Sign in Subscribe
Machine learning

Artificial Neural Networks Used to Optimize Dye Adsorption in Effluents

Strategic Insights

— 1 min read

Research conducted by the Defence Institute of Advanced Technology has found that a novel approach using Artificial Neural Networks (ANNs) can optimize the decolorization of Rhodamine B dye from effluents. The ANNs were used to model the adsorption capacity of sodium hydroxide-functionalized sugarcane bagasse biochar (NaOH-SBB), which demonstrated a strong predictive performance with a high correlation coefficient (R = 0.9531) and determination coefficient (R² = 0.9726). The study aimed to foster the United Nations' Sustainable Development Goals (SDGs) of pure water and robust health.

Key Takeaways:

  • The researchers utilized a Krill Herd algorithm-based Artificial Neural Network (ANN) model to optimize and predict the dye adsorption capacity values of NaOH-SBB.
  • The ANNs demonstrated a strong predictive performance, reflected by a high correlation coefficient (R = 0.9531) and determination coefficient (R² = 0.9726), as well as low error metrics (mean square error: 0.5669, mean absolute error: 0.3884, root mean square error: 0.7542).
  • The study aimed to foster the United Nations' Sustainable Development Goals (SDGs) of pure water and robust health by utilizing a novel approach to decolorize Rhodamine B dye from effluents.
  • The researchers examined the physicochemical characteristics of the as-prepared biochar using advanced characterization techniques, which revealed a maximum mono-layered Langmuir adsorption capacity (q) of 4.8309 mg/g at adsorbate concentrations of 15 ppm.
  • The study concluded that the results indicated a strong correlation between the empirical and ANN-prognosticated results, validating the effectiveness and reliability of the developed model.

Participating Researchers:

  • Neelaambhigai Mayilswamy, Dept. of Metallurgical and Materials Engineering, Defence Institute of Advanced Technology
  • Balasubramanian Kandasubramanian, Tushar Warjurkar, Satkirti Chame, and Saleega Shirin

Statistics:

  • The correlation coefficient (R) of the ANN model was 0.9531.
  • The determination coefficient (R²) of the ANN model was 0.9726.
  • The mean square error (MSE) of the ANN model was 0.5669.
  • The mean absolute error (MAE) of the ANN model was 0.3884.
  • The root mean square error (RMSE) of the ANN model was 0.7542.

Sources:

  • Optimization and prediction of Rhodamine B uptake onto alkali-functionalized sugarcane bagasse biochar: Krill Herd algorithm-based ANN modelling approach. Environmental Science and Pollution Research, 2025.
  • Defence Institute of Advanced Technology Reports Findings in Artificial Neural Networks (Optimization and prediction of Rhodamine B uptake onto alkali-functionalized sugarcane bagasse biochar: Krill Herd algorithm-based ANN modelling approach). Journal of Engineering. July 14, 2025; p 637.

Read more