一种粒子群算法优化支持向量机的汽车故障诊断方法研究

Research on Vehicle Fault Diagnosis Method based on Support Vector Machine Optimized by Particle Swarm Optimization

由于现代汽车技术的快速发展,使汽车整体性及可拆分性更苛刻,汽车故障诊断比较复杂且困难。提出依托粒子群算法优化层次支持向量机的故障诊断方法,这种方法故障测试时间较短,精确度更高。依托最大间隔距离原则对支持向量机模型进行优化处理,确保各节点的支持向量机具备最大分隔距离,从而减少误差累计,大大优化二叉树结构。通过桑塔纳汽车故障测试实例可知,研究提出的算法具有较高的精度,可以快速、高效率完成汽车故障检测及定位,泛化能力非常强,是一种值得推广使用的汽车故障诊断方法。

Due to the rapid development of modern automobile technology, the integrity and separability of the automobile are more demanded, and the fault diagnosis of automobile is more complex and difficult. In this paper, a fault diagnosis method based on support vector machine optimized with particle swarm optimization（PSO） algorithm is proposed, which has shorter test time and higher accuracy. In this paper, based on the principle of maximum separation distance, the support vector machine model is optimized to ensure that the support vector machine of each node has the maximum separation distance, so as to reduce the error accumulation and greatly optimize the binary tree structure. Through the example of Santana car fault test, the algorithm proposed in this paper has high accuracy, can be used to quickly and efficiently complete the vehicle fault detection and positioning, its generalization ability is very strong, thus is a kind of vehicle fault diagnosis method worthy of popularization.