基于TensorFlow的LSTM循环神经网络短期电力负荷预测

LSTM Recurrent Neural Network Short-Term Power Load Forecasting Based on Tensor Flow

随着智能电网、坚强电网的建立及人工智能领域技术的高速发展,如何对电力领域的负荷进行更高精度的预测已成为电力从业者们特别关注与研究的问题。基于TensorFlow智能学习系统的深度学习LSTM循环神经网络算法的短期电力负荷预测算法,结合某地区发电厂负荷数据设计实验,通过多次数据迭代、参数更新,进行模型训练与预测,最终的实验证明:基于TensorFlow的LSTM循环神经网络算法预测效果明显好于传统机器学习算法。随着数据量的增大,模型更显示出其良好的鲁棒性。

With the establishment of smart grids and strong grids,and the rapid development of technologies in the field of artificial intelligence,how to predict the load of the electric power more accurately has become a problem that electric power practitioners pay special attention to and study.Therefore,this paper proposes a LSTM recurrent neural network short-term power load forecasting algorithm based on TensorFlow.Experiment is designed by using actual power load data of a transformer substation.Using multiple data iterations and parameter updates to train the model and predict.The conclusions as following:Under massive data,LSTM recurrent neural network short-term power load forecasting algorithm based on TensorFlow has obviously better prediction effect than traditional machine learning algorithm.And as the amount of data increases,the model shows its good robustness.