【作者】 汪学渊；

【导师】 潘宏侠；

【作者基本信息】 中北大学 ， 检测技术与自动化装置， 2009， 硕士

【摘要】 随着人工智能技术的不断进步,诊断技术已开始进入一个新阶段,即智能化诊断阶段。将人工智能领域的各种方法加以综合利用并用于故障诊断,特别是针对大型复杂机械设备的诊断对象,将更有利于深入细致地刻画与描述故障的特征,使推理过程与客观实际更加相符,同时也克服了传统的故障诊断专家系统中所存在的知识“瓶颈”问题。将人工智能领域中的各种方法有机结合,可以大大提高故障诊断的水平和效率。本文提供了一种用于自动装填智能故障诊断的通用方案,对于类似系统的设计开发具有借鉴意义。本系统的投入使用可以提高实时故障诊断的准确性和速度,使自动装填的可靠性大大提高,同时也使我国装甲车辆自动化水平提高到了一个新的层面。在神经网络专家系统的设计中,对专家系统的知识库、推理机和解释机制分别进行全新的设计。专家系统知识库中,自动装填知识采用面向对象的表示方法存储在数据库中,而神经网络是把权值和阈值作为隐式知识库存储在数据库;专家系统和神经网络分别采用正反混合推理和正向推理;专家系统的解释机制采用预置文本与路径跟踪法,而神经网络的解释机制是把神经网络逆向推理的过程呈现在用户的面前。最后,利用VC++实现专家系统人机界面的设计。更多还原

【Abstract】 With the development of artificial intelligence, there are the diagnosis techniques in a new phase ,which is intelligent diagnosis phase .Using the various methods in the fault diagnosis,especially,in connection with diagnosis subjects of large and complex machinery and equipment, diagnosis features will be described beneficially. The inference processes fit in with the objective reality better .At the same time ,the knowledge bottleneck problems of traditional fault diagnosis expert systems will be overcome .Some kinds of AI methods integrated can improve the levels and efficiency of fault diagnosis greatly.This paper provides a universal solution which can be used to intelligently diagnose autoloader’s fault. It has the referential meaning for similar systems’designs. The system can be put into use to help improving the real-time fault diagnosis’s accuracy and speed and enhances the reliability of autoloader greatly. Meanwhile, it drives the armored vehicles’automation in our country into a new level.In the neural network expert system’s design, this paper carries out a brand-new design on the expert system’s knowledge base, inference engine and explanation facility. In the expert system’s knowledge base, autoloader knowledge is stored in the database with the object-oriented expression method. While neural network saves the weight and threshold as the implicit knowledge base in the database. The expert system and the neural network adopt backward and forward ratiocination and forward reasoning separately. The expert system’s explanation mechanism adopts preset text and path tracking method, while the neural network’s explanation mechanism presents the neural network’s reversion reasoning process to the front of the users. Finally, this paper uses VC++ to realize the expert system of man-machine interface’s design.更多还原