# EV2Gym: A Realistic EV-V2G-Gym Simulator for EV Smart Charging
<div align="center">
<img align="center" src="https://github.com/StavrosOrf/EV2Gym/assets/17108978/86e921ad-d711-4dbb-b7b9-c69dee20da11" width="55%"/>
</div>
[](https://www.python.org/downloads/release/python-360/) [](https://pypi.org/project/ev2gym/) 
---
Develop and evaluate **any type of smart charging algorithm**: from simple heuristics, Model Predictive Control, Mathematical Programming, to Reinforcement Learning!
EV2Gym is **fully customizable** and easily **configurable**! Now can also **simulate the grid**, thanks to [RL-ADN](https://github.com/EnergyQuantResearch/RL-ADN)!
The EV2Gym **Paper** can be found at: [arXiv](https://arxiv.org/abs/2404.01849) and [IEEE](https://dl.acm.org/doi/abs/10.1109/TITS.2024.3510945).
**Highly recommended** related works and repositories:
| Title | Paper | GitHub repository |
| ------------------------------------------------------------------------------------------------------------------------------ | ------------------------------------------------------------------------------------ | -------------------------------------------------------------------------------------------------- |
| Physics-Informed Reinforcement Learning for Large-Scale EV Smart Charging Considering Distribution Network Voltage Constraints | [arXiv](https://arxiv.org/abs/2510.12335) | [https://github.com/StavrosOrf/EV2Gym_PI-TD3](https://github.com/StavrosOrf/EV2Gym_PI-TD3) |
| GNN-DT: Graph Neural Network Enhanced Decision Transformer for Efficient Optimization in Dynamic Environments | [arXiv](https://arxiv.org/abs/2502.01778) | [https://github.com/StavrosOrf/DT4EVs](https://github.com/StavrosOrf/DT4EVs) |
| Scalable Reinforcement Learning for Dynamic Electric Vehicle Charging Optimization Using Graph Neural Networks | [Paper link](https://www.nature.com/articles/s44172-025-00457-8) | [https://github.com/StavrosOrf/EV-GNN](https://github.com/StavrosOrf/EV-GNN) |
| Open-source algorithms for maximizing V2G flexibility based on model predictive control | [ScienceDirect](https://www.sciencedirect.com/science/article/pii/S0378779625006704) | [https://github.com/CesarDiazLondono/MPC-G2V-V2G](https://github.com/CesarDiazLondono/MPC-G2V-V2G) |
## Installation
Install the package using pip:
```bash
pip install ev2gym
```
Run the example code below to get started ...
```python
from ev2gym.models.ev2gym_env import EV2Gym
from ev2gym.baselines.mpc.V2GProfitMax import V2GProfitMaxOracle
from ev2gym.baselines.heuristics import ChargeAsFastAsPossible
config_file = "ev2gym/example_config_files/V2GProfitPlusLoads.yaml"
# Initialize the environment
env = EV2Gym(config_file=config_file,
save_replay=True,
save_plots=True)
state, _ = env.reset()
agent = V2GProfitMaxOracle(env,verbose=True) # optimal solution
# or
agent = ChargeAsFastAsPossible() # heuristic
for t in range(env.simulation_length):
actions = agent.get_action(env) # get action from the agent/ algorithm
new_state, reward, done, truncated, stats = env.step(actions) # takes action
```
- ### For Reinforcement Learning:
To train an RL agent, using the [StableBaselines3](https://stable-baselines3.readthedocs.io/en/master/) library, you can use the following code:
```python
import gymnasium as gym
from stable_baselines3 import PPO, A2C, DDPG, SAC, TD3
from sb3_contrib import TQC, TRPO, ARS, RecurrentPPO
from ev2gym.models.ev2gym_env import EV2Gym
# Choose a default reward function and state function or create your own!!!
from ev2gym.rl_agent.reward import profit_maximization, SquaredTrackingErrorReward, ProfitMax_TrPenalty_UserIncentives
from ev2gym.rl_agent.state import V2G_profit_max, PublicPST, V2G_profit_max_loads
config_file = "ev2gym/example_config_files/V2GProfitPlusLoads.yaml"
env = gym.make('EV2Gym-v1',
config_file=config_file,
reward_function=reward_function,
state_function=state_function)
# Initialize the RL agent
model = DDPG("MlpPolicy", env)
# Train the agent
model.learn(total_timesteps=1_000_000,
progress_bar=True)
# Evaluate the agent
env = model.get_env()
obs = env.reset()
stats = []
for i in range(1000):
action, _states = model.predict(obs, deterministic=True)
obs, reward, done, info = env.step(action)
if done:
stats.append(info)
```
!!! You can develop your own reward and state functions and use them in the environment.
