Regular readers may recall that our hot water can be generated in 3 different ways: (i) conventional gas boiler, (ii) from grid electricity and (iii) from the surplus on my own solar panels. Attractions of these options are that gas is always available and stable in price, but my grid electricity is lower carbon and may at times be cheaper than gas, and my solar electricity is lowest in both carbon and cost but is subject to significant daily and seasonal variation.
The logic to sort out which source to use is managed by my HEMS. Gas is the baseline and the gas boiler is set to heat water for an hour a day in the early evening to ensure that baths etc are possible. The heating is thermostatically controlled so it doesn’t heat if the water is already hot, and that thermostat is set slightly lower than the immersion thermostat too.
The ImmerSUN normally operates automatically to divert surplus solar electricity proportionately to the immersion heater after the needs of general house load, battery charging and car charging have been taken. However if the electricity price is negative (yes, really) then the HEMS may override the ImmerSUN so that water heating is not done by free solar but instead may be delayed to allow use of paid-to-use electricity.
The final part of this triumvirate is buying electricity from the grid to heat water. Here the price of bought electricity is compared either to the price of gas and a decision made to use electricity when it is cheaper (it’s always lower CO2), or compared to the price of surplus solar (effectively zero) to buy from the grid. Both of these are obviously comparisons with a price threshold but until now the choice of threshold has been made manually – typically against gas in winter when solar output is limited and against solar in summer when more readily available. However the reality of UK weather is that this is a compromise as it may be very sunny one day but very dull the next.
The new refinement therefore is to use the existing solar forecasting integration. Solar forecasting already informs HEMS decisions about when to charge the storage battery from the grid and when to operate the wet goods (dishwasher and washing machine). The latest change is that the solar forecasting is now also use to choose whether to base a decision to buy electricity for water heating against a threshold related to the gas price or against the price of surplus solar PV.
The above schedule shows that, as a result of no significant solar production anticipated on the 4th, the HEMS has compared electricity price to gas price and thus elected to buy electricity from the grid to make hot water overnight since at 1.7640 to 2.4675 p/kWh electricity is cheaper than gas.
It can’t be very often that an energy company blogs about its customers’ achievements. Last week it happened. Octopus Energy wrote a blog entitled “How to hack your home for cheaper, greener, energy with our open API” which featured the achievements of its customers, and Greening Me got two honourable mentions.
For those not familiar with geek-speak, API is Application Programming Interface which is a mechanism by which an app, webpage or computer program may give commands to, or receive data from, another program – often a web server. Such APIs are often closed (that is that they are only available for use by the creator’s own app or webpage etc), but in this case the Octopus APIs are open so that they can be used by others (including me) to create our own apps, webpages, or other integrations to get data from Octopus. That data may be future price information for a UK electricity region or the actual consumption from a specified electricity meter for example. Octopus document their APIs and encourage others to find innovative uses for them.
Other APIs that I use were either documented privately by the manufacturers of the equipment concerned, although the manufacturer has not put the API in the public domain, or were reverse-engineered by myself or others by looking at how the manufacturer used it or at the internet traffic that it generated and working out how we could use it ourselves for a slightly different purpose. Such purposes would include controlling equipment other than by the manufacturer’s own app, or collecting data into some non-supported form.
Greening Me’s first mention in the blog came under the Smart Electric Vehicle (EV) Charging section where Octopus wrote..
One of our own smart energy pioneers, Greening Me, has used a Raspberry Pi and an add-on circuit board with our API to switch his electric car charger on/off and set the best time for his hot water immersion heater to run. He also has solar generation and so he can direct his solar power to either his smart car charger or hot water.
Later in the “I’ll do it myself (tech level 🌶🌶🌶)” section after describing a group of “smart home pioneers” Octopus wrote..
Together with Western Power Distribution, Passiv Systems have also created something similar to Greening Me’s HEMS, which is currently being trialled and evaluated as another BEIS funded research project called MADE.
