基于SSA-SMC的风光-熔盐储能系统多时间尺度优化调度

Multi-Time Scale Optimal Scheduling of Wind-Solar-Molten Salt Energy Storage System Based on SSA-SMC

摘要:

如何有效地消纳风力发电和太阳能发电产生的波动电能,提高电网的经济性和灵活性,是目前亟待解决的重要课题。本文针对熔盐储能的微电网系统提出了一种新的多级优化调度方法。在传统的麻雀寻优方法和滑模控制方法的基础上,采用了神经网络机制,对算法的主要参数进行了动态的调整和优化,使算法具有更好的鲁棒性和自适应能力。上层调度以小时为基础,在考虑经济性、负荷平稳性和可再生能源利用率等因素的基础上,利用人工神经网络辅助的麻雀搜索算法,对电力系统进行了多目标优化。下层控制则是以15分钟的实时采样数据,采用一种结合人工神经网络自适应参数的滑模控制,对电加热器和其它装置进行实时调整,提高了对负荷波动和预测误差的抗干扰能力。仿真实验证明,所提出的优化方案能够提高电力系统的整体操作效率,提高电力系统的可再生能源利用率,减少电力系统的运行费用。

Abstract:

How to effectively absorb the fluctuating energy generated by wind and solar power, while improving the economic efficiency and flexibility of the power grid, remains a key challenge. This paper proposes a novel multi-level optimal scheduling method for a microgrid system equipped with molten salt energy storage. Based on the traditional Sparrow Search Algorithm (SSA) and Sliding Mode Control (SMC), a neural network mechanism is introduced to dynamically adjust and optimize key algorithm parameters, enhancing the robustness and adaptability of the control strategy. The upper-level scheduling is performed on an hourly basis, using a neural network-assisted SSA to conduct multi-objective optimization considering economic performance, load smoothness, and renewable energy utilization. The lower-level control operates with real-time sampling data every 15 minutes and employs a sliding mode control scheme with neural network-based adaptive parameter tuning to regulate electric heaters and other devices. This improves the system’s resilience to load fluctuations and forecasting errors. Simulation results demonstrate that the proposed optimization strategy enhances the overall operational efficiency of the power system, increases renewable energy utilization, and reduces operational costs.