- Email: [email protected]
Investigation of wind power density distribution using Rayleigh probability density function
Available online at www.sciencedirect.com Available online at www.sciencedirect.com
ScienceDirect ScienceDirect
Energy Procedia 00 (2018) 000–000 Available onlineatatwww.sciencedirect.com www.sciencedirect.com Available online Energy Procedia 00 (2018) 000–000
ScienceDirect ScienceDirect
www.elsevier.com/locate/procedia www.elsevier.com/locate/procedia
Energy Procedia 157 Energy Procedia 00(2019) (2017)1546–1552 000–000 www.elsevier.com/locate/procedia
Technologies and Materials for Renewable Energy, Environment and Sustainability, TMREES18, Technologies and Materials for Renewable Energy, and Sustainability, TMREES18, 19–21 September 2018,Environment Athens, Greece 19–21 September 2018, Athens, Greece
Investigation of wind power density distribution using Rayleigh The 15th Symposium on District Heating and Cooling Investigation of International wind power density distribution using Rayleigh probability density function probability density function Assessing the feasibility of using the heat demand-outdoor Lizica-Simona Paraschiva, Spiru Paraschiva*, Ion V. Iona a Lizica-Simona Paraschiv , Spiru Paraschiv Ion V. Iona temperature function for a long-term districta*,heat demand forecast “Dunarea de Jos” University of Galati, Domneasca street, no. 47, Romania a a
“Dunarea de Jos” University of Galati, Domneasca street, no. 47, Romania
I. Andrića,b,c*, A. Pinaa, P. Ferrãoa, J. Fournierb., B. Lacarrièrec, O. Le Correc a IN+ Center for Innovation, Technology and Policy Research - Instituto Superior Técnico, Av. Rovisco Pais 1, 1049-001 Lisbon, Portugal Abstract b Veolia Recherche & Innovation, 291 Avenue Dreyfous Daniel, 78520 Limay, France Abstract c Département et Environnement - IMTdistributions Atlantique, 4 prove rue Alfred Kastler, 44300 Nantes, France When modelling the wind Systèmes speed inÉnergétiques a given location, the probability to be a useful tool. In this paper, the wind power modelling density in the an urban and a suburban Constanta county, Romania, havetobeen during thepaper, periodthe January When wind speed in a givenlocation location,inthe probability distributions prove be aanalyzed useful tool. In this wind 2017 December the hourly measured mean wind speed data. Inhave this been study, the Rayleigh density power- density in an2017, urbanbased and aon suburban location in Constanta county, Romania, analyzed during probability the period January function was used 2017, to calculate powermeasured density for eachwind location in data. orderIn to this classify them terms ofprobability wind energy. The 2017 - December based the on wind the hourly mean speed study, the inRayleigh density Abstract Rayleigh was distribution function has from the data and is fitted to the measured probability distribution on function used to calculate the been wind derived power density foravailable each location in order to classify them in terms of wind energy. The yearly basis. The power densities for thefrom two stations varieddata fromand 295.39 to 194.5 / m2. Rayleigh distribution function hasreported been derived the available is fitted to theWmeasured probability distribution on District heating networks are commonly addressed in the literature as one of the most effective solutions for decreasing the yearly basis. The power densities reported for the two stations varied from 295.39 to 194.5 W / m2. greenhouse gas emissions from the building sector. These systems require high investments which are returned through the heat © 2018 The Authors. Published by Elsevier Ltd. sales. Due to the changed climate conditions and building renovation policies, heat demand in the future could decrease, © 2019 The Authors. by Ltd. This is an open accessPublished article under the CC BY-NC-ND license (https://creativecommons.org/licenses/by-nc-nd/4.0/) © 2018 The Authors. Published by Elsevier Elsevier Ltd. This is an open articlereturn underperiod. the CC BY-NC-ND license (https://creativecommons.org/licenses/by-nc-nd/4.0/) prolonging theaccess investment Selection under responsibility of the scientific of Technologies and Materials for Renewable Energy, This is an and openpeer-review access article under the CC BY-NC-ND license committee (https://creativecommons.org/licenses/by-nc-nd/4.0/) Selection peer-review under responsibility of the scientific of Technologies Materialsfunction for Renewable The mainand scope of this paper is to assess the feasibility of usingcommittee the heat demand – outdoorand temperature for heatEnergy, demand Environment and Sustainability, TMREES18. Selection and peer-review under responsibility of the scientific committee of Technologies and Materials for Renewable Energy, Environment Sustainability, TMREES18. forecast. Theand district of Alvalade, located in Lisbon (Portugal), was used as a case study. The district is consisted of 665 Environment and Sustainability, TMREES18. buildings that vary in both construction period and typology. Three weather scenarios (low, medium, high) and three district Keywords: Wind energy; Wind speed; Rayleigh distribution functions; Wind power density; Probability density function; renovation scenarios were developed (shallow, intermediate, deep). To estimate the error, obtained heat demand values were Keywords: Wind energy; Wind speed; Rayleigh distribution functions; Wind power density; Probability density function; compared with results from a dynamic heat demand model, previously developed and validated by the authors. The results showed that when only weather change is considered, the margin of error could be acceptable for some applications 1. Introduction (the error in annual demand was lower than 20% for all weather scenarios considered). However, after introducing renovation 1. Introduction scenarios, the error value increased up to 59.5% (depending on the weather and renovation scenarios combination considered). Due to recent advances, the windwithin energy in 3.8% Europe experienced and sustainable The value of slopetechnological coefficient increased on average the sector range of up has to 8% per decade,rapid that corresponds to the Due over toinrecent technological advances, the driver wind energy sector in Europe has need experienced rapidto and growth thenumber last 20ofyears. important of this rapid growth is the benefit fromand a decrease the heatingAn hours of 22-139h during the heating season (depending onfor the Europe combination of sustainable weather growth overscenarios the lastconsidered). 20 driver of thisintercept rapid is for thesecond need major for per Europe benefit from a secure energy supply thatyears. is not An dependent foreign sources of oilgrowth and gas. A driver the significant renovation On important the otheronhand, function increased 7.8-12.7% decadetois(depending on the secure energy supply that is notsuggested dependent on be foreign sources oil function and gas.parameters A second for major is the significant coupled scenarios). The values could used to modifyofthe the driver scenarios considered, and improve the accuracy of heat demand estimations.
