| The exploration and exploitation of oil in sea areas are much more complicated than in the land. Oceanic disaster is one of the factors that result in the complexity. The main oceanic disaster is typhoon in South China Sea, but is sea ice in Bohai Sea and Huanghai Sea in the north part of China, among which the northern part of Liaodong Bay is the severest area.The distribution of sea ices in different size and strength in a sea area determines the general degree of damage from floating ices to the structures. Whether an ice can damage a structure seriously depends on its energy and strength. We can collect the floating ices data in a typical area of a typical year for the ice condition, and do a clustering analysis on the data according to some of their property indexes, then classify all the sea-ices into a few classes with different level of damage and calculate the amount of each class. The result should be important for the design of the against-ice measures. Therefore, we provided a method of classifying the floating sea-ices with FCM clustering algorithm, in which a proper understanding of the type of structure and the characteristics of the sea area is essential.In the study on damage of floating ices to the sloped artificial island in the northern part of Liaodong Bay, we use the FCM clustering algorithm to do a clustering analysis on the ices in a typical sea area of a heavy ice-conditioned year(2000), according to their area and thickness, and then classify them into 15 classes with different level of damage. On the one hand, we got a general assessment on the impact from floating ices to the sloped artificial island in the sea area. On the other hand, it proves that it is feasible we use the FCM clustering algorithm to classify sea ices.The piling up of sea-ice in front of the structure and its damage relate to many factors such as the sum of sea-ice, the thickness of ice block, the drifting speed of ice, the velocity and direction of wind and the height of the protecting wall of structure etc. Whether an ice can move cross the protecting wall to damage the facilities is determined by its size, thickness and velocity, etc. Due to the changes of floating ices, the complexity of ices distribution as well as the different damaging process for various structures, it is difficult to build an accurate mathematical model to describe the process. However, after doing an analysis on the material from practical investigations and experiments, and studying the basic mechanics principles in the damaging process, we can build a few proper fuzzy inference rules to assess the impact from the ice. Here, we provided the detailed steps to build a fuzzy inference system for the assessment of floating ices, in which it is essential to get a genuine understanding of the type of structure and the characteristics of the sea area.According to the structure that is the sloped artificial island and the meteorological and hydrological characteristics of the sea area as well as the phenomenon and conclusion from investigation and experiments, we made a few fuzzy inference rules and built a fuzzy inference system, with which we can assess the damaging level of an ice on the basis of its diameter, thickness and speed. In addition, we have examined the system with phenomenon observed at an offshore artificial island. Therefore, the fuzzy inference system is conducive to the against-disaster works in the offshore exploitation of petroleum during winter, which also indicates the feasibility of solving the similar problems with fuzzy inference. |
PDF Full Text Request