Researchers are pioneering new methods in wastewater treatment from land-based aquaculture, a sector crucial to meeting increasing global fish demand. By 2030, aquaculture is expected to supply 62% of all consumed fish. The industry, however, produces significant effluent that includes suspended particles, nitrates, ammonia, and phosphorus, posing environmental challenges. A recent study, explores the potential of artificial intelligence (AI) in enhancing the electrocoagulation/flocculation (ECF) process. This research highlights AI’s role in improving this vital environmental cleanup technique.
Electrocoagulation and flocculation unveiled
This study delves into the mechanics of the electrocoagulation and flocculation processes, crucial for wastewater treatment in aquaculture. Electrocoagulation, an electrochemical technique, uses electrical charges passed through a metal anode (such as iron, aluminum, copper, or stainless steel) to release metal ions. These ions act as coagulants, essential in destabilising and aggregating particles.
Simultaneously, flocculation aids in clustering these destabilised particles, forming a cohesive approach to remove contaminants from aquaculture effluent. Among various electrodes, aluminum is notable for its cost-effectiveness and availability. Operational variables in the electrocoagulation process can be adjusted, as demonstrated in Figure 1. The combination of electrocoagulation and flocculation is efficient, environmentally friendly, fast, and reliable. It ensures safety with minimal sludge production. This method is emerging as an effective solution for treating complex wastewater from fish farms.
AI at work as the silent architect: Modelling and optimisation
This research stands out for its integration of AI to enhance the ECF process in wastewater treatment. The study utilises two AI techniques: artificial neural networks (ANN) and adaptive neuro-fuzzy inference systems (ANFIS). Researchers have demonstrated the exceptional predictive capabilities of Artificial Neural Networks (ANN) and Adaptive Neuro-Fuzzy Inference Systems (ANFIS) in predicting treatment efficiency.
AI emerges as a silent architect, revolutionising aquaculture waste cleanup. By integrating advanced technologies (particularly AI), this breakthrough study paves the way for a sustainable eco-friendly future in the aquaculture industry, offering a promising solution for efficient and reliable aquaculture practices.Chinenye Adaobi Igwegbe
Comparing models: RSM, ANN, and ANFIS go head-to-head
Scientific inquiry is often about pushing boundaries, and the researchers did not shy away from this ethos. A comprehensive comparison of three modelling approaches—Response Surface Methodology (RSM), ANN, and ANFIS—was conducted. Surprisingly, ANFIS emerged as the front-runner (with R2: 0.9990), outperforming both RSM (R2: 0.9790) and ANN (R2: 0.9807) in terms of accuracy.
The statistical analysis, including metrics like Average Absolute Deviation (AAD) and Root Mean Square Error (RMSE), pointed to ANFIS as the most reliable predictor. It demonstrated a strong correlation between experimental and predicted values, indicating a more effective and reliable approach for treating aquaculture effluent. This not only solidifies its role in the scientific realm but also paves the way for advancements in water treatment within the context of aquaculture.
Optimising the process: ANFIS-GA and ANN-GA take the lead
The study highlights the use of Genetic Algorithms (GA) to optimise the ECF process in wastewater treatment. The application of ANFIS-GA and ANN-GA technologies led to high Turbidity Reduction removal efficiencies, achieving 98.98% and 97.81%, respectively. The ANFIS-GA optimised conditions were pH 4, current intensity 3A, electrolysis time 7.2 minutes, settling time 23 minutes, and temperature 43.8°C. Under these conditions, the effluent’s turbidity was significantly reduced from 404 NTU to 31 NTU, well below the World Health Organization’s limit of 50 NTU. Additionally, the treatment effectively reduced the organic load and total dissolved solids, along with decreasing the concentrations of calcium, iron, NH3-N, and total phosphorus. This highlights the potential for reusing treated wastewater in aquaculture.
The practical implications of transforming aquaculture effluent
Researchers have made significant strides in enhancing the treatment of aquaculture effluent, with their study revealing optimised conditions for this process. Utilising AI-guided ECF methods, they achieved high efficiency in Turbidity Reduction, surpassing regulatory standards. This achievement is not limited to laboratory conditions but extends to real-world applications, indicating a potential for reusing treated effluent. Such reuse could substantially reduce the environmental impact of land-based aquaculture, paving the way for a more sustainable and eco-friendly future in the industry.
A greener future for the aquaculture industry
The recent study combining electrocoagulation/flocculation with AI marks a significant advancement for the aquaculture industry toward a more sustainable future. The approach, grounded in scientific precision and a blend of technology and environmental care, offers a promising solution for the efficient and reliable treatment of aquaculture effluent. Focusing on environmental sustainability, the integration of AI with wastewater treatment technologies could be pivotal in establishing new, sustainable practices in aquaculture.
This research article exemplifies the dynamic interplay between scientific innovation and environmental sustainability. It highlights the integration of advanced technology, particularly AI, with the urgent requirement for eco-friendly practices in aquaculture. This combination not only demonstrates progress in laboratory research but also underscores its importance in guiding industries toward sustainability and responsible practices. Additionally, the research suggests exploring algorithms beyond GA for further optimisation of the study.
Igwegbe, C. A., Obi, C. C., Ohale, P. E., Ahmadi, S., Onukwuli, O. D., Nwabanne, J. T., & Białowiec, A. (2023). Modelling and optimisation of electrocoagulation/flocculation recovery of effluent from land-based aquaculture by artificial intelligence (AI) approaches. Environmental Science and Pollution Research, 30(27), 70897-70917. https://doi.org/10.1007/s11356-023-27387-2