© 2017 The Authors. Published by Elsevier Ltd. * Corresponding author. Tel.: +40721320403. Peer-review under responsibility of the Scientific Committee of The 15th International Symposium on District Heating and E-mail address:author. [email protected] * Corresponding Tel.: +40721320403. Cooling. E-mail address: [email protected] 1876-6102 © 2018 The Authors. Published by Elsevier Ltd. Keywords: Heat demand; Forecast; Climate change This is an open access under the CC BY-NC-ND license (https://creativecommons.org/licenses/by-nc-nd/4.0/) 1876-6102 © 2018 Thearticle Authors. Published by Elsevier Ltd. Selection under responsibility of the scientific of Technologies and Materials for Renewable Energy, Environment This is an and openpeer-review access article under the CC BY-NC-ND licensecommittee (https://creativecommons.org/licenses/by-nc-nd/4.0/) and Sustainability, TMREES18. Selection and peer-review under responsibility of the scientific committee of Technologies and Materials for Renewable Energy, Environment and Sustainability, TMREES18. 1876-6102 © 2017 The Authors. Published by Elsevier Ltd. 1876-6102 © 2019 The Authors. Published by Elsevier Ltd. Peer-review under responsibility of the Scientific Committee of The 15th International Symposium on District Heating and Cooling. This is an open access article under the CC BY-NC-ND license (https://creativecommons.org/licenses/by-nc-nd/4.0/) Selection and peer-review under responsibility of the scientific committee of Technologies and Materials for Renewable Energy, Environment and Sustainability, TMREES18. 10.1016/j.egypro.2018.11.320
2
Lizica-Simona Paraschiv et al. / Energy Procedia 157 (2019) 1546–1552 Author name / Energy Procedia 00 (2018) 000–000
1547
contribution that wind energy can make to reducing greenhouse gas emissions. As a clean and renewable resource, wind energy is already playing - and will continue to play an important role in mitigating climate change, bringing at the same time, benefits in terms of reducing emissions of air pollutants and associated cooling water consumption with many of the conventional energy generation technologies. The calculation of wind resources for a given location and the corresponding energy production is based on the wind potential assessment by anemometric measurement, followed by wind data processing to calculate the expected wind power yield for the proposed site. Typical resource assessment methods involve measuring wind speed and recording the arithmetic mean every 10 minutes or one hour for at least one year. These wind speed data are then used to obtain a relative frequency distribution to which a probability density function is fitted. This function and specific wind turbine power curve are required to calculate the amount of available energy and the electrical power generated in a specific region. 2. Site Location and Data Collection This study aimed to examine the wind energy potential of Constanta by finding Rayleigh distribution parameter and determining the available power density. The wind speed data used in this study were measured in Constanta in 2017 at two meteorological stations located near the Black Sea as shown in Fig.1. The meteorological stations are located in the south-east coastal region of Constanta, Romania. The daily wind speed data (in m/s) was collected using cup anemometer at the height of 10 m from ground level, with an hour interval over a period of 1 year, from January 2017 to December 2017. There were analyzed 17234 valid data from 17540 data recorded from both stations.
Fig. 1. Location of meteorological stations.
Because the wind speed tends to increase with height and depends mainly on atmospheric mixing and terrain roughness, therefore, we need to calculate the total wind energy potential, so the measured variation in wind speed with height at each site is defined by:
V z1 z1 V z2 z2
z V2 V1 1 z2
(1)
Where, α is power law wind shear exponent, V is the mean wind speed, z1 is the measurement height above ground level and z2 is the turbine height (m).
