Date of Award

2017

Publication Type

Master Thesis

Degree Name

M.A.Sc.

Department

Mechanical, Automotive, and Materials Engineering

First Advisor

Carriveau, Rupp

Second Advisor

Ting, David

Rights

info:eu-repo/semantics/openAccess

Creative Commons License

Creative Commons Attribution-Noncommercial-No Derivative Works 4.0 License
This work is licensed under a Creative Commons Attribution-Noncommercial-No Derivative Works 4.0 License.

Abstract

Wind Turbine condition monitoring can detect anomalies in turbine performance which have the potential to result in unexpected failure and financial loss. This study examines common Supervisory Control And Data Acquisition (SCADA) data over a period of 20 months for 21 pitch regulated 2.3 MW turbines and is presented in three manuscripts. First, power curve monitoring is targeted applying various types of Artificial Neural Networks to increase modeling accuracy. It is shown how the proposed method can significantly improve network reliability compared with existing models. Then, an advance technique is utilized to create a smoother dataset for network training followed by establishing dynamic ANFIS network. At this stage, designed network aims to predict power generation in future hours. Finally, a recursive principal component analysis is performed to extract significant features to be used as input parameters of the network. A novel fusion technique is then employed to build an advanced model to make predictions of turbines performance with favorably low errors.

Share

COinS