Generation of experimental data for model training to optimize fouling prediction

Article Engineering, Chemical

Comprehensive Parameter Screening for the Investigation of Particulate Fouling in Pipe Fittings

Niklas Jarmatz et al.

Summary: Despite the large number of pipe fittings in production plants, they have received little attention in terms of fouling. This research investigates the influence of process parameters and geometry factors on fouling in pipe fittings under isothermal conditions, aiming to improve our understanding of their fouling behavior. The results demonstrate that variations in fitting type and size, run time, and particle mass fraction have the greatest impact on fouling mass, while temperature does not influence the outcome.

CHEMIE INGENIEUR TECHNIK (2023)

Add to Collection

Article Computer Science, Artificial Intelligence

Prediction of the parameters affecting the performance of compact heat exchangers with an innovative design using machine learning techniques

Sinan Uguz et al.

Summary: The innovative compact heat exchanger (CHE) designed and manufactured using metal additive manufacturing technology was studied numerically and experimentally. The results showed that the numerical and experimental results of the CHE are compliant at the given boundary conditions. It was observed that the heat transfer rate of the CHE with lower volume is approximately 47.7% higher than that of standard brazed plate heat exchangers (BPHEs) produced by traditional methods.

JOURNAL OF INTELLIGENT MANUFACTURING (2022)

Add to Collection

Article Thermodynamics

Machine learning algorithms for improving the prediction of air injection effect on the thermohydraulic performance of shell and tube heat exchanger

Emad M. S. El-Said et al.

Summary: Improved prediction methods based on supervised machine-learning algorithms were proposed in this study to predict the effect of air injection and transverse baffles on the thermohydraulic performance of shell and tube heat exchangers. Experimental data was used to train and test four different machine-learning algorithms, with the random vector functional link model demonstrating superior prediction capabilities compared to the other models in terms of outlet temperature and pressure drop values. The root mean square error and mean relative error for RVFL results were significantly lower than SMO, SVM, or KNN.

APPLIED THERMAL ENGINEERING (2021)

Add to Collection

Article Thermodynamics