新能源混合电力系统负荷频率控制  

Load Frequency Control of Multi-area Interconnected Power System With New Energy Sources

作　　者：王晓燕 王莹 马鑫源 刘飞 杨哲 Wang Xiaoyan;Wang Ying;Ma Xinyuan;Liu Fei;Yang Zhe(State Grid Guyuan Power Supply Company,Guyuan,Ningxia 756000,China)

机构地区：[1]国网固原供电公司,宁夏固原756000

出　　处：《现代科学仪器》2022年第5期187-192,共6页Modern Scientific Instruments

基　　金：国网宁夏电力有限公司科技项目(NX202101001).

摘　　要：为提升电力系统负荷频率控制自适应性及稳定性,提出一种含新能源的混合电力系统负荷频率控制方法。搭建包含三种新能源的混合电力系统模型,采用基于云模型算法与神经网络算法结合创建的云-神经网络PI算法,通过确定负荷频率控制器的输入和输出变量,结合自学习云规则并利用神经网络实现各变量之间的非线性投射,云处理层的输出及神经网络的共同输出即为系统负荷频率控制的最佳输出。实验结果表明,该方法在阶跃负荷干扰、瞬时负荷干扰、风电随机干扰三种情况下表现出优秀的控制性能及自适应性,能够有效控制电力系统负荷频率。In order to improve the adaptability and stability of load frequency control in power system,a load frequency control method for hybrid power system with new energy is proposed.A multi-area interconnected power system model of three new energy sources is established.The Pi Algorithm of cloud-neural network based on cloud model Algorithm and neural network algorithm is used to determine the input and output variables of load frequency controller,combined with self-learning cloud rules and using neural network to realize nonlinear projection among variables,the output of cloud processing layer and the common output of neural network are the optimal output of load frequency control.The experimental results show that this method can effectively control the load frequency of power system under the condition of step load disturbance,instantaneous load disturbance and wind power random disturbance.

关 键 词：新能源 混合 电力系统 负荷频率控制 云神经网络 负荷干扰 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]