Using Genetic Algorithm and Particle Swarm Optimization BP Neural Network Algorithm to Improve Marine Oil Spill Prediction
2022
Cheng, Xueyan | Hu, Xupeng | Li, Zhenzhen | Geng, Chuanhui | Liu, Jiaxing | Liu, Mei | Zhu, Baikang | Li, Qian | Chen, Qingguo
Numerical oil spill models, which predict the transport and behavior of oil spills, are an essential tool for risk assessment and clean-up during an actual accident. The existing numerical oil spill models are mainly applied to large-scale oil spills, while few models on small-scale oil spills exist. Therefore, this study focuses on the prediction model of small-scale oil spills. Oil diffusion experiments in seawater using different oil types, including heavy oil, light oil, and gasoline, at different addition amounts under various kinds of wind were carried out, and these diffusion processes were recorded by a camera. The experimental images were processed to obtain the spread oil film area. The oil film edge processing based on genetic algorithm (GA) and back propagation artificial neural network optimized by a particle swarm optimization (PSO-BP) is proposed. Numerical prediction models were then constructed using the BP artificial neural network, the genetic algorithm-optimized back propagation neural network (GA-BP), and the PSO-BP. Among the three methods, the PSO-BP has the fastest convergence speed and the highest stability, which can usually achieve the goal. The PSO-BP reduces the possibility of the BP-ANN and the GA-BP falling into a local optimum instead of reaching global optimization. The prediction performance evaluation data are R² = 1 and MSE = 3.58e⁻⁹ – 8.87e⁻⁸. Results show that the GA and the PSO-BP provide a new approach to small-scale oil spill prediction.
Afficher plus [+] Moins [-]Mots clés AGROVOC
Informations bibliographiques
Cette notice bibliographique a été fournie par National Agricultural Library
Découvrez la collection de ce fournisseur de données dans AGRIS
