Wind Resource Assessment: Optimizing the Weibull Distribution Curve

Abstract

In this paper, it was analyze the wind speed characteristics and provide projections of optimal wind resources by comparing data at different hub heights using wind power law. We have acquired data from an operational project site in the western part of India using data points acquired from the site from August 2017 to February 2018.Weibull distribution probability distribution function and cumulative distribution function for wind speed and other parameters involved in an accurate wind resource assessment methodology. To predict wind regimes, the Weibull probability function is commonly used. A comparison of alternative methods for estimating Weibull parameters of a wind regime is presented in this research. For the estimation, five alternative methods are discussed and employed. The study examined wind speed time series data at two sites over the course of a year. Although all five methods produce similar results when plotting error versus wind speed, the standard deviation method produces the best results.

Introduction

I. INTRODUCTION

According to the Energy Outlook Report of 2021 by IEA India is the world’s third-largest energy-consuming country [1]. According to the same report, it was also observed that energy use in India has increased almost two-fold since the early 2000s[1]. While this is a positive indicator of economic growth it also indicates that India, with a background of depleting coal reserves and high thermal power dependency, stands at the brink of a massive energy crisis. If India were to face such a crisis one of two things might ensue; a massive rise in the per-unit cost of electricity or long hours of load shedding, both due to the economics of demand and supply. In such a scenario and with India’s goal of attaining carbon neutrality by the year 2070[2] the renewable energy sector

should be utilized to the maximum potential.

Resource assessment is the preliminary check for project viability at an identified potential project site. In the case of solar power plants, this process is relatively simple and most of the parameters do not have unprecedented variations and more or less follow a pattern that is predictable. Wind resource assessment, however, has been posing problems in the recent past. It has been observed that wind patterns are changing and becoming unpredictable primarily owing to climate change which we shall discuss through the length of this paper. In light of this, wind resource assessment has become particularly tedious and error-prone, project commissioning has become tougher and the existing units are largely underdelivering in terms of generation. Due to these existing utilities are incurring a loss and new companies are facing major bottlenecks in assessing project viability. To overcome these roadblocks and for wind power to emerge as a dependable power source, it is imperative that the wind resource methodology be re-evaluated and optimization techniques should be employed to improve its reliability. Through the course of this paper, we shall try to analyze various optimization techniques for wind resource assessment optimization and outline the best possible route for the Indian industry scenario. 

II. WIND SCENARIO IN INDIA

According to prem chaurasia et al [2], India has the second to largest wind market across Asia and fourth largest across the globe. It is also common knowledge that India has pioneered the most ambitious energy expansion initiatives. Wind power generation has seen a considerable boom in recent years in the country. As of 28 February 2021, the sum total of the installed capacity in India for wind power generating units was 38.79 GW which was the fourth largest installed wind capacity in the world according to The Ministry of New and Renewable Energy (MNRE), Government of India [4]. According to studies conducted by the National Institute of wind energy (NIWE), most of (95%) India’s wind potential exists in 7 windy states of the country; Gujrat, Rajasthan, Maharashtra, Tamil Nadu, Madhya Pradesh, Karnataka and Andhra Pradesh.

With talks about repowering of existing wind turbines (refer to Policy for Repowering of the Wind Power Projects-2016, MNRE) [5] with a background of a considerable change in wind patterns, it becomes that the existing and abundant wind resource be exploited to its full potential. This would not only ensure a sustainable shift from thermal power to renewable energy sources but also keep the national grid safe by injecting reactive VARs into the grid and saving it from a disastrous blackout seen in Chicago due to overdependency on the solar resource which brought forth to the iconic duck curve study [6].

III. RECENT CHALLENGES IN WIND POWER GENERATION

Westerlies also known as westerly winds impact the weather and climate on both a local and global scale. They have a high influence on precipitation patterns, ocean circulation and steering tropical cyclones, thus discovering their change in pattern as the climate warms is extremely important.

