基于CMAC的非线性逆滤波改善传感器的动态特性

在测量系统中许多传感器存在着严重的非线性静态特性和响应滞后的动态特性,当被测量对象的变化率高于传感器的响应速度时,测量结果与真值之间存在较大的误差.为了补偿这个测量误差,采用了一个由无限响应的IIR滤波器和静态非线性环节构成的非线性滤波器去改善传感器的特性.IIR滤波器的系数通过实验数据得到,非线性静态环节采用单输入单输出小脑神经网络(SISO CMAC)实现.SISO CMAC具有学习简单、收敛速度快、函数逼近精度高等特点.最后,通过对热敏电阻动态测量误差的补偿,验证了该方法的有效性.

Improving sensor''s dynamic characteristics by nonlinear inverse filtering of CMAC

The sensors have serious static nonlinear characteristics and slow dynamic responses.When the rate of change of the measured object is higher than that of the response of the sensor,there is an error in the measured value.In order to improve the sensor's dynamic characteristics,a nonlinear inverse filter is employed.It is composed of an infinite impulse response(IIR) filter and a static nonlinear block.The coefficient of IIR filter is obtained via experiment.The static nonlinear block is a kind of Cerebral Model Articulation Controller with Single-Input-Single-Output(SISO CMAC).SISO CMAC has characteristics of simple learning,fast convergence and high accuracy in function approximation.The method is proved to be effective by the experimental results that the temperature is measured by thermistor.