论文部分内容阅读
本文提出了用于估计Arrhenius反应速度函数中动力学参数的神经元微型网络模型及相应的学习方法,并例举了其在一发生于CSTR反应器中特定反应过程的参数估计的成功应用。值得注意的是,此应用例子在化学动力学的背景下,显示了文[6]中提出的神经元网络中元作用函数的可学习性概念的合理性。
In this paper, a neural network model and its corresponding learning method for estimating the kinetic parameters in the Arrhenius reaction rate function are presented. The successful application of the parameter estimation in a specific reaction process in a CSTR reactor is illustrated. It is noteworthy that this application example shows the rationality of the concept of the learnability of the meta-function in the neuronal network proposed in [6] in the context of chemical kinetics.