基于小波神经网络的电容称重传感器非线性补偿研究

Research on the Non-linearity compensation Based on Wavelet Neural Network for Capacitance weighing Sensor

为了解决电容称重传感器的非线性问题,提出了补偿其非线性的小波神经网络方法。该方法以电容称重传感器实验数据为基础,通过小波神经网络训练来确定传感器非线性补偿网络。介绍电容称重传感器非线性补偿原理,分析网络的拓扑结构,给出网络参数训练方法。结果表明,采用小波神经网络进行电容称重传感器非线性补偿具有好的鲁棒性,网络训练速度快、精度高,并能在线补偿,在测试领域有实用价值。

A method used to the capacitance weighing sensor non-linearity compensation is applied based on wavelet neural network（WNN） to settle its non-linear problem.In this method,a non-linear compensation network can be set up according to measurement data of capacitance weighing sensor.The compensating principle is introduced and the geometrical structure of the network is analyzed and the algorithms of network parameters training is given.The results show that the proposed WNN has strong robustness,and on-line correction ability,and fast network training speed and high precision when used in the capacitance weighing sensor non-linearity compensation and practical value in measurement field.