UC知道急用~请帮忙翻译啊！
来源：百度知道 编辑：UC知道 时间：2024/05/27 11:23:51
5. Conclusions
An offline modeling study of a SOFC stack using a GARBF
neural network is reported in this paper. It is shown that
the GA-RBF model is an attractive modeling solution in that
it avoids using complicated differential equations to describe
the stack, and the input–output characteristics can be achieved
quickly by GA-RBF estimation. In our study, training a GARBF
with the optimized parameters only needs 10 s on a PIII
256MHz computer and the time for prediction is no more than
5 s. Besides, compared with the BP neural network approach,
the simulation results show that the GA-RBF approach yields
higher prediction accuracy. Hence, it is feasible to establish the
model of the SOFC by using GA-RBF.
Among all the operating parameters that have an effect on
the SOFC performance, only current density and temperature
are included in our model. In the future we will incorporate
other operating parameters
An offline modeling study of a SOFC stack using a GARBF
neural network is reported in this paper. It is shown that
the GA-RBF model is an attractive modeling solution in that
it avoids using complicated differential equations to describe
the stack, and the input–output characteristics can be achieved
quickly by GA-RBF estimation. In our study, training a GARBF
with the optimized parameters only needs 10 s on a PIII
256MHz computer and the time for prediction is no more than
5 s. Besides, compared with the BP neural network approach,
the simulation results show that the GA-RBF approach yields
higher prediction accuracy. Hence, it is feasible to establish the
model of the SOFC by using GA-RBF.
Among all the operating parameters that have an effect on
the SOFC performance, only current density and temperature
are included in our model. In the future we will incorporate
other operating parameters
5. 结论 SOFC堆的An离线塑造的研究使用GARBF的
neural网络在本文报告。 显示那
the GA-RBF模型是一种有吸引力的塑造的解答由于
it避免使用复杂的微分方程描述
the堆和input–output特征可以是达到的由GA-RBF估计的quickly。 在我们的研究中，训练GARBF
with优化参量只需要10 s在PIII
256MHz计算机和预言的时期是没有比
5 s。 其外，比较BP神经网络方法，
the模仿结果表示， GA-RBF方法产生
higher预言准确性。 因此，是可行的建立SOFC的model通过使用GA-RBF的。
Among有一个作用的所有使用参数
the SOFC表现、仅电流密度和温度 在我们的模型包括的are。 将来我们将合并
other使用参数到GA-RBF模型里和基于
on这个GA-RBF模型，一些控制方案研究，例如
predictive控制和鲁棒控制将被开发。
Acknowledgement 国民863科学项目支持This工作
Development资金(第2003AA517020)， PR中国。