Yarn quality prediction for spinning production using the improved apriori algorithms

Abstract

With the advent of the era of big data, intelligent manufacturing is becoming the developing direction of textile and clothing industry. Many textile enterprises are exploring the intelligent techniques to improve production efficiency and product quality. This paper focuses on yarn quality prediction based on the big data of spinning production. First, the association rules algorithms, Aproiori and I_Apriori algorithms, which are commonly used for intelligent prediction in spinning production, are analyzed. Then, aiming to overcome their disadvantages such as low efficiency, time consuming, big error of prediction results under big data, a global optimization strategy based on genetic algorithm is proposed. This strategy optimizes the global search process of Aprioir algorithm for pruning the frequent itemsets by introducing the genetic algorithm, which can avoid the local optimal solution in the search process. Finally, based on the big real production data collected from the spinning factory, the effectiveness of the proposed Apriori algorithm are investigated and compared with the normal Apriori algorithm. The result indicates that the improved algorithm has better efficiency and more accurate prediction results and is good at dealing with the big data environment.