A Dynamic Marine Oil Spill Prediction Model Based on Deep Learning

Wang, Renda; Zhu, Zhen; Zhu, Wei; Fu, Xianping; Xing, Shengwei

doi:10.2112/jcoastres-d-20-00080.1

Cited by 7 publications

(3 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Leak detection has been the major subject of many studies. Wang et al [7] performed the use of a real-time processing process obtained through video frame monitoring of oil spills, such as midpoint, area, diffusion rate, edge circumference features, etc. They used a recurrent neural network's long-term memory network to find the relationship between features and influencing factors.…”

Section: Related Workmentioning

confidence: 99%

Deep Learning Approach for Oil Pipeline Leakage Detection Using Image-Based Edge Detection Techniques

Obaid,

Hamad

2023

JESA

View full text Add to dashboard Cite

Natural gas and oil are one of the mainstays of the global economy. However, many issues surround the pipelines that transport these resources, including aging infrastructure, environmental impacts, and vulnerability to sabotage operations. Such issues can result in leakages in these pipelines, requiring significant effort to detect and pinpoint their locations. The objective of this project is to develop and implement a method for detecting oil spills caused by leaking oil pipelines using aerial images captured by a drone equipped with a Raspberry Pi 4. Using the message queuing telemetry transport Internet of Things (MQTT IoT) protocol, the acquired images and the global positioning system (GPS) coordinates of the images' acquisition are sent to the base station. Using deep learning approaches such as holistically-nested edge detection (HED) and extreme inception (Xception) networks, images are analyzed at the base station to identify contours using dense extreme inception networks for edge detection (DexiNed). This algorithm is capable of finding many contours in images. Moreover, the CIELAB color space (LAB) is employed to locate black-colored contours, which may indicate oil spills. The suggested method involves eliminating smaller contours to calculate the area of larger contours. If the contour's area exceeds a certain threshold, it is classified as a spill; otherwise, it is stored in a database for further review. In the experiments, spill sizes of 1m 2 , 2m 2 , and 3m 2 were established at three separate test locations. The drone was operated at three different heights (5 m, 10 m, and 15 m) to capture the scenes. The results show that efficient detection can be achieved at a height of 10 meters using the DexiNed algorithm. Statistical comparison with other edge detection methods using basic metrics, such as perimage best threshold (OIS = 0.867), fixed contour threshold (ODS = 0.859), and average precision (AP = 0.905), validates the effectiveness of the DexiNed algorithm in generating thin edge maps and identifying oil slicks.

show abstract

Section: Related Workmentioning

confidence: 99%

Deep Learning Approach for Oil Pipeline Leakage Detection Using Image-Based Edge Detection Techniques

Obaid,

Hamad

2023

JESA

View full text Add to dashboard Cite

show abstract

“…For example in wildfire-spread modeling, the gated recurrent unit (GRU) and long short-term memory (LSTM) [34], convolutional neural network (CNN) [35], and CNN and LSTM [36] models have been developed based on elevation, wind direction, wind speed, temperature, humidity, precipitation, and drought information. In oil-spill modeling, LSTM [37] and CNN [38] models have been developed to study the movement of oil slicks. The remarkable performance of CNNs in the above dynamic processes makes them a promising tool for dust-storm pathway prediction.…”

Section: Introductionmentioning

confidence: 99%

Predicting Dust-Storm Transport Pathways Using a Convolutional Neural Network and Geographic Context for Impact Adaptation and Mitigation in Urban Areas

Yarmohamadi,

Alesheikh,

Sharif

et al. 2023

Remote Sensing

View full text Add to dashboard Cite

Dust storms are natural disasters that have a serious impact on various aspects of human life and physical infrastructure, particularly in urban areas causing health risks, reducing visibility, impairing the transportation sector, and interfering with communication systems. The ability to predict the movement patterns of dust storms is crucial for effective disaster prevention and management. By understanding how these phenomena travel, it is possible to identify the areas that are most at risk and take appropriate measures to mitigate their impact on urban environments. Deep learning methods have been demonstrated to be efficient tools for predicting moving processes while considering multiple geographic information sources. By developing a convolutional neural network (CNN) method, this study aimed to predict the pathway of dust storms that occur in arid regions in central and southern Asia. A total of 54 dust-storm events were extracted from the modern-era retrospective analysis for research and applications, version 2 (MERRA-2) product to train the CNN model and evaluate the prediction results. In addition to dust-storm data (aerosol optical depth (AOD) data), geographic context information including relative humidity, surface air temperature, surface wind direction, surface skin temperature, and surface wind speed was considered. These features were chosen using the random forest feature importance method and had feature importance values of 0.2, 0.1, 0.06, 0.03, and 0.02, respectively. The results show that the CNN model can promisingly predict the dust-transport pathway, such that for the 6, 12, 18, and 24-h time steps, the overall accuracy values were 0.9746, 0.975, 0.9751, and 0.9699, respectively; the F1 score values were 0.7497, 0.7525, 0.7476, and 0.6769, respectively; and the values of the kappa coefficient were 0.7369, 0.74, 0.7351, and 0.6625, respectively.

show abstract

“…Considerable efforts have been made previously to model the fate and transport of oil spills on large scales. 14,15 Most of these studies have used Lagrangian numerical models such as COZOIL, 16 SINTEF OSCAR2000, 12 MOTHY, 17 MOHID, 18 GNOME, 19 MEDSLIKII, 20 and OpenDrift 21 for simulating oil spills. Except for OSCAR, none of these models have been used to model oil spills in icy waters.…”

Section: Introductionmentioning

confidence: 99%

A numerical model for the simulation of oil–ice interaction

Abbasi

Lubbad

2021

Physics of Fluids

View full text Add to dashboard Cite

Accurate modeling of the interaction between oil and sea ice is essential for predicting oil spill fate and transport in ice-infested waters. A three-dimensional numerical model based on the smoothed particle hydrodynamics (SPH) method is incorporated to model such interactions. The effects from air and water are well captured using suitable force components and without explicit inclusion of air and water phases. This reduces the four-phase SPH model into a two-phase model, significantly reducing computational costs and potentially enabling the use of this model for large-scale simulations. We validate the model against experimental data recently available in the literature on oil–ice interactions. The experiments studied the interaction in a flume between an ice floe and oil slick for different types of crude oils. The current velocities were varied and the thicknesses of the oil slicks were measured. The validation results show that our SPH model can adequately simulate the interaction between oil slicks and ice floes. The simulated average thicknesses fit well with the measured thicknesses despite the considerable difference in the viscosity of the tested crude oil. Moreover, the effects of oil density, surface tension, viscosity, and current velocity on oil slick accumulation in front of the ice floe are studied. The higher current velocities and higher oil density lead to thicker oil slick thickness next to the ice floe. The surface tension effect on oil slick thickness is not significant. Finally, we provide estimates for the minimum oil slick thickness for a finite range of oil viscosities.

show abstract

A Dynamic Marine Oil Spill Prediction Model Based on Deep Learning

Cited by 7 publications

References 25 publications

Deep Learning Approach for Oil Pipeline Leakage Detection Using Image-Based Edge Detection Techniques

Deep Learning Approach for Oil Pipeline Leakage Detection Using Image-Based Edge Detection Techniques

Predicting Dust-Storm Transport Pathways Using a Convolutional Neural Network and Geographic Context for Impact Adaptation and Mitigation in Urban Areas

A numerical model for the simulation of oil–ice interaction

Contact Info

Product

Resources

About