# Fuzzy and Neural Network Approaches to Wind Turbine Fault Diagnosis

^{1}

^{2}

^{*}

^{†}

## Abstract

**:**

## 1. Introduction

- The paper proposes two data-driven techniques for the FDI of a WT system;
- the developed FDI strategies are based on the direct reconstruction of the fault functions affecting the monitored process. In this way, the diagnostic residuals are represented by the estimated fault signals;
- the residual generators are organised into a bank structure in order to accomplish the fault isolation task. Their structures rely on Fuzzy Systems (FSs) in the form of TS models and dynamic NNs;
- a fault sensitivity analysis enhances the design of the residual generators, by a proper selection of their inputs;
- the developed FDI schemes are applied to a WT benchmark first; moreover, the verification of their robustness and reliability features is performed using a HIL tool representing the realistic behaviour of a WT process.

## 2. Wind Turbine Benchmark

#### 2.1. WT Benchmark Model

#### 2.2. WT Fault Scenario

#### 2.3. Fault Sensitivity Analysis

## 3. Data-Driven Strategies for Fault Diagnosis

#### Fault Estimators via FS and NN Tools

## 4. Simulation Results, Experimental Validation, Comparisons and Discussion

#### 4.1. HIL Validation

#### 4.2. Comparative Analysis

#### 4.3. Discussion and Final Remarks

## 5. Conclusions

## Author Contributions

## Funding

## Institutional Review Board Statement

## Informed Consent Statement

## Data Availability Statement

## Conflicts of Interest

## Nomenclature

Acronym | Definition |

AFTC | Active Fault-Tolerant Control |

NLGA | NonLinear Geometric Approach |

ARX | Auto-Regressive eXogenous |

NLGA-AF | NLGA Adaptive Filter |

FDI | Fault Detection and Isolation |

NN | Neural Network |

FTC | Fault-Tolerant Control |

O&M | Operation & Maintenance |

FIS | Fuzzy Inference System |

PFTC | Passive Fault Tolerant Control |

FS | Fuzzy System |

RFS | Recursive identification of Fuzzy System |

GK | Gustafson-Kessel |

RMSE | Root Mean Square Error |

HIL | Hardware-In-the-Loop |

SMO | Sliding Mode Observer |

LPV | Linear Parameter Varying |

TS | Takagi-Sugeno |

MLP | Multi-Layer Perceptron |

WECS | Wind Energy Conversion System |

NARX | Nonlinear Auto-Regressive eXogenous |

WT | Wind Turbine |

## References

- Bianchi, F.D.; Battista, H.D.; Mantz, R.J. Wind Turbine Control Systems: Principles, Modelling and Gain Scheduling Design, 1st ed.; Advances in Industrial Control; Springer: Berlin/Heidelberg, Germany, 2007; ISBN 1-84628-492-9. [Google Scholar]
- Liserre, M.; Cardenas, R.; Molinas, M.; Rodriguez, J. Overview of multi-mw wind turbines and wind parks. IEEE Trans. Ind. Electron.
**2011**, 58, 1081–1095. [Google Scholar] [CrossRef] - Odgaard, P.F.; Stoustrup, J. A Benchmark Evaluation of Fault Tolerant Wind Turbine Control Concepts. IEEE Trans. Control Syst. Technol.
**2015**, 23, 1221–1228. [Google Scholar] [CrossRef] - Odgaard, P.F.; Stoustrup, J. Results of a Wind Turbine FDI Competition. In SAFEPROCESS 2012, Proceedings of the 8th IFAC Symposium on Fault Detection, Supervision and Safety of Technical Processes, Mexico City, Mexico, 29–31 August 2012; Verde, C., Astorga Zaragoza, C.M., Molina, A., Eds.; IFAC: Mexico City, Mexico, 2012; Volume 8, pp. 102–107. [Google Scholar] [CrossRef][Green Version]
- Chen, J.; Patton, R.J. Robust Model–Based Fault Diagnosis for Dynamic Systems; Kluwer Academic Publishers: Boston, MA, USA, 1999. [Google Scholar]
