基于LS-SVM及嵌入式技术的力敏传感器温度补偿

摘要

The least squares support vector machine (LS-SVM)combined with simulated annealing algorithm(SA) is transplanted into embedded intelligent instrument,then the power-sensed sensor temperature error compensation is realized by embedded technology.The sensor nonlinear model is constructed by LS-SVM,the global optimization for LS-SVM regularization parameter and kernel width is accomplished by the improved SA,and the method is proved by designing embedded system platform.The result shows that,the method has many advantages such as easy realize,high precision etc,the method has practical significance for those intelligent instruments based on embedded technology.%将最小二乘支持向量机(LS-SVM)融合改进模拟退火算法(SA)移植于嵌入式智能仪表中,结合嵌入式技术实现了对力敏传感器的温度补偿.由LS-SVM构建力敏传感器的非线性模型,利用改进的模拟退火算法对LS-SVM中的正则化参数和核宽度进行全局寻优,并通过设计嵌入式软硬件平台对该方法进行了验证.试验结果表明,该方法具有易实现,补偿精度较高等特点,对基于嵌入式智能仪表的传感器温度补偿有一定的实际意义.