Artificial Neural Network (ANN) Modeling of Biofloc and Polyethylene Microplastic Effects on Water Quality, Floc Dynamics, and Nile Tilapia Production

نویسندگان

1 Department of Chemistry, Faculty of Mathematics and Natural Science, Andalas University, Padang, Indonesia

2 Department of Chemistry, Faculty of Mathematics and Natural Science, Andalas University, Padang, Indonesia

3 Department of Medical Laboratory Technology, Syedza Saintika University, Padang, Indonesia

4 Department of Chemistry, Faculty of Mathematics and Natural Science, Andalas University, Padang, Indonesia

5 Department of Chemistry, Faculty of Mathematics and Natural Science, Andalas University, Padang, Indonesia

doi
10.48309/ajca.2026.555535.1956
چکیده

Biofloc technology (BFT) has been widely promoted as a sustainable aquaculture strategy, yet its interaction with polyethylene (PE) microplastics remains insufficiently understood. This study examined the combined effects of biofloc and PE microplastics on water quality, nitrogen cycling, protein synthesis, and production performance of Nile tilapia (Oreochromis niloticus) in a closed, zero-exchange system for 50 days. Four treatments with or without biofloc and varying microplastic concentrations were tested, and system responses were modeled using artificial neural networks (ANN). Results showed that biofloc suppressed ammonia, nitrite, phosphate, and sulfate while enhancing nitrate conversion, floc density, nitrogen retention, and crude protein, thereby improving fish growth, feed efficiency, and survival. PE microplastics provided surfaces for microbial colonization that may support nutrient absorption but also interfered with microbial processes, reducing resilience and production outcomes. Treatment B achieved the best performance, while ANN delivered high predictive accuracy. These findings underscore biofloc’s resilience under microplastic stress and its prospects as a predictive, sustainable aquaculture approach.

کلیدواژه‌ها