So it’s official – I’m a “smart energy pioneer” and a “smart home pioneer”. I also quite like the idea of being a “home hacker” in the positive sense of someone who makes their own home conform to their wishes. If you’d like to read the full blog post from Octopus Energy then you can do so here https://octopus.energy/blog/agile-smart-home-diy/.
Most of us are used to a simple world of electricity where we pay for what we consume. For most folks like myself based in the UK that’s typically a fixed price per kWh/unit consumed regardless of time of day, even through dual-rate tariffs have been around for decades – the best known being “Economy 7” tariffs. However as the grid gets smarter then there are increasing opportunities to save on, or make money from, electricity.
Conventional – pay for power.
This is the area with which most of us are most familiar. We all get the idea of paying for the power we consume. Most UK households pay a fixed price per kWh/unit regardless of the time of day. We have a competitive electricity market, so there are the choice of 70 to 80 different providers who will make different offers regarding standing charge (sometimes marketed as a subscription) and unit cost.
There’s also the opportunity to choose between a flat rate tariff or Economy 7 even on conventional meters that provide a discounted night rate for 7 hours.. These typically provide a discounted night rate, but may charge a little more during the day. They used to advertise these as ‘less than half-price electricity’ but that’s often not the case now.
Stepping up in complexity (and opportunity) smart meters provide the opportunity for a more diverse range of tariffs including different cheap night time periods, more than two rates at different times of day (in extreme 48 half-hourly rates), and a free day at the weekend (i.e. a zero rate of a weekend day) etc.
Beyond that my own tariff (Octopus Agile) not only has up to 48 different half-hourly prices/day that change daily based on that day’s market prices. That might sounds a bit scary but it can yield very cheap electricity prices – 4.48 p/kWh for me in April/May 2020 (for example) which is a third of what most people pay.
(The original version of this post wrongly had the table from my gas bill above and mistakenly claimed that I had paid “a quarter of what most people pay” rather than a third. Total consumption is untypically low at the present time due to limited miles driven.)
Agile – paid to consume
Top left on my initial diagram is Agile – paid to consume.
One of the features of the wholesale electricity market is that at times the market price for electricity goes negative. At such times the a significant excess of supply (typically because of high output from wind turbines) over demand (often but not always at night) yields a negative price so electricity companies looking to buy electricity are being paid to take it. Most electricity companies will continue to charge their customers the standard price in these circumstances but, with the octopus Agile tariff, the negative pricing is passed to the consumer so that you are paid to consume electricity. This is one of the reasons that my electricity costs are so low.
The above chart shows my electricity costs for Saturday 23rd May 2020. The blue line shows the half-hourly electricity price varying between minus 10 p/kWh and plus 15 p/kWh. The red bars show my electricity consumption in each half hour. You can see how consumption tends to be highest when the price is lowest leading to an average price paid of minus 6.22 p.kWh (i.e. they paid me to use electricity) – indeed they paid me 82.4 p to buy electricity that day.
Conventional export – paid to export
The next opportunity to make money from electricity is to sell it to the grid. Obviously that depends on having a source for the electricity typically a generating asset like solar panels or a wind turbine, possibly coupled with a storage device like a battery. It’s also possible with a battery alone, but I know no-one who does that as the economics are more challenging.
The UK currently has a scheme called Smart Export Guarantee (SEG) where you can sell your export to an electricity company. Prices vary enormously so it’s worth shopping around and not just assuming that your electricity company will give you a good offer.
There is also a smarter SEG option where Octopus offer a dynamic SEG based on market rates (Octopus Agile Export) which may at times offer a high rate, but also offers a lower rate at times, and is thus perhaps better suited to those with storage.
I myself am NOT on such a tariff as I’m on an older legacy Feed-in Tariff (FiT). Despite its name FiT is a generation incentive, not an export incentive. As a generation incentive FiT encourages self-consumption since each kWh that I consume myself does not reduce my income, whereas on SEG each kWh that I use myself (such as making hot water) would reduce export income. So, for example, if I use a kWh of electricity to make hot water that’s saved a kWh (or thereabouts) of gas at around 3 p/kWh, but if I was on SEG then I might have lost 5.5 p/kWh of export revenue to save 3 p/kWh on gas which is clearly an on-cost not a saving. There are other benefits of course because I’ve reduced my carbon footprint by using my own low CO2 electricity to replace a fossil fuel, but it’s not (in this case) improving my financial position.
