针对当前铁谱分析技术存在的不足 ,将计算机图像处理技术、数据库技术、网络通信技术相结合 ,提出了以人工智能、神经网络为核心的设计思想。介绍了基于本思想所建立的可共享的磨粒识别及磨损诊断系统WPRWDS,分析了其系统结构、实现机制、硬软件设计、功能和特点等。 Aiming at the shortcomings of the current ferrography analysis technology, this paper combines the computer image processing technology, the database technology and the network communication technology, and puts forward the design idea that takes artificial intelligence and neural network as the core. This paper introduces a shareable abrasive recognition and wear diagnosis system WPRWDS based on this idea, and analyzes its system structure, implementation mechanism, hardware and software design, functions and features.