Power Data Recovery Method Based on Time Series Model for Understanding the Operation of HVDC Near-zone Assets

• Published in: 2020 IEEE Sustainable Power and Energy Conference (iSPEC) pp. 120 - 125
• Main Authors: Zhao, Shipei, Wang, Xiaoyun, Liao, Li, Dai, Xuan, Lu, Xuefei, Jiang, Li, Wang, You
• Format: Conference Proceeding
• Language: English
• Published: IEEE 23.11.2020
• Subjects: Analytical models; Data models; Estimation; Forecasting; Interpolation; Load modeling; power data recovery; spline interpolation; Time series analysis; time series modeling
• DOI: 10.1109/iSPEC50848.2020.9350989

Aiming at the problem of the lack of power load data in the forecasting of the delivery capacity of the sending end area and the load level forecasting of the receiving end during the clean energy delivery in the southwestern region, this paper gives the power load data recovery from the perspective of time series data characteristic analysis and modeling and estimation method. Based on the analysis of the multi-scale time series characteristics of the annual, monthly, and daily fluctuations of the electric load, the multi-scale time series characteristic modeling of the electric load is established. Spline interpolation is introduced to solve the non-parametric and variable coefficient problems of the load model, and the estimation method of the key parameters in the load model is given. According to the obtained load recovery model, a recovery method for missing data of weekly power load is proposed, and the idea of recovering daily load data is given. The actual calculation examples prove that the method proposed in this paper is accurate and effective.