论文部分内容阅读
神经网络方法可用于动力学系统的参数估计，包括时域和频域建模。将单层神经网络用于单交量线性连续时不变系统的参数估计中，线性系统分别用状态方程及传递函数来描述。对于状态方程描述的线性系统，假设系统状态、状态的导数及系统输入已知，给出了相应的神经网络结构及学习算法用以估计系统参数；对于传递函数形式描述的线性系统，推导出了基于频域数据的单层神经网络结构，用以估计传递函数的参数。
The neural network method can be used to estimate the parameters of dynamical systems, including time and frequency domain modeling. The single-layer neural network is used in the parameter estimation of invariant systems of single-volume linear continuous systems. The linear systems are described by state equations and transfer functions respectively. For the linear system described by the equation of state, the system states, the derivatives of the states and the system inputs are assumed to be known. The corresponding neural network structure and learning algorithm are given to estimate the system parameters. For the linear system described by the transfer function, A single-layer neural network structure based on frequency domain data is used to estimate the parameters of the transfer function.