Load Pattern Recognition and Prediction Based on DTW K-Mediods Clustering and Markov Model

With the development of smart grid, the electricity consumption data of consumers are recorded through smart meters. Analyzing and modeling the power user data, exploring the behavior habits and characteristics of electricity use, is of great significance of demand side management, dynamic pricing, and improving energy efficiency. This study proposed DTW K-mediods clustering and Markov model for Load pattern recognition and load prediction. Firstly, K-mediods clustering and DTW-based similarity measure are used to get shape-based typical load pattern. Then, the load data of time series are coded as discrete state series based on clustering. Finally, Markov model is used to explore the rule of load mode transform, and make prediction of lode mode, which achieves good results in the test of open data sets.

[1]  Yi Wang,et al.  Clustering of Electricity Consumption Behavior Dynamics Toward Big Data Applications , 2016, IEEE Transactions on Smart Grid.

[2]  Sanjay Lall,et al.  Shape-Based Approach to Household Electric Load Curve Clustering and Prediction , 2017, IEEE Transactions on Smart Grid.

[3]  Chao Shen,et al.  A review of electric load classification in smart grid environment , 2013 .

[4]  Hui Shi,et al.  Research on power load forecasting based on combined model of Markov and BP neural networks , 2010, 2010 8th World Congress on Intelligent Control and Automation.

[5]  Omar Y. Al-Jarrah,et al.  Multi-Layered Clustering for Power Consumption Profiling in Smart Grids , 2017, IEEE Access.

[6]  Jian Liu,et al.  Analysis of customers' electricity consumption behavior based on massive data , 2016, 2016 12th International Conference on Natural Computation, Fuzzy Systems and Knowledge Discovery (ICNC-FSKD).

[7]  Yi Wang,et al.  Review of Smart Meter Data Analytics: Applications, Methodologies, and Challenges , 2018, IEEE Transactions on Smart Grid.

[8]  Heng Huang,et al.  Using Smart Meter Data to Improve the Accuracy of Intraday Load Forecasting Considering Customer Behavior Similarities , 2015, IEEE Transactions on Smart Grid.

[9]  Seema Singh,et al.  Clustering based unit commitment with wind power uncertainty , 2016 .

[10]  Gianfranco Chicco,et al.  Overview and performance assessment of the clustering methods for electrical load pattern grouping , 2012 .

[11]  Donald J. Berndt,et al.  Using Dynamic Time Warping to Find Patterns in Time Series , 1994, KDD Workshop.

[12]  Yasin Kabalci,et al.  A survey on smart metering and smart grid communication , 2016 .

[13]  N.D. Hatziargyriou,et al.  Two-Stage Pattern Recognition of Load Curves for Classification of Electricity Customers , 2007, IEEE Transactions on Power Systems.