Oops. Forgot to as GST.
Still doesn’t match if I add an extra 10%
DMO= 19.14*1.1 = 21.054 (inc GST)
Green: 0.00-0.21 (< DMO)
Yellow: 0.21-0.23 (< DMO*1.1)
Orange: 0.23-0.25 (< DMO*1.2)
Red: 0.25-3.00 (< 3.00)
In the screen shot above I have $0.20 (should be green) and $0.27 (should be red) both showing as Yellow.Let me check what is going on in our code base. I’ll get back to you shortly.
Worked well this morning to time 38 kWh of EV charging during the low price window between 7 - 11 AM. (Yes I would get a lot of benefit from a larger solar system, which is in the works.)
Quick update - I’ve just pushed price descriptors (Negative, Very Low, Low, Neutral, High, Spike) to the API.
I’ll update the component at some point in the next few days, so hopefully it’ll be available in the next release.
You’ll be able to use these to match the app exactly (I’ll update the blueprint once the new component is live)
1 Like
Update just got accepted into core, should hit next release
1 Like
Update, I’m still using these rules based scripts, but have also upgraded to EMHASS: An Energy Management for Home Assistant, which now considers; forecasts for PV, household load, Amber buy and sell costs and is optimising the scheduling of my battery cycles, EV charging, pool pumps and heaters.
Here is today’s plan which is forecasting upto $190 credits due to the price spike tonight, by cycling the battery. Discharging now, charge during the day and discharge during tonight’s price spike.
I run the optimisation every minute to ensure any changes in the Amber Electric (Australia) Custom Component prices are factored into my optimisation immediately.
I find EMHASS produces an almost identical plan to Amber SmartShift, with the added bonus that EMHASS can control my deferrable loads.
1 Like
I do not have a battery (yet!) and originally this automation was only designed to sink solar into the heated floor. Now, I also use it to control the upstairs A/C as well as a power point in the garage that charges my electric scooter. Even so, I continue to use the cost of the heated floor being turned on as an indicator to trigger my energy efficiency automations. A recently installed heat pump hot water system will be the next appliance to connect to this automation but currently it comes on around 10am each day, which actually works fine.
My floor uses 3kW and so I’ve continued to use that value in the automation below along with Amber FiT and General Use prices to calculate how much the floor actually costs to turn on given my household consumption and production. I decide the upper limit per hour I’m willing to spend (or forego in lost feed-in credits) and store this value in a input helper:
input_number.high_power_cost_rate_threshold
This value has stayed at 30c / hour for the last few months, but can adjust it easily using this slider:
Here is my central helper function which I use to trigger all my energy management automations:
- sensor:
- name: "High Power Cost Trigger"
unique_id: "high_power_cost_trigger"
unit_of_measurement: ""
state: >
{% set power_before = ((states('sensor.net_power') | int) / 1000) | float %}
{% set power_after = (((states('sensor.net_power') | int) - 3000) / 1000) | float %}
{% set general_price = states('sensor.melford_drive_general_price') | float %}
{% set feed_in_price = states('sensor.melford_drive_feed_in_price') | float %}
{% set cost_before = states('sensor.power_cost_rate') | float %}
{% if (power_after > 0 ) %}
{% set cost_after = (-power_after * feed_in_price) | float %}
{% else %}
{% set cost_after = (-power_after * general_price) | float %}
{% endif %}
{% if cost_after - cost_before < (states('input_number.high_power_cost_rate_threshold') | float) / 100 %}
true
{% endif %}
where:
- sensor:
- name: "Net Power"
unique_id: "net_power"
unit_of_measurement: "W"
state: >
{% set production = states('sensor.envoy_xxxxxxxxxxxxx_current_energy_production') | float %}
{% set consumption = states('sensor.envoy_xxxxxxxxxxxxx_current_energy_consumption') | float %}
{{ (production - consumption ) | int }}
I then use this trigger in my energy efficiency automations:
alias: Sun Floor On
description: ""
trigger:
- platform: state
entity_id:
- sensor.high_power_cost_trigger
to: "true"
for:
hours: 0
minutes: 5
seconds: 0
condition:
- type: is_temperature
condition: device
device_id: ____________________________
entity_id: sensor.hue_motion_sensor_2_temperature
domain: sensor
below: 21
enabled: false
action:
- type: turn_on
device_id: ____________________________
entity_id: light.floor_heating
domain: light
enabled: false
- type: turn_on
device_id: ____________________________
