Deep Neural Network based Automatic System for Electricity Meter Reading in Nepal
Authors: Rajiv Bishwokarma, Binay Paudyal, Pravesh Chapagain, Shirshak Bajgain, Hitendra Dev Shakya
Abstract— This work proposes a new system that Nepalese energy/utility companies can implement to obtain better accuracy of energy usage scenarios. The proposed system can also be used to reduce undesired non-technical losses in the electricity market. This reduction is achieved by replacing the middle agents with an intelligent system that can identify energy meter readings and calculate accurate billing information correctly. As part of the system, common smartphones are used for taking pictures of energy meters by the owners, who upload them to the central server. The central server then utilizes a deep neural network – YOLOv3 – to detect and identify energy meter counters and digits, from which current energy consumption is extracted. This extracted value is then used for calculating the billing information.
Keywords— meter reading, deep neural network, electricity consumption
Full Paper: to be purchased $20 (contact us at [email protected])
Published In: International Conference on Role of Energy for Sustainable Social Development in ‘New Normal’ Era (RESSD-2020)
Date of Conference: 28th-29th December 2020
Conference Location: Kathmandu, Nepal
Publisher: IEEE Power and Energy Society Nepal Chapter
Cite the paper as:
R. Bishwokarma, B. Paudyal, P. Chapagain, S. Bajgain, H. D. Shakya, “ Deep Neural Network based Automatic System for Electricity Meter Reading in Nepal”, International Conference on Role of Energy for Sustainable Social Development in ‘New Normal’ Era, 28th-29th December 2020, Kathmandu, Nepal