Skip to content

Latest commit

 

History

History
514 lines (433 loc) · 16.8 KB

README.md

File metadata and controls

514 lines (433 loc) · 16.8 KB

weewx-photovoltaics

Extension to WeeWX for processing data of the photovoltaics system E3/DC

Display example:

Prerequisites

pip install pye3dc

Installation instructions

Copy photovoltaics.py to the user directory of WeeWX and edit weewx.conf as follows:

[DataBindings]
    ...
    [[pv_binding]]
        database = pv_sqlite
        table_name = archive
        manager = weewx.manager.DaySummaryManager
        schema = user.photovoltaics.schema

[Databases]
    ...
    [[pv_sqlite]]
        database_name = photovoltaics.sdb
        database_type = SQLite

[Engine]
    [[Services]]
        prep_services = ..., user.photovoltaics.E3dcUnits
        data_services = ..., user.photovoltaics.E3dcService

[E3DC]
    [[S10EPRO]]
        protocol = RSCP
        host = replace_me
        username = replace_me
        password = replace_me
        api_key = replace_me
        query_interval = 1           # optional
        #mqtt_topic = "e3dc/weewx"   # normally not required
    [[ACTHOR]]
        protocol = MyPV
        host = replace_me
        #mqtt_topic = "acthor/weewx" # normally not required
    [[MQTT]]
        protocol = MQTT
        enable = true
        topic = "e3dc/weewx"

Restart WeeWX.

There is an installer in the base directory, but it is alpha and untested. Use it on your own risk. Make a backup of WeeWX before using the installer. If you are not sure, copy photovoltaics.py to the user directory of WeeWX manually, and also edit weewx.conf manually.

Oberservation types

This extension provides several additional observation types holding readings from the PV inverter.

To access these observation types you need to specify data_binding = pv_binding

Battery storage

Tag Unit Alternative Unit Description
emsBatteryPower W kW actual charge (postive) or discharge (negative) power
emsBatteryCharge % charge level
emsBatteryChargePower W kW actual charge power
emsBatteryDischargePower W kW actual discharge power
sumBatteryChargeEnergy Wh kWh calculated charge energy
sumBatteryDischargeEnergy Wh kWh calculated discharge energy

Supply

Tag Unit Alternative Unit Description
emsGridPower W kW actual grid power, positive values mean consumption
emsGridPurchasePower W kW actual grid power purchased
emsGridFeedinPower W kW actual power feeded into grid
emsSolarPower W kW actual solar production power
emsAddPower W kW actual power of an additional supply
sumSolarEnergy Wh kWh solar energy produced
sumGridPurchaseEnergy Wh kWh energy received from grid
sumGridFeedinEnergy Wh kWh energy feeded into grid
sumAddEnergy Wh kWh energy produced by additional supply

Consumption

Tag Unit Alternative Unit Description
emsHousePower W kW inhouse consumption power
emsWallPower W kW wallbox consumption power
emsSelfConsumption % self-consumption level
emsAutarky % autarky level
sumHouseEnergy Wh kWh calculated inhouse consumption energy
sumWallEnergy Wh kWh calculated wallbox consumption energy

Static values

Tag Unit Alternative Unit
installedPVPeakPower W kW
installedBatteryCapacity Wh kWh
pvMaxAcPower W kW
pvMaxBatChargePower W kW
pvMaxBatDischargePower W kW

PM

Readings of the power meter included in the E3/DC system

Tag Unit Description
pmGridPowerL1 W actual power at phase L1
pmGridPowerL2 W actual power at phase L2
pmGridPowerL3 W actual power at phase L3
pmGridVoltageL1 V actual voltage at phase L1
pmGridVoltageL2 V actual voltage at phase L2
pmGridVoltageL3 V actual voltage at phase L3
pmGridEnergyL1 kWh electricity meter at phase L1 *)
pmGridEnergyL2 kWh electricity meter at phase L2 *)
pmGridEnergyL3 kWh electricity meter at phase L3 *)

*) upwards counting for getting energy from the grid, downwards counting for feeding energy into the grid, even negative values are possible

