Off-grid Calculator
Pick a location
0-90
Each dot indicates the energy storage level on that day (1/1-2005 to 12/31-2015)
100%
75%
50%
25%
Jan
Feb
Mar
Apr
May
Jun
Jul
Aug
Sep
Oct
Nov
Dec
Annual average
Energy Deficit
kWh
Empty Storage
days
Energy Surplus
kWh
Full Storage
days
Based on solar radiation acquired via PVGIS. See below remarks for further details
m2
%
W/°C
kWh
The math behind it all
The simulation is built on an extremely simple model with only a handful of inputs.
It is based on solar radiation data acquired via PVGIS and fetched through a Cloudflare Worker. See the PVGIS users manual for details on the used databases. PVGIS gets data on the Americas from NSRDB as collected by NREL
Each retrieved dataset consists of 96,408 datapoints covering the hours between 1/1-2005 and 12/31-2015. After retrieval, the data is remapped into 24 hour intervals. Each datapoint has two values for that 24 hour period:
- The total solar insolation (based on global irradiance) at the chosen angle at the chosen location
- The average outdoor temperature at a height of 2 meters
The two values are used to calculate the energy-gain and energy-loss for that day.
Energygain
Wh
=
Collector‑area
m2
×
Total‑solar‑insolation
Wh
m2
×
Collector‑efficiency
Energyloss
Wh
=
Whole‑house‑heat‑transfer‑coefficient
W
°C
×
(
Average‑outdoor‑temperature
−
Indoor‑temperature
)
°C
×
24
s Wh
J
The indoor temperature is set at 22 °C
At the simulation outset, the energy-storage is full. The net-gain (or loss) for each day is then added to the storage, either increasing or decreasing its level. The storage level is not allowed to exceed the chosen maximum or fall below 0 kWh
Storagetomorrow
Wh
=
Energygain
Wh
+
Energyloss
Wh
+
Storagetoday
Wh
Due to the model's simplicity there's naturally a long list of factors which it does not account for.
To name just a few:
- Energy losses from the storage tank or battery
- Conversion losses between collector and tank or battery
- Solar heat gain through windows
- Internal gains not powered by the collectors
As the model is based on historical solar and weather data without any additional statistical treatment it cannot make any predicitions on future conditions
Github