| Summary: | To accurately evaluate the influence of the uncertainty and correlation of photovoltaic (PV) output and load on the running state of power system, a probabilistic optimal power flow (POPF) calculation method based on adaptive diffusion kernel density estimation is proposed in this paper. First, based on the distribution characteristics of PV output, the adaptive diffusion kernel density estimation model of PV output is constructed, which can transform the kernel function into a linear diffusion process to achieve self-tuning of the bandwidth of the nuclear density estimation. This model can fit the distribution of arbitrary distribution of PV power, improve the local adaptability of PV output model, and reflect the uncertainty and volatility of PV output more accurately. Therefore, it can provide more accurate input for POPF calculation. Second, the Kendall rank correlation coefficient and the least Euclidean distance are used as correlation measure and index of fitting to select the optimal Copula function, and the joint probability distribution model of PV output and load is constructed. After extracting the correlated PV output and load samples, a POPF calculation method considering the correlation of PV output and load is proposed by using genetic algorithm (GA), which takes the lowest fuel cost of power generation as the objective function. Finally, simulation studies are conducted with the measured data of a PV power plant of China and the IEEE 30-bus power system. The results show that considering the correlation between PV output and load can improve the accuracy of POPF calculation and effectively reduce the power generation cost of the power system.
|
|---|
