论文部分内容阅读
神经网络的非线性映射特性、信息的分布存储、并行处理和全局优化能力,特别是其高度的自组织和自学习能力,使其成为故障诊断的一种有效方法,已在许多实际系统中得到了成功的应用。神经网络技术的出现,为故障诊断问题提供了一个新的解决途径,特别是对于在实际中难以建立数学模型的复杂系统,神经网络更显示了其独特的作用。应用神经网络进行故障诊断主要是应用神经网络产生的残差并进行残差分析以及用神经网络进行故障模式识别。
The nonlinear mapping characteristics of neural networks, the distributed storage of information, parallel processing and global optimization ability, especially its high degree of self-organization and self-learning ability, make it an effective method for fault diagnosis, which has been obtained in many practical systems The successful application. The emergence of neural network technology provides a new solution to the problem of fault diagnosis. Especially for complex systems, which are difficult to set up mathematical models in practice, neural network shows its unique function. Application of neural network fault diagnosis is mainly the application of neural network residuals and residual analysis and neural network fault pattern recognition.