A further area of research by others is V2X (V2H and V2G) – taking electricity stored in an electric vehicle and using that within the home (V2H) or exporting it to the grid (V2G).
Export penalty – penalised for export
A logical consequence of this smart grid that I’ve outlined is being penalised for export. If there are times when the market price for electricity is negative then if I were part of that market then I might expect to be penalised for export. This doesn’t actually exist in the UK, as the only model that links SEG payments to the market price, Octopus Agile Export, protects its customers from negative pricing.
Should consumers be exposed to this risk then a logical behaviours would be:
To manage self-consumption into the negative export periods, and potentially thus increase export in the positive export periods. For example disable diversion to an immersion heater or car when export price is positive, and then maximise self-consumption when the export price (and presumably the import price also) is negative.
To disable the generating asset to avoid the export penalty.
Some people like myself will find developments in the smart energy sector a fascinating and engaging topic with opportunities both the save money and engage in creating a cleaner and greener electricity system.
However given that many choose not to even participate in the competitive market for electricity supply created when the regional electricity companies were privatised in late 1990 (i.e. 30 years ago) then there will be a significant number who are not so motivated.
This then creates opportunity for a wider variety of smart offers. Some products, at the Agile Octopus end of the spectrum, giving the consumer the opportunity to benefit from their own decision making, while others look more like a traditional dumb tariff with a very simple price structure but potentially making the energy company a more active manager of the home appliances so that the consumer hopefully plays a lower unit rate while the energy company takes responsibility for managing the assets within the home.
Back in late 2018 I purchased a Hildebrand Glow Stick Consumer Access Device (CAD) to monitor my electricity consumption. A CAD is a consumer device that can be paired with domestic smart meters to provide the consumer with a means of reading the meter. All UK smart meters are supplied with a dedicated in-home display (IHD) to display energy consumption, which is also an example of a CAD. The Glow Stick pairs with the meters like the IHD but shares the data to the cloud from where it can be read either via an app (Bright) or another device using APIs.
Each smart meter effectively has two interfaces – a Wide Area Network (WAN) connection used for metering and billing and a Home Area Network (HAN) used for connection between meters (electric and gas), hub (embedded within the electric meter) and IHD. The HAN is also available for smart home devices.
“Network hub” including (from top to bottom):
Network switch providing additional hardwired connections to the internet, placed behind..
TalkTalk router providing WiFi and 4 hardwired connections to the external internet, placed above..
Network storage, placed above..
Immersun bridge (left) and Glow Stick (right and forwards)
When I initially installed the Glow Stick it provided a very useful tool to see current and historic energy consumption, but the equivalent cost displays were incorrect (at no fault of Hildebrand) because the CAD correctly read the meter costs, but the meter was not sufficiently sophisticated to store the complex Agile tariff (where unit cost changes every 30 minutes).
I recently learned that Hildebrand now had the ability to take the tariff directly from Octopus Energy via API, bypassing the incorrect tariff data in the meter. A quick support email to Hildebrand confirmed that this was not only possible, but also that the cost data would be corrected back to when I bought the Glow Stick back in 2018. A few days later and the conversion was complete.
These two views show today’s part-complete data:
The screenshot on the left shows today’s part-complete energy data. That on the right shows the equivalent cost data. Had the unit rate been constant throughout the day then the two profiles would have been proportional, but instead the screenshots show the magnifying impact of the higher unit rates in the four to seven PM window with equivalent consumption to the late afternoon resulting in rather higher costs.
I should emphasise however that my average unit rate is very low as I usually have much higher consumption in low cost periods than I do in high cost periods.
