Conductor temperature estimation using the hadoop mapreduce framework for smart grid applications

Smart grid has become a popular issue on power system applications in recent years. By using the information and communication technology (ICT), the concept of smart grid aims to make power systems more intelligent. In smart grid, conductor temperature is an important variable for power line transmission. It dominates the limitation of the maximum current, called 'ampere capacity'. In this paper, we estimate all of the conductor temperatures on extra-high-voltage (EHV) transmission grids to monitor the ampere capacity in Taiwan. Following the IEEE 738-2007 standard and using a great amount of information from the national central weather bureau, we estimate some weather parameters in the nearest grid using a k-d tree algorithm and apply them to a Hadoop MapReduce framework to establish a conductor temperature estimation system. The proposed system is found to efficiently estimate the conductor temperature. By using the Hadoop MapReduce framework, this system can create new models by using a large amount of data related to a smart grid, and new functions can also be easily added to the system. For the future research, this system will be extended to the electricity dispatch. © 2014 IEEE.