Case 3 Simulation Results for Power Load Forecasting Models

Simulation results from a study proposing a graph-based method for high-accuracy power load forecasting. The 5.5 KB Excel file, authored by Mengze Gu and last updated in April 2026, compares the performance of an Edge Graph Attention Network (EGAT) against traditional models like GRU, MLP, and LSTM.

Use Cases

• Benchmarking graph neural networks against recurrent and feed-forward architectures for load forecasting.
• Studying the effect of converting time series data into graph features on prediction accuracy.
• Analyzing the trade-off between model performance and computational cost in attention-based graph models.

Strengths

• Results are from a direct comparison of four distinct model architectures (EGAT, GRU, MLP, LSTM).
• Dataset is openly licensed under CC-BY-4.0, permitting reuse and redistribution.
• File size is 5.5 KB, ensuring quick download and inspection.

Limitations

• Row count and column-level documentation are unknown; field semantics must be inferred after download.
• The description notes the proposed EGAT method has increased computational cost and performance sensitivity to graph design.
• The dataset's scope is limited to a single case simulation, which may not represent broader applicability.

Provenance

Source

figshare, authored by Mengze Gu.

Collection Method

Likely contains simulated or experimental results from a proposed machine learning methodology.

Time Range

null

Freshness

Last updated 2026-04-23 17:25:29; freshness should be verified.

Geography

null