torch_geometric.datasets.OPFDataset

class OPFDataset(root: str, split: Literal['train', 'val', 'test'] = 'train', case_name: Literal['pglib_opf_case14_ieee', 'pglib_opf_case30_ieee', 'pglib_opf_case57_ieee', 'pglib_opf_case118_ieee', 'pglib_opf_case500_goc', 'pglib_opf_case2000_goc', 'pglib_opf_case6470_rte', 'pglib_opf_case4661_sdetpglib_opf_case10000_goc', 'pglib_opf_case13659_pegase'] = 'pglib_opf_case14_ieee', num_groups: int = 20, topological_perturbations: bool = False, transform: Optional[Callable] = None, pre_transform: Optional[Callable] = None, pre_filter: Optional[Callable] = None, force_reload: bool = False)[source]

Bases: InMemoryDataset

The heterogeneous OPF data from the “Large-scale Datasets for AC Optimal Power Flow with Topological Perturbations” paper.

OPFDataset is a large-scale dataset of solved optimal power flow problems, derived from the pglib-opf dataset.

The physical topology of the grid is represented by the "bus" node type, and the connecting AC lines and transformers. Additionally, "generator", "load", and "shunt" nodes are connected to "bus" nodes using a dedicated edge type each, e.g., "generator_link".

Edge direction corresponds to the properties of the line, e.g., b_fr is the line charging susceptance at the from (source/sender) bus.

Parameters:

  • root (str) – Root directory where the dataset should be saved.

  • split (str, optional) – If "train", loads the training dataset. If "val", loads the validation dataset. If "test", loads the test dataset. (default: "train")

  • case_name (str, optional) – The name of the original pglib-opf case. (default: "pglib_opf_case14_ieee")

  • num_groups (int, optional) – The dataset is divided into 20 groups with each group containing 15,000 samples. For large networks, this amount of data can be overwhelming. The num_groups parameters controls the amount of data being downloaded. Allowed values are [1, 20]. (default: 20)

  • topological_perturbations (bool, optional) – Whether to use the dataset with added topological perturbations. (default: False)

  • transform (callable, optional) – A function/transform that takes in a torch_geometric.data.HeteroData object and returns a transformed version. The data object will be transformed before every access. (default: None)

  • pre_transform (callable, optional) – A function/transform that takes in a torch_geometric.data.HeteroData object and returns a transformed version. The data object will be transformed before being saved to disk. (default: None)

  • pre_filter (callable, optional) – A function that takes in a torch_geometric.data.HeteroData object and returns a boolean value, indicating whether the data object should be included in the final dataset. (default: None)

  • force_reload (bool, optional) – Whether to re-process the dataset. (default: False)