Home Assistant BCHydro Sensor

🚧 In development, but you can copy sensor.py down to your HASS install: just copy-paste sensor.py into a file name the same inside your HASS custom_components/bchydro directory. Edit bchydro_username and bchydro_password to match your BCHydro account.

Installation

Place these files inside a directory in your custom_components directory:

cd config/custom_components
git clone git@github.com:emcniece/hass-bchydro.git

Then edit your configuration.yml to include this under sensors:

sensors:
  - platform: bchydro
    username: !secret bchydro_username
    password: !secret bchydro_password

Then edit your secrets.yml to include your BCHydro account details:

bchydro_username: myemail@domain.com
bchydro_password: mypassword

Then restart your HASS installation and configure your new entites!

BCHydro Data Formats

Several calls are made to the BCHydro website for varying purposes.

Account Data

The URL_ACCT_INFO URL is used to obtain the user account slid which is later used for fetching usage data. This endpoint has a lot of other info that might be useful for consumption at a later time - here's a sample of the JSON response:

{
  "accountType": "residential",
  "evpSlid": "0001111111",
  "evpAccount": "000011111111",
  "evpAccountId": "AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA",
  "evpProfileId": "bchusername",
  "evpRateCategory": "BCRSE1101",
  "evpBillingClass": "Residential",
  "evpRateGroup": "RES1",
  "evpReadRoute": "BAAAAAAA",
  "evpBillingStart": "2020-06-06T00:00:00-07:00",
  "evpBillingEnd": "2020-08-06T00:00:00-07:00",
  "evpBilledStart": "2020-04-07T00:00:00-07:00",
  "evpBilledEnd": "2020-06-05T00:00:00-07:00",
  "evpValidityStart": "2020-06-29T00:00:00-07:00",
  "evpValidityEnd": "9999-12-31T00:00:00-08:00",
  "evpEnablementDate": "2012-12-23T23:00:00-08:00",
  "evpHeatingType": "N",
  "evpPremiseType": "10",
  "evpPostalCode": "V1V 1V1",
  "evpBillingArea": "99",
  "evpConsToDate": "584kWh",
  "evpCostToDate": "$67",
  "yesterdayPercentage": "97",
  "evpStepThreshold": "1375.89",
  "evpDaysInBillingPeriod": "62",
  "evpStepThresholdDate": "",
  "evpEstConsCurPeriod": "594",
  "evpEstCostCurPeriod": "68",
  "evpCurrentDateTime": "2020-08-06T23:30:46-07:00",
  "evpRole": "user",
  "evpCsrId": "bchusername",
  "evpComLastBillingPeakDemand": "",
  "evpComLastBillingPowerFactor": "",
  "enableSMSAlerts": "true",
  "nonWan": "False",
  "timezone": "GMT",
  "isEnDateinCBP": "True"
}

Usage Data

Points may be one of INVALID, ACTUAL, ESTIMATED(?). While evidence of ESTIMATED has been captured it seems to occur infrequently. This is something to plan for in the future and is noted in the ToDo.

After logging in, the usage URL_GET_USAGE URL returns data like this:

<Data>
  <Rates rateGroup="RES1" bpStart="Jun 6" bpEnd="Aug 6, 2020" daysSince="61" cons2date="584kWh" cost2date="$67" estCons="594" estCost="68"/>
  <Series name="Consumption data" step2Value="1375.89" isEnDateinCBP="true" evpCurrentDateTime="2020-08-06T22:41:06-07:00" blockStatus="0" nonWan="false">
    <Point type="SMI" quality="ACTUAL" dateTime="2020-07-30T00:00:00-07:00" endTime="2020-07-30T00:00:00-07:00" value="16.88" cost="0.00"/>
    <Point type="SMI" quality="ACTUAL" dateTime="2020-07-31T00:00:00-07:00" endTime="2020-07-31T00:00:00-07:00" value="12.36" cost="0.00"/>
    <Point type="SMI" quality="ACTUAL" dateTime="2020-08-01T00:00:00-07:00" endTime="2020-08-01T00:00:00-07:00" value="18.31" cost="0.00"/>
    <Point type="SMI" quality="ACTUAL" dateTime="2020-08-02T00:00:00-07:00" endTime="2020-08-02T00:00:00-07:00" value="16.27" cost="0.00"/>
    <Point type="SMI" quality="ACTUAL" dateTime="2020-08-03T00:00:00-07:00" endTime="2020-08-03T00:00:00-07:00" value="12.46" cost="0.00"/>
    <Point type="SMI" quality="ACTUAL" dateTime="2020-08-04T00:00:00-07:00" endTime="2020-08-04T00:00:00-07:00" value="14.66" cost="0.00"/>
    <Point type="SMI" quality="ACTUAL" dateTime="2020-08-05T00:00:00-07:00" endTime="2020-08-05T00:00:00-07:00" value="16.31" cost="0.00"/>
  </Series>
</Data>

In the sensor API, each of these <Point /> elements is processed like so:

for point in root.findall('Series')[0].findall('Point'):
    print(point.items())
    # [
    #   ('type', 'SMI'),
    #   ('quality', 'ACTUAL'),
    #   ('dateTime', '2020-07-30T00:00:00-07:00'),
    #   ('endTime', '2020-07-30T00:00:00-07:00'),
    #   ('value', '16.88'),
    #   ('cost', '0.00')
    # ]

ToDo

• Unit tests + CI
• Figure out how to read secrets
• Implement HASS config flow for browser-based entry of secrets
• Hass.io, integration, Supervisor, HACS compatibility
• Add more sensors for days_since_billing, consumption_to_date, cost_to_date, estimated_consumption, estimated_cost
• Handle ESTIMATED datapoints
  • Parse dates and only take latest_usage from days matching the current timestamp?

References

• HASS Community discussion