Application of Distributed Machine Learning Model in Fault Diagnosis of Air Preheater

Existing monitoring systems for the current operational status of power equipment and fault diagnosis detection systems mostly use serial computing methods, and less parallel distributed processing algorithms are used. With the development of intelligent work of power systems, more and more test data of power plant equipment is becoming more and more complex, which puts new demands on the implementation of data processing and the ability of data calculation. In this study, by using spark, two distributed machine learning models for state detection and fault diagnosis are established for the air preheater, and the confusion matrix is used for evaluation. The results show that the random forest model can effectively diagnose the faults of the air preheater.