- 上一篇：资产重组问题与应对策略开题报告
- 下一篇：塑料注塑工艺英文文献和翻译
GAs consist of four main stages: evaluation, selection, crossover, and mutation [26]. The evaluation procedure measures the fitness of each inpidual solution in the population and assigns it a relative value based on the defining optimization (or search) criteria. Typically, in a nonlinear programming scenario, this measure reflects the objective value of the given model. The selection procedure randomly selects inpiduals of the current population to develop the next generation. Various alternative methods have been proposed, but all follow the idea that the fittest have a greater chance of survival. The crossover procedure takes two selected inpiduals and combines them about a crossover point thereby creating two new inpiduals. Simple reproduction can also occur which replicates a single inpidual into the new population. The mutation procedure randomly modifies the genes of an inpidual subject by a small mutation factor, introducing further randomness into the population.
Step 1 Set initial values X ; decide a convergence parameter,; let the initial symmetric positive matrix, M(0), be the identity matrix, I; compute the gradient vector as
Step 2 _Compute the norm of the gradient vector as c(k). If c(k)< ɛ, then stop the iteration process. If not, the iteration process continues. k is the number of iterations.
Step 3 Calculate the search direction of k iterations:
Step 4 Calculate the optimum step size αk =α. Any one-dimensional search procedure can be used to minimize f（X（k）+αD（k））and obtain the α value.
Step 5 Update the design variables as
Step 6 Update the symmetric positive matrix M(k) as follows:
and superscript T denotes transposition of a matrix.
Step 7 Set k=k+1 and go to step 2.
3 Parameter optimization system for MISO plastic injection molding processes
This section presents the process parameter optimization system for MISO PIM under four process control factors and one response. The proposed optimization system integrates mold flow analysis, the Taguchi method, ANOVA, BPNNs, GAs, and the DFP method. The product is a plastic injection-molded push-button housing component. The injection time (IT), velocity pressure switch position (VP), packing pressure (PP), and injection velocity (IV) were selected as process control factors. Product weight was selected as the only response for the case study. The proposed optimization system herein has two stages. In the CAE simulations, the preliminary process parameter settings were determined using a mold flow analysis. In the process parameter optimization, the Taguchi method was used to arrange an L25 orthogonal array experiment and reduce the number of set test cycles. Subsequently, the S/N ratio and ANOVA were used to determine the initial process parameter settings that have minimal sensitivity to noise with consideration of the major quality characteristic. The experimental data of the Taguchi method were used to effectively train and test the BPNN that finely maps the relationship between the input process control factors and the output response. Then, the BPNN was inpidually combined with the DFP method and GAs to determine the final optimal parameter settings. Finally, three confirmation experiments were performed to confirm the effectiveness of the final optimal process parameter settings. The statistical averages, standard deviations, and process capability indices were compared in order to judge the best approach for determining the final optimal process parameter settings. The flow chart of the proposed
parameter optimization system is shown in Fig. 1. The procedures of the proposed system consisted of two stages and are given as follows.
Step 1 Identify the feasible quality response as the target requirement of the experiment. The response must be confirmed to have significant influences on the final product quality.
Step 2 Determine the feasible control parameters and levels that influence the performance of the quality characteristic. The number of control parameters which should be included in the experiment and the number of levels for each parameter can be decided using experience, preliminary studies, or brain storming.