Along the planet's middle latitude, westerlies blow from West to east. However, the westerlies have been migrating poleward over the last few decades. But there has been continuous debate on whether this poleward migration of westerlies will continue since the release of atmospheric carbon dioxide is increasing and leading to increased temperature. But, due to our limited knowledge about the westerlies, this scientific debate is still on thus posing the question of its movement. The flow of air is affected by the difference in temperature between the neighbouring regions or between the ocean and land areas that are nearby. This impact is supported by an instance in which the researchers discovered that the speed of wind is reduced along the Northern Hemisphere when some parts of tropical Atlantic, western Pacific and Greenland experience higher temperatures.

Human activities have highly affected the climate and thus the temperature is rising at a rapid rate. But it has also come to notice that in larger, long-term warming patterns, temperature cycles are going back and forth in between the warmer and cooler periods and this change can last for even decades. Researchers also found some reductions in wind power potential in both China as well India even though the change is small (around 1% in China and 2% in India).

IV. WIND RESOURCE ASSESSMENT

The process by which wind power developers predict the future energy production of a wind farm is known as wind Resource assessment. The most significant phase in developing a community wind project is wind resource evaluation, which serves as the foundation for evaluating initial feasibility and cash flow predictions, as well as obtaining finance. Anemometers and wind direction vanes, which are sensors that sense wind speed and direction, a data recorder, and a meteorological mast, or tower, are the three basic components of the wind resource assessment apparatus.

V. WEIBULL DISTRIBUTION

In a general data-oriented regime wind is most appropriately represented by Weibull probability distribution function and using this has now become a common academic practice for optimisation research purposes [7]. Although some experts criticise the 2-parameter Weibull model for its theoretical inadequacies in representing real-world wind velocity patterns, everyone agrees that it is the industry standard [8].

The Weibull probability distribution function can be defined as follows:

(Source: Danish Wind Industry Association)

The Weibull parameters can be estimated using a variety of methods. Even with the same set of wind data, the parameter values obtained differ depending on the approach utilised. As a result, a comparison of different ways of evaluating Weibull parameters is possible. In this work, an attempt has been made to suggest the optimum approach based on such a comparison.

VI. EXPERIMENTAL DATA SAMPLES

Wind speed at hub heights of 100m, 80m, 50m, 20m at 10-minute intervals, coordinates of the site, Temperature at 10-minute intervals, pressure at 10-minute intervals data are acquired for the course of this simulation from a site in Kaythar, Tamil Nadu a place with high wind potential according to the wind energy atlas [].

VII. DATA VALIDATION

According to NREL’s wind resource assessment handbook [7] to ensure that the data being used for analysis is free from errors and accurate, cleaning and validation of data are required. This is to ensure optimal results free of random errors.

The data is evaluated on the following criteria for data validation:

1. If the data is complete
2. If the data is reasonable
3. If the data is free of errors

The results of data validation on the available data set are as follows: -

a. General System checks

• Data records: Expected data fields for each parameter = 52,560 (144 entries per day)

Number of available data sets = 52,505

• Time Sequence: All expected time sequences are present at 10-minute intervals.

b. Measured Parameters check: These tests form the basis of data validation and can be broadly classified as:

• Range test
• Relational Test
• Trend Tests

The tables appended below give the criteria of all these tests for wind-related data used in the course of this study.

shape and scale factors, which are parameters of Weibull distribution conclude the wind speed that is most suitable for a highly efficient wind energy conversion system as well as the speed range over which operation is expected.

XI. ACLKNOWLEDGEMENTS

We would like to extend our gratitude and sincere thanks to all those who helped in timely completion of this research paper. It has been a great honour and privilege to work along with our fellow project mates and a special thanks to our project guide Dr. Surabhi Chandra, and head of the department Dr. D.S. Bankar without whose guidance, stimulating suggestions and constant encouragement, this project would not have turned out to be this successful. We would also like to acknowledge the crucial role of the National Institute of Wind Energy (NIWE) Chennai who provided us with all the required data that helped in the stimulation of our project. At last, we would express our special thanks to our parents for cooperating and helping us with our project.