entity_id: light.on_off_light_1_2
domain: light
- service: climate.set_hvac_mode
data:
hvac_mode: heat_cool
target:
entity_id: climate.melford_drive_ac
- if:
- condition: numeric_state
entity_id: sensor.office_sensor_hue_temperature
below: 21
then:
- type: turn_on
device_id: ____________________________
entity_id: light.floor_heating
domain: light
mode: single
I use a seperate automation to turn devices off as it turns out to be more reliable and flexible than combining both automations (which I found out the hard way!):
alias: Sun Floor Off
description: ""
trigger:
- platform: state
entity_id:
- sensor.high_power_cost_trigger
from: "true"
for:
hours: 0
minutes: 5
seconds: 0
- platform: state
entity_id:
- binary_sensor.melford_drive_price_spike
from: "off"
to: "on"
condition: []
action:
- type: turn_off
device_id: ____________________________
entity_id: light.on_off_light_1_2
domain: light
- service: climate.set_hvac_mode
data:
hvac_mode: "off"
target:
entity_id: climate.melford_drive_ac
mode: single
The result of all this is a cost focussed approach which simplifies the conditions required to react to your household’s net power consumption alongside dynamic tariffs. I’ve set for: to 5 minutes to smooth out some of the bumps in load and tariff prices.
So, how does it perform?
Here are the last 7 days of my devices being triggered to maintain household temperature while keeping a lid on my power bills by exporting or using available energy in a cost efficient manner. Any spikes in cost rate now only occur around dinner time if we are using the oven.
Finally, I keep track of daily cost by integrating my power cost rate:
With only a few days left of November at the time of writing, the month’s costs are as follows:
Total home energy usage (including solar): 968kWh
Total energy cost: $61.37
Average cost / kWh: 6.3c/kWh
Total solar exports: 351kWh
Total energy credits: $11.40
Average price / kWh: 3.2c/kWh
1 Like
This looks quite interesting. I am currently in the “try before I buy” phase. My energy price is static and I don’t have direct access to the dynamic prices. I am thinking about going with Tibber here in Germany, but for now I have copied a nodeRED flow that gives me the 24h forecast of the EPEX Spot prices.
Sadly I currently haven’t found a way to create a “forecast” entity, all I could manage is a “current” price. The forecast data is there somewhere, I just can’t access it. So for now I am somewhat limited, but already getting some interesting insight at least. I bookmarked this project and will surely be “borrowing” a lot of data from here…
it’s likely you’re using the flow by @Jpsy posted here or something based on it
so the upcoming data will be in the attributes of the sensor - put an entity-card on the dashboard, click it, there you’ll see it in the dropdown below the graph. didn’t bother making a graph out of it yet, but shouldn’t be too hard…
1 Like
Yes, I found it in the attributes, but am still looking for a way to extract that data and work with it. The ideal end result would be a Template that contains the cheapest price for the day which I could hand off to my charging automation.
So…I have made a little Progress but still have a long way to go
I can now reliably get data via Nordpool API and have written an Automation that should trigger when the current energy price reaches the “lowest” pricepoint. I did this with a Template Trigger, and it seems to be working perfectly (tested with the Template tool) but didn’t actually trigger in my first trial run.
More research needed I guess…
This is all theoretical work at the moment, should I decide to switch to Tibber for my electricity I will need a lot more conditions and stuff, but I am a noob at Home Assistant stuff and not very talented with yaml, so this is a “slow and steady” project for me.
Would suggest to create a template sensor instead and trigger based on entity state. Easier to debug. This forum contains many examples, just search for nordpool or entso-e for the right thread.
Yes, that is what I plan to do. Sadly most of the Templates I found didn’t work due to Syntax Changes in HA, but I will get there eventually. I am going to test my current Automation for another day or so and play with some variables like the “Charge from Grid” and “Forcible Charge” options for my Inverter.
How have you found the DETA outdoor plugs? That was what I was looking at installing for my pool pump and heater.
Also with the move the EMHASS did you add any energy monitors to dynamically get the deferable load values or have you found they are pretty static and can just be approximated with a nominal values?
