# A Study of Wind Turbine Performance Decline with Age through Operation Data Analysis

^{1}

^{2}

^{3}

^{*}

## Abstract

**:**

## 1. Introduction

## 2. The Test Case

^{TM}and Opti-speed

^{TM}control mechanisms. The control mechanisms aim to maximise energy capture at wind speeds below the rated power wind speed and to fix the power output to rated power at wind speeds above the rated power wind speed. In normal turbine operation the blade pitch angle is always below 20 degrees. In a fault condition or a pause/stop state, the blade pitch angle is fixed to approximately 86 degrees. Time series data of a number of turbine parameters are logged by the wind turbine SCADA system in 10-minute mean values. These parameters include: wind speed, wind speed standard deviation, absolute wind direction, relative wind direction, rotor RPM, generator RPM, blade pitch angle and power output. A number of 10-minute mean temperature parameters are also logged and include: gearbox oil temperature, gearbox bearing temperatures, generator bearing temperatures, internal nacelle temperature and external ambient air temperature at hub height.

## 3. Methods

- Wind turbines are typically equipped with cup anemometers mounted behind the rotor and the undisturbed wind speed is ex-post reconstructed through a nacelle transfer function;
- The power curve has non-trivial seasonal and ambient conditions dependence.

#### 3.1. Support Vector Machine Regression

- Undisturbed wind speed average, estimated by the control system through the nacelle transfer function;
- Turbulence intensity, estimated as the ratio between the standard deviation and the average of the undisturbed wind speed;
- Yaw angle;
- Blade pitch angle;
- Rotor speed;
- Generator speed.

#### 3.2. The Data Sets Arrangement and the Performance Monitoring

- ${\mathrm{D}}_{2008}$;
- ${\mathrm{D}}_{2012}$;
- ${\mathrm{D}}_{2013}$;
- ${\mathrm{D}}_{2014}$;
- ${\mathrm{D}}_{2017}$;
- ${\mathrm{D}}_{2018}$.

- ${\mathrm{D}}_{2008}$ is randomly divided in two subsets: D0 (a random selection of $\frac{2}{3}$ of the data set) and D1 (the remainder $\frac{1}{3}$ of the data set). D0 is used for training the regression, D1 is used for testing the regression. The convergence of model training is obtained through the MATLAB
^{®}$fitrsvm$ routine. - A data set posterior to ${\mathrm{D}}_{2008}$ is named as D2 and is used to quantify the performance deviation with respect to 2008.

#### 3.3. Power Curve Analysis

^{®}$fitdist$ function that is based on the statistical method of moments.

