Genetic Algorithm Optimization In Injection Mold Cooling System Problems In The Study

In the injection-molded field, the cavity temperature of injection mold and its uniformity affect significantly productivity and the quality of the final molded part. Although the temperature of injection mold is affected by many factors, control and adjustment basically depend on the cooling system. In order to gain more uniform temperature distribution than the initial design. It is important for cooling system design to analyze truly these design variables which affect molding temperature. The optimization of injection cooling system is based on Numerical Simulation of cooling system and Genetic Algorithm(GA). The former let us get some data easily, such as the temperature of cavity at different time and cooling time. These data provide the direct condition to outspread GA. Using computer program, physical system are given through a configuration of particles or a vector of parameters. These will be denoted as individuals. The population is a set of different individuals. According to the survival of the fittest, individuals which are better equipped have a higher probability of Staying alive . Hence, only by reproduction, crossover and mutation, did the structure of the creatures become more and more complex,so that they can adapted better and better to the requirements.In order to optimize injection cooling system , Genetic Algorithm and numerical simulation are integrated .Design variables which are related to the mold cooling system design were determined. The main work is as follows:1.Based on the analysis of mold cooling process, considering uniformity of cavity temperature, we construct mathematic model(objective function and constraint) according to factual problem and simplify it properly.2. On the basis of numerical simulation of cooling process , we select GA to solve the problem.In the process of caculating, the parameters of cooling system are represented by a sequence of genes according to actual situation. Then we choose appropriate fitness function , and the binary coded mode is used to code chromosome. At last, through a series of operation, such as reproduction, crossover and mutation, we can get the best solution. At the same time , in order to avoid prematurity, we adjust pcross and pmutation.3. The program of optimization was written by C++ language. Using two illustrative examples, we find the cavity temperature of optimal design is more uniform than the initial one. At one time , radius of each cooling channel and location of each cooling channel were determined .