The western U.S. has tremendous solar potential. However, the variability of power generation from solar plants presents an operational challenge for grid system operators. Experience in power grids with significant penetration of variable renewable generation (both solar and wind power) has shown that the operating flexibility of the balance of the generation portfolio is a key element in secure and economic operation. This paper presents an overview of the variable characteristics of solar power, as well as the accompanying grid performance and operational economics for a system with significant solar generation. The paper will show results of economic operational simulations of a very high solar generation future for the western half of the United States. The evaluated system is subject to significant dynamic operating constraints; and the analytical methods used account for the critical interrelationships between system security, solar variability, and imperfect solar forecasts.
The large-scale access of distributed generation has a great impact on load forecasting in substation-area, which means the load curve in the substation-area cannot reflect the real load of users. Firstly, this paper considers the impact of distributed generation on the load curve, and proposes a load forecasting model based on data cleaning and deep learning in the substation-area. Secondly, considering the data missing during communication and transmission, the KNN algorithm is adopted to complete the missing data before the data input. And then, Pearson correlation coefficients are used for the correlation analysis of the input factors related to distributed generation, and the data is trained through Long-short Term Memory in deep learning. Finally, verified by load data of some substation-area in Jiangsu Province, the prediction model established in this paper has good prediction accuracy and stability.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.