基于K-Means聚类算法的高速公路疑似假轴车辆判别

Discriminating Vehicles with Dummy Axle Based on K-Means Clustering

针对高速公路实行联网收费之后产生的海量通行记录数据,通过采用大数据技术和统计分析算法,为收费稽查业务提供新的技术手段。采用K-均值(K-Means)聚类统计结合其他数据分析方法,计算货车轴重载荷的分布情况,寻找载荷分布规律,求取不同超限范围内的轴重载荷分布曲线,得到甄别假轴车辆的新方法。通过运用该方法,为高速公路收费稽查提供数据支撑,进而为提高收费稽查工作的效率和准确度,减少高速公路运营损失提供参考。

The truck axle load data accumulated in the highway toll collecting systems are processed with K-Means clustering and analyzed to find the distribution law of the axle load of trucks.The axle load distributions of trucks with excessive freight of different degrees are studied.Apparent deviation from the typical distribution may suggest the use of a dummy axle.The new method of detecting dummy axles will improve highway toll auditing efficiency and reduce the loss of highway operation.