One of my recent electricity bills had an average of 3.05 p/kWh ex-VAT. Half-hourly rates varied between around minus 10 p/kWh (I.e. I was paid to use electricity) to plus 25 p/kWh. A low average price was achieved by shifting electricity consumption to when the price was lowest.
My next step is likely to be to use the API to get the real time household load for load management as an increasing number of electrical consumers (potentially a second car charger) risks overloading my supply fuse if all loads were on simultaneously.
For some time now my Home Energy Management System (HEMS) has been managing many of my domestic electricity consumers including:
home storage battery
water heating via immersion heater
The overarching strategy has been to:
maximise use of my own solar energy (rather than consume from the grid)
prioritise consumers for best value within the constraint of available solar generation
when power is needed from the grid to optimise the purchase price by shifting consumption to the cheapest periods (my price changes every half an hour)
For some consumers such as car charging and water heating this has resulted in those consumers switching between two modes:
self-consumption when they are enabled to use the ‘free’ electricity from my own solar panels (subject to device prioritisation) with some proportional control
boost when they run at full power drawing some if not all of the required power from the grid at the lowest available price
However the quite exceptional stress being put on the grid in the UK has prompted some expansion of capability. It’s normal once in a while that my electricity prices go negative, commonly caused by the overlap of large amounts of renewable electricity on the grid (e.g. excess solar output on sunny afternoons and/or high windfarm output due to wind conditions) and low demand (summer nights without heating demand, summer afternoons, bank holidays etc) which is exacerbated by the current corona situation. The current corona situation has made this more common and the plunge pricing events more extreme with multiple hours of negative pricing today some of which are into double digits (which I think is unprecedented). For car charging and water heating this has resulted in a new control mode.
The new control mode is a disabled status where the the device neither self-consumes nor is forced on. In the short term this increases export to the grid, but the mode is intended to help balance the grid disabling consumption for now to enable more consumption when the grid is under most extreme stress from an excess of generation over demand. Or to think of it another way, it passes up the opportunity to use free electricity now in order to be paid to use electricity later, responding to the price incentive to support balancing the grid.
Thus on a normal day these devices switch half-hourly between self-consumption (free electricity) and Boost (paid for electricity), but on a price plunge day then they switch half hourly between the new disabled state (no electricity, increased export) and the existing Boost (but now paid-to-consume electricity).
The full availability of modes is thus:
mode / other
Make and model
Mixed API and relay
Mode – Boost
Powervault “Force charge”
via HEMS relay (Ch 1)
Via HEMs relay (Ch 2) to Immersun “External Boost” input.
Mode – Self-consumption only
Priority #1 Powervault “Charge only”
Priority #2 via Immersun relay output (Ch 3)
Priority #3 Immersun default behaviour
N/A – Available in API but not used by HEMS.
via HEMS relay (Ch 4)
Immersun “Holiday” mode via API
Mode – Both self-consumption and self-discharge available
Powervault “Normal” via API
N/A – No reverse flow from car to home available (not V2X capable)
Major device modes available to HEMS
We should thus be better equipped to support the grid in the current circumstances.
The chart above shows the resulting behaviour. In particular the large negative currents through the morning to early afternoon show that much of the normal self-consumption has been disabled. Then from mid-afternoon the import shows the effect of enabling multiple consumers simultaneously. Here the behaviour of the car charging and water heating was boosted as this point, while the dishwasher’s and washing machine’s existing behaviour added to load.
Not that it has anything to do with the revised controls, but the spikiness of the export during the day shows the highly variable nature of the export through the day being a function of both variable generation through passing clouds and variable consumption with kettle boils and the like. Thus it’s important that consumers for self-consumption have automated closed loop control since manual control of an immersion heater or car charger to achieve high self-consumption with minimal import would have been almost impossible even with a level of human intervention wholly at odds with the scale of savings achieved – small savings hour-by-hour add up over the day, weeks and months but their value is relatively tiny compared to the labour to attempt equivalent control manually.
