This monitoring system can be used for gathering and visualization of car consumption data, especially for hybrid cars.
The intent of the project was to verify supplier information on electricity and fuel consumption under every-day conditions and to monitor realistic costs of electric versus gas engine.
The system uses two sources of input:
- The car charging system, which is, in its simplest version, a metered power outlet from which charging power over time is obtained.
- Car and trip data from the car manufacturer's cloud services, which needs to provide information on trip start time, trip duration, mileage, electric energy consumption as well as fuel consumption. Furthermore, fuel as well as battery fill level are also expected.
From these data, the monitor calculates and aggregates various consumption KPIs and allows tracking these over an arbitrary period of time.
A number of prerequisites are required for this system which are graphically shown in the Architecture Overview. This documentation assumes a specific configuration, which will be detailed below, but the general approach is not limited to this and can be extended to other configurations.
Monitoring the charging process requires a charging system with an interface through which access to the current charging power is available.
In the current setup, a Volkswagen VW Passat GTE (model 2018) is exclusively charged through a normal power outlet. A smart plug Fritz!DECT210 is used in connection with a Fritz!Box to track the charging power over time.
External charging without automatic access to charging data is currently not considered but could be included using the available data structures.
Most car manufacturers provide access to car data through specific cloud services.
The current setup for Volkswagen uses access through WE Connect for which a registration is required. (The above link is for Germany. Similar links are available for other countries)
The server hosts installations of the time series database InfluxDB as well as the analytics and monitoring solution Grafana. For both tools, cloud solutions as well as local docker or native installations are available.
In the current setup, both tools run in docker containers on a Synology DS220+
The data collection system hosts observer services for car charging as well as for car live data.
In the current setup, these are implemented in Python and run on a Rspberry Pi 4