This is a home-assistant custom component - place the custom_components folder in your configuration directory (or add its contents to an existing custom_components folder).
This component provides a simulated sensor that generates a time-varying signal V(t) given by the function:
V(t) = M + A sin((2 pi (t - t_0) / w) + P) + N(s)
where:
- M = the mean value, default 0
- A = the amplitude of the sine wave, default 1
- t = the time when a value is generated
- t_0 = the time when the sensor is started
- w = the time period for a single cycle of the sine wave, default value 60 seconds
- P = the phase offset to add, default value 0 degrees
- N(s) = the random Gaussian noise with spread s, defaults value 0
A simulated sensor with default values can be added to home-assistant and displayed using a history_graph using the following config:
sensor:
- platform: simulated
history_graph:
simulated_sensor_graph:
name: 'simulated'
hours_to_show: 1
entities:
- sensor.simulated
To give an example of simulating real world data, we can create a simulated relative humidity sensor (in %) using the config below:
- platform: simulated
name: 'simulated relative humidity'
unit: '%'
amplitude: 0 # Turns off sine wave
mean: 50
spread: 10
seed: 50
Here the sine wave contribution has been set to zero by configuring an amplitude of 0, and a Gaussian noise signal with spread 10 is generated about a mean value of 50. A graph of the data generated is below. Bear in mind that due to statistics, approximately 30% of values are outside of the configured spread of 10.
Configuration variables:
- name string (Optional): The name of the sensor
- unit string (Optional): The unit to apply
- amplitude number (Optional): The amplitude of the periodic components
- mean number (Optional): The mean level
- period number (Optional): The number of seconds for the period
- phase number (Optional): The phase offset (in degrees) to apply to the periodic component
- spread number (Optional): The spread (FWHM) of the Gaussian noise
- seed number (Optional): The seed value for the random component