## Table of Contents
- [Installation](#Installation)
- [Overview](#Overview)
- [Configuration File](#Configuration-File)
- [File Structure](#File-Structure)
- [Citing](#Citing-EV2Gym)
- [License](#License)
- [Contributing](#Contributing)
<!-- Bullet points with all the benefits -->
## Overview
- The simulator can be used to evaluate any type of algorithm to gain insights into its efficiency.
- The “gym environment” can readily support the development of RL algorithms.
- Replays of simulations are saved and can be solved optimally using the Gurobi Solver.
- Easy to incorporate additional functionality for any use-case.
- Now, also simulates grid for voltage magnitude!
- The number and the topology of Transformers, Charging stations, and Electric Vehicles are parameterizable.
- The user can import custom data.
- Uses only open-source data:
- EV spawn rate, time of stay, and energy required are based on realistic probability distributions *ElaadNL* conditioned on time, day, month and year.
- *Pecan Street* data is used for the load profiles.
- *Renewables Ninja* data is used for the PV generation profiles.
- EV and Charger characteristics are based on real EVs and chargers existing in NL (*RVO Survey*).
- Charging/ Discharging prices are based on historical day-ahead prices from *ENTSO-e*.
Focused on **realistic** parameters and **fully customizable**:
- **Power Transformer** model:
- Max Power Limit
- Inflexible Loads, PV, Capacity Reduction events
- **Charging Stations** model:
- Min and Max charge/discharge power/ Current
- Voltage and phases, AC or DC
- List of connected transformers
- **Electric Vehicle** model:
- Connected charging station and port
- Min and Max battery energy level
- Time of arrival and departure
- Energy at arrival/ desired energy at departure
- Min and Max power levels
- Charge and discharge efficiency
- Constant-Current/ Constant-Voltage load-curve option
- **Battery Degradation** model:
- Cyclic aging
- Calendar aging
<div align="center">
<img align="center" src="https://github.com/StavrosOrf/EV2Gym/assets/17108978/d15d258c-b454-498c-ba7f-634d858df3a6" width="90%"/>
</div>
An EV2Gym simulation comprises three phases: the configuration phase, which initializes the models; the simulation phase, which spans $T$ steps, during which the state of models like EVs and charging stations is updated according to the decision-making algorithm; and finally, in the last phase, the simulator generates evaluation metrics for comparisons, produces replay files for reproducibility, and generates real-time renders for evaluation.
## Configuration File
The configuration file is used to set the parameters of the simulation. The configuration file is a YAML file that contains the following parameters:
```yaml
# This yml file is used to configure the evsim simulation
##############################################################################
# Simulation Parameters
##############################################################################
timescale: 15 # in minutes per step
simulation_length: 96 #90 # in steps per simulation
##############################################################################
# Date and Time
##############################################################################
# Year, month,
year: 2022 # 2015-2023
month: 1 # 1-12
day: 17 # 1-31
# Whether to get a random date every time the environment is reset
random_day: True # True or False
random_hour: False # True or False
# Simulation Starting Time
# Hour and minute do not change after the environment has been reset
hour: 5 # Simulation starting hour (24 hour format)
minute: 0 # Simulation starting minute (0-59)
# Simulate weekdays, weekends, or both
simulation_days: weekdays # weekdays, weekends, or both
# EV Spawn Behavior
scenario: public # public, private, or workplace
spawn_multiplier: 5 # 1 is default, the higher the number the more EVs spawn. Play somewhere between 3-7. With 1 often only 1/5 EVs show up.
##############################################################################
# Prices
##############################################################################
discharge_price_factor: 1 # how many times more abs(expensive/cheaper) it is to discharge than to charge. Are similar, discharging cost could be a bit higher.
