CAJ | 학술논문

大型装备制造企业产品规模大、结构复杂，通常按订单生产，产品变型设计频繁，造成产品期量标准的制定非常复杂，准确性差，大大影响了ERP的实施效果。针对该问题，提出了期量标准的智能化解决方案。利用BP神经网络及其变形网络“识别”历史数据中最相似的“零件模型”，对新型零件的提前期进行“预测”。在此基础上提出了详细设计方案，开发出了相应的计算机系统，运用BP神经网络结合梯度下降法对变型零件的期量标准进行估算。
대형장비제조기업산품규모대、결구복잡，통상안정단생산，산품변형설계빈번，조성산품기량표준적제정비상복잡，준학성차，대대영향료ERP적실시효과。침대해문제，제출료기량표준적지능화해결방안。이용BP신경망락급기변형망락“식별”역사수거중최상사적“령건모형”，대신형령건적제전기진행“예측”。재차기출상제출료상세설계방안，개발출료상응적계산궤계통，운용BP신경망락결합제도하강법대변형령건적기량표준진행고산。
The large-scale production and the variety of complex orders make it necessary to frequently redesign the types of products, which adds to the complexity of the generation of period quantity standards. The imprecise data also affect the application of ERP. And intelligent solution to the problem of period and quantity standards was introduced. It utilized the BP neural network and its transmutation network to identify the nearest part model in the history data, and to estimate the lead time of the redesigned parts. Based on this theory, a detailed design proposal was put forward, and corresponding compute programs were developed. The BP neural network and gradient decent method were used to estimate the period and quantity standards of the redesigned parts.