With the rapid development of power grid, prediction of electric quantity changes has become increasingly important. High-performance power grid systems can improve economic effectiveness and operational efficiency through accurate prediction. This paper proposes a prediction model based on temperature, humidity, time, and the number of people. On account of the standards of support vector machine (SVM) and the HBase platform, we have implemented a forecasting model and designed simulative experiments. The experimental results show that time and variation in the number of people has a remarkable influence on prediction, while temperature and humidity hardly have any effects.
|Publication status||Published - Dec 2016|
|Event||The 2016 IEEE International Conference on Progress in Informatics and Computing - Shanghai, China|
Duration: 23 Dec 2016 → 25 Dec 2016
|Conference||The 2016 IEEE International Conference on Progress in Informatics and Computing|
|Abbreviated title||PIC 2016|
|Period||23/12/16 → 25/12/16|
CitationYang, H., Li, P., Masood, A., Xiao, Y., Sheng, B., & Yu, Q. (2016, December). Smart grid data analysis and prediction modeling. Paper presented at The 2016 IEEE International Conference on Progress in Informatics and Computing (PIC-2016), Shanghai Guang Dong Hotel, Shanghai, China.
- Power prediction
- Data mining
- HBase data storage
- Machine learning