My motivation for doing this project was to make life easier to all those aquaponics enthusiasts by automating the system so they can have a better control and monitoring of the system, and to allow them to experiment more easily with different settings to see which ones give best results.Aquaponic setup
I used a 20 liter aquarium as a fish tank and pvc pipes to house 10 small plant containers. Its a small experimental setup but the controller can be used on a much bigger system. It could even run a small aquaponic greenhouse or if you already have an aquaponic NFT system you can just hook it up on that.Controler
The system uses Intel Edison as a brain. Controller is made to be plug and play, you just plug in regular aquarium water pump, lights, air pump and fan to the controller box and attach the sensors. Controller can be hidden somewhere as everything can be viewed and controlled in the webpage on the local area network. Webpage displays the sensor data and enable users to change system settings, sensor data is refreshed using websockets so it has a good response to system changes.
To control water flow through the gullies water pump(AC) and water flow sensor are used. When certain amount of water passes through the gullies system can pause for some time to allow better aeration of the roots. If the water flow stops (pipes clogging or water pump malfunction) an alarm is sounded and LED indicates what is the problem.
You can set which period of the day you want your plants to receive light (eg. 10am to 10pm), if the ambient light is strong enough "artificial" light will be turned off to save power and when ambient light drops under user defined value artificial light will be turned off. There is a second light sensor that measures system light, so if the system is not getting enough light even when the artificial light is on an alarm is sounded with LED indicating what is the problem.
Aeration pump can be turned on or off by system settings, and fan can be turned on if temperature goes over user defined value.
Alternating current circuit is completely isolated from the Intel Edison circuit by using 4 channel relay module with optocouplers.Future Development
- First I'm planning to add a DIY automated fish feeder with adjustable settings (already in process)
- Water level sensor that will sound the alarm if the water level in aquarium drops too low.
- Webcam to take a picture of the plants each day and store it on a SD card, so you can review your plant progress.
- Next is to add pH, DO and EC sensors to have a better overview of the water chemistry. DO (dissolved oxygen) sensor readings can then be used to turn the air pump on if DO value drops too low.
- End goal is to store all these sensor and event data in a central database, that way we can feed it with data from multiple controllers. A web app on top of database that can display all the data and make statistics. I know there are cloud services like Salesforce IoT Cloud and Intel IoT Analytics that can achieve a similar effect, but I want to do it locally so i have full control, and can tailor the webapp (and db) the way I want and make its simple enough for everyday use.
I am very interested in setting up a project like this. I was thinking of using a Raspberry Pi Zero W. With the sensors for water chemistry, and Camera to periodically send a pic of the system. I am working with aquaponic systems in 4 schools right now and would love to be able to monitor the systems at the schools remotely. It would also be very cool for the kids to monitor the system and learn about the importance of data.