PVI DC

solar tracker readings

Tag Unit Description
pviDCpowerT0 W actual power of tracker 1
pviDCpowerT1 W actual power of tracker 2
pviDCvoltageT0 V actual voltage of tracker 1
pviDCvoltageT1 V actual voltage of tracker 2
pviDCcurrentT0 A actual current of tracker 1
pviDCcurrentT1 A actual current of tracker 2
pviDCenergyT0 Wh earned energy so far for tracker 1
pviDCenergyT1 Wh earned energy so far for tracker 2

PVI AC

PV inverter AC output

Tag Unit Description
pviACpowerL1 W AC power L1
pviACpowerL2 W AC power L2
pviACpowerL3 W AC power L3
pviACapparentPowerL1 VA AC apparent power L1
pviACapparentPowerL2 VA AC apparent power L2
pviACapparentPowerL3 VA AC apparent power L3
pviACreactivePowerL1 var AC reactive power L1
pviACreactivePowerL2 var AC reactive power L2
pviACreactivePowerL3 var AC reactive power L3
pviACvoltageL1 V AC voltage L1
pviACvoltageL2 V AC voltage L2
pviACvoltageL3 V AC voltage L3
pviACcurrentL1 A AC current L1
pviACcurrentL2 A AC current L2
pviACcurrentL3 A AC current L3
pviACenergyL1 Wh energy L1
pviACenergyL2 Wh energy L2
pviACenergyL3 Wh energy L3

myPV ACTHOR

Tag Unit
heataccuTemp °C
heataccuVoltage V
heataccuPower W
heataccuMainsVoltage V
heataccuMainsCurrent A
heataccuMainsFrequency Hz

Note: The field unixtime in data.jsn is bogus. MyPV acknowledged that bug and announced to remove it with the next release.

Sun

Tag Unit Description
solarAzimuth ° solar azimuth (compass direction of the sun)
solarAltitude ° solar altitude
solarPath % percentage of the time elapsed between sunrise and sunset

What is the difference between solarAzimuth, solarAltitude and $almanac.sun.az, $almanac.sun.alt, respectively?

  • solarAzimuth and solarAltitude are output to MQTT and thus allow live updates on web sites.
  • As solarAzimuth and solarAltitude are observation types, they can be saved to the database (as they indeed are) and displayed in diagrams.

The values are calculated by the Almanac class that provides the $almanac tag.

All those values honour temperature (outTemp) and pressure (pressure).

Database

Readings are saved to photovoltaics.sdb

MQTT

Readings can be output to MQTT. You need an MQTT broker for that.

Usage in skins

Display values (CheetahGenerator)

The observeration types described above can be used as any observation type in WeeWX.

Examples:

$current.emsBatteryPower
$day($data_binding='pv_binding').emsGridPower.max
$week($data_binding='pv_binding').sumSolarEnergy.sum
$month($data_binding='pv_binding').emsHousePower.energy_integral

To get the energy for a time span, there are two ways:

  • observation types sum...Energy together with aggregation type sum
  • observation types ems...Power together with aggregation type energy_integral

The latter of them two requires the weewx-GTS extension to be installed.

Please note: The sum...Energy observation types integrate the respective power over the archive interval. That results in a higher accuracy than the E3/DC display shows, as there 15 minutes averages are used for calculation. And it results in little differences in the readings.

Full load hours

Often people are interested in a measure of the degree of utilisation. That measure is called full load hour (in german: Vollaststunde, spezifischer Ertrag) and measured in kWh/kWp or simply hours.

The following example shows how you could calculate und display full load hours for various time periods within a template:

