Comparative Research Two Kinds Of Simulation Methods In Electrochemical Machining Process Design

As technology advances, more and more objective phenomenon perceived by humans, but unknown areas are becoming increasingly complex. Due to the limit of the body,the human brain is developed only 10%, it can't solve the problem of excessive computation,Such as the simulate the complex systems,store huge amounts of data and find the law in them etc. Computer with its powerful computing ability become one of the most important auxiliary tool of human,and the computer simulation has become a very important part of the industrial production.The first time large-scale development of the computer simulation is simulating nuclear explosions in the Manhattan project. From then on, humans discovered the importance of computer simulation, then it has been set up into a separate discipline and widely used.This article uses two different simulation methods for the electrochemical machining. The first method is computational fluid dynamics(CFD). CFD is useing discrete mathematics to solve the mathematical model, and we will get a discrete solution for the complex system. The second method is big data and machine learning. This methods can help us to automatically find the hidden relationship of the complex system. Nowadays, world is covered with dense computer network, which contains the numerous data, human hope to dig up the valuable content from the vast amounts of data, so, the big data and machine learning was born. It utilize computer programming to simulate the ability of human for solving problems, so that the machine like a human to solve the problem. The process is entering the huge amounts of data into the machine, and then the machine can use anthropomorphic thinking to find out valuable content in the data.The CFD method based on mathematical modeling, its simulation process : establish the virtual model(apply the law of physics, chemistry) ?gain the solution from virtual model. CFD simulation system runs stably, and calculates exactly, it's a mature simulation system and widely used.Its defect is only can express part of the system characteristic. Big data method's simulation process: get more groups of data? establish the virtual model(apply the machine learning methods). This method is a new kind of simulation method. Through continuous learning system data, it can acquire all features of the system. But it is very dependent for the ability of the machine learning system. The disadvantage is that: machine learning system is still in infancy now, it has high error rate and learning ability is not strong. it still on the low intelligence level. Two method's process is opposite, but they have its distinguish point each other.Nowadays they are the widespread simulation methods.In this paper, the first chapter introduces the significance of electrochemical machining simulation, several kinds of common simulation method and the process parameters of electrochemical machining. The second chapter introduces three physical field's mechanism of the pulse electrochemical machining, and established pulse electrochemical machining flow field model. In the third chapter,I design a flow field of electrochemical machining for the group pore structure, and then analyzed three kinds of flow of group pore structure by CFD. The fourth chapter uses two kinds of new machine learning method to simulate the pulse electrochemical machining system, establish a system model, and analyze the results.