Improve readme, add screenshots and application examples

[jpulakka]

[jpulakka]

Application examples:

• FIR vs. nonlinear filters, when to use (plot each to compare)
• Using sigmoid function to scale output of FIR

[drogfild]

README.md installation instructions at "Option 1: HACS" lacks sensor configuration.
Maybe it could be:

• Option 1
• Option 2
• Common for both options: Sensor installation

[jpulakka]

README.md installation instructions at "Option 1: HACS" lacks sensor configuration. Maybe it could be:

* Option 1

* Option 2

* Common for both options: Sensor installation

@drogfild do you mean that after using HACS, installation should proceed from what is currently Option 2 point 4, modifying configuration.yaml?

@robsonke kindly added HACS support, but I don't use HACS so I'm not quite up to date what exactly it does and doesn't do.

[robsonke]

It's nothing special, it's just an automated way of checking out git repos of plugins and keeping them up to date in case you release a new version.

[jpulakka]

Ok, modified README, does it make sense now?

[overas]

Thanks for the addon. I have some feature request for the documention after using nordpool_diff two weeks.

The output is most suitable for fine-tuning continuously adjustable things (thermostats), or it can be thresholded to control binary things that can be switched on/off anytime, such as water heaters.

It has been nice if you could publish example automation for how to use the different sensors for adjusting termostats for heating floor and heating pump.

I use now a threshold sensor "binary_sensor.dyr_strom" with the "sensor.nordpool_diff_triangle_10" which are on if the value is above 0 and off when the value is below 0. I use the threshold sensor in a automation for heat or cool the heating cables or heat pump. Sometimes the sensor is off to long when the price is high in the daytime. How can I adjust the time the sensor is off?

`alias: Thermostat bad1etg
description: ""
trigger:

• platform: state
  entity_id:
  • group.somebody_home
  • binary_sensor.billig_strom
  • schedule.thermostat_schedule
  • binary_sensor.regulator_padrag_under_3
    condition: []
    action:
• if:
  • condition: state
    entity_id: group.somebody_home
    state: home
  • condition: state
    entity_id: binary_sensor.dyr_strom
    state: "off"
  • condition: state
    entity_id: schedule.thermostat_schedule
    state: "on"
  • condition: state
    entity_id: binary_sensor.regulator_padrag_under_3
    state: "off"
    then:
  • service: climate.set_temperature
    data:
    temperature: 23
    hvac_mode: heat
    target:
    entity_id: climate.varmekabel_bad1etg
    else:
  • service: climate.set_temperature
    data:
    temperature: 20
    hvac_mode: heat
    target:
    entity_id: climate.varmekabel_bad1etg
    mode: single`

[jpulakka]

You're welcome @overas !

Thresholding at 0 is reasonable way to control binary things. However, as written in README,

there are no guarantees about how many hours per day the output will stay above some threshold, even if typical price variations may make the output typically behave this or that way most of the time.

Setting the threshold to lower (slightly negative) value, or trying out rank or interval filters may help a bit, but so far there's no perfect way to work around

Sometimes the sensor is off to long when the price is high in the daytime.

because the output is just price_of_electricity(t) fed through a linear filter. So price_of_electricity directly affects the output, and it can vary in arbitrary ways even though it often follows more or less similar shape over 24h.

BTW if you try thresholding the linear filters at nonzero boundary, it's a good idea to use some form of normalize.

Personally I'm using nordpool_diff to control target temperature of an air heat pump and I think it works very well there. I'll publish an example of that automation soon when I have time. Meanwhile, I'm using settings

- platform: nordpool_diff
  filter_length: 15
  filter_type: triangle
  normalize: max_min_sqrt_max