论文部分内容阅读
本文提出在ＧＭＤＨ中利用“新息贡献”准则选择关键变量（包括初始变量和中间变量）、自动构成最优部分多项式，并且给出了递推算法以避免重复运算，从而优化了模型结构，提高了精度，简化了建模过程。最后用仿真例子验证了新方法的有效性。
In this paper, we propose to select the key variables (including initial variables and intermediate variables) in GMDH using the “new contribution” criterion to automatically form the optimal partial polynomial and recursive algorithm is given to avoid repeated operations, so as to optimize the model structure and improve Accuracy simplifies the modeling process. Finally, a simulation example is used to verify the effectiveness of the new method.