摘要
针对现场实际负荷成分的时变性、随机性、复杂性、多样性和非线性的特点,在考虑实际负荷暂态特性的基础上,提出一种基于神经网络的配电网受控电流源负荷模型的方法。该方法根据负荷群在稳态运行条件下的电压和电流暂态特性,通过神经网络学习负荷群的电压电流特性,用受控电流源代替实际的负荷群,受控电流源的电流大小受神经网络控制,并利用Matlab/Simlink对所提出的负荷模型建立方法进行仿真验证。仿真结果表明所建立的负荷模型在单相短路、两相短路和三相短路条件下均具有良好的稳定性和准确性。
Considering the characteristics of time-variables,randomness,complexity,diversity and nonlinearity in actual load,a new load model with a controlled current source based on the transient characteristics of the load in distribution network is proposed.The transient voltage and current characteristics of the loads in the steady-state operation are studied by a BP neural network according to the proposed method.The actual loads are replaced by a controlled current source model,which the output current is controlled by the BP neural network.The proposed load model is simulated in the Matlab/Simlink.The simulation results suggest that the proposed load model be quite stable and accurate under the single-phase short faulty,two-phase short faulty and three-phase short faulty.
出处
《电网与清洁能源》
2010年第12期1-7,共7页
Power System and Clean Energy
基金
国家自然科学基金资助项目(60971077)
山东省自然科学基金资助项目(ZR2009FM061)
关键词
负荷模型
暂态特性
神经网络
受控电流源
load model
transient characteristics
neural network
controlled current source