- Zhang, T.; Kong, X. Adaptive fault–tolerant sliding mode control for high–speed trains with actuator faults under strong winds. IEEE Access
**2020**, 8, 143902–143919. [Google Scholar] [CrossRef] - Kuhne, P.; Poschke, F.; Schulte, H. Fault estimation and fault–tolerant control of the fast NREL 5-MW reference wind turbine using a proportional multi–integral observer. Int. J. Adapt. Control Signal Process.
**2018**, 32, 568–585. [Google Scholar] [CrossRef] - Kim, D.; Lee, D. Fault Parameter Estimation Using Adaptive Fuzzy Fading Kalman Filter. Appl. Sci.
**2019**, 9, 3329. [Google Scholar] [CrossRef][Green Version] - Malik, H.; Almutairi, A. Modified Fuzzy-Q-Learning (MFQL)-Based Mechanical Fault Diagnosis for Direct-Drive Wind Turbines Using Electrical Signals. IEEE Access
**2021**, 9, 52569–52579. [Google Scholar] [CrossRef] - Qu, F.; Liu, J.; Zhu, H.; Zhou, B. Wind turbine fault detection based on expanded linguistic terms and rules using non-singleton fuzzy logic. Appl. Energy
**2020**, 262, 114469. [Google Scholar] [CrossRef] - Guo, P.; Fu, J.; Yang, X. Condition Monitoring and Fault Diagnosis of Wind Turbines Gearbox Bearing Temperature Based on Kolmogorov-Smirnov Test and Convolutional Neural Network Model. Energies
**2018**, 11, 2248. [Google Scholar] [CrossRef][Green Version] - Yu, W.; Huang, S.; Xiao, W. Fault Diagnosis Based on an Approach Combining a Spectrogram and a Convolutional Neural Network with Application to a Wind Turbine System. Energies
**2018**, 11, 2561. [Google Scholar] [CrossRef][Green Version] - Dybkowski, M.; Klimkowski, K. Artificial Neural Network Application for Current Sensors Fault Detection in the Vector Controlled Induction Motor Drive. Sensors
**2019**, 19, 571. [Google Scholar] [CrossRef] [PubMed][Green Version] - Zare, S.; Ayati, M. Simultaneous fault diagnosis of wind turbine using multichannel convolutional neural networks. ISA Trans.
**2021**, 108, 230–239. [Google Scholar] [CrossRef] - Zhou, Q.; Xiong, T.; Wang, M.; Xiang, C.; Xu, Q. Diagnosis and Early Warning of Wind Turbine Faults Based on Cluster Analysis Theory and Modified ANFIS. Energies
**2017**, 10, 898. [Google Scholar] [CrossRef][Green Version] - Cho, S.; Choi, M.; Gao, Z.; Moan, T. Fault detection and diagnosis of a blade pitch system in a floating wind turbine based on Kalman filters and artificial neural networks. Renew. Energy
**2021**, 169, 1–13. [Google Scholar] [CrossRef] - Babuška, R. Fuzzy Modeling for Control; Kluwer Academic Publishers: Boston, MA, USA, 1998. [Google Scholar]
- Simani, S.; Fantuzzi, C.; Rovatti, R.; Beghelli, S. Parameter Identification for Piecewise Linear Fuzzy Models in Noisy Environment. Int. J. Approx. Reason.
**1999**, 1, 149–167. [Google Scholar] [CrossRef][Green Version] - Korbicz, J.; Koscielny, J.M.; Kowalczuk, Z.; Cholewa, W. (Eds.) Fault Diagnosis: Models, Artificial Intelligence, Applications, 1st ed.; Springer: London, UK, 2004; ISBN 3540407677. [Google Scholar]
- Odgaard, P.F.; Stoustrup, J.; Kinnaert, M. Fault–Tolerant Control of Wind Turbines: A Benchmark Model. IEEE Trans. Control Syst. Technol.
