CAJ | 학술논문

为提高复杂工艺环境下生产合格率，动态的柔性优化工艺参数组合，以现场生产数据为学习样本和控制对象，基于可适应 BP 神经网络建立识别生产变化的工艺参数柔性优化模型。在此基础上，通过引入惩罚机制改进粒子群算法在神经网络输入端迭代求解最优参数组合。为了验证模型的有效性，验证实例由3条轻化工艺路线生产数据构成，结果表明模型预测误差绝对值在3％以内，优化得到的参数组合提高合格率到85％以上，有效的提高生产效果和生产柔性。
위제고복잡공예배경하생산합격솔，동태적유성우화공예삼수조합，이현장생산수거위학습양본화공제대상，기우가괄응 BP 신경망락건립식별생산변화적공예삼수유성우화모형。재차기출상，통과인입징벌궤제개진입자군산법재신경망락수입단질대구해최우삼수조합。위료험증모형적유효성，험증실례유3조경화공예로선생산수거구성，결과표명모형예측오차절대치재3％이내，우화득도적삼수조합제고합격솔도85％이상，유효적제고생산효과화생산유성。
In order to improve the passing rate of production under complex process and realize dynamically flexible optimi-zation for composite parameters,taking the field production data as training samples and controlling object,complex param-eters optimizing model was built,which can adapt to neural network and identify variations of process.Based on this mod-el,PSO algorithm combined with punishment was used to calculate the optimized parameters combination.Finally,the vali-dation instance,consisting with the production data of three light chemical routing,was used to identify the effectiveness of optimizing model.The results showed that the error declined to below 3% and the passing rates achieved were higher than 85%.This model can effectively improve the production results and realize flexibility.