当今，电子系统越来越复杂，结构更为紧凑，用传统的方法进行故障检测已显得困难，而且费用不断上升。本文旨在探讨一种基于磁场映像的无接触式测试方法。依据是：在一定运行状态下，存在故障的电路板，其磁场分布与正常的电路板是不同的。利用人工神经网络（ＡＮＮｓ），可实现对磁场映像的自动识别、解释和故障定位。 Nowadays, the electronic system is more and more complicated and the structure is more compact. It has become difficult and costly to carry out the fault detection by the traditional method. This paper aims to explore a non-contact test method based on magnetic field mapping. Basis is: In a certain state of operation, the existence of fault circuit board, the magnetic field distribution and the normal circuit board is different. Artificial neural networks (ANNs) enable the automatic identification, interpretation and fault location of magnetic field images.