摘要
整经轴数是色织生产中由整经工艺确定的一个品种内待加工的整经轴数目,该参数是制定色织生产作业计划所需的重要特征量。由于整经工艺制定过程非常复杂,其通常需要较长时间才能完成,因而该参数在调度时难以有效获取。为解决复杂色织生产过程调度所需的整经轴数预测问题,提出了一种智能预测算法。该算法针对实际问题特点,将支持向量机与神经网络技术相结合,由支持向量机对与整经轴数相关的参数进行特征提取,并将特征提取后的特征属性作为神经网络的输入,相应的特征属性权重被用以指导神经网络的学习过程。数值计算及实际制造企业应用效果表明该算法是有效的,能满足面向实际色织生产过程的整经轴数预测需要。
The trim beam number is the number of trim beams to be processed inside a pattern, which is decided by the trim technique. The parameter is important in the production planning of the colored weaving manufacturing process. To the difficulty of obtaining the trim beam number before scheduling, a predictive algorithm based on the combination of artificial neural network （ANN） and support vector machine （SVM） is presented. SVM is adopted for the feature selection while ANN for the learning process guided by the output of SVM. Numerical com- putational results from the practical manufacturing production data show that the proposed algorithm is effective and can satisfy the requirements of the production process.
出处
《控制工程》
CSCD
2005年第6期523-526,共4页
Control Engineering of China
基金
国家重点基础研究计划资助项目(2002CB312200)
国家自然科学基金资助项目(60004010
60274045
60443009)
关键词
支持向量机
神经网络
整经轴数
预测
调度
support vector machine
artificial neural network
trim beam number
prediction
scheduling
