Figure - available from: Journal of Signal Processing Systems
This content is subject to copyright. Terms and conditions apply.
Predicted values vs Actual for a LSTM b WT-LSTM c ANN d WT-ANN. The prediction is hour-ahead for a year worth of data, and the horizontal axis is presented in hours starting on August 15th, 2018.
Source publication
Renewable energy resources have gathered substantial interest, and several nations are striving to use them as the dominant power resource. However, the power output from these energy sources is inherently uncertain due to their reliance on natural forces like wind, sunlight, tides, geothermal, etc. An accurate estimation of expected consumer load...
Similar publications
The accuracy of electric load forecasting is a most influential factor in electricity market for reliable and economic power supply. The load demand is influenced by the weather condition. This paper demonstrates the performance of hybrid 3-level wavelet transform extreme learning machine for short term (day ahead) and medium term (season ahead) lo...
Citations
... Generally, load prediction mainly includes ultra-shortterm, short-term, medium-term, and long-term predictions. Specifically, the ultra-short-term load prediction indicates the load prediction within 1 h; the short-term mainly shows the daily and weekly load prediction; the medium-term ranges from months to years; and the long-term refers to the load prediction in the next 3 to 5 years or even longer [5,6]. As the power system scale increases, the type of load data is more complex and diverse in the smart grid, and the existence of various external influencing factors aggravates the randomness and nonlinear of the power load [7]. ...
Deep learning technology creates the condition for the optimization of the smart grid, and the big data analytical technique has the most efficient way to analyze and share the power load spatio-temporal data in the smart grid. Utilizing the graph-based method to learn the structure of load date distribution and load prediction has become hot-spot research. This paper proposes EnGAT-BiLSTM, an enhanced graph neural networks framework to realize short-term load prediction. The EnGAT-BiLSTM model aims to improve the prediction accuracy of the load and solve the sampled data sparsity in the short-term prediction. In this model, the Box-Cox transformation technology is used to denoise and obtain the effective load sampled data set; a dynamic load knowledge graph (DLKG) is designed to map the internal attribute of the various electrical load and the correlation of the external influencing factors; the graphic attention mechanism is introduced in the local network structure of graph neural network (GNN) to extract the high-quality load spatio-temporal features; the deep bi-directional long short-term memory (BiLSTM) framework is used for the lifelong learning of the load prediction. The extensive load-sampled datasets in the real world are employed to evaluate our method. The experimental results indicate that EnGAT-BiLSTM improves significantly in load prediction accuracy and has good robustness. The model will provide a valuable theoretical basis and guidance for the smart grid IoT system.
