Ship collision risk degree estimation method based on improved BP neural network

• Main Authors: YUAN YUQI, WANG NANJUN, WANG TING, GUO FEI, LIN XIAOGONG
• Format: Patent
• Language: Chinese; English
• Published: 18.08.2020
• Subjects: CALCULATING; COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS; COMPUTING; COUNTING; DATA PROCESSING SYSTEMS OR METHODS, SPECIALLY ADAPTED FORADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL, SUPERVISORYOR FORECASTING PURPOSES; PHYSICS; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE,COMMERCIAL, FINANCIAL, MANAGERIAL, SUPERVISORY OR FORECASTINGPURPOSES, NOT OTHERWISE PROVIDED FOR

The invention relates to a ship collision risk degree estimation method based on an improved BP neural network. The ship collision risk degree estimation method comprises the steps of (1) improving atraditional BP neural network; (2) establishing a ship collision risk degree estimation model; and (3) estimating the ship collision risk degree. According to the ship collision risk degree estimationmethod, the establishment of a traditional BP neural network mathematical model is completed; and on one hand, the mathematical model is conveniently improved, and on the other hand, the traditionalBP neural network mathematical model is conveniently compared with the improved BP neural network mathematical model. According to the ship collision risk degree estimation method, an adaptive learning rate algorithm is provided, and the learning rate of each node is optimized to the maximum extent, and the convergence rate of the network is further improved. In order to prevent the network from falling into a required erro