摘要
配网侧多种分布式电源的接入将对系统的安全性、可靠性、稳定性产生重大影响,含有多种分布式电源的广义负荷建模是一种机遇与挑战。首先在综合负荷模型(SLM)的虚拟母线上增加一台异步发电机和两个有功功率源,构造一套含有固体氧化物燃料电池(SOFC)并网系统、三相单机光伏并网系统以及恒速异步风力发电机三种分布式电源的广义负荷模型结构,并对该模型初始化。然后提出一种模型参数的辨识策略:将低灵敏度且非时变的参数固定为聚合值,对于高灵敏且时变的参数,在聚合值附近采用遗传算法进行辨识,即只辨识电动机比例、定子电抗、风力发电相对纯负荷的比例、电动机初始负载率、配电网络的电阻、电抗、光伏发电相对纯负荷比例等7个模型参数。通过对三种不同负荷水平进行仿真,获取实测数据样本并进行参数辨识和适应性分析,验证了含多种分布式电源的广义负荷模型描述能力强,泛化能力好,参数辨识结果分散性较小。
Penetration of a large amount of distributed generators （DG） connected to distribution network will have notable effects on power system security, reliability, and stability. Generalized load modeling with a large amount of distributed generation considered is an opportunity as well as a challenge. This paper proposes a generalized load model structure of distribution network which contains SOFC systems, three-phase single stage photovoltaic systems and constant speed asynchronous generator, which can be obtained by adding two active power sources and one asynchronous generator to the fictitious bus of SLM model. For the generalized load model, an induction generator should be added to the fictitious bus of SLM model. Low sensitivity and non time-varying parameters for polymerization are fixed value, for high sensitive and time-varying parameters, in polymerization near value by using the Genetic Algorithm to identify. Only seven parameters, that is, the leakage inductance of stator, the portion of induction motor, the load factor, the relatively pure load proportion of wind generators, the relatively pure load proportion of photovoltaic generators, the distribution network resistor, and the distribution network inductance need to be identified. The ability of characterization and generalization, and the stability of identified parameters of the proposed generalized load model are validated by three simulation cases with different load levels on the distribution network.
出处
《电力系统保护与控制》
EI
CSCD
北大核心
2013年第4期105-111,共7页
Power System Protection and Control
关键词
广义负荷建模
分布式电源
参数辨识
generalized load modeling
distributed generator
parameters identification
