.png){kind=icon}
.png){kind=icon}
Return
Short-term load forecasting with dense average network
.png){kind=icon}
OA
AI
DOI：10.1016/j.eswa.2021.115748.png)
.png){kind=icon}
.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
As an important part of the power system, power load forecasting directly affects the national economy. Small improvements in power load forecasts can save millions of dollars for the power industry. Therefore, improving the accuracy of power load forecasting has always been the pursuing goal for a power system. Based on this goal, this paper proposes a novel connection, the dense average connection, in which the outputs of all preceding layers are averaged as the input of the next layer in a feed-forward fashion. Dense average connection can alleviate the problem of gradient explosion without introducing new parameters. Based on dense average connection, we construct the dense average network (DaNet) for power load forecasting. On two public datasets (ISO-NE dataset and NAU dataset), we use MAPE, MAE and RMSE to evaluate the performance of DaNet. The predictions of DaNet are better than those of existing benchmarks. On this basis, this paper uses the ensemble method to reduce the peak value of prediction bias, which helps to alleviate the dispatching problem caused by unexpected loads. To verify the reliability of the model predictions, the robustness is analyzed and verified by adding input disturbances. The experimental results show that the proposed model is effective and robust for power load forecasting.
Keywords:
Short-term load forecasting
Deep learning
Dense average network
Robustness
AI Summary
Key information extracted from the uploaded paper, including a brief overview, abstract, background, key highlights, visual analysis, and future outlook.
Journal
IF:
7.5
Papers:
2.9W
Citations:
10.2W