Quantitative evaluation method for the complementarity of wind–solar–hydro power and optimization of wind–solar ratio

Complementarity between wind power, photovoltaic, and hydropower is of great importance for the optimal planning and operation of a combined power system. However, less attention has been paid to quantify the level of complementarity of wind power, photovoltaic and hydropower. Therefore, this paper proposes a complementarity evaluation method for wind power, photovoltaic and hydropower by thoroughly examining the fluctuation of the independent and combined power generation. To begin with, a series of evaluation indices are presented from two perspectives: random fluctuation between adjacent time slots and ramp within continuous time windows. Secondly, a method for screening the typical meteorological day is proposed by combining t-distributed stochastic neighbor embedding with K-means clustering algorithm to improve the evaluation of daily complementarity. This method can deal with the feature extraction of high-dimension and correlated data sequences. A case study is conducted for the validation of the proposed method, and the evaluation result of the proposed method was compared with that of the Kendall’ rank correlation coefficient based on the Gaussian copula function. It is found that the proposed method has higher accuracy because it describes the complementarity from two aspects including fluctuation and ramp. It is also found from the study case that the optimum complementarity level for a certain case can be achieved by changing the ratio of photovoltaic and wind power. This work will provide reference for the optimization of power grid dispatch and power supply planning of combined power stations.