论文部分内容阅读
测试系统的非线性动态补偿是仪器技术的一个重要方面.采用BP神经网络对测试系统进行动态补偿.BP神经网络的结果决定于网络输入、隐层和输出节点.由于其非线性映射特性,BP神经网络完全能够反映测试系统的动态响应特性.采用了收敛速度较快的递推预报误差算法训练神经网络.试验结果表明,BP神经网络的特性完全能够满足测试系统的动态补偿要求,表明本文的方法是有效的.
The nonlinear dynamic compensation of the test system is an important aspect of the instrument technology. The BP neural network is used to dynamically compensate the test system. The result of BP neural network is determined by the network input, hidden layer and output node. Because of its nonlinear mapping characteristics, BP The neural network can fully reflect the dynamic response characteristics of the test system, and the neural network is trained by recursive prediction error algorithm with fast convergence rate.The experimental results show that the characteristics of the BP neural network can fully satisfy the dynamic compensation requirements of the test system, The method is valid.