Research of Diagnosing Causes of Turbine Faults Based on PSO-LightGBM Algorithm

Based on detecting the categories of turbine faults, diagnosing causes of turbine faults is proposed, which has contributions to the field of intelligent operation and maintenance. To improve the diagnostic accuracy and practical efficiency, a hybrid model based on particle swarm optimization-light gradient boosting machine (PSO-LightGBM) algorithm is designed. LightGBM is used as a classifier to diagnose causes of turbine faults. PSO is used to automatically optimize the parameters during the LightGBM training. Following diagnostic results, maintenance methods are automatically linked from the knowledge base. Compared with five state-of-the-art intelligent classifiers, the experimental results demonstrate that the proposed method has higher accuracy and efficiency in diagnosing causes of turbine faults.