**2013**, 21, 1168–1182. [Google Scholar] [CrossRef][Green Version] - Simani, S.; Farsoni, S.; Castaldi, P. Data-Driven Techniques for the Fault Diagnosis of a Wind Turbine Benchmark. Int. J. Appl. Math. Comput. Sci. AMCS
**2018**, 28, 247–268. [Google Scholar] [CrossRef][Green Version] - Simani, S.; Castaldi, P. Data-Driven Design of Fuzzy Logic Fault Tolerant Control for a Wind Turbine Benchmark. In SAFEPROCESS 2012, Proceedings of the 8th IFAC Symposium on Fault Detection, Supervision and Safety of Technical Processes, Mexico City, Mexico, 29–31 August 2012; Astorga-Zaragoza, C.M., Molina, A., Eds.; Instituto de Ingeniería, Circuito escolar, Ciudad Universitaria, IFAC: Mexico City, Mexico, 2012; Volume 8, pp. 108–113. ISBN 978-3-902823-09-0. [Google Scholar] [CrossRef]
- Simani, S.; Farsoni, S.; Castaldi, P. Residual Generator Fuzzy Identification for Wind Farm Fault Diagnosis. In IFAC’14, Proceedings of the 19th World Congress of the International Federation of Automatic Control, Cape Town, South Africa, 24–29 August 2014; IFAC & South Africa Council for Automation and Control, IFAC: Cape Town, South Africa, 2014; Volume 19, pp. 4310–4315. [Google Scholar] [CrossRef][Green Version]
- Simani, S.; Farsoni, S.; Castaldi, P. Active Fault Tolerant Control of Wind Turbines Using Identified Nonlinear Filters. In Proceedings of the 2013 Conference on Control and Fault-Tolerant Systems (SysTol), Nice, France, 9–11 October 2013; pp. 383–388. [Google Scholar] [CrossRef]
- Simani, S.; Farsoni, S.; Castaldi, P. Residual Generator Fuzzy Identification for Wind Turbine Benchmark Fault Diagnosis. Machines
**2014**, 2, 275–298. [Google Scholar] [CrossRef][Green Version] - Simani, S. Application of a Data–Driven Fuzzy Control Design to a Wind Turbine Benchmark Model. Adv. Fuzzy Syst.
**2012**, 2012, 50436. [Google Scholar] [CrossRef][Green Version] - Simani, S.; Patton, R.J. Fault diagnosis of an industrial gas turbine prototype using a system identification approach. Control Eng. Pract.
**2008**, 16, 769–786. [Google Scholar] [CrossRef] - Fantuzzi, C.; Simani, S.; Beghelli, S.; Rovatti, R. Identification of piecewise affine models in noisy environment. Int. J. Control
**2002**, 75, 1472–1485. [Google Scholar] [CrossRef] - Odgaard, P.F.; Stoustrup, J.; Kinnaert, M. Fault Tolerant Control of Wind Turbines—A Benchmark Model. In Proceedings of the 7th IFAC Symposium on Fault Detection, Supervision and Safety of Technical Processes, IFAC, Barcelona, Spain, 30 June–3 July 2009; Volume 1, pp. 155–160. [Google Scholar] [CrossRef]
- Rovatti, R.; Fantuzzi, C.; Simani, S. High-speed DSP-based implementation of piecewise-affine and piecewise-quadratic fuzzy systems. Signal Process. J.
**2000**, 80, 951–963. [Google Scholar] [CrossRef] - Simani, S.; Castaldi, P. Active Actuator Fault Tolerant Control of a Wind Turbine Benchmark Model. Int. J. Robust Nonlinear Control
**2014**, 24, 1283–1303. [Google Scholar] [CrossRef][Green Version] - Simani, S.; Castaldi, P. Data-Driven and Adaptive Control Applications to a Wind Turbine Benchmark Model. Control Eng. Pract.
**2013**, 21, 1678–1693. [Google Scholar] [CrossRef] - Tan, C.P.; Edwards, C. Sliding mode observers for robust detection and reconstruction of actuator and sensor faults. Int. J. Robust Nonlinear Control
**2003**, 13, 443–463. [Google Scholar] [CrossRef] - Kwon, H.; Yoon, H.; Choi, D. Restricted Evasion Attack: Generation of Restricted-Area Adversarial Example. IEEE Access
**2019**, 7, 60908–60919. [Google Scholar] [CrossRef] - Kwon, H.; Yoon, H.; Park, K.W. Acoustic-decoy: Detection of adversarial examples through audio modification on speech recognition system. Neurocomputing
**2020**, 417, 357–370. [Google Scholar] [CrossRef]

