Table 1 Merits and Demerits of the Existing Water Quality Prediction Approaches.
Author [citation] | Methodology | Features | Challenges |
|---|---|---|---|
Nemade and Shah17 | XGBoost and the random forest | ● It makes it possible to control nutrient concentrations in aquaponics farms more effectively, guaranteeing that fish and plants get the right ratio of nutrients for healthy growth | ● There may be errors in this model because of a variety of things, including dynamic systems, biological relationships, and climate |
Cao et al.18 | GBRT and GRU | ● It can increase system effectiveness by regulating factors like pH, temperature, and oxygen concentration by forecasting water quality | ● It can be difficult to gather accurate and adequate data for predicting the water quality of aquaponics pond water, particularly if there are any gaps in surveillance or restricted availability of past data |
Monirul et al.19 | Embedded model | ● It supports aquaponics practices that are both ecologically friendly and economical by reducing the possibility of imbalances in nutrients and excessive release into the atmosphere | ● The precision of water quality estimates can be impacted by outside variables such as variations in fish and vegetation populations, temperature swings, and fertilization availability |
Rodriguez et al.20 | Bagging and Boosting Ensemble Technique | ● It can assist in lowering losses brought on by illnesses of the fish or plants, increasing the financial viability of aquaponics activities | ● It became complex to forecast modifications to water quality with accuracy in aquaponics farms because of the intricate relationships that exist among fish, plants, and microbes |
Karimanzira and Rauschenbach21 | SCADA, ERP and MES | ● It has the potential to simplify the process of monitoring, decreasing the need for manual inspection and enabling operators to concentrate on other elements of system administration | ● It is challenging and laborious to calibrate forecasting models for aquaponics water from ponds, necessitating substantial data gathering and analysis |
Cao et al.22 | GRU and K-Means clustering | ● It enables accurate water control, reducing waste and encouraging effective water usage in aquaponics systems | ● Finding the right algorithms for precise forecasts is difficult because of the absence of established ways of predicting the water quality of aquaponics pond water |
Ahmed et al.23 | LSTM | ● It can aid in the prevention of infections and illnesses between both organisms in systems for aquaponics through early detection of water quality variations | ● The movements of aquaponics farms remain under investigation, and forecasts may not be as accurate due to gaps in the understanding of the various components influencing water quality |
John et al.24 | RNN | ● It lessens fluctuations and gives both creatures a more constant atmosphere by assisting in the maintenance of predictable circumstances inside the system | ● It may have incorrect or erroneous warnings, which could prompt needless actions or system disturbances |