Lizica-Simona Paraschiv et al. / Energy Procedia 157 (2019) 1546–1552 Author name / Energy Procedia 00 (2018) 000–000
1548
3
The wind shear exponent depends on the complexity and roughness of the terrain for the specific site. 3. Rayleigh distribution Rayleigh's distribution can be used to describe the wind variations in a wind regime with an acceptable level of accuracy without the need for data collected over short time intervals, as is the case of the Weibull distribution, because in many cases, such information cannot be available. Rayleigh's distribution is a simplified case of the Weibull distribution, which is derived by assuming the shape factor as 2. Because of its simplicity, this distribution is widely used for wind energy modeling. The average velocity is expressed as:
1 Vm c 1 k
(2)
Taking k = 2 in Eq. (2), obtain:
Vm c
2Vm 3 c 2
(3)
The probability density function indicates the fraction of time (or probability) for which the wind is at a given velocity V and it is given by:
f (V )
V 2 Vm2
e
V 2 4 Vm
(4)
Cumulative distribution is given by:
F (V ) 1 e
V 2 4 Vm
(5)
The validity of Rayleigh distribution in wind energy analysis has been established by comparing the Rayleigh generated wind pattern with long term field data. The probability of wind velocity to be between V1 and V2 is:
P(V1 V V2 ) e
V 2 1 4 Vm
e
V 2 2 4 Vm
(6)
and the probability of wind to exceed a velocity of VX is given by: V 2 X 4 Vm
P(V VX ) e
(7)
In order to check how accurately the theoretical probability density function fits with observation data the root mean square error (RMSE) parameter test is performed. Best results are obtained when these values are close to zero.
4
Author name / Energy Procedia 00 (2018) 000–000
Lizica-Simona Paraschiv et al. / Energy Procedia 157 (2019) 1546–1552
1 RMSE N
1
N
y x i 1
1549
i
i
2
2
(8)
Where N is the number of observations, 𝑦𝑦𝑖𝑖 is the frequency of observation, 𝑥𝑥𝑖𝑖 is the frequency of Rayleigh.
4. Results and Discussion
The data of wind speed in the present calculation was obtained in 2017 from weather stations in Constanta. The hourly mean wind values measured throughout the total year at the two stations are shown in Fig. 2 and 3. It can be seen in Fig. 2 and 3, that the highest wind speed of 26.6 m / s at the CT-3 station and 24.3 m / s at the CT-6 station occurs in January.
Wind speed [m/s]
30.0
26.6
25.0
Vm=5.6 m/s
20.0 15.0 10.0 5.0 1 252 503 754 1005 1256 1507 1758 2009 2260 2511 2762 3013 3264 3515 3766 4017 4268 4519 4770 5021 5272 5523 5774 6025 6276 6527 6778 7029 7280 7531 7782 8033 8284 8535
0.0
Hours ‐ 2017
Fig. 2. Hourly variation of wind speed at CT-3 (2017)
25.0
24.3
Vm=4.7 m/s
20.0 15.0 10.0 5.0 0.0 1 245 489 733 977 1221 1465 1709 1953 2197 2441 2685 2929 3173 3417 3661 3905 4149 4393 4637 4881 5125 5369 5613 5857 6101 6345 6589 6833 7077 7321 7565 7809 8053 8297 8541
Wind speed [m/s]
30.0
Hours ‐ 2017
Fig. 3. Hourly variation of wind speed at CT-6 (2017)
Lizica-Simona Paraschiv et al. / Energy Procedia 157 (2019) 1546–1552 Author name / Energy Procedia 00 (2018) 000–000
1550
5
The probability density distribution obtained from the Rayleigh model was compared to the measured distributions to study their suitability. The annual comparison shows that the Rayleigh model fit well with the measured probability density distribution as shown in Fig. 4 and 5. 0.14 0.12
RMSE=0.012429
Frequency
0.1
k=1.458497
0.08 0.06 0.04 0.02 0
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25
Wind speed [m/s] Observed
Rayleigh column
Rayleigh line
Fig. 4. Comparison of observed and predicted wind speed frequencies at CT-3 (2017)
0.18 0.16
RMSE=0.013499
0.14
k=1.364391
Frequency
0.12 0.1 0.08 0.06 0.04 0.02 0
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25
Wind speed [m/s] Observed
Rayleigh column
Rayleigh line
Fig. 5. Comparison of observed and predicted wind speed frequencies at CT-6 (2017)
6
Author name / Energy Procedia 00 (2018) 000–000 Lizica-Simona Paraschiv et al. / Energy Procedia 157 (2019) 1546–1552
1551
Analysis of probability density function presented in Fig. 4 and Fig 5 shows the Rayleigh probability and cumulative distributions for all speed ranges observed during the 2017. The real data is presented on the same diagrams so that a comparison can be made between real data and resulted probability density functions. In Fig. 4, it is observed that most of the wind speeds are in the range of 1-7 m/s during all seasons. Rayleigh seems to give a reasonable fit to observed wind speed data as distribution of observed data is not largely deviated from estimated Rayleigh during entire year. Fig. 4 and 5 contain the values of RMSE for Rayleigh distribution and the values of correlation coefficient in both figures are very low (nearly 0), which indicates that the predictions made by Rayleigh are extremely accurate. Errors in calculating the power densities using the distribution Rayleigh model in comparison to those using the measured probability density distribution are presented in Fig. 6. The highest error values occur for wind speed of 1-4 m/s with 4.03% for CT-3 and 3.16% for CT-6 respectively. 0.04
CT‐3
0.03
CT‐6
0.02
Error [%]
0.01 0 ‐0.01
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25
‐0.02 ‐0.03 ‐0.04 ‐0.05
Wind speed [m/s]
Fig. 6. Error values in calculating the wind power density obtained from the Rayleigh models in reference to the wind power density obtained from the measured data.