Here is a similar import-only half-hourly view from the smart meter WAN side:
The screenshot above clearly shows those periods of import being targeted at the periods with the most negative prices. Since different consumers need power for different periods of time (for example it takes about 7 hours to charge the battery, but only 2 or 3 for a tank of hot water) then consumption rises as cost falls. My consumption-weighted average cost was -6.22 p/kWh yesterday. However the same price point during the day or night has delivered different consumption from the grid as the output of the solar panels must be consumed before consumption from the grid starts. We are still some way from the point where it becomes economically attractive to disable the solar panels.
Finally the above image shows the last 28 days of electricity average cost in p/kWh. Although some other days included some periods of negative pricing, the quite exceptional pricing yesterday is amply illustrated with the combination of extreme prices and the new load management mode delivering revenue (i.e. negative cost) of 82.4 pence through consumption of 13.252 kWh at an average 6.22 p/kWh.
This is something of a zero sum game in terms of consumption as I don’t artificially increase consumption to improve income – such as leaving the oven on with the door open during a summer day – this is all about shifting consumption that would have happened anyway. However we have not only supported the grid when it is most stressed but also reduced our energy costs significantly (to the point of being significantly negative) by moving consumption from being predominately self-consumption (i.e. from our own solar panels) to being predominantly grid consumption.
For some time now I’ve been thinking about creating a real time display which pulls together data from a variety of sources around the home to provide an overview of what’s going on without the need to visit multiple web pages or apps. Until the last 10 days or so that involved little more than thoughts of how I might evolve the existing immersun web page with more content (I don’t have the skills to write my own app), but then about 10 days ago I saw an online gauge that someone else had created to show energy price and inspiration struck. Ten days later I have my monitor working, albeit not complete:
The monitor pulls together information from:
My electricity tariff for p/kWh
My immersun for power data (to/from: grid, solar, water, house)
My storage battery for power in/out and state of charge
My HEMS for electricity cost thresholds between different battery modes.
The gauge consists of two parts: (i) an upper electricity cost part and (ii) a lower power part.
The upper electricity cost part is effectively a big price gauge from 0 p/kWh to 25 p/kWh with a needle that moves each half hour as the price changes. It has various features:
The outer semi-circular ring (blue here) shows today’s relationship between battery mode and electricity price. Today is very sunny, so no electricity was bought from the grid to charge the battery, and this part is all blue for normal battery operation. If the days was duller and electricity was to be bought to charge the battery, then two further sectors would appear:
a dark green sector from zero upwards showing the grid prices at which the battery would be force charged from the grid, and
a light green sector showing when the battery is not permitted to discharge but may continue to charge from solar.
In inner semi-circular ring (green / yellow / red here) currently just colour-codes increasing electricity price, but will be used to show today’s prices at which car charging and water heating are triggered from the grid.
The current price per kWh is taken from Octopus’s price API, while the current cost per hour is derived both from this and the grid power from the immersun.
The needle grows from a simple dot indicating the price per kWh only when no power is drawn from the grid to a full needle when the electricity cost is 10 pence per hour or more.
The lower power part is effectively a power meter ranging from 5,000 Watts of export to the left to 5,000 Watts of import to the right. It updates every few seconds. It has various features:
The outer semi-circular ring (orange /maroon / green here) shows how power is being consumed:
orange – shows consumption by the house less specified loads
maroon – shows battery charging
blue (not shown) – shows water heating
green – shows export to the grid
The inner semi-circular ring (yellow here) shows the source of power. The sum of the sources should equal the sum of the consumers. The sources are:
maroon (not shown) – shows battery discharge
yellow – shows solar power
red (not shown) – shows grid power
The power value shows the net import or export from / to the grid, while SoC refers to the state of charge of the battery (0-100%). The combination of import power and electricity price gives the cost per hour in the top gauge.
The needle position shows net import (to the right) or next export (to the left). The needle should thus be to the left of the green sector, or to the right of the (unseen) red sector. Needle length show the full power being handled and is thus proportionate to the angle of the sector including all the colours in the lower gauge and extends from 0 to 5 kW.
