Impact of Energy Storage Lifespan on System Operation Strategy in Full-Horizon Operation Simulation by Machine Learning

As the penetration rate of renewable energy sources in the power grid increases, it is essential to conduct an in-depth study on the scenarios where energy storage systems are used in conjunction with renewable energy sources to mitigate the uncertainty of renewable energy output and enhance the availability and economic viability of renewable energy generation within the system. During the operation of energy storage systems, there is a cost associated with their lifespan. Therefore, it is worth investigating how to improve the economic performance of hybrid wind-storage power stations during their operation. Conducting continuous operation simulations over the entire year would result in excessively high computational costs; however, machine learning methods can effectively extract typical scenarios, reducing computational load while retaining the characteristics of system operation status. In this paper, machine learning techniques are employed to extract typical scenarios and analyze the impact of considering the lifespan cost of energy storage on the dispatch strategy of an improved IEEE 6-bus system.