The CT-3 station reported an average power density of 295.4 W/m2 and the CT-6 station reported a value of 194.5 W/m2 as shown in Fig. 7.
Wind power density [W/m^2]
295.4 194.4999828
300.0 200.0 100.0 0.0
CT‐3
CT‐6
Fig. 7. Average power density
1552
Author name / Energy Procedia 00 (2018) 000–000 Lizica-Simona Paraschiv et al. / Energy Procedia 157 (2019) 1546–1552
7
In agreement with the commercially international system of classification for wind power scale, the CT-6 location correspond to the wind power class 3, since the density value is lower than 200 W/m2. Therefore, this urban location present wind characteristics that are not suitable for wind power development. The CT-3 location corresponds to the wind power class 4 which is generally accepted as a feasible class for commercial wind power development. 5. Conclusion This study analyzed wind characteristics and wind energy potential in Constanta, Romania. For this, wind speed data was analyzed at two monitoring stations over a one-year period at one-hour intervals to predict the wind behavior as accurately as possible. Based on this analysis, the following conclusions can be drawn: 1. The average wind speed was 5.6 m for CT-3 and 4.7 for CT-6, respectively. 2. RMSE for Rayleigh distribution was 0.012429 for CT-3 and 0.013499 for CT-6, respectively. 3. The average wind energy density based on the average wind speed approach was 295.4 W / m2 for CT-3 and 194.49 W/ m2 for CT-6, respectively. Wind power density calculations have indicated that the CT-3 region is in Class 4, which is a suitable area for wind turbine installation. References [1] Sathyajith Mathew, “Wind Energy-Fundamentals, Resource Analysis and Economics”, Springer Berlin Heidelberg New York, (2006). [2] Trevor M. Letcher, Wind Energy Engineering - A Handbook for Onshore and Offshore Wind Turbines, Academic Press, (2017) [3] Sajid Ali, Sang-Moon Lee, Choon-Man Jang, “Statistical analysis of wind characteristics using Weibull and Rayleigh distributions in Deokjeok-do Island e Incheon, South Korea”, Renewable Energy 123 (2018): 652-663, doi: 10.1016/j.renene.2018.02.087. [4] S. A. Ahmeda, H. O. Mahammeda, “A Statistical Analysis of Wind Power Density Based on the Weibull and Ralyeigh models of "Penjwen Region" Sulaimani/ Iraq”, Jordan Journal of Mechanical and Industrial Engineering, 6 (2), (2012): 135 – 140. [5] Chong Li, Youying Liu, Gang Li, Jianyan Li, Dasheng Zhu, Wenhua Jia, Guo Li, Youran Zhi, Xinyu Zhai, “Evaluation of wind energy resource and wind turbine characteristics at two locations in China”, Technology in Society 47, (2016): 121-128, http://dx.doi.org/10.1016/j.techsoc.2016.09.003 [6] K.S.R. Murthy, O.P. Rahi, “A comprehensive review of wind resource assessment”, Renewable and Sustainable Energy Reviews 72 (2017): 1320–1342, http://dx.doi.org/10.1016/j.rser.2016.10.038 [7] Ahmad Hemami, “Wind turbine technology”, Cengage Learning, (2012). [8] Ahmad Sedaghat, Arash Hassanzadeh, Jamaloddin Jamali, Ali Mostafaeipour, Wei-Hsin Chen, “Determination of rated wind speed for maximum annual energy production of variable speed wind turbines”, Applied Energy 205 (2017): 781–789, http://dx.doi.org/10.1016/j.apenergy.2017.08.079. [9] Cristina Herrero-Novoa, Isidro A.Pérez, M. Luisa Sánchez, M. Ángeles García, Nuria Pardo, Beatriz Fernández-Duque, “Wind speed description and power density in northern Spain”, 138 (1) (2017): 967-976. [10] Paraschiv Spiru, Paraschiv Lizica-Simona, “Technical and economical analysis of a PV/wind/diesel hybrid power system for a remote area” Energy Procedia 147 (2018): 343-350, doi:10.1016/j.egypro.2018.07.102.
