-
Notifications
You must be signed in to change notification settings - Fork 1
Home
With new legislation in The Netherlands, starting from 2020, it is likely that solar panel owners in the Netherlands will receive less money for energy that they feed into the electricity grid then what they pay for energy that they retrieve from that same electricity grid.
Currently, solar panel owners can use the electricity grid as one big solar battery with a 100% efficiency. The value of the ‘generate’ meter will be deducted from the ‘usage’ meter, meaning that both meters are considered to be equal in value per kWh. However, the biggest portion of the kWh price is (energy) taxes, but also partially usage of the electricity grid. Considering the value of generated and used electricity equal in value can be considered as subsidies. Unfortunately for solar panel owners the Dutch government wants to use those subsidies to develop other opportunities for clean energy, since solar energy is considered to be mature by now.
Since this forum is not meant for political topics, but for technical solutions, let’s move on to counter measures, how to optimize our systems for maximum financial / electrical profit. With a solar system it is inevitable that generation and usage of electricity is not balanced. A battery, like for instance the Tesla Powerwall, will be a good solution to store the surplus of generated energy locally ‘behind the meter’. There where the solar panel owner is in charge.
The question is, what size battery should I buy? It is not recommended to be too small, it will be full and empty much too fast. Not too big either, it doesn’t make sense to fill the battery on a sunny summer day to the top, use it for your nightly consumption and find the battery the next morning still half full, leaving only half the battery available for charging on day 2. Ideally the battery tips the ‘full’ and the ‘empty’ point each day without overshoot in both directions.
Well, Domoticz can be used for making the right choice. With 3 virtual devices the presence of a solar battery behind the meter can be simulated. It is up to the user to tweak the capacity of the virtual battery to find the sweet spot for max use of the battery through the year. In my case I started with a 4000 Wh virtual battery. Already in the spring I saw that the battery reached ‘full’ and ‘empty’ point on a lot of days in a row. I therefore increased the battery (for free 😉 ) to 5000Wh, in order to store more solar energy and consume from it as well. In my case the 5000 Wh battery works much better than the 4000Wh one. In the attached graph it can be seen that the ‘purchase’ of the additional 1000 Wh was done in the middle of August of this year.
The script doesn’t take into consideration the cost/reward for a bigger/small battery. The meter logs can be exported to Excel for more precise financial calculation.
An existing energy meter in Domoticz. This may be a P1-meter or 2 seperate energy meters for consumption from the grid and production to the grid.
3 virtual devices need to be added
- Custom sensor that only displays the current value of the solar battery: important to see min, avg and max values over the days
- P1-meter that shows within 1 device power and energy usage of the solar battery in both directions
- Electricity meter that shows the lost power and energy due to a full and empty battery
2 Domoticz user variables need to be added
- The capacity of the the simulated battey
- The maximum power of the simulated inverter for the battery
- Download the script
- Place the script in the folder: \domoticz\scripts\dzVents\scripts
- Update the script regarding the self-chosen device names (as is explained in the script text)
- Choose to enable / disable counting 'lost energy' when energy is consumed while the battery is empty
- Choose a script interval: some P1-meters update every second, others every 10 seconds. For accuracy it is not important to update the battery devices with a high frequency
Although a virtual battery is 'free', a physical one isn't, both in size and cost. In countries with a clear winter and summer season and PV as the energy generator, it might be better to choose a smaller battery:
- Winter months: PV generates only little energy. Most of this is consumed instantly, only some can be stored in the battery. In these months there won't be enough energy generated to even fill the battery
- summer months: PV generates lots of energy, but the energy consumption is rather low: less lights on due to 'longer' days. Less usage of appliances due to spending time outdoor, less heating related usage, summer holidays away from home etc. The battery will be full in a short time, but it won't get emptied over night. The next day it can only be 'topped up'. Think about it, in this situation it doesn't make sense whether you have 5kWh battery, or 50kWh. That next day you can only 'top it up' with the same amount!
Battery inverters cost money, the bigger, the more expensive. Check peak usage (peak production is usually known, based on the installation) if it makes sense to cover all usage by the battery or use the grid for these peaks. After all, a small peak won't translate into a lot of kWh and won't cost a lot either.
Normally the situation is that the more measuring points you have, the more accurate the results are. The flip side is that it takes more CPU-time and database capacity to store all information.
For the solar battery, 2 things have to be taken into consideration:
- When the script interval is small, every usage and consumption Wh will be 'caught'. With a higher interval this will be missed, but check out this example: with a small interval it is measured that in the 1st interval there is a usage of 10Wh. In the second interval there is a return of 10Wh. The result is 0Wh. With a bigger interval the result would be the same. Only when the battery is nearly full, or nearly empty, it would make a difference for a moment in time: the produced / used energy would just fit / not fit the battery. The results will be minor, since most times the consumption or production is quite stable and than passing the full or empty mark of a battery will be counted correctly anyway
- P1-meters only output in Wh. This is the smallest amount. When a small 'virtual battery inverter' is used, there is a chance that with a small script interval there will be inaccuracies. A small example: a 1800W inverter will only deliver 0,5Wh / second (0,5Wh/sec x 3600sec = 1800Wh). With a 1 second script interval, the 1st second will register 1Wh, the 2nd second 0Wh, the 3rd second 1Wh etc. By increasing the script interval, the accuracy will increase: 10 second interval will give 5Wh (= 5 +/-0.5Wh accuracy = +/-10%), a 20 second interval reduces this to 10+/-0,5Wh = +/-5%
Change the log level within the script to LOG_ERROR to have the least amount of logging
Change the dzvents settings in the Domoticz settings (Tab 'Other')