论文部分内容阅读
针对配料过程原料质量参数存在的不确定性,以原料消耗成本最小为优化目标,将不确定质量参数以随机数的形式引入质量指标约束中,建立了一种配料过程随机优化模型.考虑传统蒙特卡洛抽样方法的不足,采用一种更高效的Hammersley sequence sampling(HSS)技术,获得随机优化模型对应的期望值优化模型.将HSS技术用于遗传算法的种群初始化和交叉、变异操作,以保证种群分布的均匀性,实现随机优化问题的有效求解.工业应用实验结果表明,所提方法不仅能够有效降低原料的消耗成本,而且能够保证产品质量指标满足生产要求,优化结果具有较好的鲁棒性,为配料过程的随机优化控制提供了一个优化模式.
According to the uncertainty of raw material quality parameters in ingredients process, taking the minimum raw material cost as the optimization objective, the uncertain quality parameters are introduced into the quality index constraints in the form of random numbers, and a stochastic optimization model of ingredients process is established. Carlo sampling method is not enough, a more efficient Hammersley sequence sampling (HSS) technology is used to obtain the expected value optimization model corresponding to the stochastic optimization model.Using HSS technology for population initialization, crossover and mutation operation of genetic algorithm to ensure the population Distribution uniformity to achieve stochastic optimization problem.Experimental results show that the proposed method not only can effectively reduce the cost of raw materials, but also can ensure the product quality indicators to meet the production requirements, the optimization results have better robustness , Which provides an optimization mode for stochastic optimization control of ingredients process.