1. Real-time Hosting Capacity Assessment for Electric Vehicles: A Sequential Forecast-then-Optimize Method
- Author
-
Zhuang, Yingrui, Cheng, Lin, Qi, Ning, Wang, Xinyi, and Chen, Yue
- Subjects
Mathematics - Optimization and Control - Abstract
Hosting capacity (HC) assessment for electric vehicles (EVs) is of great significance to ensure the safe integration of EVs and the reliable operation of power systems. Existing HC assessment methods mainly focus on a long-term perspective (e.g., system planning), and model the EV charging demands as scalar values. However, HC estimated from a long-term perspective may be inaccurate and unreliable for real-time operation, since long-term peak estimation differs from the real-time stochasticity of EV charging demands. In this regard, this paper proposes a real-time HC assessment method for EVs through a three-step process of real-time probabilistic forecasting, real-time risk analysis and probabilistic optimization. Specifically, we first conduct real-time probabilistic forecasting through an adaptive spatio-temporal graph convolutional network to describe the stochasticity in EV charging demands across multiple charging stations. This model leverages adaptive spatial feature extraction, attention-based temporal feature extraction, and second-order graph representation to capture the spatio-temporal features of EV charging demands. Subsequently, based on the probabilistic forecasting, we propose a real-time risk analysis method, which is achieved by calculating the probabilistic power flow based on an improved Gaussian mixture model. Furthermore, we provide a real-time formulation of the HC of EVs and propose an optimization model to assess it. Numerical experiments on a real-world dataset demonstrate that the proposed forecasting model outperforms state-of-the-art forecasting methods by achieving the lowest RMSE value of 0.0442, the proposed real-time risk analysis method maintains over 96.6% accuracy with 99.99% reduction in computational complexity, and the real-time HC is improved by 66.3% compared to long-term assessment., Comment: This is a manuscript submitted to Applied Energy
- Published
- 2024