SENSOR FAILURE IDENTIFICATION AND SEGREGATION USING WAVELET PERFORMANCE ANALYSIS FOR WSN BASED STATUS SURVEILLANCE SYSTEM OF A WIND TURBINE

Authors

  • Saravanan Muthampatty Sengottaiyan Department of Electrical and Electronics Engineering, Mahendra College of Engineering Tamilnadu, India
  • Jeyabharath Rajaiah Department of Electrical and Electronics Engineering, K S R Institute for Engineering and Technology, Tamilnadu, India
  • Ravivarman Shanmugasundaram Department of Electrical and Electronics Engineering, Vardhaman College of Engineering, Telangana, India
  • Jamuna Ponnusamy Department of Electrical and Electronics Engineering, Nandha Engineering College, Tamilnadu, India

DOI:

https://doi.org/10.23055/ijietap.2023.30.3.8917

Keywords:

Current sensors, Wireless Sensor Node , Wavelet transient approach, Cross correlation technique, Wind turbine, Fault identification, Invariable anomaly fault, Interference fault

Abstract

One of the most useful renewable energy sources is wind, from which electrical power can be generated using a turbine system for long periods. The reliability of a wind turbine mainly depends on the maintenance work carried out at the site. The Status Surveillance System (SSS) is an important factor for wind turbines to guarantee uninterrupted power supply to the end user. However, the condition monitoring system based on the Wireless Sensor Node (WSN), housed with the current sensor node, is more vulnerable to failure due to circumstantial faults. Due to sensor faults, the data used for decision-making on maintenance are corrupted. This paper devises a robust and reliable mechanism called Sensor Failure Identification and Segregation (SFIS) to detect and detach corrupted data to effectively perform work related to wind turbine failure detection. The short-circuit fault is addressed by a wavelet transient approach to restore the corrupted data, while the invariable anomaly fault is analyzed with the help of the cross-correlation method. Hence, the interference fault can be analyzed using a dynamic time-warping approach. The proposed mechanism is compared with the existing Adaptive Neuro-Fuzzy Inference System (ANFIS) method that uses Supervisory Control and Data Acquisition (SCADA) to prove its reliability and robustness. SFIS offers a reliable and cost-effective solution for wind turbine maintenance work.

Published

2023-06-16

How to Cite

Saravanan, S. M., Rajaiah, J., Shanmugasundaram, R., & Ponnusamy, J. (2023). SENSOR FAILURE IDENTIFICATION AND SEGREGATION USING WAVELET PERFORMANCE ANALYSIS FOR WSN BASED STATUS SURVEILLANCE SYSTEM OF A WIND TURBINE. International Journal of Industrial Engineering: Theory, Applications and Practice, 30(3). https://doi.org/10.23055/ijietap.2023.30.3.8917

Issue

Section

Information System and Technology