Optimized Artificial Neural Network-Based Mathematical Model and Software Application for Predicting Raw Mix Lime Saturation Factor for High-Quality Cement Production

This study develops LSF predictive models by employing artificial neural networks (ANN) optimized with particle swarm optimization (PSO), Levenberg–Marquardt (LM), and genetic algorithms (GA), using model dataset of two thousand four hundred and sixty data points. Dependable variables selected were lime, silica, alumina, and iron oxide. To enhance the practicality and ease of use, the models were converted into mathematical equations and further integrated into software application. The models' performance was compared using verification dataset of one hundred data points and LM-ANN model presented the best performance and strongly recommended for LSF estimation.本研究以包含2460个数据点的建模数据集为基础，采用经粒子群优化（Particle Swarm Optimization, PSO）、莱文贝格－马夸特（Levenberg–Marquardt, LM）算法以及遗传算法（Genetic Algorithm, GA）优化的人工神经网络（Artificial Neural Network, ANN），构建了LSF预测模型。所选用的可靠输入变量为生石灰、二氧化硅、氧化铝与氧化铁。为提升模型的实用性与易用性，本研究将所开发的模型转化为数学方程，并进一步集成至软件应用中。本研究通过包含100个数据点的验证数据集对各模型的性能开展对比，结果表明LM-ANN模型表现最优，强烈推荐用于LSF估算。