## 4. Results

#### 4.1. Regression Results

#### 4.2. Analysis before and after Gearbox Replacement

#### 4.3. Power Curve Analysis and Production Assessment

## 5. Conclusions

## Author Contributions

## Acknowledgments

## Conflicts of Interest

## References

- Kumar, Y.; Ringenberg, J.; Depuru, S.S.; Devabhaktuni, V.K.; Lee, J.W.; Nikolaidis, E.; Andersen, B.; Afjeh, A. Wind energy: Trends and enabling technologies. Renew. Sustain. Energy Rev.
**2016**, 53, 209–224. [Google Scholar] [CrossRef] - He, D.; Li, Y. Overview of Worldwide Wind Power Industry. In Strategies of Sustainable Development in China’s Wind Power Industry; Springer: Singapore, 2020; pp. 29–60. [Google Scholar]
- Byon, E.; Ntaimo, L.; Singh, C.; Ding, Y. Wind energy facility reliability and maintenance. In Handbook of Wind Power Systems; Springer: Berlin/Heidelberg, Germany, 2013; pp. 639–672. [Google Scholar]
- Artigao, E.; Martín-Martínez, S.; Honrubia-Escribano, A.; Gómez-Lázaro, E. Wind turbine reliability: A comprehensive review towards effective condition monitoring development. Appl. Energy
**2018**, 228, 1569–1583. [Google Scholar] [CrossRef] - Echavarria, E.; Hahn, B.; Van Bussel, G.; Tomiyama, T. Reliability of wind turbine technology through time. J. Sol. Energy Eng.
**2008**, 130. [Google Scholar] [CrossRef] - Carroll, J.; McDonald, A.; McMillan, D. Failure rate, repair time and unscheduled O&M cost analysis of offshore wind turbines. Wind Energy
**2016**, 19, 1107–1119. [Google Scholar] - Yang, W.; Court, R.; Jiang, J. Wind turbine condition monitoring by the approach of SCADA data analysis. Renew. Energy
**2013**, 53, 365–376. [Google Scholar] [CrossRef] - Zaher, A.; McArthur, S.; Infield, D.; Patel, Y. Online wind turbine fault detection through automated SCADA data analysis. Wind Energy Int. J. Prog. Appl. Wind Power Convers. Technol.
**2009**, 12, 574–593. [Google Scholar] [CrossRef] - Butler, S.; Ringwood, J.; O’Connor, F. Exploiting SCADA system data for wind turbine performance monitoring. In Proceedings of the 2013 IEEE Conference on Control and Fault-Tolerant Systems (SysTol), Nice, France, 9–11 October 2013; pp. 389–394. [Google Scholar]
- Lapira, E.; Brisset, D.; Ardakani, H.D.; Siegel, D.; Lee, J. Wind turbine performance assessment using multi-regime modeling approach. Renew. Energy
**2012**, 45, 86–95. [Google Scholar] [CrossRef] - Long, H.; Wang, L.; Zhang, Z.; Song, Z.; Xu, J. Data-driven wind turbine power generation performance monitoring. IEEE Trans. Ind. Electron.
**2015**, 62, 6627–6635. [Google Scholar] [CrossRef] - Staffell, I.; Green, R. How does wind farm performance decline with age? Renew. Energy
**2014**, 66, 775–786. [Google Scholar] [CrossRef] [Green Version] - Kurz, R.; Brun, K. Degradation of gas turbine performance in natural gas service. J. Nat. Gas Sci. Eng.
**2009**, 1, 95–102. [Google Scholar] [CrossRef] - Olauson, J.; Edström, P.; Rydén, J. Wind turbine performance decline in Sweden. Wind Energy
**2017**, 20, 2049–2053. [Google Scholar] [CrossRef] - Dai, J.; Yang, W.; Cao, J.; Liu, D.; Long, X. Ageing assessment of a wind turbine over time by interpreting wind farm SCADA data. Renew. Energy
**2018**, 116, 199–208. [Google Scholar] [CrossRef] [Green Version] - Smola, A.J.; Schölkopf, B. A tutorial on support vector regression. Stat. Comput.
**2004**, 14, 199–222. [Google Scholar] [CrossRef] [Green Version] - Lee, G.; Ding, Y.; Xie, L.; Genton, M.G. A kernel plus method for quantifying wind turbine performance upgrades. Wind Energy
**2015**, 18, 1207–1219. [Google Scholar] [CrossRef] - Hwangbo, H.; Ding, Y.; Eisele, O.; Weinzierl, G.; Lang, U.; Pechlivanoglou, G. Quantifying the effect of vortex generator installation on wind power production: An academia-industry case study. Renew. Energy
**2017**, 113, 1589–1597. [Google Scholar] [CrossRef] - Astolfi, D.; Castellani, F.; Terzi, L. Wind Turbine Power Curve Upgrades. Energies
**2018**, 11, 1300. [Google Scholar] [CrossRef] [Green Version] - International Electrotechnical Commission. Wind Energy Generation Systems—Part 12-1: Power Performance Measurements of Electricity Producing Wind Turbines; Technical Report, IEC 61400-12-1: 2017; International Electrotechnical Commission: Geneva, Switzerland, 2007. [Google Scholar]
- Sequeira, C.; Pacheco, A.; Galego, P.; Gorbeña, E. Analysis of the efficiency of wind turbine gearboxes using the temperature variable. Renew. Energy
