Significance
With the global call to mitigate and reduce the environment pollution through emissions of greenhouse gases, a lot of diversification have been experienced in the energy sector with the focus currently shifting to the use of renewable energy sources like wind energy, away from well-known fossils fuels which have been known for ages. This has led to the rapid growth of the wind energy sector due to the increased use of wind energy and advanced researchers and exploration of new technologies which has led to creation of more advanced wind turbines in this sector.
However, the current high cost of operation and maintenance incurred in the wind sector has not been beneficial, and there is a considerable need for cost reductions. Just like any other industrial or engineering and technological field, the wind energy sector is currently using age-based modelling strategies for their daily operations and maintenance processes. This is, however, bound to fail or to be of least helpful in the future, as they are focused mainly on corrective and preventive measures. Instead, a predictive strategy will be of more benefit. For example, being able to predict the likelihood of occurrence of some components failures in the wind turbines and the possible causes of such failure will pave the way for corrective planning measures to prevent such shortcomings rather than waiting.
Wind turbines are used for the generations of wind energy. These turbines are susceptible to various environmental and weather conditions. Although, some of the weather conditions may favor efficient wind energy production, most of them are the primary cause of such failures. Predicting wind turbine failures resulting from the environmental conditions may be complex and requires the use of sophisticated models.Nonetheless, it is very crucial as it will provide the information that may be used in facilitating making the various decisions concerning operations and maintenance.
A group of researchers from the University of Zaragoza in Spain, Maik Reder, Nurseda Yürüşen and Professor Julio Melero addressed this essential problem by proposing a framework for analyzing the various environmental conditions and their contributions to the failures experienced in the components of the wind turbines. The analysis involved the use of two sets of data mining approaches, supervised and unsupervised techniques, as well as an apriori algorithm in relating the failures and the environmental factors. Their work is published in the journal, Reliability Engineering and Systems Safety.
The authors pointed out a direct relationship between the various failures in the wind turbine components such as gearbox and frequency converter to weather condition, which included wind speed, relative humidity among others.
As a significant contribution of this research, this framework will help analyzing the various weather conditions and to give an appropriate prediction of the type of failure they are likely to cause on the components of the wind turbines. With this information, the performance techniques of the turbines can be effectively evaluated. The strategy does not only apply to wind turbines but can also be borrowed and applied in a variety of tasks for predicting and preventing failures that may occur in other sectors. As a result, it will considerably reduce the operation and maintenance costs of wind turbines or the other structures in the industries.
Reference
Reder, M., Yürüşen, N., & Melero, J. (2018). Data-driven learning framework for associating weather conditions and wind turbine failures. Reliability Engineering & System Safety, 169, 554-569.
Go To Reliability Engineering & System Safety