The enclosed datasets are generated from freshwater aquaponics catfish ponds. The datasets are generated automatically at 5 seconds intervals using the following water quality sensors driven by the ESP 32 microcontroller: Dallas Instrument Temperature sensor (DS18B20), DF Robot Turbidity sensor, DF Robot Dissolved Oxygen sensor, DF Robot pH sensor V2.2, MQ-137 Ammonia sensor, and MQ-135 Nitrate sensor.
The project is funded by the Lacuna Award for Agriculture in Sub-Saharan Africa 2020 under the management of the Meridian Institute Colorado, USA.
The datasets and results in this section were sensor readings from the end of May to the end of June 2021. All sensor units were not deployed at the same time. Unit 1 (for Pond 1) was deployed 3 weeks earlier than the rest, so it has generated over 50,000 datasets while the others are about 20,000 as at the time of this report. The datasets are downloaded at intervals, cleaned, and labeled. The datasets are also downloadable from Kaggle.com.
|Pond Number||Pond Description||Link|
|1||IoT Pond 1||Download|
|2||IoT Pond 2||Download|
|3||IoT Pond 3||Download|
|4||IoT Pond 4||Download|
|5||IoT Pond 5||Download|
|6||IoT Pond 6||Download|
|7||IoT Pond 7||Download|
|8||IoT Pond 8||Download|
|9||IoT Pond 9||Download|
|10||IoT Pond 10||Download|
|11||IoT Pond 11||Download|
|12||IoT Pond 12||Download|
This dataset is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
The datasets can be used for machine learning tasks of prediction and classification. For example, each sensor data can be modeled against times of the day, against other sensors, etc. The predictions may include determining optimal values of sensors at given times in a day, rate of water change, the best times to feed the fishes, feed consumption rate, etc.