紧急控制状态下储能提高电力系统暂态稳定性研究

Research on Energy Storage to Improve Transient Stability of Power Systems Under Emergency Control Conditions

针对电力系统在面对突发负荷变化和可再生能源波动时的暂态稳定性问题,提出了一种融合轻量梯度提升机(LightGBM)、长短期记忆网络(LSTM)和信任区域策略优化(TRPO)的基于储能系统的紧急控制策略,称为LightGBM-LSTM-TRPO模型,旨在提高电力系统的稳定性和可靠性。首先,分析了电力系统暂态稳定性的重要性及当前面临的挑战,强调了传统控制方法在紧急控制状态下的局限性。随后,使用LightGBM算法预测电力系统的状态变化趋势,包括负荷变化和可再生能源波动。接着,使用LSTM算法处理电力系统中的时间序列历史数据,以捕捉电力系统动态变化的复杂性和长期依赖关系。最后,使用TRPO算法优化储能系统在紧急控制状态下的响应策略,以维持电网的暂态稳定性。实验的仿真结果表明,与传统的LSTM和改进的LightGBM-LSTM等算法相比,本文所提的LightGBM-LSTM-TRPO模型能准确预测出电力系统的储能输出量,其紧急控制状态下的切机量和切负荷量更少,响应速度更快,电压恢复程度更高,可快速缓解电网在突发事件下的压力。

A emergency control strategy based on energy storage system,called LightGBM-LSTM-TRPO model,is proposed to address the transient stability issues of power systems in the face of sudden load changes and renewable energy fluctuations.The strategy integrates lightweight gradient boosting machine(LightGBM),long short-term memory network(LSTM),and trust region strategy optimization(TRPO),aiming to improve the stability and reliability of power systems.Firstly,the importance of transient stability in power systems and the current challenges faced were analyzed,emphasizing the limitations of traditional control methods in emergency control states.Subsequently,the LightGBM algorithm is used to predict the trend of state changes in the power system,including load fluctuations and renewable energy fluctuations.Next,the LsTM algorithm is used to process time series historical data in the power system to capture the complexity and long-term dependencies of dynamic changes in the power system.Finally,the TRPO algorithm is used to optimize the response strategy of the energy storage system in emergency control states,in order to maintain the transient stability of the power grid.The simulation results of the experiment show that compared with traditional LSTM and improved LightGBM-LSTM algorithms,the LightGBM-LSTM-TRPO model proposed in this paper can accurately predict the energy storage output of the power system.Its emergency control state has fewer machine and load shedding,faster response speed,higher voltage recovery degree,and can quickly alleviate the pressure on the power grid in case of emergencies.