Study Of Intelligent Cylindrical Traverse Grinding Expert System

Intelligent manufacturing (IM) is a result for combination of the artificialintelligence (AI) technique and manufacturing technique. Its target is to replace ofparts of person's intelligence mental work, and the ultimate target is to realize theoverall intelligence of the organizational skill and manufacturing environment ofthe manufacturing process. The customer requests the performance of the productperfect and dependable, and request to make qualified once possibly .Therefore, itrequest not only to make the system have the higher production efficiency, muchlower and original material and energies to consume, but also request thecomponents to have the higher accuracy, roughness of surface, integrity of surfaceand strict manufacturing consistency. Grinding usually is the latest process of themechanical processing, the working quality affects the service result of theworkpiece directly. The main parameter which characterize the quantity of theworkpiece includes of surface roughness, surface burn, roundness and the value ofdeforming of the workpiece etc.. But the grinding process is very complicated,there is a lot of influencing factors, and the relation of the factors are complicated,it's very hard to establishing mathematics model precisively, the selection ofgrinding parameters influence the finish quantity of workpiece directly. In thereality grinding process, it depends on the experience of operators mostly, andneeds a great deal of experience knowledge of expert in this field. Spoke up fromthe traditional meaning, these works are completed all by the experienced operator.The existing state has become one of the key techniques which hindrance thedevelopment of some advanced manufacturing technique. The inexorable trend ofthe development of the grinding technique is to make it intelligent.Currently expert system has already played an important role in many areas ofmodern science technique. The expert system is a knowledge engineering softwarethat act as mankind experts in some area, that is ,an expert system includes theknowledge related with a certain special problem area, and has the ability to appliesthese knowledge to resolve problems of this area. This paper established Expertsystem of intelligent cylindrical traverse grinding based on the research of theprecision finishing method ?cylindrical longitudinal grinding which widely appliedby the mechanical manufacturer, it is a part of intelligent cylindrical traversegrinding system ,used to provide initial optimization grinding parameters for thegrinding control system.Expert system of intelligent cylindrical traverse grinding is a combined expertsystem that associate knowledge based on heuristics and grinding optimizationmodel , and use the evolution strategy to optimize the model. Expert system ofintelligent cylindrical traverse grinding is mainly composed by the man-machineinterface,knowledge base, inference machine ,knowledge obtain subsystem,database and explanation subsystem.1. Man-machine interfaceThe program of the man-machine interface is based on Visual C++ 6.0??thevisual development environment and the object-oriented programming method.The system has friendly man-machine interface, and it made operating of customerconvenience and flexible.2. Knowledge baseThe knowledge base of this system gathers the rule database, the modeldatabase and the instance database as an integrality, making expression ofknowledge more generalization. The rule database stores the expert knowledgebased on the rule expression and the rule that to hunt the knowledge information;The model database stores grinding force and power model , surface roughnessmodel, workpiece dimension model and grinding temperature model, all thesemodels above were verified through experiment, there also stores grinding processoptimization model used for grinding parameters' optimization;The instancedatabase uses to store a large number of grinding instances.3. Inference machineInference machine is mainly used for receiving treatment request and initialtreatment condition from the customer, after receiving the treatment request,inference machine then extracts the corresponding instance based on the full inputparameters, if there exists the correspond instance that the treatment request andinitial treatment condition is in accordance with the input by customer, the grindingparameters of this instance will be the inferential result. Otherwise, inferencemachine will change to the rule database and reason out the scope values ofgrinding parameters, then call the corresponding model from model database tooptimize according to given optimization target, the result will used as grindingparameters for grinding process.4. Knowledge obtain subsystemKnowledge obtain subsystem can abundant and renew the knowledge of theknowledge base. We use two kinds of methods to obtain the knowledge: Theartificial method is to input the new knowledge or modify the old knowledgethrough man-machine interface by customer or knowledge engineer;The automaticway is to store the grinding instance into instance database by system itself , whenthe grinding result is satisfied.5. Database and Explanation subsystemDatabase is used to storage the initial processing conditions,used rules ofinference process and inference result. Explanation subsystem is used to explainthe inference process to the users.Besides, we use ( μ , λ )-ES strategy of the evolution strategy to solute thegrinding process optimization model. since cylindrical traverse grinding processbased on model is very complex, the parameter influence the grinding quantity isso many, the relationship of the parameters is complex and nonline, and there hasnot only continuous variable but also discrete variable, we brought forward to usean improved ( μ , λ )-ES strategy algorithm.Our study of expert system of intelligent cylindrical traverse grinding is basedon a renovated MMB 1320 precise semiautomatic cylindrical grinder. Visual C++and SQL Server is used to realize the software, we use Visual C++ to program theinterface and SQL Server to maintain and manage the database, they connect withODBC API. This system is convenient, practical, reliable, and has good interactionand transplantation, so it has the widely foreground.