基于BP神经网络的继电器质量特性分析方法

利用BP神经网络对继电器输入特性与输出特性的关系进行建模,预测任意输入特性参数组合方案下的输出特性参数值。研究了单一输入特性参数波动对继电器输出特性的影响。试验结果表明,基于BP神经网络的质量特性分析方法快速、准确,对继电器产品的设计制造过程具有一定的理论和实际意义。

Quality Characteristic Analysis Method of Relay Based on BP Neural Network

The BP neural network was adopted to model the relationship between the input and output characteristic parameters and predict the output characteristic parameters under the circumstances of arbitrary combinations of multiple input characteristic parameters. Based on the method, this paper shows how one single input characteristic parameter influences the output characteristics when other input characteristic parameters are constant. The experimental results show that the quality characteristic analysis method based on BP neural network is fast and accurate and it has theoretical and practical significance to the design and manufacture process of electro magmatic relay products to some extent.