##############################################################################
# Distribution Network Settings
##############################################################################
v2g_enabled: True # True or False
number_of_charging_stations: 150 # Play somewhere 5-20
number_of_transformers: -1
number_of_ports_per_cs: 1
# Provide path if you want to load a specific charging topology
# *(where chargers are located and what are their characteristics),
# else write None for a randomized one with the above parameters
charging_network_topology: None #./config_files/charging_topology_10.json
simulate_grid: True # True or False
network_info:
vm_pu: 1.0 # Voltage magnitude in per unit
s_base: 1000 # Base power in kVA
load_multiplier: 1 # Load multiplier for the grid
pv_scale: 80 # Percentage% of maximum PV power with respect to the total load of the bus
bus_info_file: './ev2gym/data/network_data/node_34/Nodes_34.csv'
branch_info_file: './ev2gym/data/network_data/node_34/Lines_34.csv'
# bus_info_file: './ev2gym/data/network_data/node_123/Nodes_123.csv'
# branch_info_file: './ev2gym/data/network_data/node_123/Lines_123.csv'
pf_solver: 'Laurent' # 'Laurent' or 'PandaPower'
##############################################################################
# Power Setpoints Settings
##############################################################################
# How much the power setpoints can vary in percentage compared to the nominal power
# The higher the number the easier it is to meet the power setpoints, the opposite for negative numbers
power_setpoint_enabled: True # True or False
power_setpoint_flexiblity: 80 # (in percentage +/- %)
##############################################################################
# Inflexible Loads, Solar Generation, and Demand Response (Not compatible with simulate_grid = True)
##############################################################################
# Whether to include inflexible loads in the transformer power limit, such as residential loads
tr_seed: -1 # seed for the random number generator of transformer loads(-1 for random seed)
inflexible_loads: # Offices, homes
include: False # True or False
inflexible_loads_capacity_multiplier_mean: 1 # 1 is default, the higher the number the more inflexible loads
forecast_mean: 30 # in percentage of load at time t%
forecast_std: 5 # in percentage of load at time t%
# PV solar Power
solar_power:
include: False # True or False
solar_power_capacity_multiplier_mean: 1 # 1 is default, the higher the number the more solar power
forecast_mean: 20 # in percentage of load at time t%
forecast_std: 5 # in percentage of load at time t%
# Whether to include demand response in the transformer power limit
demand_response:
include: False # True or False
events_per_day: 1
#How much of the transformer power limit can be used for demand response
event_capacity_percentage_mean: 35 # (in percentage +/- %) reduction
event_capacity_percentage_std: 5 # (in percentage +/- %)
event_length_minutes_min: 60
event_length_minutes_max: 60
event_start_hour_mean: 12
event_start_hour_std: 2
# How many minutes ahead we know the event is going to happen
notification_of_event_minutes: 60
##############################################################################
# EV Specifications
##############################################################################
heterogeneous_ev_specs: False #if False, each EV has the same specifications
# such as battery capacity, charging rate, etc.
ev_specs_file: ./ev2gym/data/ev_specs_v2g_enabled2024.json # path to the file with the EV specifications
##############################################################################
# Default Model values
##############################################################################
# These values are used if not using a charging network topology file or
# if the EV specifications are not provided
# Default Transformer model
transformer:
max_power: 200 # in kW
# Default Charging Station model
charging_station:
min_charge_current: 0 # Amperes
max_charge_current: 32 # Amperes
min_discharge_current: 0 # Amperes (actual value <=0)
max_discharge_current: -32 # Amperes (actual value <=0)
voltage: 400 # Volts
phases: 3 # 1,2, or 3
# Default EV model
ev:
#The following values are used if "heterogeneous_ev_specs = False"
battery_capacity: 70 # in kWh
max_ac_charge_power: 22 # in kW
min_ac_charge_power: 0 # in kW
max_dc_charge_power: 50 # in kW
max_discharge_power: -22 # in kW
min_discharge_power: 0 # in kW
ev_phases: 3
charge_efficiency: 1 # 0-1 (0% - 100%)
discharge_efficiency: 1 # 0-1 (0% - 100%)
transition_soc: 1 # 0-1 (0% - 100%)
#The following values are also used if "heterogeneous_ev_specs = True"
min_battery_capacity: 15 # in kWh
min_time_of_stay: 200 # in minutes
min_emergency_battery_capacity: 15 # in kWh
desired_capacity: 1 # in (0-1) (0% - 100%) #Keep at 100% for now
#if trasition_soc is < 1, the curve of the line is affected by:
transition_soc_multiplier: 50 # default 1 (the higher the number the shorter the effect of CCCV region)
```
## File Structure
The file structure of the EV2Gym package is as follows:
```bash
├── ev2gym
│ ├── baselines
│ │ ├── gurobi_models/
│ │ ├── mpc/
│ │ ├── heuristics.py
│ ├── data/
│ ├── models
│ │ ├── ev2gym_env.py
│ │ ├── ev.py
│ │ ├── transformer.py
│ │ ├── ev_charger.py
│ │ ├── replay.py
│ │ ├── grid.py
│ ├── rl_agent
│ │ ├── reward.py
│ │ ├── state.py
│ ├── utilities
│ │ ├── loaders.py
│ │ ├── utils.py
│ │ ├── arg_parser.py
│ ├── example_config_files
│ │ ├── BusinessPST.yaml
│ │ ├── PublicPST.yaml
│ │ ├── V2GProfitPlusLoads.yaml
│ ├── visuals
│ │ ├── plots.py
│ │ ├── renderer.py
│ ├── scripts/
```
Class Diagram of the EV2Gym Environment:
<div align="center">
<img align="center" src="https://github.com/StavrosOrf/EV2Gym/assets/17108978/8ca5bf11-6ed4-44f6-9faf-386382609af1" width="55%"/>
</div>
## Citing EV2Gym
If you use this code in your research, please cite as:
```bibtex
@ARTICLE{10803908,
author={Orfanoudakis, Stavros and Diaz-Londono, Cesar and Emre Yılmaz, Yunus and Palensky, Peter and Vergara, Pedro P.},
journal={IEEE Transactions on Intelligent Transportation Systems},
title={EV2Gym: A Flexible V2G Simulator for EV Smart Charging Research and Benchmarking},
year={2025},
volume={26},
number={2},
pages={2410-2421},
keywords={Vehicle-to-grid;Smart charging;Optimization;Benchmark testing;Batteries;Data models;Schedules;Reinforcement learning;Prediction algorithms;Power transformers;Electric vehicle optimization;gym environment;reinforcement learning;mathematical programming;model predictive control (MPC)},
doi={10.1109/TITS.2024.3510945}}
```
## License
This project is licensed under the MIT License - see the [LICENSE.md](LICENSE) file for details.
