Study On Environmental Risk Assessment Of Oil Spill Accidents Based On RBF Artificial Neural Network And Damage Valuation

By virtue of its special geographical position,the richness of species and uniqueness of the ecological landscape,the gulf provides a variety of services,including material and spiritual levels,for human activities,which is generally called the ecological service function of the gulf ecosystem.At the same time,the gulf has become a high incidence area of oil-spill accidents due to its special geographical location.With the increasing demand for offshore oil exploitation and international petroleum transportation,it also brings more risks to the occurrence of oil-spill accidents at sea.On the other hand,the semi-enclosed waters characteristic of the gulf also greatly increase the degree of damage to the gulf ecosystem caused by the oil-spill,which in turn has significant adverse effects on human production and activities.The monetary assessment of ecosystem damage caused by oil-spills in the gulf will help to solve the impact of oil spills on the ecological functions of marine ecosystems as well as provide scientific data support for rapid emergency response.Therefore,the research of this paper has practical significance for environmental risk assessment and emergency disposal of gulf oil spill.In view of the identification of the gulf ecosystem's service functions,on the basis of consulting relevant literature both at home and abroad and comparing various theories and methods,this paper explores the damage of marine ecosystem service function caused by oil-spill through using relevant basic theoretical methods such as artificial neural network,environmental economics,ecological economics.Through constructing RBF artificial neural network training model,seven representative index parameters were selected as input data to simulate the oil spill pollution scope(Sweep area of oil film)after the oil spill accident correspondingly.On this basis,this paper constructs a monetary assessment method of oil-spill damage to the gulf ecosystem services and applies this systematic method to the environmental risk assessment of the oil-spill accident in Panjin port 300,000 ton crude oil terminal.The results show that the RBF artificial neural network training model based on seven influencing parameters as an input unit can quickly predict the scope of oil film diffusion after oil spill.By the monetization assessment of the damage to the gulf ecosystems,we can know the degree of ecological damage to the gulf under different scenarios caused by oil spills and can provide better supporting data for the emergency response to the gulf oil spill.