为准确实现对在航船舶的风险评估,建立船舶安全航行系统风险分级ANFIS模型.该方法分层列出4级影响船舶风险值的风险因素,通过模糊推理系统初步定义各项风险因素的隶属度模型.根据建立的模型和典型在航船舶历史数据,运用模糊神经网络的自学习性对模型进行修正,最终实现对船舶风险的客观评估.得到的数据对比图及误差图分析表明,该方法能够使典型数据充分加入,有效克服建模中的主观影响,并在合理的误差范围内较客观地评估在航船舶整体风险.
To achieve risk assessment on sailing ships accurately,a risk ranking ANFIS model of ship safety navigation system is established. Four-ranking risk factors affecting ship risk values are listed hierarchically by this method and the membership degree model of various risk factors is preliminarily defined by the fuzzy inference system. According to the established model and the typical sailing ship history data,the model is modified by the self-learning habit of the fuzzy neural network. Finally,the objective assessment of ship risk is obtained. The analysis on the contrast and error figures from the obtained data shows that this method can fully use the typical data,effectively overcome the subjective influence of modeling,and objectively assess the overall risk of sailing ships within reasonable error.