基于神经网络的瓦斯传感器非线性校正

Nonlinear Compensation of Methane Sensor Based on Neural Network

瓦斯传感器输出非线性问题的原因很多,其中包括敏感元件温度变化所带来的影响,但即使使用恒温瓦斯检测技术,非线性误差仍然存在,影响了瓦斯检测的准确性。在简述用神经网络进行非线性校正的原理的基础上,探讨了用BP神经网络实现瓦斯传感器非线性校正的模型、MATLAB算法和实现程序,通过计算机仿真,其结果表明采用该方法能够得到令人满意的结果,且结构简单,准确度高。

There are a lot of reasons for nonlinear characteristics of methane sensor,including the changing temperature of sensitive components.Although a kind of new measuring technique of thermostatic methane detection was used,the nonlinear error was still there,which has decreased the accuracy of the methane detection.Based on a brief introduction of the general principles of nonlinear calibration using neural network,this paper discusses the model,MATLAB algorithm and realized program for nonlinear calibration of methane sensor using BP neural network.After computer simulations,the ideal result was obtained through the approach and the construction of the neural network is simple and the precision is good.