Performance Comparison of Different Machine Learning Algorithms on the Prediction of Wind Turbine Power Generation

Eyecioğlu, Önder; Hangun, Batuhan; Kayışlı, Korhan; Yeşilbudak, Mehmet

doi:10.1109/icrera47325.2019.8996541

Cited by 25 publications

(16 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A first-order model is used for modeling the generator. The generator torque is described [29] as (6) where Tg-rf is the generator torque reference and γg is the generator time constant. The power generated Pg by the generator is expressed as (7) where ηg is the efficiency of the generator.…”

Section: Generatormentioning

confidence: 99%

See 1 more Smart Citation

Pitch Control of a Digital Hydraulics Pitch System for Wind Turbine Based on Neuro-Fuzzy Digital Pitch Controller

Narayanan

Ramakrishnan

2021

IJRER

View full text Add to dashboard Cite

Digital Hydraulics has been implemented in the hydraulic pitch system of Wind Turbine (WT) which aids in uniform power generation at above-rated wind speeds. A novel Digital Hydraulic Pitch System (DHPS) model which uses a hydraulic motor as the end actuator was established and a wind generator system model was developed. Further, a Neuro-Fuzzy Digital Pitch Controller (NFDPC) with two distinct neuro-fuzzy controllers was employed in the developed models. Two comparative studies are conducted and in these studies, the proposed NFDPC was compared with Proportional-Integral-Derivative Cascade Controller (PIDCC) where PID is implemented in the inner and outer loop. In the first comparative study (considering custom wind speed), results revealed that the proposed controller maintains rated generator speed than PIDCC. The Root Mean Square Error (RMSE) (between rated speed and generated speed) of the generator was 1.23 and 3.73 for the proposed and PIDCC, respectively. Thus, the proposed controller generates uniform power. Consequently, in the second case study (real-time effective wind speed), the proposed controller maintains optimal generator speed with a minimal RMSE of 0.66 when compared to PIDCC with an RMSE of 3.55. Thus, uniform power was developed by the proposed controller. Therefore from these results, the proposed controller generates uniform power at above-rated wind speed with large wind fluctuations.

show abstract

Section: Generatormentioning

confidence: 99%

“…One of the primary research work in WT is to generate uniform power. Moreover, to generate optimal power at above-rated wind speeds, pitch control is mostly preferred [4][5][6]. In general, two types of pitch systems are used in the WT such as Electro-mechanical Pitch System (EPS) and Hydraulic Pitch System (HPS).…”

Section: Introductionmentioning

confidence: 99%

Pitch Control of a Digital Hydraulics Pitch System for Wind Turbine Based on Neuro-Fuzzy Digital Pitch Controller

Narayanan

Ramakrishnan

2021

IJRER

View full text Add to dashboard Cite

show abstract

“…The wind power density describes the wind resource available at a site, i.e., the capacity of a specific site for the wind energy production. It can be calculated using the relation [6]: Months from Jan to dec (1)(2)(3)(4)(5)(6)(7)(8)(9)(10)(11)(12) where P is the wind power per unit area (A). Equation 9 can also be expressed as follows [10]:…”

Section: Wind Power Densitymentioning

confidence: 99%

Feasibility study of distributed wind energy generation in Jumla Nepal

Dhakal

Yadav

Koirala

et al. 2020

Preprint

View full text Add to dashboard Cite

Renewable energy production needs serious attention in highly traditional, inefficient, and energy-dependent countries like Nepal. Moreover, the option of an effective renewable energy technology that is economically feasible and environmentally acceptable is a topic of interest due to the availability of various types of renewable energy sources in Nepal. Among other renewable energy sources like micro hydro, solar, biogas etc, very few studies had been conducted on wind energy sources in Nepal and those few studies also focuses mostly on large scale wind farming. So, this study analyzes the suitability of distributed wind energy production in Tila village of Jumla district in the western part of Nepal. Five-year (2015-2019) wind speed data were examined to obtain wind power density and energy density. Two-parameter Weibull probability density function was used to evaluate these two quantities. The annual Weibull parameters k and c of 1.73 and 4.21 m/s were obtained to calculate 43.79 W/m2 power density and 378.37 kWh/m2 energy density. This study also provides the economic evaluation of a 100 kW distributed wind energy system, and the technical and economic aspects of the proposed system are compared with the corresponding characteristics of the existing renewable energy systems, i.e., micro hydropower and solar power. The study shows that when there is not enough sunlight for the solar PV system and not enough water flow coupled with maintenance problems in the micro hydropower system, the distributed wind energy system may function as a substitute system.

show abstract

“…Multiple models, like Gradient Boosted Trees [5], SVMs [6], and Artificial Neural Networks [7]- [9] have been used in these cases. Regression based models have also been analyzed [10] in the estimation of wind speeds. While metrics like RMSE, MAE, and R-Square have been studied in this angle, it is easy to observe that these are not conflicting.…”

Section: Introductionmentioning

confidence: 99%

Benefits of Multiobjective Learning in Solar Energy Prediction

Kannan¹

2023

Preprint

View full text Add to dashboard Cite

While the space of renewable energy forecasting has received significant attention in the last decade, literature has primarily focused on machine learning models that train on only one objective at a time. A host of classification (and regression) tasks in energy markets lead to highly imbalanced training data. Say, to balance reserves, it is natural for market regulators to have a choice to be more/less averse to false negatives (can lead to poor operating efficiency and costs) than to false positives (can lead to market shortfall). Besides accuracy, other metrics like algorithmic bias, RMBE (in regression problems), inferencing time, and model sparsity are also very crucial. This paper is amongst the firsts in the field of renewable energy forecasting that attempts to present a Pareto frontier of solutions (tradeoffs), that answers the question on handling multiple objectives by means of using the XGBoost model (Gradient Boosted Trees). Our proposed algorithm relies on using a sequence of weighted (uniform meshes) single objective model training routines. Real world data examples from the Amherst (Massachusetts, United States) solar energy prediction panels with both triobjective (focus on accuracy) and biojective (focus on fairness/bias) classification instances are considered. Numerical experiments appear promising and clear advantages over single objective methods are seen by observing the spread and variety of solutions (model configurations).

show abstract

Performance Comparison of Different Machine Learning Algorithms on the Prediction of Wind Turbine Power Generation

Cited by 25 publications

References 15 publications

Pitch Control of a Digital Hydraulics Pitch System for Wind Turbine Based on Neuro-Fuzzy Digital Pitch Controller

Pitch Control of a Digital Hydraulics Pitch System for Wind Turbine Based on Neuro-Fuzzy Digital Pitch Controller

Feasibility study of distributed wind energy generation in Jumla Nepal

Benefits of Multiobjective Learning in Solar Energy Prediction

Contact Info

Product

Resources

About