The Deta sockets have been pretty solid. They were a bit of a pain to setup with localtuya but I haven’t been able to find any other outdoor rated sockets. Sometimes the wifi is a little fiddle as the pool equipment is away from the house.
My pool heater is a big three phase heat pump and I was worried about how I could get a big enough relay to switch. In the end the heat pump has a flow switch that turns on and off with a smaller flow pump that could plug into a 10A GPO, I use the other socket for the 10A filter pump.
For the three phase heat pump I use a Shelly 3EM with CT clamps for power monitoring. The 2x 10A sockets report combined power via local tuya. So I can monitor the pumps and heat pump separately and control as two different loads in EMHASS. I set the run hours for the heat pump to the hours of extremely_low (< 5¢/ kWh) prices for the day. I set the filter pump to run for a certain daily energy or a function of the solar pv forecast for the day.
- name: def_total_hours_pool_filter
state: >-
{{
(is_state('automation.p_deferable0_automation', 'on') | int(0)) *
(
max(
(4 - states('sensor.pool_daily_energy') | int(0) / 1000) | int(0),
(states('sensor.solcast_forecast_today') | float(0) / states('sensor.solcast_peak_forecast_today') | float(0)*1000)
)|int(0)
)
}}
- name: def_total_hours_pool_heatpump
state: "{{is_state('automation.p_deferable1_automation','on')|int
* max(0,((state_attr('sensor.amber_feed_in_forecast', 'forecasts')|selectattr('descriptor','eq','extremely_low')|list|count)/2)
)|int(0)}}"
Some details on how I have configued the Power Flow Plus Card
type: custom:power-flow-card-plus
entities:
battery:
entity: sensor.apf_battery_entity
state_of_charge: sensor.gateway_battery
invert_state: true
color_icon: true
color_circle: true
display_state: one_way
state_of_charge_unit_white_space: false
color_state_of_charge_value: true
display_zero_tolerance: 100
grid:
entity: sensor.apf_grid_entity
invert_state: true
display_zero_tolerance: 150
display_state: one_way
color_icon: true
color_circle: true
use_metadata: false
secondary_info:
entity: sensor.solaredge_imported_energy
unit_of_measurement: kWh
icon: ''
color_value: false
unit_white_space: true
display_zero: true
solar:
entity: sensor.apf_generation_entity
display_zero_state: false
color_value: true
color_icon: true
invert_state: false
use_metadata: false
secondary_info:
display_zero: true
entity: sensor.apf_generation_entity_energy_daily
unit_of_measurement: kWh
color_value: true
home:
entity: sensor.apf_real_house_load
color_icon: true
color_value: true
secondary_info:
entity: sensor.apf_house_entity_energy_daily
unit_of_measurement: kWh
color_value: true
display_zero: true
override_state: false
individual2:
entity: sensor.ev_charging
secondary_info:
entity: sensor.m3p_battery
unit_of_measurement: '%'
color_value: true
template: >-
{{states('sensor.m3p_battery') + '% ' + states('sensor.my_battery')
+'%'}}
display_zero: true
color_value: true
color_icon: true
display_zero_state: false
calculate_flow_rate: true
display_zero_tolerance: 50
show_direction: false
inverted_animation: false
name: EV x2
icon: mdi:car-2-plus
color:
- 79
- 158
- 79
individual1:
entity: sensor.hvac_power
name: HVAC
icon: mdi:hvac
secondary_info:
unit_of_measurement: kWh
template: ''
color_value: true
unit_white_space: true
decimals: 0
display_zero: true
display_zero_tolerance: 5
entity: sensor.heating_cooling_daily
color_icon: true
color_value: true
color:
- 64
- 121
- 255
display_zero: true
display_zero_state: false
calculate_flow_rate: false
inverted_animation: false
use_metadata: false
display_zero_tolerance: 200
fossil_fuel_percentage:
display_zero: false
color_icon: true
display_zero_state: false
display_zero_tolerance: 1
color_value: true
use_metadata: false
clickable_entities: true
display_zero_lines:
mode: hide
transparency: 50
grey_color:
- 189
- 189
- 189
use_new_flow_rate_model: true
w_decimals: 0
kw_decimals: 1
transparency_zero_lines: 0
min_expected_power: 1000
max_expected_power: 10000
max_flow_rate: 10
min_flow_rate: 1
Some details on how I have configured the Sankey Chart Card
height: 200