**2019**, 135, 465–472. [Google Scholar] [CrossRef] - Byrne, R.; Hewitt, N.J.; Griffiths, P.; MacArtain, P. Observed site obstacle impacts on the energy performance of a large scale urban wind turbine using an electrical energy rose. Energy Sustain. Dev.
**2018**, 43, 23–37. [Google Scholar] [CrossRef] - Choukulkar, A.; Pichugina, Y.; Clack, C.T.; Calhoun, R.; Banta, R.; Brewer, A.; Hardesty, M. A new formulation for rotor equivalent wind speed for wind resource assessment and wind power forecasting. Wind Energy
**2016**, 19, 1439–1452. [Google Scholar] [CrossRef] - Scheurich, F.; Enevoldsen, P.B.; Paulsen, H.N.; Dickow, K.K.; Fiedel, M.; Loeven, A.; Antoniou, I. Improving the accuracy of wind turbine power curve validation by the rotor equivalent wind speed concept. J. Phys. Conf. Ser. IOP Publ.
**2016**, 753, 072029. [Google Scholar] [CrossRef] [Green Version] - Pandit, R.; Infield, D. Comparative analysis of binning and support vector regression for wind turbine rotor speed based power curve use in condition monitoring. In Proceedings of the IEEE 2018 53rd International Universities Power Engineering Conference (UPEC), Glasgow, UK, 4–7 September 2018; pp. 1–6. [Google Scholar]
- Pandit, R.K.; Infield, D. Comparative assessments of binned and support vector regression-based blade pitch curve of a wind turbine for the purpose of condition monitoring. Int. J. Energy Environ. Eng.
**2019**, 10, 181–188. [Google Scholar] [CrossRef] [Green Version] - Marvuglia, A.; Messineo, A. Monitoring of wind farms’ power curves using machine learning techniques. Appl. Energy
**2012**, 98, 574–583. [Google Scholar] [CrossRef] - Kusiak, A.; Verma, A. Monitoring wind farms with performance curves. IEEE Trans. Sustain. Energy
**2012**, 4, 192–199. [Google Scholar] [CrossRef] - Song, Z.; Zhang, Z.; Jiang, Y.; Zhu, J. Wind turbine health state monitoring based on a Bayesian data-driven approach. Renew. Energy
**2018**, 125, 172–181. [Google Scholar] [CrossRef] - Zhang, J.; Jiang, N.; Li, H.; Li, N. Online health assessment of wind turbine based on operational condition recognition. Trans. Inst. Meas. Control
**2019**, 41, 2970–2981. [Google Scholar] [CrossRef] - Ding, Y. Data Science for Wind Energy; CRC Press: Boca Raton, FL, USA, 2019. [Google Scholar]
- Astolfi, D.; Castellani, F.; Garinei, A.; Terzi, L. Data mining techniques for performance analysis of onshore wind farms. Appl. Energy
**2015**, 148, 220–233. [Google Scholar] [CrossRef] - International Electrotechnical Commission. Power Performance Measurements of Electricity Producing Wind Turbines; Technical Report 61400-12; International Electrotechnical Commission: Geneva, Switzerland, 2005. [Google Scholar]
- Vapnik, V. The Nature of Statistical Learning Theory; Springer Science & Business Media: Berlin/Heidelberg, Germany, 2013. [Google Scholar]
- Astolfi, D.; Castellani, F. Wind turbine power curve upgrades: Part II. Energies
**2019**, 12, 1503. [Google Scholar] [CrossRef] [Green Version] - Pandit, R.K.; Infield, D. Comparison of binned and Gaussian Process based wind turbine power curves for condition monitoring purposes. J. Maint. Eng.
**2018**, 2. [Google Scholar] - Astolfi, D.; Castellani, F.; Terzi, L. Fault prevention and diagnosis through scada temperature data analysis of an onshore wind farm. Diagnostyka
**2014**, 15. [Google Scholar] - Tomporowski, A.; Flizikowski, J.; Kasner, R.; Kruszelnicka, W. Environmental control of wind power technology. Rocz. Ochr. Srodowiska
**2017**, 19, 694–714. [Google Scholar]