#from weewx.units import ValueTuple, ValueHelper
...
  <table class="table-striped">
    <tbody>
    <tr>
      <td></td>
      <td>Heute</td>
      <td>Gestern</td>
      <td>Diese Woche</td>
      <td>Diesen Monat</td>
      <td>Dieses Jahr</td>
    </tr>
#set $kwp=$current.installedPVPeakPower.kilowatt.raw
#if $kwp is not None and $kwp>0
#set $hday=$day(data_binding="pv_binding").emsSolarPower.energy_integral.kilowatt_hour.raw/$kwp*3600.0
#set $hday_vh=ValueHelper(ValueTuple($hday,'second','group_deltatime'),formatter=$station.formatter)
#set $hyday=$yesterday(data_binding="pv_binding").emsSolarPower.energy_integral.kilowatt_hour.raw/$kwp*3600.0
#set $hyday_vh=ValueHelper(ValueTuple($hyday,'second','group_deltatime'),formatter=$station.formatter)
#set $hweek=$week(data_binding="pv_binding").emsSolarPower.energy_integral.kilowatt_hour.raw/$kwp*3600.0
#set $hweek_vh=ValueHelper(ValueTuple($hweek,'second','group_deltatime'),formatter=$station.formatter)
#set $hmonth=$month(data_binding="pv_binding").emsSolarPower.energy_integral.kilowatt_hour.raw/$kwp*3600.0
#set $hmonth_vh=ValueHelper(ValueTuple($hmonth,'second','group_deltatime'),formatter=$station.formatter)
#set $hyear=$year(data_binding="pv_binding").emsSolarPower.energy_integral.kilowatt_hour.raw/$kwp*3600.0
#set $hyear_vh=ValueHelper(ValueTuple($hyear,'second','group_deltatime'),formatter=$station.formatter)
    <tr>
      <td>Vollaststunden</td>
      <td style="text-align:right">$hday_vh.format("%(hour)d:%(minute)02d:%(second)02d h")</td>
      <td style="text-align:right">$hyday_vh.format("%(hour)d:%(minute)02d:%(second)02d h")</td>
      <td style="text-align:right">$hweek_vh.format("%(hour)d:%(minute)02d:%(second)02d h")</td>
      <td style="text-align:right">$hmonth_vh.format("%(hour)d:%(minute)02d:%(second)02d h")</td>
      <td style="text-align:right">$hyear_vh.format("%(hour)d h")</td>
    </tr>
#end if
    </tbody>
  </table>

Energy consumption during day and night time

During the light day you can use the PV energy for yourself and reduce energy costs. When it is dark you need energy from the grid or a battery pack. To think about a battery you need to know how much energy is consumed by night. So this is an example how to calculate it for the day:

#from weewx.units import ValueTuple, ValueHelper
...
#set $n1=$daylight(horizon=2,use_center=1).emsHousePower.energy_integral.kilowatt_hour.raw-$day(data_binding="pv_binding").heataccuPower.energy_integral.kilowatt_hour.raw
#set $n1_vh=ValueHelper(ValueTuple($n1,'kilowatt_hour','group_energy'),formatter=$station.formatter)
#set $n2=$day.emsHousePower.energy_integral.kilowatt_hour.raw-$daylight(horizon=2,use_center=1).emsHousePower.energy_integral.kilowatt_hour.raw
#set $n2_vh=ValueHelper(ValueTuple($n2,'kilowatt_hour','group_energy'),formatter=$station.formatter)
<p>heating: $day(data_binding="pv_binding").heataccuPower.energy_integral.kilowatt_hour</p>
<p>daylight consumption (without heating): $n1_vh</p>
<p>night consumption: $n1_vh</p>

To calculate those values for the week is more difficult:

#set $weekEd=0.0
#set $weekEn=0.0
#for $span in $LMTweek(data_binding="pv_binding").days
#for $pp in $span.daylights(horizon=1,use_center=1)
#set $houseday=$pp.emsHousePower.energy_integral.kilowatt_hour.raw
#end for
#set $heat=$span.heataccuPower.energy_integral.kilowatt_hour.raw
#set $house=$span.emsHousePower.energy_integral.kilowatt_hour.raw
#if $houseday is not None and $heat is not None and $house is not None
#set $weekEd+=$houseday-$heat
#set $weekEn+=$house-$houseday
#end if
#end for
#set $weekEd_vh=ValueHelper(ValueTuple($weekEd,'kilowatt_hour','group_energy'),formatter=$station.formatter)
#set $weekEn_vh=ValueHelper(ValueTuple($weekEn,'kilowatt_hour','group_energy'),formatter=$station.formatter)
<p>week heating: $week(data_binding="pv_binding").heataccuPower.energy_integral.kilowatt_hour</p>
<p>week daylight consumption: $weekEd_vh</p>
<p>week night consumption: $weekEn_vh</p>

To use those examples weewx-GTS neeeds to be installed. It provides the $daylight and .daylights time spans.

Those examples assume it is light enough for the PV modules to produce energy if the altitude of the sun is above 2°.

