Optimization Of Process Parameters For Thin Plastic Shell Injection Molding

Injection molding is a preferred method to produce the plastic products in batch. But it is a great perplexing problem to reduce and prevent the warping phenomenon during injection molding. The problem can't be solved effectively by experience, technical knack or repetitious test. Therefore, a new thought has been put forward that we can apply the CAE technology in this field. By CAE we can forecast the warping to optimize the parameter of injection molding. Then, the producing efficiency and the quality of the plastic products will be improved correspondingly.The principal contents studied in this thesis are as follows:1. Several process parameters influencing the warp distortion are synthetically studied. Taking general thin plastic shell as research object, we arrange the combination of process parameters in orthogonal experiment and get result table from numerical simulation of the injection molding. In addition, the influences of various process parameters are analysed, and the optimized parameter combination is obtained to minimize the warp distortion quantity. A useful conclusion then be reached: several other parameters besides the injection time, such as injection temperature, pack time, also influence warp distortion strongly.2. Non-linear relationship between the injection process parameters and the degree of warp distortion is established through the neural network which is testified by the test samples.3. Optimization of process parameters based on the neural network and orthogonal test is carried out. Within the range of process parameters, ANN model, as the substitute of CAE numerical simulation test, combined with orthogonal test is introduced to further optimize the process parameters to decrease the warp distortion. The work in this thesis shows: the combination of network, orthogonal test and numerical simulation may obviously lessen the time of optimizing process parameters and improve the efficiency of process design. On condition that the time of numerical simulation is the same, More precise conclusion and less warp distortion quantity can be obtained via using both the ANN model and orthogonal experiment, comparing with the orthogonal test only.4. The prediction of the degree of warp is studied according to the on-linear relationship of the ANN model, within the range of specimen of experiment, a...