## Contributing
EV2Gym is an open-source project and welcomes contributions! Please get in contact with us if you would like to discuss about the simulator.
Raw data
{
"_id": null,
"home_page": null,
"name": "ev2gym",
"maintainer": null,
"docs_url": null,
"requires_python": ">=3.6",
"maintainer_email": null,
"keywords": "gym, Reinforcement Learning, V2X, V2G, G2V, EVs, ev2gym, Electric Vehicles, Electric Vehicle Simulator",
"author": "Stavros Orfanoudakis",
"author_email": null,
"download_url": "https://files.pythonhosted.org/packages/92/db/aa3071ff50ba1eb4f895262c968b0a6c2f55d5519cded54c289f994265df/ev2gym-2.0.0.tar.gz",
"platform": null,
"description": "\r\n# EV2Gym: A Realistic EV-V2G-Gym Simulator for EV Smart Charging\r\n\r\n<div align=\"center\">\r\n<img align=\"center\" src=\"https://github.com/StavrosOrf/EV2Gym/assets/17108978/86e921ad-d711-4dbb-b7b9-c69dee20da11\" width=\"55%\"/>\r\n</div>\r\n\r\n[](https://www.python.org/downloads/release/python-360/) [](https://pypi.org/project/ev2gym/) \r\n---\r\n\r\nDevelop and evaluate **any type of smart charging algorithm**: from simple heuristics, Model Predictive Control, Mathematical Programming, to Reinforcement Learning!\r\n\r\nEV2Gym is **fully customizable** and easily **configurable**! Now can also **simulate the grid**, thanks to [RL-ADN](https://github.com/EnergyQuantResearch/RL-ADN)!\r\n\r\nThe EV2Gym **Paper** can be found at: [arXiv](https://arxiv.org/abs/2404.01849) and [IEEE](https://dl.acm.org/doi/abs/10.1109/TITS.2024.3510945).\r\n\r\n**Highly recommended** related works and repositories:\r\n| Title | Paper | GitHub repository |\r\n| ------------------------------------------------------------------------------------------------------------------------------ | ------------------------------------------------------------------------------------ | -------------------------------------------------------------------------------------------------- |\r\n| Physics-Informed Reinforcement Learning for Large-Scale EV Smart Charging Considering Distribution Network Voltage Constraints | [arXiv](https://arxiv.org/abs/2510.12335) | [https://github.com/StavrosOrf/EV2Gym_PI-TD3](https://github.com/StavrosOrf/EV2Gym_PI-TD3) |\r\n| GNN-DT: Graph Neural Network Enhanced Decision Transformer for Efficient Optimization in Dynamic Environments | [arXiv](https://arxiv.org/abs/2502.01778) | [https://github.com/StavrosOrf/DT4EVs](https://github.com/StavrosOrf/DT4EVs) |\r\n| Scalable Reinforcement Learning for Dynamic Electric Vehicle Charging Optimization Using Graph Neural Networks | [Paper link](https://www.nature.com/articles/s44172-025-00457-8) | [https://github.com/StavrosOrf/EV-GNN](https://github.com/StavrosOrf/EV-GNN) |\r\n| Open-source algorithms for maximizing V2G flexibility based on model predictive control | [ScienceDirect](https://www.sciencedirect.com/science/article/pii/S0378779625006704) | [https://github.com/CesarDiazLondono/MPC-G2V-V2G](https://github.com/CesarDiazLondono/MPC-G2V-V2G) |\r\n\r\n\r\n## Installation\r\n\r\nInstall the package using pip:\r\n```bash\r\npip install ev2gym\r\n```\r\n\r\n Run the example code below to get started ...\r\n```python\r\nfrom ev2gym.models.ev2gym_env import EV2Gym\r\nfrom ev2gym.baselines.mpc.V2GProfitMax import V2GProfitMaxOracle\r\nfrom ev2gym.baselines.heuristics import ChargeAsFastAsPossible\r\n\r\nconfig_file = \"ev2gym/example_config_files/V2GProfitPlusLoads.yaml\"\r\n\r\n# Initialize the environment\r\nenv = EV2Gym(config_file=config_file,\r\n save_replay=True,\r\n save_plots=True)\r\nstate, _ = env.reset()\r\nagent = V2GProfitMaxOracle(env,verbose=True) # optimal solution\r\n# or \r\nagent = ChargeAsFastAsPossible() # heuristic\r\nfor t in range(env.simulation_length):\r\n actions = agent.get_action(env) # get action from the agent/ algorithm\r\n new_state, reward, done, truncated, stats = env.step(actions) # takes action\r\n\r\n```\r\n- ### For Reinforcement Learning:\r\nTo train an RL agent, using the [StableBaselines3](https://stable-baselines3.readthedocs.io/en/master/) library, you can use the following code:\r\n```python\r\nimport gymnasium as gym\r\nfrom stable_baselines3 import PPO, A2C, DDPG, SAC, TD3\r\nfrom sb3_contrib import TQC, TRPO, ARS, RecurrentPPO\r\n\r\nfrom ev2gym.models.ev2gym_env import EV2Gym\r\n# Choose a default reward function and state function or create your own!!!