A Construction Method of Textile Process Recommendation Model Based on Multi-Model Ensemble

Abstract

The invention discloses a method for constructing textile process recommendation models based on a multi-mode set. According to the method, firstly, a sub-model is designed and constructed for the process of each step. Secondly, each sub-model is trained. In this way, according to the order specification information, textile process recommendation models (such as rotating speed, spindle speed, front roller diameter and the like) are respectively designed for eight main steps including scotching, cotton carding, drawing, roving and the like. Meanwhile, a production line corresponding to the order information is automatically formed. Therefore, each step is clearly defined. Meanwhile, the steps are described uniformly in the spinning process description language.

Description

technical field [0001] The invention relates to the textile field, in particular to a method for constructing a textile process recommendation model based on a multi-model set. Background technique [0002] In terms of process design, many textile enterprises in my country mainly use handwriting, and some enterprises use Excel for editing, which usually requires the cooperation of 4 people / day to complete. When encountering customer order changes or complaints, the entire process needs to be recalculated and designed. In particular, when encountering small-batch and multi-variety orders, the ex post facto nature of process design data can easily lead to the loss of data information between spinning samples, processes and orders, making it impossible to accurately reconcile accounts, not only causing a lot of waste of raw materials, but also It leads to the interruption of subsequent processes (plan release, production processing, plan scheduling, etc.). Therefore, how to s...

AI Technical Summary

Problems solved by technology

Moreover, there is a lack of data sharing with the manufacturing level, and the entire process design process needs manual corrections and supplements; Donghua University has developed a set of intelligent process design software platform systems suitable for the textile industry based on the National Science and Technology Support Project

The main function of the system is to automatically design the textile process with the help of historical process parameters (such as vehicle speed, weft density, roller diameter, etc.) , but it is still insufficient in the closed-loop operation of process data between multiple processes and the data sharing with the manufacturing level

In addition, from the perspective of engineering application, some companies have developed an intelligent process design system suitable for domestic textile equipment by considering various indicators of equipment performance (such as rotational speed, spindle speed, roller diameter, etc.), but the problem is that my country The equipment of textile enterprises is mainly imported, so the process design system is difficult to implement in China

Images