Efficient High Heating Value estimation using Latin Hypercube Sampling and Artificial Neural Network-based approach.

Area of Science:

• Environmental Engineering
• Energy Recovery
• Waste Management

Background:

• Accurate estimation of Municipal Solid Waste (MSW) High Heating Value (HHV) is crucial for maximizing energy recovery in Waste-to-Energy (WTE) systems.
• Uncertainty in elemental composition of MSW poses a challenge for precise HHV prediction.

Purpose of the Study:

• To develop and validate a robust method for forecasting the HHV of MSW.
• To identify key elemental parameters influencing MSW HHV.
• To enhance the accuracy and reliability of WTE system operations.

Main Methods:

• Integration of Artificial Neural Network (ANN) model with Latin Hypercube Sampling (LHS) for uncertainty handling.
• Utilizing elemental composition (Carbon, Hydrogen, Nitrogen, Sulfur, Oxygen) as input parameters for the ANN model.
• Employing a synaptic weight approach to determine parameter significance.

Main Results:

• The proposed ANN-LHS model achieved a minimum Mean Absolute Percentage Error (MAPE) of 2.18%.
• Statistical performance metrics include MSE of 0.012, RMSE of 0.107, and R² of 0.767.
• Carbon (C) was identified as the most significant parameter influencing the HHV of MSW.

Conclusions:

• The combined ANN-LHS approach provides an accurate, economical, and robust method for MSW HHV estimation.
• This methodology strengthens MSW management and WTE conversion processes by addressing compositional uncertainties.
• The findings support improved efficiency and effectiveness in waste-to-energy applications.