基于GM(1,1)模型和BP神经网络的四川省用电量预测

Production of Electricity Consumption Based on GM(1,1) and BP Neural Networkin Sichuan Province, China

随着中国经济的快速增长,人民对能源的需求也在逐步增长。电能作为居民基本的生活能源,其是否能够充足稳定的供应决定了社会经济的运行。当在面对突发事件无法及时满足供电需求时,对于用电量的预测就成为了一种未雨绸缪的重要方法。本文基于GM(1,1)模型和BP神经网络建立了一种新的GM-ABP模型。我们将GM-ABP模型用于预测中国四川省年用电总量。本文提出的模型在GM(1,1)模型的基础上加以修正,利用Adam算法优化的BP神经网络对残差进行拟合。预测结果显示GM-ABP模型的预测精度均优于GM(1,1)模型和Adam-BP神经网络模型,且预测精度有较大提升。最后通过GM-ABP模型给出了四川省2021~2023年用电总量的预测值。

With the rapid growth of China’s economy, people’s demand for energy is also gradually increasing. Electric energy is the essential energy of life;its sufficient and stable supply determines the opera-tion of the social economy. When the power supply demand cannot be timely met in the face of emergencies, the forecast of electricity consumption has become a meaningful way to prepare for that. This paper establishes a new GM-ABP model based on GM (1,1) model and BP neural network. We use the GM-ABP model to predict the total annual electricity consumption in Sichuan Province, China. The model proposed in this paper is modified based on the GM (1,1) model, and the residual is fitted by BP neural network optimized by the Adam algorithm. The prediction results show that the prediction accuracy of the GM-ABP model is better than that of the GM (1,1) model and Adam-BP neural network model, and the prediction accuracy is greatly improved. Finally, the GM-ABP model is used to predict the total electricity consumption of Sichuan Province in 2021~2023.