摘要
注塑成型过程系统动力学的不确定性和复杂性,限制了传统优化技术的运用。将智能优化技术引入注塑成型工艺优化,以成型过程的主要工艺参数为设计变量,利用BP网络获得工艺参数对各质量指标的近似计算关系式,以各质量指标的模糊综合评价函数为遗传算法的适应度函数,建立了解决多因素作用和多指标约束的注塑成型工艺的混合智能优化模型。通过对比实例的正交仿真试验结果,验证了该模型能快速、自动地实现非线性和不确定系统的优化求解。
Uncertainty and complexity of kinetics in injection molding system hampers the application of traditional optimization technology. The optimization model with multi-factor effect and multiindex restriction of injection molding processing was created to which hybrid intelligent technology was introduced. The main processing parameters were used as design variables. Mathematic relation between processing parameters and quality indexes was obtained by BP neural network. Comprehensive valuation formula of fuzzy quality indexes was used as the fitness function of genetic algorithm. In contrast with the results of orthogonal numerical simulation experiment, it was proved that this model could find the optimal solution of non-linear and uncertain system rapidly and automatically.
出处
《农业机械学报》
EI
CAS
CSCD
北大核心
2006年第6期138-143,共6页
Transactions of the Chinese Society for Agricultural Machinery
基金
浙江省教育厅2005年度高校科研立项(项目编号:20050016)
关键词
注塑成型
混合智能
优化
Injection molding, Hybrid intelligence, Optimization
