Optimized Membrane Fouling Prediction and Mitigation for Improved Water Treatment: a Review DOI Creative Commons

Olufunmilola Oyenike Ajayi,

Thabo Falayi

International Journal of Chemical Engineering and Materials, Год журнала: 2024, Номер 3, С. 162 - 180

Опубликована: Дек. 31, 2024

This review article presents recent advancements in membrane filtration technologies, particularly focusing on fouling mechanisms affecting reverse osmosis (RO) membranes. It a comprehensive analysis of various studies conducted over the past two decades, highlighting complexities caused by natural organic matter (NOM), particulate matter, and biofouling. The also examines innovative modelling approaches to predict behaviour, including development Membrane Fouling Index-Ultrafiltration (MFI-UF) method application advanced characterization techniques such as optical coherence tomography (OCT) Near-Edge X-ray Absorption Fine Structure (NEXAFS) spectroscopy. Additionally, it discusses effectiveness pre-treatment strategies, coagulation flocculation mitigating enhancing performance. Finally, integration artificial intelligence (AI) predicting behaviour is highlighted, with emphasis its potential optimize operational parameters systems.

Язык: Английский

Effect of Hyperparameter Tuning on the Performance of YOLOv8 for Multi Crop Classification on UAV Images DOI Creative Commons
Oluibukun Gbenga Ajayi, Pius Onoja Ibrahim,

Oluwadamilare Samuel Adegboyega

и другие.

Applied Sciences, Год журнала: 2024, Номер 14(13), С. 5708 - 5708

Опубликована: Июнь 29, 2024

This study investigates the performance of YOLOv8, a Convolutional Neural Network (CNN) architecture, for multi-crop classification in mixed farm with Unmanned Aerial Vehicle (UAV) imageries. Emphasizing hyperparameter optimization, specifically batch size, study’s primary objective is to refine model’s size improved accuracy and efficiency crop detection classification. Using Google Colaboratory platform, YOLOv8 model was trained over various sizes (10, 20, 30, 40, 50, 60, 70, 80, 90) automatically identify five different classes (sugarcane, banana trees, spinach, pepper, weeds) present on UAV images. The assessed using accuracy, precision, recall aim identifying optimal size. results indicate substantial improvement classifier from 10 up while significant dips peaks were recorded at 70 90. Based analysis obtained results, Batch 60 emerged best overall automatic Although F1 score moderate, combination high makes it most balanced option. However, Size 80 also shows very precision (98%) (84%), which suitable if focus achieving precision. findings demonstrate robustness identification highlighting impact tuning appropriate performance.

Язык: Английский

Процитировано

5

Optimized Membrane Fouling Prediction and Mitigation for Improved Water Treatment: a Review DOI Creative Commons

Olufunmilola Oyenike Ajayi,

Thabo Falayi

International Journal of Chemical Engineering and Materials, Год журнала: 2024, Номер 3, С. 162 - 180

Опубликована: Дек. 31, 2024

This review article presents recent advancements in membrane filtration technologies, particularly focusing on fouling mechanisms affecting reverse osmosis (RO) membranes. It a comprehensive analysis of various studies conducted over the past two decades, highlighting complexities caused by natural organic matter (NOM), particulate matter, and biofouling. The also examines innovative modelling approaches to predict behaviour, including development Membrane Fouling Index-Ultrafiltration (MFI-UF) method application advanced characterization techniques such as optical coherence tomography (OCT) Near-Edge X-ray Absorption Fine Structure (NEXAFS) spectroscopy. Additionally, it discusses effectiveness pre-treatment strategies, coagulation flocculation mitigating enhancing performance. Finally, integration artificial intelligence (AI) predicting behaviour is highlighted, with emphasis its potential optimize operational parameters systems.

Язык: Английский

Процитировано

0