The gauge scales to fill the smallest of screen height or width and translates to be centrally positioned regardless of screen size. My intention is to display it on an old mobile phone as an energy monitor, but I can also access it on any web browser on any device within the home.
As regular readers will recall I’ve recently updated my battery controls with solar predication. For some years now my battery storage has automatically charged during sunshine and later discharged in the evening, and over the last year I added the HEMS to manage buying electricity including for the battery when the electricity cost was cheapest as there’s little solar in the winter, but now the HEMS has the ability to automatically adjust how much power is bought from the grid depending on how much output is expected from the solar panels in order to deliver a charged battery by the end of the day when the electricity price rises significantly. The battery charging can thus swing from all solar to all grid and all points in between entirely automatically based on the solar forecast.
Now however I’ve also added the capability to adjust when the wet goods – dishwasher and washing machine – run as a function of expected solar panel output.
As with the battery controls, I don’t attempt to match generation half hour by half hour with device operation because of uncertainty around how precisely a solar forecast from the evening before will match actual solar production up to twelve hours later. Instead my solar algorithm extracts the number of hours of significant solar production and the earliest start time of that production. For the battery that information is used to modify the number of hours of battery charging to be bought from the grid and the end of the window during which those hours of charging may be bought. Now I’ve updated the controls for the wet goods similarly.
The existing wet good controls look to start the machine at the time when the electricity cost for the cycle is cheapest within a time window set by the user, so the user defines the earliest and latest acceptable start times and the algorithm finds the cheapest start time within that window. The updated wet goods controls assess the number of hours of solar charging available and if both (i) the window exceeds a certain size and (ii) the user’s time window includes the solar window then the user window is narrowed to start at the start of the solar window. The resulting start time between the start of the solar window and the end of the user window may not be absolutely the cheapest grid cost but my assumption is that the solar contribution (which could be up to 100%) will in practice make this the cheapest grid cost as any power needed from the grid will be at a relatively good price. If this start time in the solar window does not reflect absolutely the cheapest grid cost then a second start time may also be identified which is the cheapest grid cost.
In the above example the start times are both within the solar window and the cheapest energy price, so no cheaper alternative is also offered. Start times within the solar window tend to be in the afternoon as the grid energy costs are lower. This also improves the probability that the battery may briefly discharge if the total load exceeds the solar output. However in the above example the energy price is so low that the battery is also force-charging (it’s dark green) so any surplus demand will come from the grid.
In the above example the cheapest time to run each cycle from the grid is at night, although given the availability of solar during the day then any small saving in grid costs at night is very likely wiped out by the ability to run some (if not all) of the cycle during the day from the solar surplus rather than the grid. Both start times are available – the absolutely lowest grid cost and the lowest grid cost during the solar window.
It’s probably also worth mentioning the implications of running the appliance alongside the battery status being different colours:
Blue battery and state of charge being at or over 80% – appliance is prioritised over battery allowing battery to discharge to meet appliance needs as required – least likely to draw anything from grid (but likely highest cost to draw from grid)
Light green battery or (blue battery and state of charge being below 80%) – appliance is prioritised over battery charging alone (so battery may not discharge to support appliance) – a little more likely to draw anything from grid (but likely mid cost to draw from grid)
Dark green battery – battery charging and appliance are equal priority – most likely to draw something from grid (but likely cheapest cost to draw from grid, even to the point of being paid to draw from grid at times)
Today we are increasingly reliant upon the internet. During the current crisis when many are confined to home the internet provides opportunities unimaged to previous generations. Whether it’s online shipping, entertainment, using zoom or other online meeting tools to maintain family, society or church connections; the internet provides the means to supply us with our material needs, entertainment and social contacts in ways unimagined previously.
However our home, and other smart homes, may be particularly vulnerable should the internet become unavailable either in general or a particular service upon which we rely becomes unavailable. I thus thought that for this post I’d reflect upon the services our home relies upon for normal operation and what the impact of their absence would be. I will also reflected upon other failures of the service where such failures have been previously observed.
