Open AccessBeijing-Tianjin-Hebei Energy Demand Combination Forecast Analysis
Author(s) -
Li Jian,
Xiaoqian Zhang
Publication year - 2021
Publication title -
iop conference series. earth and environmental science
Language(s) - English
Resource type - Journals
eISSN - 1755-1307
pISSN - 1755-1315
DOI - 10.1088/1755-1315/631/1/012104
Subject(s) - beijing , autoregressive integrated moving average , artificial neural network , gray (unit) , computer science , econometrics , time series , geography , china , economics , artificial intelligence , machine learning , medicine , archaeology , radiology
Energy demand forecasting is the basis for responding to high-quality economic development requirements and targeted adjustments to the Beijing-Tianjin- Hebei energy structure. This paper selects five main factors that affect energy demand, constructs a combined forecasting model of a combination of multi-factor gray neural network and ARIMA-BP neural network, and introduces the idea of chaos optimization on this basis to simulate and analyse data from 2012 to 2016, and predict the energy demand in the Beijing-Tianjin-Hebei region in 2020 and 2025.The results show that: 1. Compared with the CGA-ARIMA-BP model and the CGA-GNN model, the CGA-GNN-ARIMA-BP model has higher prediction accuracy; 2. It is estimated that in 2020 and 2025, the energy demand in the Beijing-Tianjin-Hebei region will reach 493 and 552 million tons of standard coal.