An approach to real-time fault detection in health monitoring of offshore wind-farms

Wind is a clean resource for electricity generation. However, its power is difficult to harness due to the effect of environmental parameters which cause fault occurrences in the wind-turbine towers and decrease the efficiency of the system. We propose Flexible Threshold Selection and Fuzzy inference system based Fault Detection System (FTSFFDS), which allows autonomous and continuous monitoring of the wind farm in real-time and gives warnings in case of fault occurrences. This is a simple and efficient method which forms the basis for the structural health monitoring of Offshore wind-farms. The system uses WSN in offshore wind-farm to sense the parameters and transmit the readings to the remote observer. The method utilizes adaptive thresholds to cater to the readings at different times of the day. The decision for new threshold is made when the correlation between the datasets is low. The proposed method is efficient in reducing the overall size of the data packets transmitted by the sensor nodes and thus saving energy in the WSN. The simulations confirm that the energy efficiency of the system is increased by nearly ten times compared to standard protocols e.g. LEACH. This paper is an extension to our previous research-work which proposed an application-specific clustering and routing protocol NETCRP (Network Lifetime Enhancing Clustering and Routing Protocol).