\r\nfrom ev2gym.rl_agent.reward import profit_maximization, SquaredTrackingErrorReward, ProfitMax_TrPenalty_UserIncentives\r\nfrom ev2gym.rl_agent.state import V2G_profit_max, PublicPST, V2G_profit_max_loads\r\n\r\nconfig_file = \"ev2gym/example_config_files/V2GProfitPlusLoads.yaml\"\r\nenv = gym.make('EV2Gym-v1',\r\n config_file=config_file,\r\n reward_function=reward_function,\r\n state_function=state_function)\r\n# Initialize the RL agent\r\nmodel = DDPG(\"MlpPolicy\", env)\r\n# Train the agent\r\nmodel.learn(total_timesteps=1_000_000,\r\n progress_bar=True)\r\n# Evaluate the agent\r\nenv = model.get_env()\r\nobs = env.reset()\r\nstats = []\r\nfor i in range(1000):\r\n action, _states = model.predict(obs, deterministic=True)\r\n obs, reward, done, info = env.step(action)\r\n\r\n if done:\r\n stats.append(info)\r\n```\r\n!!! You can develop your own reward and state functions and use them in the environment.\r\n\r\n\r\n## Table of Contents\r\n\r\n- [Installation](#Installation)\r\n- [Overview](#Overview)\r\n- [Configuration File](#Configuration-File)\r\n- [File Structure](#File-Structure)\r\n- [Citing](#Citing-EV2Gym)\r\n- [License](#License)\r\n- [Contributing](#Contributing)\r\n\r\n<!-- Bullet points with all the benefits -->\r\n## Overview\r\n\r\n\r\n\r\n- The simulator can be used to evaluate any type of algorithm to gain insights into its efficiency.\r\n- The \u201cgym environment\u201d can readily support the development of RL algorithms.\r\n- Replays of simulations are saved and can be solved optimally using the Gurobi Solver.\r\n- Easy to incorporate additional functionality for any use-case.\r\n- Now, also simulates grid for voltage magnitude!\r\n- The number and the topology of Transformers, Charging stations, and Electric Vehicles are parameterizable.\r\n- The user can import custom data.\r\n- Uses only open-source data:\r\n - EV spawn rate, time of stay, and energy required are based on realistic probability distributions *ElaadNL* conditioned on time, day, month and year.\r\n - *Pecan Street* data is used for the load profiles.\r\n - *Renewables Ninja* data is used for the PV generation profiles.\r\n - EV and Charger characteristics are based on real EVs and chargers existing in NL (*RVO Survey*).\r\n - Charging/ Discharging prices are based on historical day-ahead prices from *ENTSO-e*.\r\n\r\nFocused on **realistic** parameters and **fully customizable**:\r\n\r\n- **Power Transformer** model:\r\n - Max Power Limit\r\n - Inflexible Loads, PV, Capacity Reduction events\r\n- **Charging Stations** model:\r\n - Min and Max charge/discharge power/ Current\r\n - Voltage and phases, AC or DC\r\n - List of connected transformers\r\n- **Electric Vehicle** model:\r\n - Connected charging station and port\r\n - Min and Max battery energy level\r\n - Time of arrival and departure\r\n - Energy at arrival/ desired energy at departure\r\n - Min and Max power levels\r\n - Charge and discharge efficiency\r\n - Constant-Current/ Constant-Voltage load-curve option\r\n- **Battery Degradation** model:\r\n - Cyclic aging\r\n - Calendar aging\r\n\r\n\r\n<div align=\"center\">\r\n<img align=\"center\" src=\"https://github.com/StavrosOrf/EV2Gym/assets/17108978/d15d258c-b454-498c-ba7f-634d858df3a6\" width=\"90%\"/>\r\n</div>\r\n\r\nAn EV2Gym simulation comprises three phases: the configuration phase, which initializes the models; the simulation phase, which spans $T$ steps, during which the state of models like EVs and charging stations is updated according to the decision-making algorithm; and finally, in the last phase, the simulator generates evaluation metrics for comparisons, produces replay files for reproducibility, and generates real-time renders for evaluation.