Modeling and experimental study on drilling rig anti-jamming valve with BP neural network

Authors

  • Wei Ma School of Mechanical Engineering, University of Science and Technology Beijing
  • Fei Ma School of Mechanical Engineering, University of Science and Technology Beijing

Keywords:

drilling rig, anti-jamming valve, Back-propagation (BP) neural network, hidden layers, rotation pressure control feed (RPCF)

Abstract

An effective anti-jamming system is significant in improving the working reliability of hydraulic drilling rigs. An anti-jamming valve is a core component of an anti-jamming system. A back-propagation (BP) neural network model was established for an anti-jamming valve by analyzing the structure and working principle of an anti-jamming valve on a drilling rig and by utilizing the theory of neural network. The established model was employed and applied with the raw information to a stone pit to perform structural topology optimization and obtain the optimal model. The theoretical feed pressure was determined through calculation and analysis after the rock drill became completely jammed. A comparison between the relative error of computed results and the experimental values shows that the forecast data obtained with the BP model (8%) are closer to the experimental values than those obtained with the theoretical formula (14%). The highly nonlinear characteristic of a BP neural network provides a fresh insight into the intellectualization of a drilling rig anti-jamming system.

Downloads

Published

2016-04-08