another [9,17,19]. Sine the eonomi feasibility of the wind energy development depends on its ability to generate eletriity at a low operating ost per unit energy, aurate estimate of all the osts involved in generating eletriity over the life span of the system is essential. Different methods are generally used to estimate the operating ost of a unit energy produed by the wind energy onversion system. The most ommonly used method however, is the levelised ost of eletriity (LCOE) [17]. The LCOE is a measure of the marginal ost of eletriity over a period of time and it is ommonly used to ompare the eletriity generation osts from various soures [18]. In addition, the LCOE gives the average eletriity prie needed for a net present value of zero when a disounted ash flow analysis is performed. The determination of the ost of unit energy involves three basi steps: (i) estimation of energy generated by the wind turbine over a given period (e.g. year); (ii) estimate the total investment ost of the projet; and (iii) divide the ost of investment by the energy produe by the system. The unit ost of energy using LCOE method an be estimated using the following expression [16]: CRF LCOE ¼ C I þ C omðesþ 8760P R C f ost=kwh where C I is the total investment ost, 8760P R C f is the annual energy output of wind turbine in kwh, CRF and C om(es) are the apital Table 3 Range of speifi ost of wind turbines based on the rated power. Wind turbine size (kw) Speifi ost/kw Average speifi ost/kw < > ð7þ ð9þ reovery fator and present worth of the annual ost throughout lifetime of the wind turbines expressed as () and (11) [17]: CRF ¼ ð1 þ eþn e ð1 þ eþ n 1 C omðesþ ¼ C om 1 1 þ e om e e om 1 þ e n ost=year ðþ ð11þ where C om,e om,nand e are the operation and maintenane ost for the first year, esalation rate of operation and maintenane osts, useful lifetime of turbine, and disount rate, respetively. The disount rate an be orreted for inflation rate (r) and inflation esalation rate (e) using the following expressions [9]: e a ¼fð1 þ eþð1 þ rþg 1 ð12þ where e a is alled the apparent esalation rate. The disount rate an be determined from: e ¼ ð1 þ iþ 1 ð13þ ð1 þ e a Þ 5. Results and disussion 5.1. Performane of seleted wind turbines For the wind turbine performane assessment and eonomi analysis, five of the Polaris Ameria ommerial wind turbine models with rated power range from 20 kw to 500 kw are seleted [20]. The seleted wind turbine models and their harateristi properties are given in Table 4. The seleted wind turbines are designed to operate at different hub heights. In this study, however, the highest hub heights for eah model were used. For eah loation, the annual energy output and apaity fator based on Weibull distribution funtion parameters at their respetive hub height are determined. The annual energy output from the seleted wind turbine models at all the loations is presented in Fig. 1. The annual energy output ranges from about 7.53 MWh in Bida with P-20 model to MWh in Minna using model. Among the 50 kw model turbines, model produed more power than P This is beause model have lower ut-in wind speed and rated wind speed when ompared with P17-50 model. Regardless of the loation, the wind turbine model produe highest quantity of annual energy output while P-20 model generated least energy. For P-20, the annual energy output vary from 7.53 MWh in Bida to MWh in Minna while in the ase of, the annual energy output varies between 205 MWh in Bida and MWh in Minna. Based on the amount of eletriity produed, model is the best hoie for all the loations onsidered in this study. The seleted wind turbine models apaity fators are presented in Fig. 2. The model has the highest value among the models onsidered for all the sites. This is beause its hub height is highest among the models onsidered in this study. The C f values for this model are 14.78%, 4.70%, 26.34%, 8.60%, 24.73% and 36.% for Abuja, Bida, Ilorin, Lokoja, Makurdi and Minna, respetively. Furthermore, the P17-50 model has the least apaity fators for eah site. This is due to its ut-in wind and rated wind speeds whih are highest ompared with other low rated power (650 kw) wind turbine models. The apaity fators for P-20, and P50-70 are equal and greater than the suggested value reommended before an investment an be onsidered worthwhile in Ilorin and Makurdi. In addition, the P19-0 model an be marginally onsidered for wind energy development in Ilorin and