**Figure 3.**The average difference R between power measurement Y and estimation $\widehat{Y}$ (Equation (14)). Data sets, respectively: 2008 vs. 2012, 2013, 2014, 2015, 2017, 2018.

**Figure 5.**The average difference R between power measurement Y and estimation $\widehat{Y}$ (Equation (14)). Data sets: 2017–2018 and 2019.

**Figure 6.**The average difference R between power measurement Y and estimation $\widehat{Y}$ (Equation (14)). Data sets: 2008 and 2019.

Specification | Data |
---|---|

Model | PLH-400V52 |

Rated Power | 935 kW |

Rated RPM (Low speed shaft) | 26 min${}^{-1}$ |

Gearing ratio | 61.799 |

Weight | 5400 kg |

Parameter | Logging Interval | Years Available |
---|---|---|

Wind Speed (m/s) | 10 min. | 2008, 2014, 2017–2018 (12 months), 2019 (6 months) |

Wind Speed Standard Deviation (m/s) | 10 min. | 2008, 2014, 2017–2018 (12 months), 2019 (6 months) |

Wind Direction (deg) | 10 min. | 2008, 2014, 2017–2018 (12 months), 2019 (6 months) |

Ambient Temperature (°C) | 10 min. | 2008, 2014, 2017–2018 (12 months), 2019 (6 months) |

Rotor Speed (RPM) | 10 min. | 2008, 2014, 2017–2018 (12 months), 2019 (6 months) |

Blade Pitch Angle (deg) | 10 min. | 2008, 2014, 2017–2018 (12 months), 2019 (6 months) |

Generator Speed (RPM) | 10 min. | 2008, 2014, 2017–2018 (12 months), 2019 (6 months) |

Power (kW) | 10 min. | 2008, 2014, 2017–2018 (12 months), 2019 (6 months) |

Gear oil Temperature (°C) | 10 min. | 2008, 2014, 2017–2018 (12 months), 2019 (6 months) |

Data Set | $\mathsf{\Delta}$ |
---|---|

2012 vs. 2008 | −0.1% |

2013 vs. 2008 | −0.7% |

2014 vs. 2008 | −0.4% |

2015 vs. 2008 | −1.1% |

2017 vs. 2008 | −6.2% |

2018 vs. 2008 | −4.0% |

**Table 4.**Estimate of the performance change before and after gearbox replacement. (below rated power).

Data Set | $\mathsf{\Delta}$ |
---|---|

2019 vs. 2017–2018 | +1.9% |

**Table 5.**Estimate of the performance change: 2008 vs. 2019 after gearbox replacement. (below rated power).

Data Set | $\mathsf{\Delta}$ |
---|---|

2008 vs. 2019 | −2.9% |

Year | Uave (m/s) | Weibull A (m/s) | Weibull k | Power Density (W/m${}^{2}$) |
---|---|---|---|---|

2008 | 6.2 | 6.95 | 1.91 | 280.5 |

2014 | 5.8 | 6.54 | 1.93 | 227.2 |

2017–2018 | 5.6 | 6.32 | 2.04 | 196.6 |

Wind Year | Power Curve | AEP (kWh) | $\mathsf{\Delta}$ AEP (kWh) | % $\mathbf{\Delta}$ AEP |
---|---|---|---|---|

2008 | 2008 | 1,837,900 | - | - |

2014 | 1,810,700 | −27,200 | −1.5% | |

2017–2018 | 1,759,600 | −78,300 | −4.3% | |

2019 | 1,782,200 | −55,700 | −3.0% | |

2014 | 2008 | 1,598,300 | - | - |

2014 | 1,571,500 | −26,800 | −1.7% | |

2017–2018 | 1,520,200 | −78,100 | −4.9% | |

2019 | 1,539,900 | −58,400 | −3.6% | |

2017–2018 (12 months) | 2008 | 1,434,500 | - | - |

2014 | 1,407,300 | −27,200 | −1.9% | |

2017–2018 | 1,354,400 | −80,100 | −5.6% | |

2019 | 1,372,000 | −62,500 | −4.4% |

© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

## Share and Cite

**MDPI and ACS Style**

Byrne, R.; Astolfi, D.; Castellani, F.; Hewitt, N.J.
A Study of Wind Turbine Performance Decline with Age through Operation Data Analysis. *Energies* **2020**, *13*, 2086.
https://doi.org/10.3390/en13082086

**AMA Style**

Byrne R, Astolfi D, Castellani F, Hewitt NJ.
A Study of Wind Turbine Performance Decline with Age through Operation Data Analysis. *Energies*. 2020; 13(8):2086.
https://doi.org/10.3390/en13082086

**Chicago/Turabian Style**

Byrne, Raymond, Davide Astolfi, Francesco Castellani, and Neil J. Hewitt.
2020. "A Study of Wind Turbine Performance Decline with Age through Operation Data Analysis" *Energies* 13, no. 8: 2086.
https://doi.org/10.3390/en13082086