Bachibouzouk
commented
2 years ago @MArpogaus, @FredericHirschmueller, @c-moeller this is the summary of what needs to be done in order to implement the generation of load profile. @MArpogaus, feel free to comment on or ask questions :)
This issue's aim is to describe how such a feature could be implemented following our discussion with @MArpogaus
The (python) program which would generate a load profile should take its inputs from a dict (or json format) and return its output in json format. It would be good if the program is installable as a package.
This package will then be installed inside a docker container (following rl-institut/wam-server-api structure) with task queue and workers assigned.
That way once the service is deployed, not only open-plan.rl-institut.de can access the service and one could serve other apps.
The communication between open_plan online tool and the load profiles generator will happen via http requests
In open_plan online tool we would create a new page to gather the necessary input parameters from the users, then send a http request to the loadprofile server with those parameters and expect a timeseries in return, which we would save in a database for the user to then assign to demand assets in their scenarios.
The input parameters should be described with the following fields
One input parameter should be the wished length of the timeseries (default automatic with number of timesteps)
The format of the returned timeseries should be simply an array or floats and the unit of the timeseries