5 Author's personal opy 3366 M.S. Adaramola et al. / Energy Conversion and Management 52 (2011) Table 4 Charateristis of the seleted wind turbines [20]. Polaris P-20 Polaris Polaris P17-50 Polaris P19-0 Polaris Rated power (kw) Hub height (m) Rotor diameter (m) Cut-in wind speed (m/s) Rated wind speed (m/s) Cut-out wind speed (m/s) Annual average energy output (kwh/year) P-20 P17-50 P19-0 Abuja Bida Ilorin Lokoja Makurdi Minna Loations Fig. 1. The annual power output from the seleted wind turbine models for all the loations. Capaity fator (%) P-20 P17-50 P19-0 Abuja Bida Ilorin Lokoja Makurdi Minna Loations Fig. 2. The apaity fator for the seleted wind turbine models for all the loations. Makurdi. In addition, all the models are good hoie for wind energy development in Minna. The apaity fators for the models for Abuja, Bida and Lokoja indiate that they may not be good sites for wind energy development for eletriity generation but small sale appliations. However, by re-design the seleted wind turbine models to operate at higher hub heights and lower rated wind speed ompared with their urrent design parameters both the annual energy output and apaity fator in these sites (espeially, Abuja) ould signifiantly be improved. Table 5 Cost analysis for seleted wind turbines ($/kwh). WT-model LCOE min LCOE max LCOE ave Abuja P P P Bida P P P Ilorin P P P Lokoja P P P Makurdi P P P Minna P P P Eletriity ost The eonomi analysis of the seleted wind turbine models was arried out using LCOE method. The ost estimation per unit kwh of energy produed by models was estimated based on the following assumptions: i. The lifetime (n) of eah of the Polaris wind turbine used in this study is 20 year [20]. ii. The interest rate (i), and inflation rate (r) were to be 16% and 4%, respetively; while the inflation esalation rate (e) was assumed to vary between 0% and 5%. iii. Operating and maintenane ost (C om ) was assumed to be 25% of the annual ost of the wind turbine (system prie/lifetime) and the esalation rate of operation and maintenane (C om(es) ) is assumed varying between 0% and %. iv. Other initial osts inluding that for land, installation, and grid integration are assumed to be 30% of the wind turbine ost. v. It is further assumed that the wind turbine produes the same amount of energy output in eah year during its useful lifetime.

6 Author's personal opy M.S. Adaramola et al. / Energy Conversion and Management 52 (2011) Unit energy ost ($/kwh) P19-0 P19-0 models (in Minna) ould also be eonomially worthwhile. In general, the operating and maintenane osts inrease as the turbine lifetime dereases and the hanges in inflation rate reveal a strong effet on both ost of eletriity and disount rate or return on investment. The effet of operating and maintenane esalation rate on the average LCOE for, P19-0 and models in Minna are presented in Fig. 3. When the operating and maintenane esalation rate inreases from 0% to %, there is a gradual inrease in the unit energy ost by about 7.24%, 7.23% and 7.30%, respetively, for, P19-0 and models. However, with inreasing inflation esalation, LCOE for these models dereases by about 29.0% (Fig. 4), while the disount rate or return on investment dereases from 11.54% to 6.23% Operating and maintenane esalation rate (%) Fig. 3. Effet of operating and maintenane esalation rate on LCOE for three seleted wind turbines in Minna. The results of the LCOE without operating and maintenane and inflation esalation rates in all the sites for seleted wind turbines are shown in Table 5. It an be observed that LCOE depends on the speifi ost of eah wind turbine and site wind harateristis (represented by the turbine apaity fator). For a given wind turbine, the least LCOE is obtained for Minna while the highest LCOE is alulated for Bida. For the lowest speifi ost of eah turbine, the least ost of unit energy per kwh is obtained with model for Abuja, Bida, Ilorin, Lokoja, Makurdi and Minna as $/kwh, $/kwh, $/kwh, $/kwh, $/kwh and $/kwh, respetively. For the highest speifi ost of eah turbine the LCOE vary between $/kwh in Minna for P model to $/kwh in Bida for P17-50 model. The urrent ost of eletriity in Nigeria is about N.00 [21] or $/kwh (1$ N153, [22]). From this result, it an be inferred that based on the average LCOE value for eah wind turbine models and under the assumed onditions in this study, the LCOE for model in Minna shows to be most eonomially viable option. However, if subsidy of about N6.00 (or $0.0400) per kwh is provided by the government to enourage investment in wind energy development, model (in Ilorin and Makurdi), and and Unit energy ost ($/kwh) P Inflation esalation rate (%) Fig. 4. Effet of inflation esalation rate on LCOE for three seleted wind turbines in Minna. 