Abstract
In order to quickly and accurately recognize and classify the fault modes of hydraulic end of fracturing pump according to the real-time test data of fracturing pump, this paper proposes a k-medoids clustering algorithm based on feature reduction. First, FRMB algorithm is used to reduce the features of the fracturing pump test data set, remove the redundant features from the data set, calculate the correlation evaluation of each subset, and assign the weight value of the corresponding feature subset according to the importance. Finally, k-medoids algorithm is used to realize clustering and identify the fault mode of hydraulic end of fracturing pump. The experimental results show that compared with the classical clustering algorithm, the k-medoids clustering algorithm based on feature reduction can ensure high accuracy while improving the fault diagnosis speed of fracturing pump.
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call