Discussion: A more precise definition of time series in H2Integrate

When H2Integrate is run for an annual simulation with hourly time steps, the operational time series information for many of the subsystems is given as 8760-length array. In many cases, one can make a reasonable assumption about what an element of that array represents, but in some cases it is not so obvious and I think it would be valuable to have a common understanding for future model development and system level control efforts.

Examples:

• If I take the first element of the solar electricity generation array and find a value of 10 kW (solar_electricity_out[0] = 10), then it feels like a safe assumption to say that the solar generation subsystem was producing 10 kW at 12:00 AM, Jan. 1 and continued to produce an average of 10 kW until 12:59 AM, Jan. 1.
• Similarly, if I look at the battery charging and see battery_charge[0] = 10, then I will assume the battery was charging itself with 10 kW at 12:00 AM, Jan. 1, and continued to do so until 12:59 AM, Jan. 1.
• Now, if I look at the battery state of charge, it isn't so obvious. The stored energy in the battery is essentially the integral of charging over time so what does battery_soc[0] represent? Is it
  • the soc at 12:00 AM, Jan. 1 (before the effect of battery_charge[0])
  • the soc at 12:59 AM, Jan. 1 (including the effect of battery_charge[0])
  • some average value during that time period?

No files, tests, or entry points are named. Start by tracing how hourly operational arrays and battery state of charge are produced for an annual simulation, then document and agree on a common time-series convention; done means the timing semantics for quantities such as generation, charging, and SOC are explicitly defined.