Fault Diagnosis in Benchmark Process Control System Using Stochastic Gradient Boosted Decision Trees
Tarun Chopra1, Jayashri Vajpai2
1Tarun Chopra, Ph.D. Scholar, Department of Electrical Engineering, Faculty of Engineering, J.N.V. University, Jodhpur, India.
2Dr. Jayashri Vajpai, Associate Professor, Department of Electrical Engineering, Faculty of Engineering, J.N.V. University, Jodhpur, India.
Manuscript received on June 24, 2011. | Revised Manuscript received on June 30, 2011. | Manuscript published on July 05, 2011. | PP: 98-101 | Volume-1 Issue-3, July 2011. | Retrieval Number: C068061311
Open Access | Ethics and Policies | Cite | Mendeley
© The Authors. Published By: Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: Decision trees create an easily understood structure for evaluating complex decisions. Tree Boost models often have a degree of accuracy that cannot be obtained using a large, single-tree model. Tree Boost models are adaptable, easy to interpret and often equal to or superior to any other predictive functions including neural networks. In this paper, the performance of the proposed approach based on Stochastic Gradient Boosted Decision Trees based method is demonstrated on the DAMADICS benchmark problem. An attempt has been made to improve the performance of fault diagnosis task on DAMADICS benchmark.
Keywords: Fault Diagnosis, Stochastic Gradient Boosted Decision Trees, DAMADICS.