Optimal Power System Scheduling Considering Uncertainty-Induced Generation Unit Output Curtailment

Wei Hu,Ning Mi,Hailiang Zhong,Weihan Chen,Kun Ma,Tianxiang Geng,Yun Wang,Bo Su,Long Yang

DOI: https://doi.org/10.1109/ccssta62096.2024.10691794

2024-01-01

Abstract:Under the call of the national “dual carbon” goals, the growth of the proportion of new energy generation has become an inevitable trend in energy structure adjustment. The “static parameter” model, which traditional thermal power units rely on, can no longer accurately reflect their output uncertainty, urgently requiring a new optimization scheduling strategy to cope with this change. This study proposes an optimized scheduling model for power generation systems that considers the output curtailment and unplanned outages of thermal power units. It employs non-parametric kernel density estimation techniques to model the uncertainty of thermal power unit output and uses chance constraints to define the fluctuation range of generation output and the system’s reliability. The model adopts a scenario generation method based on Copula theory to construct output scenarios and further employs reduction techniques to optimize the number of scenarios. Moreover, this paper introduces Conditional Value at Risk as a key indicator to measure the cost of system uncertainty risk, thereby providing a more comprehensive perspective in risk assessment. Simulation experiments on the IEEE 30-node standard system demonstrate that compared to traditional optimization scheduling models relying on static parameters, the proposed model can more effectively balance the output fluctuations of different thermal power units, significantly reducing the risk of curtailment due to supply shortages. This study offers a new theoretical basis and practical guidance for optimizing the scheduling of power systems under the context of large-scale integration of new energy and economic fluctuations.