Anomaly Detection of Dust Removal System Through Gradient Boosting Decision Tree Algorithm

In the process of modern industrial production, the breakdown of industrial equipment will bring serious economic loss, making the diagnosis of equipment a research hotspot. Among numerous equipment, the dust removal system contains many sub-equipment and parameters, which make its fault diagnosis a challenging problem. This paper introduces a solution to the anomaly detection problem of dust removal fans based on the improved K-means combined with the gradient boosting decision tree (GBDT) model. This study uses K-means to divide the obtained original data into different classes, then utilizes the model to complete the classification and prediction tasks for anomaly detection. It was found that the model used is more suitable for the fault diagnosis of the dust removal fan.