基于L-M算法的反向传播网络的湿度传感器输出误差补偿研究

针对湿度传感器的输出非线性问题,提出了基于L-M算法建立BP神经网络进行补偿校正,实现电阻型湿度传感器的输入与输出非线性补偿,并与共轭梯度算法、拟牛顿算法所建立的神经网路模型进行对比,重点比较了模型迭代性能、标准偏差;最后发现当神经网络用L-M算法进行训练模拟时在迭代性能、标准偏差等方面具有更优异的表现,更适合湿度传感器的非线性特性的补偿校正。

Research on Output error Compensation of Humidity Sensor Based on L-M Algorithm and BP Neural Network[HS)]

Aiming at the nonlinear output of humidity sensor, a nonlinear compensation scheme based on BP neural network is proposed. BP neural network is established based on L-M algorithm, and the input and output nonlinear compensation correction of the resistance humidity sensor is realized, Compared with the conjugate gradient algorithm and the BP neural network model proposed by the quasi Newton algorithm, the model error performance and convergence speed are compared. The results show that the BP neural network model based on L-M algorithm has more efficient performance in convergence speed, error performance and so on, its compensation correction is more suitable for nonlinear characteristics of humidity sensors.