基于SVR的多传感器数据融合处理方法

摘要

To improve the measurement accuracy and stability of the sensor, a new sensor data fusion method based on support vector machine for regression(SVR) is presented. Experimental results show that the proposed method can decrease cross-sensitivity factor of pressure sensor and improve the measurement accuracy of pressure and temperature validly. And judging from it's stabilty of the zero point ,this new method proves to be much better than multidimensional regression analysis method and artificial neural network method.%为了提高传感器的稳定性和每个参量的测量精度,本文提出了一种新的基于支持向量机回归估计(SVR)的多传感器数据融合处理方法.仿真实验结果表明,该方法能有效降低压力传感器的交叉灵敏度系数,提高压力和温度的测量精度.而且压力传感器的零位压力相对温度变化的稳定性明显优于多维回归分析和人工神经网络的数据融合处理方法.