Analysis of Real Time Fault Data of Multi Terminal Transmission System using Python Learning Tools
Gyanesh Singh1, A.Q. Ansari2, Md. Abul Kalam3
1Gyanesh Singh, Department of Electrical & Electronics Engineering, IMS Engineering College Ghaziabad Uttar Pradesh India.
2A.Q. Ansari, Electrical Engineering Department Faculty of Engineering JMI New Delhi Jamia Millia Islamia New Delhi India
3Md. Abul Kalam, Department of Electrical Engineering, JSSATE NOIDA Uttar Pradesh India.

Manuscript received on 18 April 2019 | Revised Manuscript received on 21 May 2019 | Manuscript published on 30 May 2019 | PP: 517-523 | Volume-8 Issue-1, May 2019 | Retrieval Number: F2513037619 /19©BEIESP
Open Access | Ethics and Policies | Cite | Mendeley | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (

Abstract: Transmission line faults are common difficulties in today’s world. Faults mainly depend on types of load and their nature. Transmission lines are divided in different zones and hence cannot predict easily the faults and their types. Several protective devices have incorporated in the past few years to classification of fault but none guaranteed the accuracy. In this paper author proposes an advanced machine learning algorithm to classification of faults and provided some future protection technique to minimize faults and reliability of supply to the consumer. In this python learning tools author compare the two algorithm namely using KNeighbors Classifier and Using Multinomial Logistic Regression. The data for experiments are obtained from BBMB Punjabi Bagh 220 KV substation New Delhi. The real experiments validates the improvements and future forecasting regarding faults in lines and analysis of results with their accuracy are elaborated in this paper. A discussion of real data and their implication in future conservation of energy.
Index Terms: Power System Real Faults, Python, KNeighbors Classifier and Using Multinomial Logistic Regression Algorithm. Python and MS Excel. Python Libraries Numpy, Pandas, Matplotlib, Seaborn, IDEs Jupyter Notebook.

Scope of the Article: E-Learning