Research On Reliability And Risk Analysis Method Of Autonomous Ship Intelligent Navigation System

With the advent of the era of industry 4.0,ships that the main tool of water transportation have received extensive attention in terms of safety,green and digital development.At the same time,under the impetus of the new wave of science and technology,ships based on digitalization and aiming at autonomy have become a new stage of ship development.As one of the typical functions of autonomous ships,intelligent navigation is a new trend in the development of shipbuilding industry and a research hotspot in the ship field.The intelligent navigation system combines software and hardware and various sensors to replace the original human control,and realizes the advanced technical content of information perception and fusion,situation cognition and learning,intelligent decision-making and control.In the course of navigation,various new risks will inevitably occur.Therefore,it is necessary to improve the reliability of intelligent navigation system equipment,redundant design of low reliability equipment,improve the reliability of autonomous ship intelligent navigation system.At the same time,real-time risk monitoring of autonomous ships in navigation is carried out to predict possible risk events and take emergency measures in advance,so as to reduce the probability of accidents and improve the safety of autonomous ships.In this thesis,the GO-Bayesian method is used to analyze the reliability of the autonomous ship intelligent navigation system.Firstly,the system is reasoned to identify the weak links of the system,and the reliability of the system and equipment is redundantly designed.Secondly,the bowtie-dynamic Bayesian network method is used to calculate the real-time risk probability of autonomous ships,and the risk prediction system of intelligent navigation operation process of autonomous ships is developed.Finally,using the reliability analysis results of the intelligent navigation system and the analysis of the risk probability,the risk event is proposed,and the emergency plan is prepared for the risk event.The main research contents of this thesis are summarized as follows :(1)The structure of autonomous ship intelligent navigation system is analyzed,and the composition principle diagram of autonomous ship intelligent navigation system is formed.The autonomous ship intelligent navigation system replaces the original human action and realizes the integration process of ’ perception,cognition,decision-making and control’.According to the required functions,it is divided into situational awareness system,ship maneuvering system and remote monitoring and communication system.The situation awareness system is responsible for the integration of external information,including target recognition and positioning.The ship maneuvering system is responsible for decision-making,control and power.It is the core part of the autonomous ship intelligent navigation system.The remote monitoring and communication system is responsible for information interaction and communication support between the ship and the shore.At the same time,the logical relationship between the specific equipment is determined,and the composition schematic diagram of the autonomous ship intelligent navigation system is formed.(2)The reliability analysis of the autonomous ship intelligent navigation system is carried out to identify the weak links of the system.This forms the reliability analysis method theory of autonomous ship intelligent navigation system.According to the functions required for intelligent navigation,the system is divided into situation awareness system,ship maneuvering system and remote monitoring and communication system,which constitutes the schematic diagram of autonomous ship intelligent navigation system.On this basis,according to the characteristics of intelligent navigation system equipment,the reliability analysis model of autonomous ship intelligent navigation system is established by using GO and Bayesian coupling analysis method,and the weak links of the system are identified by Bayesian quantitative reasoning.This provides decision-making basis for the redundant design of system equipment and improves the safety of autonomous ships.(3)The risk analysis of the intelligent navigation operation process of the autonomous ship is carried out,and the risk probability in the navigation is calculated in real time.This forms the risk analysis method of autonomous ship intelligent navigation operation process and develops the risk prediction system of autonomous ship intelligent navigation operation process.By collecting the risk events and risk factors that may occur in the process of intelligent navigation,the risk evolution path is determined,and the intelligent navigation bow analysis model is constructed.In order to facilitate the calculation,the bow-tie model is transformed into a dynamic Bayesian network model through the transformation rules,and the state of each node is input as dynamic evidence.The model is used to calculate the risk value in navigation.Through the calculation of risk analysis,emergency measures can be taken in time to avoid losses.At the same time,based on the above analysis,the intelligent navigation risk analysis of autonomous ships is divided into a monitoring module of risk factors,an update module of node status in the risk analysis model,and a calculation and display module of risk probability.(4)For the risk events that may occur during the intelligent navigation of autonomous ships,an emergency plan is prepared.This forms the risk emergency process of autonomous ship intelligent navigation and lays the foundation for the subsequent risk control system.Based on the reliability analysis results of the autonomous ship intelligent navigation system and the real-time risk probability in the intelligent navigation process,the ALARP principle is used to calculate the upper and lower limits of the risk standard value in the navigation process,so as to judge the size of the risk probability.At the same time,the risk events that exceed the upper limit of the risk standard value,that is,the events that cause the autonomous ship to fail to sail normally,are divided into four special risk events according to the cause of the fault,and the emergency measures are explained one by one.This method not only predicts the risks in navigation,but also takes effective measures to minimize environmental pollution and property losses before the risks come.