摘要
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．
Taking hydroelectric generator unit as a quasi-linear homo-tructural variable-parameter system,the paper analyzes its some characteristics and gives the concept of neural network identificability. And it proofs the neural network identificability for hydroelectric generating unit which is BIBS stable. After studying the identification results, it is shown that the network has good approach ability and strong fault tolerant.
出处
《水电能源科学》
1997年第2期17-23,共7页
Water Resources and Power
基金
国家教委博士点基金
关键词
神经网络
水轮发电机组
智能控制
自动化
neural network identificability, hydroelectric generating unit, quasi-linear homo-structural variable-parameter system
