【摘要】 文中提出了一种基于免疫原理的新型径向基函数(RBF—Radial Basis Function)神经网络模型。该模型利用人工免疫系统的记忆、学习和自组织调节原理,进行RBF神经网络隐层中心数量和位置的选择,并采用递推最小二乘算法确定网络输出层的权值。将这种新型的RBF神经网络应用于建立热工过程的非线性模型。仿真研究表明,这种建模方法不仅计算量较小,而且精度高,并有较好的泛化能力。更多还原

【Abstract】 A novel RBF neural network model based on immune principles is presented in this paper. The memory, learning and self-organization abilities of artificial immune system are introduced to the selecting of the number and positions of hidden layer RBF centers, the output layer weights are decided with the recursive least squares algorithm. The nonlinear model of thermal processes is estabilished with the novel RBF neural network, simulation study proves that the method has less calculation, high precision and good generalization ability.更多还原