What if you had a “crystal ball” for farming?

As farmers, we have to always be prepared for the future. For thousands of years, we have relied on our gut feeling or instincts to guide our decisions based on our past experiences. In this new era of Agriculture, equipped with the extraordinary power of computers and AI technologies we are changing the way we farm. The following paper describes the preliminary development of a powerful system that utilizes deep learning and mathematical models to predict future concentrations of pH. While small, this and other contributing works are driving us towards the future of farming.

“Aquaponic systems provide a reliable solution to grow vegetables while cultivating fish (or other aquatic organisms) in a controlled environment. The main advantage of these systems compared with traditional soil-based agriculture and aquaculture installations is the ability to produce fish and vegetables with low water consumption. Aquaponics requires a robust control system capable of optimizing Fish and plant growth while ensuring a safe operation. Intending to support a control system, this work explores the design process of Deep Learning models based on Recurrent Neural Networks to forecast one hour ahead of pH values in small-scale industrial Aquaponics. This implementation guides us through the machine learning life cycle with industrial time-series data, i.e. data preprocessing, feature engineering, model architecture selection, training, and validation.”

Read the full pdf

Cardenas-Cartagena, J., Elnourani, M. and Beferull-Lozano, B., 2022, March. Forecasting Aquaponic Systems Behaviour With Recurrent Neural Networks Models. In Proceedings of the Northern Lights Deep Learning Workshop (Vol. 3).

FAQs

Q1: How often should aquaponic systems be forecasted using RNN models?

The frequency of forecasting depends on the system's dynamics and goals. Regular forecasting, such as daily or weekly, is common for optimal results.

Q2: Can RNN models adapt to changes in aquaponic system parameters over time?

Yes, RNN models are designed to adapt to changing parameters over time, providing flexibility and accuracy in forecasting as the system evolves.

Q3: Are there privacy concerns with the data collected for aquaponic forecasting?

Ensuring data privacy is paramount. Implementing secure data storage, anonymization practices, and obtaining user consent addresses privacy concerns in aquaponic forecasting.