Artificial intelligence technique development for energy-efficient waste-to-energy: A case study of an incineration plant

The increasing heat value and complexity of waste types in Taiwan's incineration plants have led to reduced facility capacity and utilization rates. Traditional control systems struggle to manage rapid and irregular fluctuations in waste heat values, often resulti ng in poor stability and prolonged response times. This study introduces an artificial intelligence-based heat value prediction and combustion control system that enhances system efficiency and stability without equipment upgrades. The system predicts future waste heat trends, enabling precise operational adjustments. This results in shorter response times, improved combustion stability, and higher energy recovery efficiency, effectively replacing traditional control systems for more accurate waste management. Our system evaluates waste input uniformity to ensure consistent feed and employs a Long Short-Term Memory neural network architecture to predict waste combustion heat values, greatly enhancing combustion stability. The model's R 2 value of 0.96 allows for optimized control parameters that reduce system response times. Field tests show a 70 % reduction in response time and a 20 % improvement in power system stability. Compared to past operational experiences at incineration plants, it significantly improves stable operations and enhances operational stability by 44 %, confirming the practical utility of artificial intelligence in waste management.