6. Conlusion In this study, the wind energy potential and eonomi analysis in seleted six loations in north entral part of Nigeria were investigated. In addition, the performane of seleted ommerial wind turbine models designed for eletriity generation loated in these sites were examined. The findings from this study an be summarized as follows: The annual mean wind speeds are 3.61 m/s, 2.75 m/s, 4.39 m/s, 3.16 m/s, 4.58 m/s and 4.29 m/s for Abuja, Bida, Ilorin, Lokoja, Makurdi and Minna, respetively and the respetive annual mean power densities for these loations are W/m 2, W/m 2, W/m 2, W/m 2, W/m 2 and W/m 2. The apaity fator was found to vary between 2.86% in Bida with P17-50 model and 36.% in Minna for model. The eonomi analysis showed that the average LCOE varies between $/kwh for in Minna and $/ kwh for P17-50 in Bida. In term of energy prodution and apaity fator, Minna is most promising site for wind energy development follow by Makurdi and Ilorin among the loations onsidered. However, based on unit ost of eletriity, Minna is the most viable site. Inreasing the operating and maintenane esalation rate from 0% to % for, P19-0 and models in Minna inrease the estimated LCOE by about 7%. However, by inreasing the inflation esalation rate from 0 to 5% lead to derease in the LCOE by about 29% but return on investment is found to derease from 11.54% to 6.23%. Referenes [1] Ojosu JO, Salawu RI. A survey of wind energy potential in Nigeria. Solar Wind Tehnol 1990;7: [2] Adekoya LO, Adewale AA. Wind energy potential of Nigeria. Renew Energy 1992;2:35 9. [3] Fagbenle RO, Katende J, Ajayi OO, Okeniyi JO. Assessment of wind energy potential of two sites in North-East, Nigeria. Renew Energy 2011;36: [4] Fadare DA. The appliation of artifiial neural networks to mapping of wind speed profile for energy appliation in Nigeria. Appl Energy 20;87: [5] Adaramola MS, Oyewola OM. On wind speed pattern and energy potential in Nigeria. Energy Poliy 2011;39: [6] Ohunakin OS. Assessment of wind energy resoures for eletriity generation using WECS in north entral region, Nigeria. Renew Sustain Rev 2011;5: [7] Manwell JF, MGowan JG, Rogers AL. Wind energy explained: Theory, design and appliation. 2nd ed. Chihester: John Wiley and Sons Ltd; [8] International Standards Organisation. Guide to the expression of unertainty in measurement. 1st ed.; [9] Mathew S. Wind energy: fundamentals, resoure analysis and eonomis. Heidelberg: Springer; [] Ilina A, MCarthy E, Chaumel J-L, Retiveau J-L. Wind potential assessment of Quebe Provine. Renew Energy 2003;28:

7 Author's personal opy 3368 M.S. Adaramola et al. / Energy Conversion and Management 52 (2011) [11] Li M, Li X. Investigation of wind harateristis and assessment of wind energy potential for Waterloo region, Canada. Energy Convers Manage 2005;46: [12] Meel J, Smulders P. Wind pumping: a handbook, World Bank Tehnial Paper No. 1. The World Bank, Washington, DC. USA; [13] Akpinar EK, Akpinar S. An assessment on seasonal analysis of wind energy harateristis and wind turbine harateristis. Energy Convers Manage 2005;46: [14] Balouktsis A, Chassapis D, Karapantsios TD. A nomogram method for estimating the energy produed by wind turbine generators. Solar Energy 2002;72: [15] Uar A, Balo F. Evaluation of wind energy potential and eletriity generation at six loations in Turkey. Appl Energy 2009;86: [16] Justus CG, Hargraves WR, Mikhail A, Graber D. Methods for estimating wind speed frequeny distributions. J Appl Meteor 1978;17: [17] Gökçek M, Genç MS. Evaluation of eletriity generation and energy ost of wind energy onversion systems (WECSs) in Central Turkey. Appl Energy 2009;86: [18] Hearps P, MConell D. Renewable energy tehnology ost review. Tehnial Paper Series. Melbourne Energy Institute; Marh [19] Gökçek M, Erdem HH, Bayülken A. A tehno-eonomial evaluation for installation of suitable wind energy plants in Western Marmara, Turkey. Energy Explor Exploit 2007;25: [20] Polaris Ameria. < [aessed ]. [21] Market ompetition and rates division: Tariff regulation in the Nigerian eletriity supply industry. A presentation at the NERC/NARUC workshop; July < [22] < [aessed ].