ScienceDirect ScienceDirect
Energy Procedia 00 (2018) 000–000 Available onlineatatwww.sciencedirect.com www.sciencedirect.com Available online Energy Procedia 00 (2018) 000–000
ScienceDirect ScienceDirect
www.elsevier.com/locate/procedia www.elsevier.com/locate/procedia
Energy Procedia 157 Energy Procedia 00(2019) (2017)1546–1552 000–000 www.elsevier.com/locate/procedia
Technologies and Materials for Renewable Energy, Environment and Sustainability, TMREES18, Technologies and Materials for Renewable Energy, and Sustainability, TMREES18, 19–21 September 2018,Environment Athens, Greece 19–21 September 2018, Athens, Greece
Investigation of wind power density distribution using Rayleigh The 15th Symposium on District Heating and Cooling Investigation of International wind power density distribution using Rayleigh probability density function probability density function Assessing the feasibility of using the heat demand-outdoor Lizica-Simona Paraschiva, Spiru Paraschiva*, Ion V. Iona a Lizica-Simona Paraschiv , Spiru Paraschiv Ion V. Iona temperature function for a long-term districta*,heat demand forecast “Dunarea de Jos” University of Galati, Domneasca street, no. 47, Romania a a
“Dunarea de Jos” University of Galati, Domneasca street, no. 47, Romania
I. Andrića,b,c*, A. Pinaa, P. Ferrãoa, J. Fournierb., B. Lacarrièrec, O. Le Correc a IN+ Center for Innovation, Technology and Policy Research - Instituto Superior Técnico, Av. Rovisco Pais 1, 1049-001 Lisbon, Portugal Abstract b Veolia Recherche & Innovation, 291 Avenue Dreyfous Daniel, 78520 Limay, France Abstract c Département et Environnement - IMTdistributions Atlantique, 4 prove rue Alfred Kastler, 44300 Nantes, France When modelling the wind Systèmes speed inÉnergétiques a given location, the probability to be a useful tool. In this paper, the wind power modelling density in the an urban and a suburban Constanta county, Romania, havetobeen during thepaper, periodthe January When wind speed in a givenlocation location,inthe probability distributions prove be aanalyzed useful tool. In this wind 2017 December the hourly measured mean wind speed data. Inhave this been study, the Rayleigh density power- density in an2017, urbanbased and aon suburban location in Constanta county, Romania, analyzed during probability the period January function was used 2017, to calculate powermeasured density for eachwind location in data. orderIn to this classify them terms ofprobability wind energy. The 2017 - December based the on wind the hourly mean speed study, the inRayleigh density Abstract Rayleigh was distribution function has from the data and is fitted to the measured probability distribution on function used to calculate the been wind derived power density foravailable each location in order to classify them in terms of wind energy. The yearly basis. The power densities for thefrom two stations varieddata fromand 295.39 to 194.5 / m2. Rayleigh distribution function hasreported been derived the available is fitted to theWmeasured probability distribution on District heating networks are commonly addressed in the literature as one of the most effective solutions for decreasing the yearly basis. The power densities reported for the two stations varied from 295.39 to 194.5 W / m2. greenhouse gas emissions from the building sector. These systems require high investments which are returned through the heat © 2018 The Authors. Published by Elsevier Ltd. sales. Due to the changed climate conditions and building renovation policies, heat demand in the future could decrease, © 2019 The Authors. by Ltd. This is an open accessPublished article under the CC BY-NC-ND license (https://creativecommons.org/licenses/by-nc-nd/4.0/) © 2018 The Authors. Published by Elsevier Elsevier Ltd. This is an open articlereturn underperiod. the CC BY-NC-ND license (https://creativecommons.org/licenses/by-nc-nd/4.0/) prolonging theaccess investment Selection under responsibility of the scientific of Technologies and Materials for Renewable Energy, This is an and openpeer-review access article under the CC BY-NC-ND license committee (https://creativecommons.org/licenses/by-nc-nd/4.0/) Selection peer-review under responsibility of the scientific of Technologies Materialsfunction for Renewable The mainand scope of this paper is to assess the feasibility of usingcommittee the heat demand – outdoorand temperature for heatEnergy, demand Environment and Sustainability, TMREES18. Selection and peer-review under responsibility of the scientific committee of Technologies and Materials for Renewable Energy, Environment Sustainability, TMREES18. forecast. Theand district of Alvalade, located in Lisbon (Portugal), was used as a case study. The district is consisted of 665 Environment and Sustainability, TMREES18. buildings that vary in both construction period and typology. Three weather scenarios (low, medium, high) and three district Keywords: Wind energy; Wind speed; Rayleigh distribution functions; Wind power density; Probability density function; renovation scenarios were developed (shallow, intermediate, deep). To estimate the error, obtained heat demand values were Keywords: Wind energy; Wind speed; Rayleigh distribution functions; Wind power density; Probability density function; compared with results from a dynamic heat demand model, previously developed and validated by the authors. The results showed that when only weather change is considered, the margin of error could be acceptable for some applications 1. Introduction (the error in annual demand was lower than 20% for all weather scenarios considered). However, after introducing renovation 1. Introduction scenarios, the error value increased up to 59.5% (depending on the weather and renovation scenarios combination considered). Due to recent advances, the windwithin energy in 3.8% Europe experienced and sustainable The value of slopetechnological coefficient increased on average the sector range of up has to 8% per decade,rapid that corresponds to the Due over toinrecent technological advances, the driver wind energy sector in Europe has need experienced rapidto and growth thenumber last 20ofyears. important of this rapid growth is the benefit fromand a decrease the heatingAn hours of 22-139h during the heating season (depending onfor the Europe combination of sustainable weather growth overscenarios the lastconsidered). 20 driver of thisintercept rapid is for thesecond need major for per Europe benefit from a secure energy supply thatyears. is not An dependent foreign sources of oilgrowth and gas. A driver the significant renovation On important the otheronhand, function increased 7.8-12.7% decadetois(depending on the secure energy supply that is notsuggested dependent on be foreign sources oil function and gas.parameters A second for major is the significant coupled scenarios). The values could used to modifyofthe the driver scenarios considered, and improve the accuracy of heat demand estimations.
