利用模糊系统和径向高斯函数网络，设计一种具有自适应能力的模糊神经网络．用高斯函数表示模糊规则前件的隶属度函数，然后，构造一种递阶自组织在线学习算法，从输入输出样本数据中，通过学习提取模糊ＩＦＴＨＥＮ规则；在此基础上，提出一种非线性时变系统的自适应状态观测器设计和故障检测方法，并对其结构及特征进行了讨论，仿真结果表明，这种自适应状态观测器能很好地观测系统的状态，并能有效地应用于系统的故障检测． Using fuzzy system and radial Gaussian function network, a fuzzy neural network with adaptive ability is designed. Gaussian function is used to represent the membership function of the fuzzy rule antecedent. Then, a hierarchical self-organizing online learning algorithm is constructed, and the fuzzy IFTHEN rule is extracted from the input and output sample data. On this basis, a The design and fault detection of adaptive state observer for nonlinear time-varying systems are discussed. Its structure and characteristics are also discussed. The simulation results show that this adaptive observer can observe the state of the system well and is effective Applied to the system fault detection.