遗传优化支持向量机的传感器动态建模

Sensor Dynamic Modeling Based on Genetic Optimization LS-SVM Algorithm

传感器动态建模是研究传感器工作机理、评价动态性能指标及设计动态校正环节的重要途径。在分析最小二乘支持向量机回归算法的基础上,针对其可调参数的选择问题,提出了基于遗传算法进行全局优化的方法。仿真结果表明,将遗传优化最小二乘支持向量机算法应用于传感器的动态建模中,具有学习速度快和建模精度高的优点,所建模型具有较强的实用性和可靠性,为改善传感器动态性能及其在线补偿创造了条件。

Sensor dynamic modeling is important means for researching operational mechanism of sensor,evaluating dynamic performance indexes and designing dynamic correct element.On the basis of analyzing the least square support vector machine（LS-SVM） regression algorithm,in accordance with the selection for adjustable parameters,the global optimization based on genetic algorithm is proposed.The result of simulation indicates that applying genetic optimization LS-SVM algorithm in sensor dynamic modeling features advantages of fast learning speed and high accurate modeling,and the model built is more reliable and practicable,it creates favorable conditions for improving dynamic performance and online compensation of the sensors.