pimatic-smartmeter-stats

Creating statistics from smartmeter values. This plugin contains 2 devices.

The SmartmeterDegreedaysDevice uses outdoor temperature and windspeed data and heating energy consumption data (gas, etc) to create degreeday and efficiency values. They are a measure of your heating energy effciency. You can use them to compare your own energy efficiency and what for example isolation measures or lowering the indoor temparture will do.
The SmartmeterStatsDevice reads any smartmeter value that increases over time. The generated values are showing the change of the value over the last hour, day, week or month.

Installation

To enable the smartmeter-stats plugin add this to the plugins section via the GUI or add it in the config.json file.

...
{
  "plugin": "Smartmeter-stats"
}
...

Degreedays

Degreedays is a measure for heating and cooling usage. The so called Heating degreedays are typical indicators of household energy consumption for heating. This plugin works only for heating degreedays.

The degreeday calculation used in the plugin is based on:

BaseTemprature: is the outdoor reference temperature for degreeday days, the default is 18°C and can be changed.
Temperature: the outdoor and indoor temperature
Windspeed: the windspeed in m/s
Month of the year (a factor depending on the month)

To use the degreedays device you need to have the following data variables available in Pimatic:

Realtime outdoor temperature
Realtime energy total value (not the current usage but the absolute total usage value).

If available the following variables will increase the quality of the data:

Realtime windspeed
Realtime indoor temperature of the main heated room (not yet used)

Realtime means the value should be updated at least once an hour.

Degreedays device

When the plugin is installed (including restart) a SmartmeterDegreedays device can be added. Below the settings with the default values.

{
  "id": "<smartmeter-degreedays-id>",
  "class": "SmartmeterDegreedaysDevice",
  "temperature": "<name>", // The variable that hold the outdoor temperature
  "energy": "<variable>" // The variable that holds the actual total energy value. This can be gas or electricity.
  "wind": <variable> // The variable that hold the windspeed value (optional)
  "temperatureIn": "<variable>", // The variable that hold the indoor temperature (optional)
  "baseTemperature": 18°C // The temperature the heating degreedays calculation is based upon
  "energyUnit": "" // The energy unit in the frontend
  "energyLabel": "" // A custom label for energy to use in the frontend
  "energyAcronym": "" // Acronym for energy to show as value label in the frontend
  "log": true | false // Enable to have daily log of the Degreeday values and get data for regression calculation
  "test": boolean // Enable to speedup the daily proces to minutes (for testing)
}

Configuration

The statistics are calculated on a daily bases starting at midnight. The statistics variables are available per hour or per day.

Daily (update at start of day):

status (init, processing 1st day or yesterday)
temperatureDay - yesterdays average outdoor temperature
temperatureInDay - yesterdays average outdoor temperature (optional)
energyDay - yesterdays energy consumption
windspeedDay - yesterdays average windspeed (optional)
degreedaysDay - yesterdays degreeday values (00:00-23:59)
efficiencyDay - yesterdays efficiency factor: (energy consumption) / degreeday
r2 - linear regression fit-factor for the current baseTemperature
baseTemperature - baseTemperature from config
calcTemperature - starting from minimum 10 days data a optimal baseTemperature is calculated and will become visible in the GUI

The plugin needs max 2 days to get aligned for all variables except r2 and the calcTemperature. These values needs a minimum of 10 days data. On init of the plugin during the first day the status is 'init'. During the first full day the status is 'processing 1st day'. And starting the 2nd day the status is the date of yesterday and the values are from yesterday and accurate.

For longterm usage of the values, a log is used. The log will add at the start of every day the values for that passed day. To make the log standalone usable, a timestamp is added.

The daily data is added as a JSON record. The logfile is made compact and readable with one daily data row per day. The logfile is available in a directory called 'smartmeter-data', located in the pimatic home directory of the computer running Pimatic (mostly pimatic-app). The log will have the name 'device-name'-data.json.

The baseTemperature is default set at 18°C. To improve the quality of the baseTemperature a regression function is added. After a minimum of 10 days data in the log, when every day the optimum for the BaseTemperature is calculated based on regression calculation on the dataset of the daily degreedays and energy consumption. The linear regression algorithm searches for the baseTemperature with the best "Goodness-of-Fit". That value is show in the Gui (config baseTemp/calculated baseTemp ). After a while you can choose to change the baseTemperature value in the config to the calculated optimal value. You can adjust the baseTemp by creating a variable called 'baseTempInput'. Via a variablesInput device to can play with the baseTemp and see what is does for the r2 factor; what a more realistic baseTemp is for your house. This is also done automatically with the result in the calcTemp variable.

You can reset the device and set the values to 0 with the command "reset 'device name'". This command can be used in rules, so you can create a button to reset the values or reset on any other condition/event. The current log and variable data is saved to a backup file in the same directory.

Stats device

When the plugin is installed (including restart) a SmartmeterStats device can be added. Below the settings with the default values.

{
  "id": "<smartmeter-stats-id>",
  "class": "SmartmeterStatsDevice",
  "input": "<name>", // The generated name for the input value (is result of expression)
  "expression": "....." // The input variable or expression string
  "unit": "" // The unit for the values, examples kWh, m3, etc
  "statistics": ["minute", hour", "day", "week", "month"] // The timescale and name of the resulting variable
  "log": false | true // Enable to have a daily log of the Stats values
  "test": boolean
}

Configuration

The initial device exposes no variables in the GUI. You can add them in the device configuration. The available statistics variables are:

actual - the actual input value. This is the result of the expression and will be updated if one of the expression variables changes.
minute - the change of the expression input value in the last minute
hour - the change of the expression input value in the last hour
day - the change of the input value in the last day (00:00-23:59)
week - the change of the input value in the last week (monday-sunday)
month - the change of the input value in the last month

On init of the plugin the first readout is based on the time left till the next full hour, day(00:00), week (monday) or month (1ste).

For longterm usage of the values, a log can be enabled. The log will add at the start of a new day (selected in config) the values for that previous day. To make the log standalone usable, a timestamp is added. The timestamp is de date of the previous day.

You can reset the device and set the values to 0 with the command reset 'device name'. This command can be used in rules, so you can create a button to reset the values or reset on any other condition/event.

The plugin is in development. Please backup Pimatic before you are using it!