Short-term County-Level Power Load Forecasting Model Based on Improved Deep Belief Network

Konferenz: CAIBDA 2022 - 2nd International Conference on Artificial Intelligence, Big Data and Algorithms
17.06.2022 - 19.06.2022 in Nanjing, China

Tagungsband: CAIBDA 2022

Seiten: 4Sprache: EnglischTyp: PDF

Autoren:
Han, Yiming; Zhang, Bin; Jin, Panlong; Tian, Xing (State Grid Ningxia Electric Power Co., Ltd. Economic and Technological Research Institute; Yinchuan, Ningxia, China)

Inhalt:
Grid load forecasting is the prerequisite and cornerstone of power system planning, decision-making and economic operation. For the power industry, it is also a key technical project. This paper studies the short-term county-level PLF model based on the improved deep belief network, and then analyzes the PLF model and the related theories of the improved deep belief network on the basis of literature data and the short-term county-level PLF model based on the improved deep belief network. The first-level PLF model is constructed, and the constructed model is tested. Through the test results, it is concluded that the model constructed in this paper has a large forecast error in the experiment. This is because the forecast is short-term and the amount of data is not large.