Conclusion

In this research paper, we studied four different methods to determine the values of k and c i.e., shape parameter and scale parameter respectively. We observed how efficiently Weibull parameters can be calculated with minimum errors. Out of the four methods, the maximum likelihood method and power density method are more efficient as compared to the empirical method and method of moments. While calculating the Weibull parameters it was found that the maximum likelihood method gives the best Weibull distribution for a given wind speed. It was also noted that the error in computation increases with a reduction in hub height. A statistical method was also carried out to rank the methods using RMSE, chi-square and efficiency of the method- R2. According to RMSE and chi-square for the most likelihood method smallest values were calculated. The calculated values by these three methods have magnitudes very close to each other, therefore the most likelihood method (MLM) without any doubt shows the best performance.

References

[1] IEA (2021), Energy Outlook Report, International Energy Association [2] Prem Kumar Chaurasiya, Vikas Warudkar, Siraj Ahmad (2019), Wind Energy Development and Policy in India: A review, Volume 24 Science Direct Energy Strategy reviews [3] MNRE, Wind Energy Overview, Ministry of New and Renewable Energy India [4] MNRE (2016), Policy for Repowering of Wind Projects, Ministry of New and Renewable Energy (Wind Energy Division) [5] Samantha A Janko, Michael R Arnold, Nathan G Johnson (2016), Implications of High Penetration Renewables and utilities in Residential PV markets, Volume 180 Applied Energy Elsevier. [6] Gary L. Johnson (1985), Wind Energy Systems 1985, Prentice-Hall, Inc., Englewood Cliffs, New Jersey 07632. [7] J. C. Chadee and C. Sharma (2001) Wind speed distributions: a new catalogue of defined models, Wind Engineering, Vol. 25, No. 6. [8] J Ravi Chandran(2019), Probability and Random Processes for Engineers, Wiley Publications [9] IRENA, Global Atlas for Wind Energy, International Renewable Energy Agency [10] A. K. Azad, M. G. Rasul, T. Yusaf, “Statistical Diagnosis of the Best Weibull Methods for Wind Power Assessment for Agricultural Applications”, Energies ISSN 1996-1073, pp. 3057-3085, 2014. [11] A.N. Celik, “A statistical Analysis of Wind Power Density Based on The Weibull and Rayleigh Models at the southern Region of Turkey”, Renewable Energy Vol. 29, No. 2, pp. 21-33, 2006. [12] D.K. Kaoga, D. Raidandi, N. Djongyang, S.Y. Duka, “Comparison of five numerical methods for estimating Weibull parameters for wind energy applications in the district of Kousseri, Cameroon“, Asian Journal of Natural & Applied Sciences 3, pp. 72-87, 2014. [13] S. Andronopoulos, “Model description of the Rodos meteorological pre– processor”, Environmental Research Laboratory, Institute of Nuclear Technology and Radiation Protection, National Centre for Scientific Research “Demokritos”, Greece, 2009. [14] I. Munteanu, A. I. Bratcu, N. A. Cutululis, E. Ceanga, “Optimal Control of Wind Energy Systems – Towards a Global Approach”, Springer- Verlag London, 2008. [15] D. Indhumathy, C.V. Seshaiah, K. Sukkiramathi, “Estimation of Weibull Parameters for Wind speed calculation at Kanyakumari in India”, International Journal of Innovative Research in Science, Engineering and Technology ISSN: 2319-8753, pp. 8340-8345, 2014. [16] A. Costa Rocha, R.C. de Sousa, C.F. de Andrade, M.E.V. da Silva, “Comparison of seven numerical methods for determining Weibull parameters for wind energy generation in the northeast region of Brazil”, Appl. Energy 89, 395-400, 2012. [17] A. S. Glantz, B.K. Slinker, “Primer od Applied Regression and Analysis of Variance”, 1990. [18] A Gelman \"A Bayesian Formulation for Exploratory Data Analysis and Goodness-of-Fit Testing\", International Statistical Review 71, pp. 369-382, 200

Copyright

Copyright © 2022 Arundhati Dogra, Anshul Sathe, Dr. D. S. Bankar, Dr. Surabhi Chandra, Anirudh Singh Malik. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.