MPSO-SVM的压力传感器的非线性校正研究

Research on Nonlinear Correction of Pressure Sensor Based on MPSO-SVM

为了消除压力传感器受温度变化和电压波动的影响而产生的非线性特性,提出了改进粒子群优化支持向量机(MPSO-SVM)非线性校正,利用改进粒子群首先对支持向量机的参数进行搜索寻优,通过建立压力传感器输出特性与其实际电压值之间非线性映射关系的校正模型,再根据支持向量机具有逼近任意非线性函数的特点,实现压力传感器非线性校正。实验结果表明,压力传感器的最大相对波动从原来的22.2%降为0.12%,有效地消除了温度和电压波动的影响,此方法实现简单、成本低,具有实用价值。

In order to eliminate the nonlinear characteristic of the pressure sensor caused by the change of temperature and voltage,a nonlinear correction method based on Modified Particle Swarm Optimization and Support Vector Machine is presented and adjusted the pressure sensor.The parameters of support vector machine are optimized by modified particle swarm optimization,then an adjusting model based on the nonlinear mapping between outcome characteristic of the pressure sensor and its actual voltage is established.It can adjust the pressure sensor effectively because support vector machine can approach any nonlinear function.The experiment results show that the maximum relative fluctuation reduces from the initial 22.2% to 0.12%,and eliminates the effect caused by the change of temperature and voltage.This method is simple and inexpensive,and it has value in practice.