作者: Peng Guo , Nan Bai
DOI: 10.3390/EN4112077
关键词:
摘要: Condition Monitoring (CM) of wind turbines can greatly reduce the maintenance costs for farms, especially offshore farms. A new condition monitoring method a turbine gearbox using temperature trend analysis is proposed. Autoassociative Kernel Regression (AAKR) used to construct normal behavior model temperature. With proper construction memory matrix, AAKR cover working space gearbox. When has an incipient failure, residuals between estimates and measurement will become significant. moving window statistical detect changes residual mean value standard deviation in timely manner. one these parameters exceeds predefined thresholds, failure flagged. In order simulate fault, manual drift added initial Supervisory Control Data Acquisitions (SCADA) data. Analysis simulated failures shows that effective.