Diagrams (ImageGenerator)

in skin.conf:

[ImageGenerator]
    ...
    [[day_images]]
        ...
        [[[dayPV]]]
            data_binding = pv_binding
            [[[[emsSolarPower]]]]
                label = Sonne
                color = "#ffc83f"
            [[[[emsGridPower]]]]
                label = Netz
            [[[[emsBatteryPower]]]]
                label = Batterie
            [[[[emsHousePower]]]]
                label = Verbrauch

        [[[dayPVcharge]]]
            data_binding = pv_binding
            yscale = 0,100,10
            [[[[emsBatteryCharge]]]]
                label = Ladestand der Batterie

Repeat those definitions with week, month, and year

To show those diagrams add the appropriate img tags in a template:

<img src="dayPV.png" />
<img src="dayPVcharge.png />

Belchertown skin

Example, what to add to graphs.conf:

[PV-Anlage]
    title = "PV-Anlage"
    show_button = true
    button_text = "PV-Anlage"
    data_binding = pv_binding
    page_content = """
some general text"""

    [[pvGraphToday]]
        time_length = today
        tooltip_date_format = "LLL"
        gapsize = 300 
        title = PV-Leistung heute
        yAxis_label = Leistung
        exporting = 1
        [[[emsSolarPower]]]
            name = "PV-Leistung (5-Minuten-Durchschnitt)"
            color = "#ffc83f"

    [[solarRadGraph]]
        time_length = today
        tooltip_date_format = "LLL"
        gapsize = 300 
        title = "Sonnenstrahlung und UV-Index heute"
        exporting = 1
        [[[radiation]]]
            name = Sonnenstrahlung
            zIndex = 1
            color = "#ffc83f"
        [[[maxSolarRad]]]
            name = Theoretischer Maximalwert
            type = area
            color = "#f7f2b4"
            yAxis_label = "W/m&sup2;"
        [[[UV]]]
            yAxis = 1
            yAxis_min = 0
            yAxis_max = 14
            color = "#90ed7d"
            yAxis_label = "UV"
            name = "UV-Index"
            zIndex = 2

    [[pvGraphWeek]]
        time_length = 604800 # Last 7 days
        tooltip_date_format = "LLLL"
        aggregate_type = avg
        aggregate_interval = 3600 # 1 hour
        gapsize = 3600 # 1 hour in seconds
        start_at_whole_hour = true
        title = PV-Leistung Woche
        yAxis_label = Leistung
        exporting = 1
        [[[emsSolarPower]]]
            name = "PV-Leistung (Stundendurchschnitt)"
            color = "#ffc83f"

    [[PVEnergy]]
        time_length = 2592000 # Last 30 days
        tooltip_date_format = "dddd LL"
        gapsize = 86400 # 1 day in seconds
        title = "PV-Ertrag"
        aggregate_interval = 86400
        yAxis_label = Energie
        yAxis_label_unit = "Wh"
        start_at_midnight = true
        exporting = 1
        [[[emsSolarPower]]]
            name = Tagesertrag
            aggregate_type = energy_integral
            type = column
            color = "#ffc83f"

    [[PVmax]]
        time_length = 2592000 # Last 30 days
        tooltip_date_format = "dddd LL"
        gapsize = 86400 # 1 day in seconds
        title = "Tägliches Leistungsmaximum"
        aggregate_interval = 86400
        yAxis_label = Leistung
        #yAxis_label_unit = "W"
        start_at_midnight = true
        exporting = 1
        [[[emsSolarPower]]]
            name = Tagesleistungsmaximum
            aggregate_type = max
            type = spline
            color = "#ffc83f"

See skins/Belchertown/about/photovoltaics.html.tmpl for an example PV page with live update of solar power by MQTT. Please note: The example assumes the readings to be stored in weewx.sdb rather than photovoltaics.sdb. So add data_binding = pv_binding to change that if necessary.

To show a diagram about solar course you could use the following configuration:

    [[sunpath]]
        time_length = 172800
        title = "Sonnenstand"
        yAxis_label = Winkel
        [[[solarPath]]]
            zIndex = 2
            yAxis = 1
            yAxis_min = -25
            yAxis_max = 100
            yAxis_tickInterval = 25
            yAxis_label = "Zeit (%)"
        [[[solarAzimuth]]]
            yAxis = 0
            yAxis_min = -90
            yAxis_max = 360
            yAxis_tickInterval = 90
        [[[solarAltitude]]]
            yAxis = 0
            yAxis_min = -90
            yAxis_max = 360
            yAxis_tickInterval = 90

Links