【摘要】 文中给出了一种研究全球变化中数据融合的新方法———粗集神经网络结构。用改进的方法训练网络权值 ,分析和仿真结果表明 ,此种模型具有较好的泛化、学习、映射能力。说明新模型能够解决传感器输出为二值或一个范围的多传感器数据融合问题。更多还原

【Abstract】 This paper presents a new method for global change modeling and data fusion via rough neural network based on rough set model, in which the improved method is used to train the connection weights of the neural network and the architecture presented. Analyses and experiments indicate that the model is of reinforcement learning properties,mapping capabilities, reflecting ability, and can solve the data fusion problem when the sensors outputs are two values or a range of values.更多还原