基于数据驱动的孤岛微网自适应调频策略

微电网孤岛运行时，基于下垂控制的并联逆变器无法消除频率的静态偏差，必须借助二次调频来稳定频率值。在进行二次频率控制器参数设计时需要用到微网频率响应模型，然而由于微网系统的结构复杂多变以及系统内微源和负荷种类多样等原因，微网系统的数学模型难以获取,控制器参数也因此难以整定。为解决上述问题，提出一种基于数据驱动的改进无模型自适应控制的二次调频策略，该控制算法仅需要采样关键节点处的输入输出数据，利用RBF神经网络的自适应和自学习能力并按照一定的控制周期在线整定二次调频系统的无模型自适应控制器参数，从而将频率稳定在基准值。仿真结果验证了所提策略有很好的瞬态响应特性,同时具有较强的鲁棒性。

Adaptive Frequency Regulation Strategy for Islanded Microgrid Based on Data Driven Method

Parallel inverters based on droop control cannot eliminate the static frequency deviation during the operation of an islanded microgrid, and the frequency must be stabilized with the help of secondary frequency control. The frequency response model of the microgrid needs to be used in the design of secondary frequency controller parameters. However, due to the complex and changeable structure of the microgrid system and the variety of micro-sources and loads in the system, the mathematical model of the microgrid system is difficult to obtain and the controller parameters are thus difficult to set. To solve the above problems, a secondary frequency regulation strategy based on the data driven method is proposed. The control strategy only needs to use the input and output data sampled at the key nodes and uses the self-adaptation and self-learning capabilities of the radial basis function (RBF) neural network to tune the parameters of the model-free adaptive controller of the secondary frequency modulation system online according to a certain control period, thereby stabilizing the frequency at the reference value. Simulation results verify that the proposed strategy has fast transient response characteristics and strong robustness as well.