Hi @Brettus_Maximus . Thanks for joining the group. If you have an Octopus Home Mini I can include you in the early alpha testing if you have any interest in that…
@David_Piper sure, I didn’t want to bother you or anything but I’m more than happy to test if I can help. Also happy to let you get it done when it’s done if you had too much “help” 
I’ve got a home mini on intelligent go (relatively simple use case)
Hi. I do have a mini.
Agile as you probably know is pricing dependant on wholesale cost. The prices for each day are sent out 16:00/16:30 each day. There are several services which then display the costs in a graph or colour coded charts. Blue is usually negative, green is usually below 20p then yellow and red as cost increases. If these can be put into an insight graph that would be awesome. Im sure the info is all off the octopus api
Sorry if this is stuff you know! I cant really offer much more lol
That sounds like it will work with the existing device definitions. Though I have worked with a tariff that has both positive and negative prices (I have separate tariffs for import and export). But the calculations are symmetrical so it should be fine. Might need to think about re-titling some of the capabilities to be better descriptive - maybe renaming the device too to reflect the ambiguity about Import and Export (Agile might be good 
)
Some screen shots from my Import Tariff Device (click ⯈ to view)
Device capability values on the left. Pricing insight on the right.
Remember insights are backward looking rather than forward looking - this might not be what you want.
We will need to define an algorithm for price-banding (represented by the colours you describe on the Octopus chart) - this is an area where your input can be really helpful. This is a desirable feature that I have already identified (see post 25). The device has full insight into tomorrow’s prices so we can band each price-slot based on the full days pricing profile. Then we can have flow-cards that get triggered by a specific price band - triggering flows that configure your electricity use according to band. I think that will meet your need very well.
Banding
For banding, how about this as a first attempt? Compute the range of prices in the day (say from -12p to +48p - a 60p range). Then each quartile becomes colour (-12 to +2.99; +3 to +17.99; +18 to +32.99; +33 to +48). This scheme will also work with cosy octopus currently 13.58 - 41.54 - a 28 range so [nearly] 7p quartiles. 13.58 to 20.57, 20.58 to 27.57, 27.58 to 34.57 and 34.58 to 41.57. The standard cosy rate of 27.69 will lie in the yellow band which feels appropriate.
1 Like
Version 1.0.4 Alpha Testing
Version 1.0.4 has been made available to the Alpha Test group. If you want to join the early evaluation, let me know and I will add you to the group.
From a device point of view, the app is complete in alignment with initial plans. There are additional Octopus functions that can be supported that will need further devices to be defined.
For now the development focus is moving from devices and capabilities to flow cards. This will enable the data inside the existing devices to control devices in the home. For example, if the Unit Price goes down then expensive devices may be turned on; or turned off when the Unit Price rises.
Hi
I would appreciate if I could be added as I would love to test this out
Thanks
Hi I would be interested in helping test this? I have a home mini and am on IO go if that ticks any boxes?
I would be very interested to test this David.
I have Intelligent Octopus Go + Export, & a Mini.
Progress Update
It’s been a few days since I posted anything here. Things are progressing well. The small alpha testing group has uncovered some bugs (not a surprise). These are mostly fixed (one fix is still being tested). There is one capability (projected bill) to be implemented. I also want to clean up the code (eliminate redundant code; improve factoring across classes; remove debug logging). The app has been published in test mode in the Homey App Store so that it’s easy to install.
Once these hurdles have been passed I will publish the link to the app here so it can be tested more widely and close down the alpha testing group.
2 Likes
Nice Insights
Here are some nice insights from the Energy Account device. On my account a new Invoicing Period started midnight on the 21st. The charts each show the last 24 hours (21st-22nd)… First chart is Period Import kWh, second is Period Export kWh. Lastly is the Period Total Bill (so far). You can see that early on all the energy is imported (export flatlines) - and the bill amount increases. At midnight there is a vertical increase in the bill - the fixed daily charge is added. Then at 10am export starts and the bill amount starts to decrease.
Nothing more than you would expect, but it’s great to see the dynamics working out for real…
1 Like
Finally got round to installing this. Pulls in all the data from my account which is on the tracker tariff. I dont seem to get any energy usage coming through though.
Good to have you installed 
. Not sure what you mean by “energy usage coming through” - can you post some screenshots of your tariff devices, please? I am pretty sure you have an Octopus Mini - otherwise you should not have been able to instantiate any devices. If that’s not true, please let me know because you will have discovered another “feature” inserted by me 
.
Other than that, I am bemused, because having read the account data (that you imply is working), the only thing that app does is read the Octopus Home Mini data once a minute and use this to update the device capabilities… Weird.
Hi @J273. Nice observation. I too am seeing this behaviour today. Will log an Issue and investigate. Thanks for the heads-up. Logged as Bug #7.
Fix being tested.
1 Like
A fix for Bug #7 has been published as V1.0.18
Progress Update
Continuing to make progress with the small alpha-test group. Usual frustrations of debugging at a distance and without real sight of the sheer variety of data that the App is having to handle. Also the stupidity of enthusiasm - being tempted into releasing code too early (even for Alpha).
Lots of lessons learned about make the code robust (data from the API, is it there at all?, if it is there, is it a null?, if it’s not a null then is it an array, if its an array then does it contain any data elements, if it contains data elements are they the ones the App wants…)
There is one more major feature that I want to integrate before we move into a more public test with a potentially wider audience. Intelligent tariffs have the unique feature of Flexible Dispatches (basically scheduled lumps of cheap energy at expensive times to help charge an EV, for example).
If Flexible Dispatches can be integrated with appropriate flow cards then the App will be a good baseline on which other features can be built. On the future list will be Free Energy and Energy Saving Periods.
I would be very interested to test this David.
I have a Mini and would like to intergrate with homey pro
1 Like
Hi @Matt_Tugwell - thanks for responding. I will add you to the Alpha Test group. Before I do that can you let me know which tariff(s) you are currently on? Many thanks David.
I’m in the flexible octopus tariff
That’s great, I do not believe we have any other testers on Flexible Octopus, so an interesting proof of compatibility. I will add you to the group now…
