| Nowadays the land resources are increasingly scarce,the exploitation of underwater resources is of great significance to human development.The subsea parallel platform,which integrates complex network system,precise dynamic control,real-time scheduling and management,information fusion and other technologies,has significant advantages in underwater exploration,testing and other aspects,and is increasingly applied in various underwater missions.The platform can not only carry out large-scale underwater tests,but also carry out large-scale monitoring of underwater environment and objects with cameras,sonar,water quality sensors and other equipment.However,the subsea parallel platform has been working in a complex and harsh deepwater environment for a long time.Coupled with the complex control system of the platform itself,the platform is vulnerable to electromagnetic,pressure,corrosion and other potential factors.As a result,the platform often faces various failures and dangers.In order to avoid the platform failure caused by sinking or other serious consequences,a set of emergency control system is specially developed.The development of emergency control system has important research significance and practical value for improving the safety and reliability of the platform.This paper mainly studies from the following aspects:First of all,the platform emergency control system architecture and platform fault mechanism are introduced and analyzed.The communication architecture design of the platform under normal and emergency conditions is described in detail,and the three communication modes between the water emergency control unit and the lifting emergency control unit under emergency conditions are introduced.The platform faults were further refined by drawing the fault tree model.The platform models under different fault modes are established,and the simulation results are summarized.Secondly,theoretical analysis,fault modeling and fault diagnosis are carried out for the common faults of permanent magnet synchronous motor with complex mechanism.Simulink and Ansys Maxwell were used to model and simulate each fault respectively.The BP neural network,SOM-BP series neural network and PSO-SOM-BP series neural network models were established respectively,and the simulation data were brought into the three neural network models.The test results showed that the PSO-SOM-BP series neural network model had the best effect,and the accuracy of the test set was 90.67%.SOM-BP series neural network model followed,the accuracy of test set was 82.33%,and BP neural network model was poor,the accuracy of test set was 79%.Thirdly,the platform fault detection mode,emergency decision mode and fuzzy expert system are studied.The main fault detection methods of the platform are introduced in detail,such as power supply fault,communication fault,etc.,and the emergency measures of all kinds of faults are expounded to ensure that the fault can be solved in time.The platform emergency decision analysis based on fuzzy expert system is carried out,and five kinds of emergency decisions are given according to different fault degrees.The emergency decision surface diagram between different faults of the platform is presented.Simulation tests of fuzzy expert system are carried out for different kinds of faults of the platform.The test results are all in line with expectations.Finally,the hardware and software design of the platform emergency control system and the semi-physical simulation research are carried out.Including the selection of key parts of the control system,emergency console/cabinet 3D drawing,related system schematic drawing,PLC hardware configuration,software key technology introduction,software programming,debugging and other work.Finally,the fault diagnosis and emergency decision of the platform were verified by the semi-physical simulation test.The test results reached the expected standard,which proved the practicability and reliability of the platform emergency control system. |
PDF Full Text Request