**Figure 10.**Pitch sensor measurements ${\beta}_{i,mj}$ and generated power ${P}_{g,m}$ in full load conditions.

Variable | ${\beta}_{1,m1}$ | ${\beta}_{1,m2}$ | ${\beta}_{2,m1}$ | ${\beta}_{2,m2}$ | ${\beta}_{3,m1}$ | ${\beta}_{3,m2}$ |

Pitch # | 1 | 1 | 2 | 2 | 3 | 3 |

Sensor # | 1 | 2 | 1 | 2 | 1 | 2 |

Variable | ${\omega}_{r,m1}$ | ${\omega}_{r,m2}$ | ${\omega}_{g,m1}$ | ${\omega}_{g,m2}$ | ||

Speed | Rotor | Rotor | Generator | Generator | ||

Encoder # | 1 | 2 | 1 | 2 | ||

Variable | ${\tau}_{g,m}$ | ${P}_{g,m}$ | ${v}_{w,m}$ | ${\tau}_{r,m}$ | ||

Measure | Generator torque | Generated power | Wind speed | Rotor torque | ||

Model | Generator | Generator | Anemometer | Estimated |

Fault # | 1 | 2 | 3 |

Typology | Fixed value | Scaling error | Fixed value |

Sensor # | Blade 1 | Blade 2 | Blade 1 |

Fault # | 4 | 5 | 6 |

Typology | Fixed value | Scaling error | Dynamics |

Sensor # | Pitch 1 | Generator 2 | Actuator 2 |

Fault # | 7 | 8 | 9 |

Typology | Dynamics | Fixed value | Dynamics |

Sensor # | Pitch 3 | Converter | Drive-train |

Fault Case | Fault Type | Most Affected Input-Output Measurements |
---|---|---|

1 | Sensor | ${\beta}_{1,m1}$, ${\beta}_{1,m2}$, ${\omega}_{g,m2}$ |

2 | Sensor | ${\beta}_{1,m2}$, ${\beta}_{2,m2}$, ${\omega}_{g,m2}$ |

3 | Sensor | ${\beta}_{1,m2}$, ${\beta}_{3,m1}$, ${\omega}_{g,m2}$ |

4 | Sensor | ${\beta}_{1,m2}$, ${\omega}_{g,m2}$, ${\omega}_{r,m1}$ |

5 | Sensor | ${\beta}_{1,m2}$, ${\omega}_{g,m2}$, ${\omega}_{r,m2}$ |

6 | Actuator | ${\beta}_{1,m2}$, ${\beta}_{2,m1}$, ${\omega}_{g,m2}$ |

7 | Actuator | ${\beta}_{1,m2}$, ${\beta}_{3,m2}$, ${\omega}_{g,m2}$ |

8 | Actuator | ${\beta}_{1,m2}$, ${\tau}_{g,m}$, ${\omega}_{g,m2}$ |

9 | System | ${\beta}_{1,m2}$, ${\omega}_{g,m1}$, ${\omega}_{g,m2}$ |

Fault Case | 1 | 2 | 3 | 4 | 5 |
---|---|---|---|---|---|

RMSE% | 1.61% | 2.22% | 1.95% | 1.87% | 1.92% |

Sdt. Dev. | ±0.02% | ±0.03% | ±0.01% | ±0.01% | ±0.01% |

Fault Case | 6 | 7 | 8 | 9 | |

RMSE% | 2.15% | 1.76% | 2.13% | 1.98% | |

Sdt. Dev. | ±0.02% | ±0.01% | ±0.02% | ±0.01% |

Fault Case | 1 | 2 | 3 | 4 | 5 |
---|---|---|---|---|---|

RMSE % | 0.91% | 0.92% | 0.94% | 1.21% | 1.17% |

Sdt. Dev. | ±0.01% | ±0.01% | ±0.01% | ±0.02% | ±0.01% |

Fault Case | 6 | 7 | 8 | 9 | |

RMSE % | 1.61% | 0.98% | 0.95% | 1.41% | |

Sdt. Dev. | ±0.01% | ±0.01% | ±0.01% | ±0.02% |

Fault Case | 1 | 2 | 3 | 4 | 5 |
---|---|---|---|---|---|

TS FSs | 1.69% | 2.29% | 2.01% | 1.94% | 1.99% |

NARX NNs | 0.99% | 0.98% | 0.99% | 1.28% | 1.21% |

Fault Case | 6 | 7 | 8 | 9 | |

TS FSs | 2.22% | 1.81% | 2.21% | 2.03% | |

NARX NNs | 1.69% | 1.02% | 1.01% | 1.51% |

Fault Case | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 |
---|---|---|---|---|---|---|---|---|---|

TS FSs | 1.69% | 2.29% | 2.01% | 1.94% | 1.99% | 2.22% | 1.81% | 2.21% | 2.03% |

NARX NN | 0.99% | 0.98% | 0.99% | 1.28% | 1.21% | 1.69% | 1.02% | 1.01% | 1.51% |

NLGA-AF | 1.37% | 1.45% | 1.73% | 1.75% | 1.56% | 1.99% | 1.45% | 1.54% | 1.76% |

RFS | 1.99% | 2.67% | 2.44% | 2.56% | 2.67% | 2.97% | 2.23% | 2.78% | 2.82% |

SMO | 1.87% | 1.82% | 2.11% | 2.01% | 1.91% | 2.34% | 1.95% | 2.08% | 2.23% |

Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations. |

© 2021 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 (https://creativecommons.org/licenses/by/4.0/).

## Share and Cite

**MDPI and ACS Style**

Farsoni, S.; Simani, S.; Castaldi, P. Fuzzy and Neural Network Approaches to Wind Turbine Fault Diagnosis. *Appl. Sci.* **2021**, *11*, 5035.
https://doi.org/10.3390/app11115035

**AMA Style**

Farsoni S, Simani S, Castaldi P. Fuzzy and Neural Network Approaches to Wind Turbine Fault Diagnosis. *Applied Sciences*. 2021; 11(11):5035.
https://doi.org/10.3390/app11115035

**Chicago/Turabian Style**

Farsoni, Saverio, Silvio Simani, and Paolo Castaldi. 2021. "Fuzzy and Neural Network Approaches to Wind Turbine Fault Diagnosis" *Applied Sciences* 11, no. 11: 5035.
https://doi.org/10.3390/app11115035