Effect if unreachable
effect of other observed failure
In unreachable when daily schedule being generated then schedule assumes no solar production like mid-winter.
On April 7th I observed several hours around the middle of the day when there were no forecasted or estimated actuals reported, but the service continued to accept our measured readings. No impact on HEMS operation.
Agile electricity price data.
If unreachable when daily schedule being generated then schedule carries over from prior. Schedule may be generated by manual initiation of the script when the required data becomes available. If necessary operation of the car charger PLC could be suspended causing the car to charge at full power immediately and the car’s own timer used to set operating hours if needed.
If tomorrow’s price data is not yet published (i.e. it’s overdue), then the optimisation is performed using the most recent day’s price data. Since today will no longer be within the dataset provided by the solar forecast (as it’s no longer in the future) then no solar window will be found and so all the required number of hours will be bought assuming today’s (not tomorrow’s) prices, so the battery will still be charged but at a sub-optimal cost. The optimisation can be started manually once the overdue pricing data is available.
If the real time generation data wasn’t available at all then solar car charging would be disabled. Car charging based on bought electricity price would continue.
I have observed occasions when the real time data ceases to be updated (presumably because communications between immersun and cloud is lost) which then throws out the upload to the solar forecasting and the disabling of solar car charging at low generation levels. The forecasting site doesn’t seem to be phased by some erroneous data having never yet dropped below 0.96 correlation. The solar car charging has once been enabled later than it should which had some impact on battery state of charge, but was disabled slightly later via the immersun relay output.
An API is also used to switch the immersun into or out of holiday mode on days with significant periods of negatively-priced electricity. Unavailability of the API could leave the immersun locked in holiday mode and thus completely unable to heat water, or not in holiday mode when it should be causing self-consumption to continue when export would be more optimal to allow for negatively-priced import later. Vacation mode may be enabled or disabled manually via the immersun’s front panel to mitigate.
Lack of availability of the API would disable the ability to switch operating modes. This would leave the Powervault either charging from solar only or force charging, and may or may not permit discharge. The unit can be reset to normal state via repeated operation of front panel button which would disable scheduling but provide default solar operation.
Lack of availability would prevent the wet goods being turned on via API. Operation via the bushputton on the plug and indeed via Apple Homekit should remain.
The heating and other automations are run from Apple Homekit. My understanding is that the local service is provided from the two Apple TVs acting as hubs which should continue. Remote access via devices not in the home would be disabled.
I have access to an API giving data from the HAN side of my smart meter and independent current measurement, but this is not currently used for control so no operational impact.
Impact assessment for loss of different cloud services
Overall I would conclude that there’s no significant issue here. The house would continue to be heated and appliances will still work. Any impact would be around energy consumption and cost only.
Some mitigation could be arrived at by:
resetting the Powervault storage battery to restore default normal operation,
manually starting or stopping vacation mode on the immersun’s front panel, and
suspending operation of the PLC on the car charger to enable car charging.
Last night rather unexpectedly my solcast solar irradiance data tuned itself. I use this data to predict the output of my solar panels and adjust what I buy from the grid in response. I had expected that tuning would happen eventually, but my understanding was that two month’s data was required, not the two week’s data that I had so far supplied.
Prior to tuning my predictions had looked like this..
Although the system is clearly predicting output to a reasonable degree of accuracy, there are two obvious issues:
The orientation of the array from due south seems a little off, as my array starts and ends generation earlier than the prediction suggesting the the orientation should be slightly more easterly in the model.
The peak at a sustained 4 kW is overly optimistic. The panels can generate 4 kW according to my monitoring, but only relatively briefly and certainly not for more than an hour in March.
Nevertheless, the overall identification of better and worse days is clearly working.