unit_prefix: M
round: 1
min_box_height: 3
min_box_distance: 5
show_states: true
show_units: true
sections:
- entities:
- type: entity
children:
- sensor.apf_solar2batt_energy_monthly
- sensor.apf_solar2house_energy_monthly
- sensor.apf_solar2grid_energy_monthly
entity_id: sensor.apf_generation_entity_energy_monthly
color: var(--warning-color)
name: Solar
- type: remaining_child_state
children:
- sensor.apf_grid2batt_energy_monthly
- sensor.apf_grid2house_energy_monthly
entity_id: grid
color: var(--error-color)
name: Grid
- entities:
- type: entity
children:
- battery
entity_id: sensor.apf_solar2batt_energy_monthly
color: var(--warning-color)
name: Solar2Batt
- type: entity
children:
- battery
entity_id: sensor.apf_grid2batt_energy_monthly
color: var(--error-color)
name: Grid2Batt
- type: entity
children:
- sensor.apf_grid_export_energy_monthly
entity_id: sensor.apf_solar2grid_energy_monthly
color: var(--warning-color)
name: Solar2Grid
- type: entity
children:
- sensor.apf_house_entity_energy_monthly
entity_id: sensor.apf_solar2house_energy_monthly
color: var(--warning-color)
name: Solar2House
- type: entity
children:
- sensor.apf_house_entity_energy_monthly
entity_id: sensor.apf_grid2house_energy_monthly
color: var(--error-color)
name: Grid2House
- entities:
- type: remaining_parent_state
children:
- sensor.apf_batt2house_energy_monthly
- sensor.apf_batt2grid_energy_monthly
entity_id: battery
color: var(--success-color)
name: Battery Charging
- entities:
- type: entity
children:
- sensor.apf_house_entity_energy_monthly
entity_id: sensor.apf_batt2house_energy_monthly
color: var(--success-color)
name: Batt2House
- type: entity
children:
- sensor.apf_grid_export_energy_monthly
entity_id: sensor.apf_batt2grid_energy_monthly
color: var(--success-color)
name: Batt2Grid
- entities:
- type: entity
children: []
entity_id: sensor.apf_house_entity_energy_monthly
name: House
- entity_id: sensor.apf_grid_export_energy_monthly
color: var(--error-color)
name: Grid Export
type: custom:sankey-chart
show_names: true
wide: true
min_state: 0
energy_date_selection: false
title: Energy - Monthy
2 Likes
Matching your Air Conditioning to Excess Solar.
Again for those long days of negative feed in, matching your air conditioning to consume your excess solar is an excellent sink for all that very cheap energy. Your household stays cool, bonus if you work from home, have pets and it also saves having to blast the aircon in the afternoon when the prices are high.
There is a great project which uses the DRED interface on most aircon systems to limit power consumption using a set of four relays: Controlling Air conditioner using Demand Enabled Response Device per AS/NZS 4755
My premise was to change the setpoint to keep the aircon in the ball park of my excess solar consumption. If I’m drawing too much power, I raise the set point, and the aircon draws less power. if I’m generating too much excess solar I reduce the set point and the aircon draws more power.
To calculate the relevant set point automation I am using the following formula:
SetPoint = ExternalTemp - CoolingFactor/ExcessSolarPower
I was also playing around with InternalTemp as a variable, but that doesn’t seem to have much impact on the long term power consumption.
The CoolingFactor is in (W / °C) and is different for every household, depends on how big your AC is, how good is your insulation and thermal mass. For my household I find a cooling factor of 1500 W / °C works pretty well. I set up the CoolingFactor as an input_number helper.
I use a template calculate the helper to actual excess solar power available for HVAC.
Finally, I have an automation which is triggered when the excess solar power changes to change the setpoint of my air conditioner, in this automation you can see I also moderate the power by the battery state of charge, so my battery fills first and then in the afternoon once the battery is full, the aircon gets the bulk of the power to drive down the house temp.
Interested to hear how others are managing their aircon with HA and Amber low prices.
2 Likes
Old but useful thread. Bump
1 Like