Getting Started

The main interface for the Tyba Python Client is the Client class:

import os
from tyba_client.client import Client
PAT = os.environ["TYBA_PAT"]
client = Client(PAT)

As you can see in the above example, you must have a Tyba PAT (Personal Access Token) in order to use the Client. The Authentication section below explains how to store your PAT in your local environment. With the PAT stored and a client instance instantiated, we can easily simulate projects and pull energy price data.

Run a Simulation

1. Make the necessary imports:

   from generation_models import PVGenerationModel, PVSystemDesign, SingleAxisTracking
   

2. Define a simple model:

   model = PVGenerationModel(
      solar_resource=(40.6685396, -73.9345585),
      inverter="Delta Electronics: M80U_XXX [480V]",
      pv_module="Boviet Solar Technology Co._ Ltd. BVM6612M-315",
      system_design=PVSystemDesign(
          dc_capacity=2000.0,
          ac_capacity=1500.0,
          poi_limit=15000.0,
          gcr=0.38,
          tracking=SingleAxisTracking(
              rotation_limit=45.0,
              backtrack=True
          )
      ),
     project_term=1
   )
   

3. Schedule the model run and get the run id:

   res = client.schedule(model)
   id_ = res.json()["id"]
   print(res.text)
   

   You will see an id print, confirming that the run has been scheduled.

4. Wait to receive the results:

   results = client.wait_on_result(id_)
   

5. View and explore the results

   # Results is a dictionary. To review keys:
   print(results.keys())
   
   # Review System Power 8760
   results['system_power']
   

Pull Historical Price Data

1. Get a list of all the independent system operators (ISOs) represented in the dataset

   print(client.services.get_all_isos().json())
   

2. Get a list of all of the Local Market Price (LMP) nodes for a given ISO

   price_nodes = client.lmp.get_all_nodes(iso='CAISO').json()
   node_directory = pd.DataFrame(price_nodes)
   print(node_directory.head())
   

3. Get a pandas Series of LMP prices based on node id

   res = client.lmp.get_prices(
             node_ids=['20000004677'],
             market=Market.RT,
             start_year=2024,
             end_year=2024
   ).json()
   res = res['20000004677']
   df = pd.Series(res['prices'], index=pd.to_datetime(res['datetimes'])).tz_localize(None)
   print(node_directory.head())
   

Pull a Price Forecast

1. Define a time range of interest

   tz = pytz.timezone("US/Central") # Can be any timezone but results will always be returned in the node-local timezone
   start_time = tz.localize(datetime(2023, 10, 10)) # localized datetime is required to avoid ambiguity
   end_time = tz.localize(datetime(2023, 10, 17)) # localized datetime is required to avoid ambiguity
   

2. Pull the most recent forecast for an LMP node based on node name

   node_name = "HB_HOUSTON"  # Use client.lmp.get_all_nodes from the previous section to find your node name
   da_forecasts = forecast.most_recent(node_name, "da", start_time, end_time)
   # Results are a list of dicts. To see the first dict:
   print(da_forecasts[0])
   

Next…

Feel free to dig deeper with the the Project Simulation Service, Historical Price Service, and Price Forecast Service pages, which discuss the usage of their respective features in greater detail

Authentication

As mentioned above, you must have a Tyba PAT to use the Client. If you do not have or know your PAT, please contact us.

Mac or Linux instructions

Shell Environment (preferred)

Save your Tyba PAT in the shell’s environment:

1. Add TYBA_PAT="your-pat-in-quotes" to a new line in ~/.bashrc

2. To check, open a new shell/terminal and run env and confirm TYBA_PAT is present and properly set

This approach is preferred even if you are running a Jupyter notebook because it keeps the PAT from being exposed if the notebook/script is shared

Jupyter Notebook

Save your Tyba PAT in the notebook’s environment:

1. In a notebook cell, enter the following:

   %env TYBA_PAT=your_pat_without_quotes
   

2. The cell will return your key with “env: TYBA_PAT = your_pat_without_quotes”

3. To keep the PAT secure, delete the previous cell. You will need to repeat this process each time you open the notebook

Windows Instructions

Shell Environment

On a Windows machine, the simplest (but less secure) approach is to hardcode the PAT directly into your script, i.e. replace PAT = os.environ["TYBA_PAT"] with PAT="your-pat-in-quotes"

Jupyter Notebook (preferred)

See the Jupyter notebook section above.