However after tuning both issues have been resolved. The measured and predicted curves match very closely. Note that the prior predictions have been updated by the tuning process, so March 26th which is included in both the images looks subtly different. Note also that my control is based on forecasts from the evening before, not the ‘1 hour ahead forecasts’ illustrated above. You may also observe that there is a discrepancy in the morning of April first, where the measured data is zero while the estimated actuals and forecasts are quite healthy, which arises as a result of a temporary failure of the immersun server from where the data is taken.
The ability to do tuning requires an upload of data. I upload generation data continuously at 5-minute intervals (the shortest allowed interval) which may explain the early availability of the tuned results. The script that I use to achieve this takes data indirectly from my ImmerSUN and is modified from this script. I achieved a 99% correlation which is pretty good. Subsequently it seems that the tuning takes place automatically each day.
As winter turned to spring my thoughts for my HEMS turned to thinking about how to adjust the operation of the HEMS in managing battery charging to account for anticipated solar production. Previously the HEMS was configured to buy a preset number of hours of charge from the grid each day, typically overnight when the power tends to be cheapest. Through 2019 as the seasons changed I periodically adjusted this figure to create headroom to store charge from the solar panels later in the day. However I would like to make this adjustment automatically day by day.
Late last year I came across solcast a website that predicts the output of solar panels. The user creates an account, describes their PV array (location, capacity and orientation), and can then download predictions via API.
The orange line shows the predicted output for the next few days, with the light grey area showing the confidence interval from 10 to 90%. As a prediction there’s a degree of uncertainty associated with the prediction as there is with a weather forecast. The 10% line shows that 1 day in 10 the output will be lower than the grey area, while the 90% line shows that 1 day in 10 then output will be higher than the grey area.
My original prediction were based on the orange line (the 50th percentile) where output was equally likely to be above or below this amount. However my risk on an incorrect prediction is not even. If I fail to buy enough power from the grid when the price is low then I risk paying up to 35p/kWh to buy when the price is high, whereas if I buy an unnecessary cheap kWh from the grid I may spend an unnecessary 5p/kWh on average. Thus I decided to take a more conservative position on risk to obtain the lowest cost position. I opted for a 20% risk, so 1 day in 5 I might underestimate my purchase from the grid reflecting the ratio of grid prices high:low. I estimated this 20th percentile assuming a normal distribution on the low side where the 20th percentile = 0.34 * 50th percentile + 0.66 * 10th percentile.
The process is as follows:
download the data in a half-hourly format to be comparable to the half-hourly Agile electricity price data,
calculate the 20th percentile from the 10th and 50th percentiles,
count the number of half-hours in the 20th percentile data above a threshold that provides for charging the battery at full power,
establish the earliest half hour when solar would charge the battery at full power,
adjust the period for buying power from the grid by the period anticipated for solar charging (tomorrow a total of 7 hours is to be achieved by 6.5 hours from solar leaving 0.5 hours from the grid),
adjust the end time for buying power from the grid to align with the earliest hour when the battery can charge at full power from solar.
The result of these calculations can be seen above. One half-hour of lowest price battery charging has been identified overnight to meet the requirement for charging from the grid. Normal HEMS behaviour has also identified several other periods of grid charging at lower cost during the day, but these are not counted towards the target for buying from the grid due to the potential for a double count of solar and grid charging during the day. (In a similar manner there are multiple start times during the afternoon when the dishwasher or washing machine can be run on grid power more cheaply than the optimal overnight start time.)
No explicit command from the HEMS is required to enable proportional charging of the battery from the solar panels when there is a solar surplus as 2 of the 3 used battery operating modes (normal and charge-only) have this capability, and the third mode force charges the battery. If the battery is force charged during solar surplus then the source of the energy will of course be the solar panels, but any shortfall will be met from the grid.
Proportional to solar surplus
Proportional to shortfall
Used at high grid prices (today discharging enabled at > 8 p/kWh)
Proportional to solar surplus
Used at mid grid price to save stored energy for period of higher grid price (between 5 and 8 p/kWh today)
Used at lower grid price to buy from grid (today buying at < 5p/kWh). If low price occurs at time with reasonable solar generation then use of solar output will happen automatically, with only any shortfall coming from the grid.