摘要
为了解决电容称重传感器的非线性问题,应用遗传算法训练径向基函数神经网络实现其非线性补偿。介绍非线性补偿的原理和网络训练方法。从实测数据出发,建立了电容称重传感器的非线性补偿模型。该方法能同时优化网络结构和参数,具有全局寻优能力,补偿精度高、鲁棒性好、网络训练速度快、能实现在线软补偿。实验结果表明,本文所用的电容称重传感器非线性补偿方法是有效和可行的。
A method used to the capacitance weighing sensor non-linearity compensation is applied based on radial basis function neural network that is trained by genetic algorithms to settle its non-linear problem. The principle and algorithms of neural network are introduced. In this method,the configuration and parameters of non-linearity compensation model are optimized by genetic algorithm. The non-linear compensation model is set up by radial basis function neural network according to measurement data. The proposed non-linearity compensation method has high precision,strong robustness,fast network training speed,good global searching ability and on-line soft compensation ability. The experimental results show the capacitance weighing sensor non-linearity compensation method is efficient and feasible.
出处
《电气自动化》
2010年第6期10-12,共3页
Electrical Automation
基金
江苏省高等学校自然科学基础研究基金
淮阴师范学院教授基金资助项目(项目号07KJD510027
08HSJSK02)
关键词
电容称重传感器
非线性补偿
遗传算法
径向基函数神经网络
capacitance weighing sensor non-linearity compensation genetic algorithm radial basis function neural network