Research On Loom Information Collection And Warp-Out Time Prediction For Weaving Workshop

With the digitalization of production,intelligent as the core of the "Industry 4.0" gradually emerged,China’s textile industry gradually cloud-based industrial Internet of Things technology into manufacturing,in order to improve production efficiency and strengthen market competitiveness.Weaving as an important production process in the textile industry,with a variety of equipment,a large number of heterogeneous information communication and other characteristics,at this stage,the use of traditional monolithic upper cloud centralized management of weaving information mode is prone to information transmission difficulties,low reliability and other situations,so that schedulers can not be based on effective information to reasonably arrange the production plan of loom equipment,resulting in equipment production interruptions,low efficiency This leads to interruptions and low efficiency.In view of the above problems,this paper starts the research from three aspects of loom information collection,abnormal data processing and loom up time prediction,and proposes a cloud-side distributed loom information collection and up time intelligent prediction system architecture applicable to the weaving production environment,and the main work is completed as follows:(1)The implementation plan of the cloud-side distributed loom information collection and loom time prediction is designed.By analyzing the weaving production process and the dynamic data characteristics of the equipment,the construction requirements of the overall solution architecture are clarified,and the cloud-side distributed loom information collection and up-time intelligent prediction system is constructed.(2)A cloud-side distributed loom information collection and processing method is proposed.It is designed from three aspects: loom information multi-source collection,abnormal data processing and distributed hybrid storage of information.For the cloud-side distributed information collection,a unified communication module of heterogeneous terminals is designed,the cloud-side communication format is determined,and,the information communication services of the edge end and the central cloud end are designed to solve the communication heterogeneity problem of weaving equipment.In terms of cloud-side distributed abnormal data processing,the cloud-side model training and edge-side field application are adopted to achieve effective processing of weaving machine abnormal data to ensure a highly reliable and high-quality data base for the system.In terms of cloud-side distributed information storage,the weaving equipment information model is defined,and the edge-side real-time data caching scheme and the central cloud-side permanent storage scheme for historical big data are designed respectively to enhance the storage and integration capability of weaving information.(3)An intelligent application of weaving machine turnaround time prediction for weaving shops is realized.In order to solve the problem of large deviation between the theoretical calculated value and the actual loom completion time when calculating the loom completion time using predefined static parameters of planned production,a loom completion prediction method based on LSTM recurrent neural network is proposed,and the performance of the model is optimized in terms of loss degree and training time based on the dynamic loss change threshold to realize the loom completion time prediction.This method can provide accurate prediction of loom turnaround time and provide real-time and accurate scheduling basis for enterprises.In this paper,the above research results have been applied in weaving workshops of many textile enterprises.The results show that the cloud edge distributed loom information acquisition and intelligent application system designed can realize the reliable collection of loom equipment production information and the accurate prediction of loom loom time,meet the requirements of intelligent application in weaving production scenarios,and improve the production and operation efficiency of weaving workshops.