I'm not sure about what kind of sensor, IoT devices, and hardware to use, with water quality monitoring in a number of tanks (real-time) using a machine learning algorithm I hope to collect some advice, here are some requirements:
1- The tank lid must be kept closed, so the sensor, if inside, must be waterproof or able to bear high humidity.
2- The solution must work for years; I don't want a system that breaks every other month. The Arduino needs to be around 4 meters away from the tank.
4- There is a need for real-time monitoring, sampling time of one every 15 minutes is enough. So, the sensors or devices connect to could to send data.
5- There will be a display or some other device to allow me to check easily the water quality of each tank.
The water quality in real-time is very important.
Hardware shouldn't cost more than 400€ (Arduino excluded). The water in the tank is used for "irrigation uses" but not for drinking.
Any hints or help with the code are highly appreciated.
Thanks for the help,
Greetings.
I moved your topic to an appropriate forum category @alexlabuser .
In the future, please take some time to pick the forum category that best suits the subject of your topic. There is an "About the _____ category" topic at the top of each category that explains its purpose.
As as been said ; You need to decide what you mean by water quality ,and what specific things you want to measure .
Then look for industrial sensors for that property - if it is to work long term you’ll need quality industrial sensors .
I mean by "water quality" the degree of water pollution. if the water is clean or polluted. and any type of ML algorithms, but I plan to use Logistic Regression to classify the water degree ( pollution or clean).
I mean by "water quality" the degree of water pollution. if the water is clean or polluted. and any type of ML algorithms to classify the water degree ( pollution or clean).
with regard to monitoring do you have local WiFi or LoRa coverage?
can you place an antenna so that it can transmit/receive signals, i.e. not inside a metal tank
Depending on the type and degree of pollution, polluted water may be fine for irrigation.
You need to be more specific about what can and cannot be in the water and still meet your criteria as being "clean".
Now you have simply replaced one vague term "quality" with another vague term "pollution". There are no sensors for such vague concepts.
Why would this be necessary? Can you not simply base this binary classification of the level of one or more sensors? Do you have a need to include ML in your project in some way to meet some academic criteria?
It could be a multilabel classification (clean, not clean (between), pollution ), I want to use ML to classify the quality of water and then make some discussion based on this value for each tank. each time, we have new samples of each tank, and based on ML, we can classify water quality and decide to continue to irrigation plant or no.
if you have WiFi coverage and require a display have a look at LiLigo T-Display
if has an ESP32 (with on board WiFi, Bluetooth Classic and BLE) and plenty of IO functionality for your sensors etc plus a display
so long as the device has some suitable digital interface (RS232, I2C, SPI, RS485, etc) you should be able to interface to a ESP32 or similar microcontroller
think you need to experiment!
EDIT: have also looked at labkotec systems - not cheap but may give you some ideas
Yep
… water could be polluted by sewage , heavy metals, salt, nitrogen run off , dead animals, algae, micro organisms , and anything else that you can think of .
Which do you want to detect -‘all’ is not an answer ?
This is not yet an Arduino issue - find your pollutant , then research detection methods.
Whether it has Wi-Fi or what sort of display is irrelevant at this time.