基于改进神经网络的汽车ABS故障诊断

Fault Diagnosis of Automobile ABS System Based on the Improved Neural Network

针对传统故障诊断专家系统存在效率低、无法满足现代汽车智能化发展要求等问题,提出了神经网络技术完成对ABS故障诊断的设计方案。论文分析了ABS系统的工作原理,找出了ABS系统常见的故障模式和原因,并以传感器和调节器为对象建立BP神经网络模型进行训练,针对传统BP神经网络存在的局部极小值和收敛速度慢的问题,提出了基于数值优化L-M算法对BP网络进行改进,缩短了诊断时间,提高了诊断效率。

Aiming at the problem that the traditional fault diagnosis expert system is inefficient and can＇t meet the requirements of modern automobile intelligent development, the neural network technology is put forward to design the ABS fault diagnosis. This paper analyzes the working principle of ABS system, finds out common failure mode of the ABS system and the reason, and carry out training based on the establishment of BP neural network model, and the BP neural network model is taking the sensor and the regulator as the object. Aiming at the problem of local minimum and convergence speed of traditional BP neural network, this paper proposes an improved L-M algorithm based on numerical optimization, which shortens the diagnosis time and improves the diagnosis efficiency.