论文部分内容阅读
为了提高溢油预测的准确性,建立和优化溢油预测模型,提出了基于径向基函数网络模型的溢油预测方法,实现溢油预测功能。径向基函数网络模型解决了模拟预测过程中样本库巨大、函数模型收敛速度慢的问题。通过选择有效的输入参数和样本数据,建立局部逼近网络;通过径向基函数训练样本数据,利用输出值与实际值之间的误差作为约束条件调整权重因子、径向基中心和宽度,加快函数模型的收敛速度。该模型模拟了溢油的漂移、扩散过程,达到预测的目的。利用该模型,建立了溢油预测模块,并针对一次溢油事故进行预测模拟,验证了该模型的可行性,能够为应急决策提供一定的支持。
In order to improve the accuracy of oil spill prediction and establish and optimize the oil spill prediction model, an oil spill prediction method based on radial basis function network model is proposed to realize the oil spill prediction function. The radial basis function network model solves the problem of huge sample base and slow convergence of the function model in the simulation prediction process. By selecting valid input parameters and sample data, a local approximation network is established. The radial basis function is used to train the sample data, and the error between the output value and the actual value is used as a constraint to adjust the weight factor, radial base center and width, Convergence rate of the model. The model simulates the drift and diffusion of oil and reaches the forecasting goal. By using this model, an oil spill prediction module is established, and a simulation of a spill accident is carried out. The feasibility of the model is verified and it can provide some support for emergency decision-making.