Skip to content

Latest commit

 

History

History
64 lines (51 loc) · 2.76 KB

README.md

File metadata and controls

64 lines (51 loc) · 2.76 KB
Raw

SmartDSO

Documentation

The input set is as follows:

  • london_data - if actual data from london measurements should be used (see here); otherwise, standard load profiles are used
  • ev_ratio - how many of the deployed cars are of type ev, other cars will be simulated but do not consume electricity
  • pv_ratio - how many houses, which can have a PV have a PV in the simulation
  • number_consumers - option to limit the simulated consumers (0 means unlimited)
  • scenario - one of:
    1. PlugInCap
    2. MaxPvCap
    3. MaxPvSoc
    4. PlugInInf
  • scenario_name -> if it contains Flat, Flat prices are used, else Spot -> electricity price only
    • Flat: single average of the Spot Price from the simulation interval
    • Spot: hourly spot price from simulation interval
  • iteration - use the iteration index as a random seed to get different reproducible runs
  • sub_grid - calculates a different part of the grid -> used for parallelization of calculating the different grids
  • database_uri - URI of the database where the results are stored

The grid fee curve is used to calculate the grid fees which are added to the electricity price to build the final consumer price. The consumers then decide wether they need to charge, based on the current SoC (state of charge).

The cars are connected to the same simulation in each iteration, as well as scenario. Yet the household can randomly have a PV or does not have a pv. Furthermore, the car properties are assigned randomly too.

The Car name is for example S7C183_4.8_44.1 meaning:

  • S7 -> Subgrid 7
  • C183 -> Car ID 183 (fixed)
  • 4.8 -> the assigned household has 4.8kwp PV
  • 44.1 -> the battery of the EV Car has a capacity of 44.1 kWh

fixed Simulation properties

  • irradiation
  • households
  • grid properties
  • spot market price
  • Car ID

variable Simulation properties

  • Car properties
  • household PV
  • mobility behavior of the car

Scientific Paper - RMIT

'PlugInCap-PV25-PriceFlat-L', 'MaxPvCap-PV25-PriceFlat-L', 'MaxPvCap-PV80-PriceSpot-L', 'MaxPvSoc-PV80-PriceSpot-L']

  • Different Pv Scenarios are evaluated:
    1. tariff Flat+PlugInCap -> Case 1 - Status Quo (PlugInCap-PV25-PriceFlat)
    2. tariff Flat+MaxPvCap -> Case 2 - own consumption optimization (MaxPvCap-PV25-PriceFlat)
    3. tariff Spot+MaxPvCap -> Case 3 - own consumption optimization with marktsignal (MaxPvCap-PV80-PriceSpot)
    4. tariff Spot+MaxPvSoc -> Case 4 - Nutzenfunktion auf Basis des Füllstandsniveaus (MaxPvSoc-PV80-PriceSpot)

Run Simulation

  1. to start the simulation with docker run the command: sudo chmod -R o+rw . in sim_result directory.
  2. run python scenario_generator.py --case b --slp to generate the docker-compose.yml
  3. docker-compose up -d