| Nowadays, with the continuous development of China in the areas of maritime industry,the requirement on safety and comfort during the sailing is increasing. Fin and anti-rolling tank are used to reduce the rolling motion which is caused by the sailing ship, sea wind or waves. The effect of reducing the roll by using fin and anti-rolling tank is varying from the different sea conditions. When the ship is equipped with fin and anti-rolling tank at the same time, then integrated anti-rolling device can have the advantages of fin and anti-rolling and overcome the disadvantages of them, so the anti-rolling effect will be better. The study on the anti-rolling device integrated fault detection and diagnostic systems also began to receive more attention.In this paper, fault detection and diagnosis of the integrated anti-rolling device consists three modules: the underlying hardware circuit, PC monitoring interface and the man-machine dialogue interface of fault diagnosis. While the underlying hardware circuit system can achieve a variety of environmental parameters acquisition and long-distance wireless transmission, PC monitoring interface can observe the parameters of the environmental and the location information of the corresponding sensor. And then, it also can store the historical data. When the parameter threshold is exceeded, the system will alarm and diagnostics. The diagnostic system can realize the man-machine dialogue, and then diagnosis the system fault.Ultimately, it will give the conclusions of the most likely cause of the malfunction.Firstly, the fin and anti-rolling tank fault tree can be established according to the composition structure of fin and anti-rolling tanks, working principle, reference material, and expert experience. Then, it can analysis fault tree in qualitative and quantitative way. Last, the fault tree minimal cut sets and the importance of every bottom event will be determined.Then, the paper analyzes the data monitoring functional requirements of the integrated anti-rolling device and designs the hardware and software of monitoring system. CC2430 wireless module is used to establish a Zigbee wireless network. According to the different functions of each node, this paper designs the hardware circuit and writes nodes work program, and completes the PC monitor screen.Last, combining the production rules and frame type knowledge expression, according to expert diagnostic knowledge base in SQL Server database established based on the established fault tree, in the process of the knowledge creation and improvement, this paper checks the knowledge consistency. The inference engine is improved, including reasoning direction and search strategies. Using bidirectional reasoning can make the fault reasoning easier to implement in engineering. During the fault searching, the probability of bottom events occurring, the search costs and the value of information are considered which can avoid the blind reasoning and improving the effect of the system diagnostics. This paper chooses Visio Studio 2010 and Windows 7 as development platform, and uses C# and C language for system programming. Finally, a friendly human-machine interface is designed,which can make the system more convenient and easy operation for users. |