最小二乘支持向量机在动态称重系统中的应用  

Application of Least Squares Support Vector Machine in Dynamic Weighing System

作　　者：刘杨[1] 蒋庆[1] 桑迎平 

机构地区：[1]中国计量学院计量测试工程学院,浙江杭州310018

出　　处：《仪表技术与传感器》2013年第12期170-172,共3页Instrument Technique and Sensor

摘　　要：目前我国动态称重系统不管是精度还是速度水平都有待提高,通过对动态称重信号的分析处理,提高系统的称重速度和精度,对于改善称重系统性能具有现实意义。以自主研发的动态称重装置为实验平台,利用最小二乘支持向量机(LS-SVM),在MATLAB环境下,建立预测模型实现动态称重终值预报,所建预测模型的预测均方误差为0.965 0,统计量值为0.993 8,达到了较好的效果,因此,基于最小二乘支持向量机的动态称重预测模型能够更好地提高系统称重效率,改善系统称重精度.The accuracy and efficiency levels of our fruits dynamic weighing system have to be improved. It means a lot to the present situation of our dynamic weighing system that if the accuracy and efficiency of the system can be improved by analyzing the signal of the system. In this paper, firstly, the sample signal of an independent - developed separation system is used to build a fore- casting model suitable for the dynamic weighing system in the MATLAB circumstance by using LS - SVM ;The mean square error of the forecasting model is 0. 9650, the statistic value of the forecasting model is 0. 9938, achieving good results which proves that this LS - SVM forecasting model can be more helpful in improving the accuracy and efficiency of the dynamic weighing system.

关 键 词：最小二乘支持向量机 预测模型 动态称重 

分 类 号：TH273.5[机械工程—机械制造及自动化]