pypownet - documentation¶
pypownet is a simulator for power (electrical) grids able to emulate a power grid (of any size or characteristics) subject to a set of temporal injections (productions and consumptions) for discretized timesteps.
The simulator is able to simulate cascading failures, where successively overflowed lines are switched off and a loadflow is computed on the subsequent grid. It comes with an Reinforcement Learning-focused environment, which implements states (observations), actions (reduced to node-splitting and line status switches) as well as a reward signal. Also, the simulator possess a runner module for easily benchmarking new controlers, and has a renderer module that allows to observe the evolving grid and the actions of a controler.
- Agents specificities
- Using Environment information
- Example of agents