© 2017 The Authors. Published by Elsevier Ltd. * Corresponding author. Tel.: +40721320403. Peer-review under responsibility of the Scientific Committee of The 15th International Symposium on District Heating and E-mail address:author. [email protected] * Corresponding Tel.: +40721320403. Cooling. E-mail address: [email protected] 1876-6102 © 2018 The Authors. Published by Elsevier Ltd. Keywords: Heat demand; Forecast; Climate change This is an open access under the CC BY-NC-ND license (https://creativecommons.org/licenses/by-nc-nd/4.0/) 1876-6102 © 2018 Thearticle Authors. Published by Elsevier Ltd. Selection under responsibility of the scientific of Technologies and Materials for Renewable Energy, Environment This is an and openpeer-review access article under the CC BY-NC-ND licensecommittee (https://creativecommons.org/licenses/by-nc-nd/4.0/) and Sustainability, TMREES18. Selection and peer-review under responsibility of the scientific committee of Technologies and Materials for Renewable Energy, Environment and Sustainability, TMREES18. 1876-6102 © 2017 The Authors. Published by Elsevier Ltd. 1876-6102 © 2019 The Authors. Published by Elsevier Ltd. Peer-review under responsibility of the Scientific Committee of The 15th International Symposium on District Heating and Cooling. This is an open access article under the CC BY-NC-ND license (https://creativecommons.org/licenses/by-nc-nd/4.0/) Selection and peer-review under responsibility of the scientific committee of Technologies and Materials for Renewable Energy, Environment and Sustainability, TMREES18. 10.1016/j.egypro.2018.11.320
2
Lizica-Simona Paraschiv et al. / Energy Procedia 157 (2019) 1546–1552 Author name / Energy Procedia 00 (2018) 000–000
1547
contribution that wind energy can make to reducing greenhouse gas emissions. As a clean and renewable resource, wind energy is already playing - and will continue to play an important role in mitigating climate change, bringing at the same time, benefits in terms of reducing emissions of air pollutants and associated cooling water consumption with many of the conventional energy generation technologies. The calculation of wind resources for a given location and the corresponding energy production is based on the wind potential assessment by anemometric measurement, followed by wind data processing to calculate the expected wind power yield for the proposed site. Typical resource assessment methods involve measuring wind speed and recording the arithmetic mean every 10 minutes or one hour for at least one year. These wind speed data are then used to obtain a relative frequency distribution to which a probability density function is fitted. This function and specific wind turbine power curve are required to calculate the amount of available energy and the electrical power generated in a specific region. 2. Site Location and Data Collection This study aimed to examine the wind energy potential of Constanta by finding Rayleigh distribution parameter and determining the available power density. The wind speed data used in this study were measured in Constanta in 2017 at two meteorological stations located near the Black Sea as shown in Fig.1. The meteorological stations are located in the south-east coastal region of Constanta, Romania. The daily wind speed data (in m/s) was collected using cup anemometer at the height of 10 m from ground level, with an hour interval over a period of 1 year, from January 2017 to December 2017. There were analyzed 17234 valid data from 17540 data recorded from both stations.
Fig. 1. Location of meteorological stations.
Because the wind speed tends to increase with height and depends mainly on atmospheric mixing and terrain roughness, therefore, we need to calculate the total wind energy potential, so the measured variation in wind speed with height at each site is defined by:
V z1 z1 V z2 z2
z V2 V1 1 z2
(1)
Where, α is power law wind shear exponent, V is the mean wind speed, z1 is the measurement height above ground level and z2 is the turbine height (m).
