Study On Hydraulic Fault Diagnosis Expert System For Forging Hydraulic Press

The research of expert system is one of hot problems now。 The research and design of hydraulic fault diagnosis expert system have significantly realistic meanings because we apply hydraulic system extensively and its faults are relatively complex and occur frequently。 The main contents and achievements of this paper are shown as follows。After deeply analyzing the development history of expert system and the present situation of hydraulic fault diagnosis technology, this paper introduces the purpose of the research task, and points out urgent needs of developing forging hydraulic press fault diagnosis expert system。 This paper has studied the structuring methods of traditional hydraulic fault diagnosis expert systems and has analyzed their defect, and it has made optimization methods in knowledge processing system, inference technology and control strategy, explanation mechanism and man-machine interface respectively。 It has built the knowledge storehouse of 2000 tons fast forging hydraulic fault diagnosis expert system with database。 Those kinds, for example CRecordSet, CRecordView and CDatabase are used for maintenance and the management of knowledge storehouse。 It has established the storehouse's data source and its connection through Microsoft Visual FoxPro and ODBC driver。 The management is realized through ODBC kind。 We can easily modify the database knowledge records such as add or delete records。 It has compiled inference program by VC++ language and realized diagnosis inference。The way the paper adopts is using the state of fault in fault diagnosis knowledge storehouse to match the keywords of state of fault which the user inputs, and then using the bookmarks to record the location of the records which matched successfully, then repeated the diagnose inference again, finally gets the last result and displays them to the user. However since the characteristic of VC++ language, it will be easy to make deep layer knowledge storehouse and deep layer inference under the help of the field experts。 The user interface of this expert system constructed by VC++ language is very friendly。 So this kind of expert system develops prevail against LISP, PROLOG and other language system。...