Fig 1.
TransGrid-CostOpt: Architecture Diagram of a Hybrid Framework for Cost Prediction and Optimization of Distribution Network Assets.
Fig 2.
Transformer-based Multi-source Feature Fusion and Deep Representation Extraction Framework.
Fig 3.
Architecture of Time Series Prediction with Bidirectional LSTM and Contextual Information Fusion.
Fig 4.
Hierarchical Meta-Reinforcement Learning Framework for Cost Optimization Decision-Making in Distribution Networks.
Table 1.
Overview of Datasets Used in Distribution Network Load Forecasting and Cost Optimization Tasks.
Table 2.
Comparison of Evaluation Results between TransGrid-CostOpt and Baseline Models on Two Datasets.
Fig 5.
Comparative Performance of TransGrid-CostOpt and Other Models.
Table 3.
Comparison of Computational Overhead for TransGrid-CostOpt vs. Baseline Models.
Table 4.
Ablation Study Results for TransGrid-CostOpt Model on Two Datasets.
Table 5.
Ablation Study Results for TransGrid-CostOpt Model with Multiple Module Removal on Two Datasets.
Fig 6.
Overall Ablation Experiment Results for TransGrid-CostOpt.
Table 6.
Ablation Study Results on Fusion Methods.