Publications by authors named "Robert Makomere"
Precision forecasting of spray-dry desulfurization using Gaussian noise data augmentation and k-fold cross-validation optimized neural computing.
J Environ Sci Health A Tox Hazard Subst Environ Eng
March 2024
Article Synopsis
- Perceptron models are vital for pattern recognition and classification, and this study focused on applying artificial neural networks (ANN) to predict the performance of a mini-spray dryer for desulfurization processes.
- The research utilized k-fold cross-validation and trained 21 ANN models on various input conditions, including temperature, pH, stoichiometric ratio, and solid concentration, with synthetic data generated through Gaussian noise augmentation.
- The study highlighted that effective cross-validation and data augmentation are essential for improving neural network performance, with the logsig activation function and 10 hidden cells yielding the best results in accurately mapping data to actual values.
The assessment of response surface methodology (RSM) and artificial neural network (ANN) modeling in dry flue gas desulfurization at low temperatures.
J Environ Sci Health A Tox Hazard Subst Environ Eng
March 2023
Article Synopsis
- The study evaluates the efficiency of a flue gas desulfurization (FGD) system, focusing on sulfur dioxide removal and reagent conversion in dry FGD (DFGD) systems.
- Researchers used response surface methodology (RSM) and artificial neural networks (ANN) to analyze how factors like hydration time and temperature affect the system's performance.
- The results show that a specific ANN model (5-10-2 architecture) produced highly accurate predictions with R-squared values of 0.993 and 0.9986, indicating good model reliability and performance in mapping the DFGD process.