\r\n\r\n## Configuration File\r\n\r\nThe configuration file is used to set the parameters of the simulation. The configuration file is a YAML file that contains the following parameters:\r\n```yaml\r\n# This yml file is used to configure the evsim simulation\r\n\r\n##############################################################################\r\n# Simulation Parameters\r\n##############################################################################\r\ntimescale: 15 # in minutes per step\r\nsimulation_length: 96 #90 # in steps per simulation\r\n\r\n##############################################################################\r\n# Date and Time\r\n##############################################################################\r\n# Year, month, \r\nyear: 2022 # 2015-2023\r\nmonth: 1 # 1-12\r\nday: 17 # 1-31\r\n# Whether to get a random date every time the environment is reset\r\nrandom_day: True # True or False\r\nrandom_hour: False # True or False\r\n\r\n# Simulation Starting Time\r\n# Hour and minute do not change after the environment has been reset\r\nhour: 5 # Simulation starting hour (24 hour format)\r\nminute: 0 # Simulation starting minute (0-59)\r\n\r\n# Simulate weekdays, weekends, or both\r\nsimulation_days: weekdays # weekdays, weekends, or both\r\n\r\n# EV Spawn Behavior\r\nscenario: public # public, private, or workplace\r\nspawn_multiplier: 5 # 1 is default, the higher the number the more EVs spawn. Play somewhere between 3-7. With 1 often only 1/5 EVs show up.\r\n\r\n##############################################################################\r\n# Prices\r\n##############################################################################\r\ndischarge_price_factor: 1 # how many times more abs(expensive/cheaper) it is to discharge than to charge. Are similar, discharging cost could be a bit higher.\r\n\r\n##############################################################################\r\n# Distribution Network Settings\r\n##############################################################################\r\nv2g_enabled: True # True or False\r\nnumber_of_charging_stations: 150 # Play somewhere 5-20\r\nnumber_of_transformers: -1\r\nnumber_of_ports_per_cs: 1\r\n# Provide path if you want to load a specific charging topology\r\n# *(where chargers are located and what are their characteristics),\r\n# else write None for a randomized one with the above parameters\r\ncharging_network_topology: None #./config_files/charging_topology_10.json\r\n\r\nsimulate_grid: True # True or False\r\nnetwork_info: \r\n vm_pu: 1.0 # Voltage magnitude in per unit\r\n s_base: 1000 # Base power in kVA\r\n load_multiplier: 1 # Load multiplier for the grid\r\n pv_scale: 80 # Percentage% of maximum PV power with respect to the total load of the bus\r\n bus_info_file: './ev2gym/data/network_data/node_34/Nodes_34.csv'\r\n branch_info_file: './ev2gym/data/network_data/node_34/Lines_34.csv'\r\n # bus_info_file: './ev2gym/data/network_data/node_123/Nodes_123.csv'\r\n # branch_info_file: './ev2gym/data/network_data/node_123/Lines_123.csv'\r\n\r\npf_solver: 'Laurent' # 'Laurent' or 'PandaPower'\r\n\r\n##############################################################################\r\n# Power Setpoints Settings\r\n##############################################################################\r\n# How much the power setpoints can vary in percentage compared to the nominal power\r\n# The higher the number the easier it is to meet the power setpoints, the opposite for negative numbers\r\npower_setpoint_enabled: True # True or False\r\npower_setpoint_flexiblity: 80 # (in percentage +/- %)\r\n\r\n##############################################################################\r\n# Inflexible Loads, Solar Generation, and Demand Response (Not compatible with simulate_grid = True)\r\n##############################################################################\r\n# Whether to include inflexible loads