Multi-beam Sonar Elevation Data Prediction Based on Optimized BP Neural Network

The BP neural network is trained by using the elevation data of multi-beam sonar around the artificial reef. The results show that the neural network can obtain better prediction results based on the topographic trend data of the surrounding seabed. GA and PSO are used to optimize the weights and th...

Full description

Saved in:
Bibliographic Details
Main Authors: Liu, Lishou, Xiong, Haojie, Lu, Yuanhang
Format: Conference Proceeding
Language:English
Subjects:
Artificial reef
BP neural network
Data models
Estimation
Filtering
Genetic algorithm
Height data of multi-beam sonar
Meters
Neural networks
Particle swarm optimization
Predictive models
Stability analysis
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:The BP neural network is trained by using the elevation data of multi-beam sonar around the artificial reef. The results show that the neural network can obtain better prediction results based on the topographic trend data of the surrounding seabed. GA and PSO are used to optimize the weights and thresholds of BP neural network respectively, and the prediction results of traditional Kriging method, Yang Chizhong filtering and estimation method, BP neural network, GA-BP neural network and PSO-BP neural network are compared and analyzed. It is found that PSO-BP neural network prediction model is superior to other methods. The multi-beam sonar elevation data around the artificial reef is used to predict the bottom elevation data of the artificial reef, and the cubic meter is calculated. The results show that PSO-BP neural network is obviously superior to other methods, and the prediction accuracy of the data is higher and the fitting effect is better, which can be well applied to the monitoring of the cubic meter of the artificial reef.
ISSN:2693-289X
DOI:10.1109/ITOEC53115.2022.9734449