Lizica-Simona Paraschiv et al. / Energy Procedia 157 (2019) 1546–1552 Author name / Energy Procedia 00 (2018) 000–000
1548
3
The wind shear exponent depends on the complexity and roughness of the terrain for the specific site. 3. Rayleigh distribution Rayleigh's distribution can be used to describe the wind variations in a wind regime with an acceptable level of accuracy without the need for data collected over short time intervals, as is the case of the Weibull distribution, because in many cases, such information cannot be available. Rayleigh's distribution is a simplified case of the Weibull distribution, which is derived by assuming the shape factor as 2. Because of its simplicity, this distribution is widely used for wind energy modeling. The average velocity is expressed as:
1 Vm c 1 k
(2)
Taking k = 2 in Eq. (2), obtain:
Vm c
2Vm 3 c 2
(3)
The probability density function indicates the fraction of time (or probability) for which the wind is at a given velocity V and it is given by:
f (V )
V 2 Vm2
e
V 2 4 Vm
(4)
Cumulative distribution is given by:
F (V ) 1 e
V 2 4 Vm
(5)
The validity of Rayleigh distribution in wind energy analysis has been established by comparing the Rayleigh generated wind pattern with long term field data. The probability of wind velocity to be between V1 and V2 is:
P(V1 V V2 ) e
V 2 1 4 Vm
e
V 2 2 4 Vm
(6)
and the probability of wind to exceed a velocity of VX is given by: V 2 X 4 Vm
P(V VX ) e
(7)
In order to check how accurately the theoretical probability density function fits with observation data the root mean square error (RMSE) parameter test is performed. Best results are obtained when these values are close to zero.
4
Author name / Energy Procedia 00 (2018) 000–000
Lizica-Simona Paraschiv et al. / Energy Procedia 157 (2019) 1546–1552
1 RMSE N
1
N
y x i 1
1549
i
i
2
2
(8)
Where N is the number of observations, 𝑦𝑦𝑖𝑖 is the frequency of observation, 𝑥𝑥𝑖𝑖 is the frequency of Rayleigh.
4. Results and Discussion
The data of wind speed in the present calculation was obtained in 2017 from weather stations in Constanta. The hourly mean wind values measured throughout the total year at the two stations are shown in Fig. 2 and 3. It can be seen in Fig. 2 and 3, that the highest wind speed of 26.6 m / s at the CT-3 station and 24.3 m / s at the CT-6 station occurs in January.
Wind speed [m/s]
30.0
26.6
25.0
Vm=5.6 m/s
20.0 15.0 10.0 5.0 1 252 503 754 1005 1256 1507 1758 2009 2260 2511 2762 3013 3264 3515 3766 4017 4268 4519 4770 5021 5272 5523 5774 6025 6276 6527 6778 7029 7280 7531 7782 8033 8284 8535
0.0
Hours ‐ 2017
Fig. 2. Hourly variation of wind speed at CT-3 (2017)
25.0
24.3
Vm=4.7 m/s
20.0 15.0 10.0 5.0 0.0 1 245 489 733 977 1221 1465 1709 1953 2197 2441 2685 2929 3173 3417 3661 3905 4149 4393 4637 4881 5125 5369 5613 5857 6101 6345 6589 6833 7077 7321 7565 7809 8053 8297 8541
Wind speed [m/s]
30.0
Hours ‐ 2017
Fig. 3. Hourly variation of wind speed at CT-6 (2017)
Lizica-Simona Paraschiv et al. / Energy Procedia 157 (2019) 1546–1552 Author name / Energy Procedia 00 (2018) 000–000
1550
5
The probability density distribution obtained from the Rayleigh model was compared to the measured distributions to study their suitability. The annual comparison shows that the Rayleigh model fit well with the measured probability density distribution as shown in Fig. 4 and 5. 0.14 0.12
RMSE=0.012429
Frequency
0.1
k=1.458497
0.08 0.06 0.04 0.02 0
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25
Wind speed [m/s] Observed
Rayleigh column
Rayleigh line
Fig. 4. Comparison of observed and predicted wind speed frequencies at CT-3 (2017)
0.18 0.16
RMSE=0.013499
0.14
k=1.364391
Frequency
0.12 0.1 0.08 0.06 0.04 0.02 0
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25
Wind speed [m/s] Observed
Rayleigh column
Rayleigh line
Fig. 5. Comparison of observed and predicted wind speed frequencies at CT-6 (2017)
6
Author name / Energy Procedia 00 (2018) 000–000 Lizica-Simona Paraschiv et al. / Energy Procedia 157 (2019) 1546–1552
1551
Analysis of probability density function presented in Fig. 4 and Fig 5 shows the Rayleigh probability and cumulative distributions for all speed ranges observed during the 2017. The real data is presented on the same diagrams so that a comparison can be made between real data and resulted probability density functions. In Fig. 4, it is observed that most of the wind speeds are in the range of 1-7 m/s during all seasons. Rayleigh seems to give a reasonable fit to observed wind speed data as distribution of observed data is not largely deviated from estimated Rayleigh during entire year. Fig. 4 and 5 contain the values of RMSE for Rayleigh distribution and the values of correlation coefficient in both figures are very low (nearly 0), which indicates that the predictions made by Rayleigh are extremely accurate. Errors in calculating the power densities using the distribution Rayleigh model in comparison to those using the measured probability density distribution are presented in Fig. 6. The highest error values occur for wind speed of 1-4 m/s with 4.03% for CT-3 and 3.16% for CT-6 respectively. 0.04
CT‐3
0.03
CT‐6
0.02
Error [%]
0.01 0 ‐0.01
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25
‐0.02 ‐0.03 ‐0.04 ‐0.05
Wind speed [m/s]
Fig. 6. Error values in calculating the wind power density obtained from the Rayleigh models in reference to the wind power density obtained from the measured data.
