摘要
汽车气压监测过程中,气压变化呈现毫秒级变化,传感监测参数在短期内形成海量积累,大量干扰参数对关键参数形成干扰,传统的挖掘算法在对气压参数挖掘过程中,受到干扰参数的影响,无法快速定位关键信息,挖掘效果不好。为提高气压监测的准确性,提出利用贝叶斯网络关联规则算法的汽车气压监测传感参数高效挖掘方法。构建汽车气压监测参数特征提取公式,对提取的参数进行归一化处理。利用贝叶斯网络实现汽车气压监测相关参数分类,获取上述参数的支持度,并且建立监测传感参数挖掘模型,实现汽车气压监测传感参数挖掘。实验结果表明,利用改进算法建立汽车气压监测传感参数挖掘模型,能够提高挖掘的准确性,满足汽车运行安全监测的实际需求。
In the air pressure monitoring process of automobile, the air pressure presents millisecond changing, and sensing monitoring parameter forms massive accumulation in the short term, a lot of interference parameters dis- turb the key parameters. In order to improve the accuracy of air pressure monitoring, a high efficiency mining method for the sensing parameters of automobile air pressure monitoring was proposed based on the Bayesian network associa- tion rules algorithm. A feature extraction formula of automobile＇s air monitoring parameter was established, and the extracted parameter was normalized. A Bayesian network was used to achieve the related parameters＇ classification of automobile＇s air pressure monitoring. To obtain the support degree of the parameters, we set up the mining model of monitoring sensor parameters, and realized the sensor parameters mining of automobile＇s air pressure monitoring. The experimental results show that the improved algorithm used to establish the mining model of automobile＇s air pressure monitoring sensor parameters, can improve the accuracy of mining, and meet the actual demand of automobile＇s safety running momtonng.
出处
《计算机仿真》
CSCD
北大核心
2014年第8期176-179,共4页
Computer Simulation
关键词
汽车
气压监测
参数挖掘
Automobile
Air Pressure monitoring
Parameters mining
