论文部分内容阅读
在分析了水电机组作为准线性同构异参系统有关特性基础上,给出了神经网络可辨识性的定义,论证了BIBS稳定的水轮发电机的神经网络可辨识性.对仿真模型和辨识网络的计算结果的分析比较表明神经网络具有良好的映射性能及较强的容错性.
Based on the analysis of the pertinent characteristics of hydropower units as quasi-linear isomorphic systems, the definition of neural network identifiability is given and the neural network identifiability of BIBS-stabilized hydroelectric generators is demonstrated. The analysis and comparison of the simulation results of the simulation model and the identification network show that the neural network has good mapping performance and strong fault tolerance.