in the transformer power limit, such as residential loads\r\ntr_seed: -1 # seed for the random number generator of transformer loads(-1 for random seed)\r\n\r\ninflexible_loads: # Offices, homes\r\n include: False # True or False\r\n inflexible_loads_capacity_multiplier_mean: 1 # 1 is default, the higher the number the more inflexible loads\r\n forecast_mean: 30 # in percentage of load at time t%\r\n forecast_std: 5 # in percentage of load at time t%\r\n\r\n# PV solar Power\r\nsolar_power:\r\n include: False # True or False\r\n solar_power_capacity_multiplier_mean: 1 # 1 is default, the higher the number the more solar power\r\n forecast_mean: 20 # in percentage of load at time t%\r\n forecast_std: 5 # in percentage of load at time t%\r\n\r\n# Whether to include demand response in the transformer power limit\r\ndemand_response:\r\n include: False # True or False\r\n events_per_day: 1\r\n #How much of the transformer power limit can be used for demand response\r\n event_capacity_percentage_mean: 35 # (in percentage +/- %) reduction\r\n event_capacity_percentage_std: 5 # (in percentage +/- %)\r\n event_length_minutes_min: 60\r\n event_length_minutes_max: 60\r\n event_start_hour_mean: 12\r\n event_start_hour_std: 2\r\n # How many minutes ahead we know the event is going to happen\r\n notification_of_event_minutes: 60\r\n\r\n##############################################################################\r\n# EV Specifications\r\n##############################################################################\r\nheterogeneous_ev_specs: False #if False, each EV has the same specifications\r\n# such as battery capacity, charging rate, etc.\r\nev_specs_file: ./ev2gym/data/ev_specs_v2g_enabled2024.json # path to the file with the EV specifications\r\n\r\n##############################################################################\r\n# Default Model values\r\n##############################################################################\r\n# These values are used if not using a charging network topology file or \r\n# if the EV specifications are not provided\r\n\r\n# Default Transformer model\r\ntransformer:\r\n max_power: 200 # in kW\r\n\r\n# Default Charging Station model\r\ncharging_station: \r\n min_charge_current: 0 # Amperes\r\n max_charge_current: 32 # Amperes\r\n min_discharge_current: 0 # Amperes (actual value <=0)\r\n max_discharge_current: -32 # Amperes (actual value <=0)\r\n voltage: 400 # Volts\r\n phases: 3 # 1,2, or 3\r\n\r\n# Default EV model\r\nev:\r\n #The following values are used if \"heterogeneous_ev_specs = False\"\r\n battery_capacity: 70 # in kWh\r\n max_ac_charge_power: 22 # in kW\r\n min_ac_charge_power: 0 # in kW\r\n max_dc_charge_power: 50 # in kW\r\n max_discharge_power: -22 # in kW\r\n min_discharge_power: 0 # in kW\r\n ev_phases: 3 \r\n charge_efficiency: 1 # 0-1 (0% - 100%)\r\n discharge_efficiency: 1 # 0-1 (0% - 100%)\r\n transition_soc: 1 # 0-1 (0% - 100%)\r\n\r\n #The following values are also used if \"heterogeneous_ev_specs = True\"\r\n min_battery_capacity: 15 # in kWh\r\n min_time_of_stay: 200 # in minutes\r\n min_emergency_battery_capacity: 15 # in kWh\r\n desired_capacity: 1 # in (0-1) (0% - 100%) #Keep at 100% for now\r\n #if trasition_soc is < 1, the curve of the line is affected by:\r\n transition_soc_multiplier: 50 # default 1 (the higher the number the shorter the effect of CCCV region)\r\n```\r\n\r\n## File Structure\r\nThe file structure of the EV2Gym package is as follows:\r\n```bash\r\n\u251c\u2500\u2500 ev2gym\r\n\u2502 \u251c\u2500\u2500 baselines\r\n\u2502 \u2502 \u251c\u2500\u2500 gurobi_models/\r\n\u2502 \u2502 \u251c\u2500\u2500 mpc/\r\n\u2502 \u2502 \u251c\u2500\u2500 heuristics.py\r\n\u2502 \u251c\u2500\u2500 data/\r\n\u2502 \u251c\u2500\u2500 models\r\n\u2502 \u2502 \u251c\u2500\u2500 ev2gym_env.py\r\n\u2502 \u2502 \u251c\u2500\u2500 ev.py\r\n\u2502 \u2502 \u251c\u2500\u2500 transformer.py\r\n\u2502 \u2502 \u251c\u2500\u2500 ev_charger.py\r\n\u2502 \u2502 \u251c\u2500\u2500 replay.py\r\n\u2502 \u2502 \u251c\u2500\u2500 grid.py\r\n\u2502 \u251c\u2500\u2500 rl_agent\r\n\u2502 \u2502 \u251c\u2500\u2500 reward.py\r\n\u2502 \u2502 \u251c\u2500\u2500 state.py\r\n\u2502 \u251c\u2500\u2500 utilities\r\n\u2502 \u2502 \u251c\u2500\u2500 loaders.py\r\n\u2502 \u2502 \u251c\u2500\u2500 utils.py\r\n\u2502 \u2502 \u251c\u2500\u2500 arg_parser.py\r\n\u2502 \u251c\u2500\u2500 example_config_files\r\n\u2502 \u2502 \u251c\u2500\u2500 BusinessPST.yaml\r\n\u2502 \u2502 \u251c\u2500\u2500 PublicPST.yaml\r\n\u2502 \u2502 \u251c\u2500\u2500 V2GProfitPlusLoads.yaml\r\n\u2502 \u251c\u2500\u2500 visuals\r\n\u2502 \u2502 \u251c\u2500\u2500 plots.py\r\n\u2502 \u2502 \u251c\u2500\u2500 renderer.py\r\n\u2502 \u251c\u2500\u2500 scripts/\r\n```\r\n\r\nClass Diagram of the EV2Gym Environment:\r\n<div align=\"center\">\r\n<img align=\"center\" src=\"https://github.com/StavrosOrf/EV2Gym/assets/17108978/8ca5bf11-6ed4-44f6-9faf-386382609af1\" width=\"55%\"/>\r\n</div>\r\n\r\n## Citing EV2Gym\r\n\r\nIf you use this code in your research, please cite as:\r\n```bibtex\r\n@ARTICLE{10803908,\r\n author={Orfanoudakis, Stavros and Diaz-Londono, Cesar and Emre Y\u0131lmaz, Yunus and Palensky, Peter and Vergara, Pedro P.},\r\n journal={IEEE Transactions on Intelligent Transportation Systems}, \r\n title={EV2Gym: A Flexible V2G Simulator for EV Smart Charging Research and Benchmarking}, \r\n year={2025},\r\n volume={26},\r\n number={2},\r\n pages={2410-2421},\r\n keywords={Vehicle-to-grid;Smart charging;Optimization;Benchmark testing;Batteries;Data models;Schedules;Reinforcement learning;Prediction algorithms;Power transformers;Electric vehicle optimization;gym environment;reinforcement learning;mathematical programming;model predictive control (MPC)},\r\n doi={10.1109/TITS.2024.3510945}}\r\n```\r\n\r\n## License\r\n\r\nThis project is licensed under the MIT License - see the [LICENSE.md](LICENSE) file for details.\r\n\r\n\r\n## Contributing\r\n\r\nEV2Gym is an open-source project and welcomes contributions! Please get in contact with us if you would like to discuss about the simulator.\r\n\r\n\r\n",
"bugtrack_url": null,
"license": null,
"summary": "A realistic V2G simulator environment",
"version": "2.0.0",
"project_urls": null,
"split_keywords": [
"gym",
" reinforcement learning",
" v2x",
" v2g",
" g2v",
" evs",
" ev2gym",
" electric vehicles",
" electric vehicle simulator"
],
"urls": [
{
"comment_text": null,
"digests": {
"blake2b_256": "b009d335b2c7952952284ec557c36a1fb554416ce5b231d9bf34fa134ddaef63",
"md5": "9b2c85491426439a2df73012e9beae59",
"sha256": "b1f22280d9ae9da93bb26031730412d952845014cd2718f92bc030b78cddcb58"
},
"downloads": -1,
"filename": "ev2gym-2.0.0-py3-none-any.whl",
"has_sig": false,
"md5_digest": "9b2c85491426439a2df73012e9beae59",
"packagetype": "bdist_wheel",
"python_version": "py3",
"requires_python": ">=3.6",
"size": 2966176,
"upload_time": "2025-10-17T22:17:15",
"upload_time_iso_8601": "2025-10-17T22:17:15.833177Z",
"url": "https://files.pythonhosted.org/packages/b0/09/d335b2c7952952284ec557c36a1fb554416ce5b231d9bf34fa134ddaef63/ev2gym-2.0.0-py3-none-any.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": null,
"digests": {
"blake2b_256": "92dbaa3071ff50ba1eb4f895262c968b0a6c2f55d5519cded54c289f994265df",
"md5": "b504338ae89ac8a141e04b42735b29fb",
"sha256": "5a37381ad4f7f32023869c651785df80b14a078e34c7bd67c2aec16df4c911b1"
},
"downloads": -1,
"filename": "ev2gym-2.0.0.tar.gz",
"has_sig": false,
"md5_digest": "b504338ae89ac8a141e04b42735b29fb",
"packagetype": "sdist",
"python_version": "source",
"requires_python": ">=3.6",
"size": 2865324,
"upload_time": "2025-10-17T22:17:18",
"upload_time_iso_8601": "2025-10-17T22:17:18.088078Z",
"url": "https://files.pythonhosted.org/packages/92/db/aa3071ff50ba1eb4f895262c968b0a6c2f55d5519cded54c289f994265df/ev2gym-2.0.0.tar.gz",
"yanked": false,
"yanked_reason": null
}
],
"upload_time": "2025-10-17 22:17:18",
"github": false,
"gitlab": false,
"bitbucket": false,
"codeberg": false,
"lcname": "ev2gym"
}
JSON