The CT-3 station reported an average power density of 295.4 W/m2 and the CT-6 station reported a value of 194.5 W/m2 as shown in Fig. 7.
Wind power density [W/m^2]
295.4 194.4999828
300.0 200.0 100.0 0.0
CT‐3
CT‐6
Fig. 7. Average power density
1552
Author name / Energy Procedia 00 (2018) 000–000 Lizica-Simona Paraschiv et al. / Energy Procedia 157 (2019) 1546–1552
7
In agreement with the commercially international system of classification for wind power scale, the CT-6 location correspond to the wind power class 3, since the density value is lower than 200 W/m2. Therefore, this urban location present wind characteristics that are not suitable for wind power development. The CT-3 location corresponds to the wind power class 4 which is generally accepted as a feasible class for commercial wind power development. 5. Conclusion This study analyzed wind characteristics and wind energy potential in Constanta, Romania. For this, wind speed data was analyzed at two monitoring stations over a one-year period at one-hour intervals to predict the wind behavior as accurately as possible. Based on this analysis, the following conclusions can be drawn: 1. The average wind speed was 5.6 m for CT-3 and 4.7 for CT-6, respectively. 2. RMSE for Rayleigh distribution was 0.012429 for CT-3 and 0.013499 for CT-6, respectively. 3. The average wind energy density based on the average wind speed approach was 295.4 W / m2 for CT-3 and 194.49 W/ m2 for CT-6, respectively. Wind power density calculations have indicated that the CT-3 region is in Class 4, which is a suitable area for wind turbine installation. References [1] Sathyajith Mathew, “Wind Energy-Fundamentals, Resource Analysis and Economics”, Springer Berlin Heidelberg New York, (2006). [2] Trevor M. Letcher, Wind Energy Engineering - A Handbook for Onshore and Offshore Wind Turbines, Academic Press, (2017) [3] Sajid Ali, Sang-Moon Lee, Choon-Man Jang, “Statistical analysis of wind characteristics using Weibull and Rayleigh distributions in Deokjeok-do Island e Incheon, South Korea”, Renewable Energy 123 (2018): 652-663, doi: 10.1016/j.renene.2018.02.087. [4] S. A. Ahmeda, H. O. Mahammeda, “A Statistical Analysis of Wind Power Density Based on the Weibull and Ralyeigh models of "Penjwen Region" Sulaimani/ Iraq”, Jordan Journal of Mechanical and Industrial Engineering, 6 (2), (2012): 135 – 140. [5] Chong Li, Youying Liu, Gang Li, Jianyan Li, Dasheng Zhu, Wenhua Jia, Guo Li, Youran Zhi, Xinyu Zhai, “Evaluation of wind energy resource and wind turbine characteristics at two locations in China”, Technology in Society 47, (2016): 121-128, http://dx.doi.org/10.1016/j.techsoc.2016.09.003 [6] K.S.R. Murthy, O.P. Rahi, “A comprehensive review of wind resource assessment”, Renewable and Sustainable Energy Reviews 72 (2017): 1320–1342, http://dx.doi.org/10.1016/j.rser.2016.10.038 [7] Ahmad Hemami, “Wind turbine technology”, Cengage Learning, (2012). [8] Ahmad Sedaghat, Arash Hassanzadeh, Jamaloddin Jamali, Ali Mostafaeipour, Wei-Hsin Chen, “Determination of rated wind speed for maximum annual energy production of variable speed wind turbines”, Applied Energy 205 (2017): 781–789, http://dx.doi.org/10.1016/j.apenergy.2017.08.079. [9] Cristina Herrero-Novoa, Isidro A.Pérez, M. Luisa Sánchez, M. Ángeles García, Nuria Pardo, Beatriz Fernández-Duque, “Wind speed description and power density in northern Spain”, 138 (1) (2017): 967-976. [10] Paraschiv Spiru, Paraschiv Lizica-Simona, “Technical and economical analysis of a PV/wind/diesel hybrid power system for a remote area” Energy Procedia 147 (2018): 343-350, doi:10.1016/j.egypro.2018.07.102.