Low Voltage Prediction Based on Spark and Ftrl
- Chen Gao, Zhongan Ding, Shengteng Yan, Hongkun Mai
- Corresponding Author
- Chen Gao
Available Online June 2017.
- https://doi.org/10.2991/ammee-17.2017.34How to use a DOI?
- Ftrl, Low voltage, Distributed, Big data.
- "Big data" is a popular keyword in recent years, and data related to the power grid mainly character huge amount, high complexity and broad sources. As low voltage has great influence on normal daily power utilization, besides good real-time monitoring and exception handling, a model based on big-data algorithms and multi-dimensional characters is also needed for real-time prediction. This paper advances a FTRL algorithm based on the Spark framework, and establishes a highly fault-tolerant, real-time, accurate and fast low-voltage prediction system by setting up a FTRL real-time computation platform based on effective characters extracted from a huge amount of voltage data. It can be seen through analysis of experimental results that this system is able to effectively predict low voltage and give alarms, which shows great improvement over the current manual monitoring mechanism.
- Open Access
- This is an open access article distributed under the CC BY-NC license.
Cite this article
TY - CONF AU - Chen Gao AU - Zhongan Ding AU - Shengteng Yan AU - Hongkun Mai PY - 2017/06 DA - 2017/06 TI - Low Voltage Prediction Based on Spark and Ftrl BT - Advances in Materials, Machinery, Electrical Engineering (AMMEE 2017) PB - Atlantis Press SN - 2352-5401 UR - https://doi.org/10.2991/ammee-17.2017.34 DO - https://doi.org/10.2991/ammee-17.2017.34 ID - Gao2017/06 ER -