Home >

Promote The Application Of AI In The Production Field Of Textile Industry

2024/9/4 16:38:00 0

　　 Artificial intelligence (AI) is reconstructing the industrial structure and underlying base, and reshaping productivity and production relations. The development of intelligent economy has become an important opportunity for industrial transformation and upgrading and for achieving lane changing overtaking.

　　 To grasp the opportunity of AI, China's textile industry needs to return to its essence, strengthen the research in vertical fields, achieve breakthroughs in the application of key scenarios, and build a rich application ecology; To grasp the AI tuyere, we need to appropriately look ahead to the layout, integrate fragmented scenes, precipitate high-quality data, do a good job in "AI+textile", and build design, manufacturing, marketing, and brand building on "big data+big computing power+strong algorithm".

　　 In order to deeply analyze how AI participates in the development of the textile industry at present and in the future and empower the whole industry, we will launch the column "AI+Textile" in Action "to introduce the development and trend of AI in the industry and promote the development of new industrial productivity.

　　 Do a good job of "3+1" mode and promote the application of AI in the production field of textile industry

Yin Qiang, Director of Informatization Department of China Textile Industry Federation

　　 In recent years, with the application and development of a new generation of information technology, artificial intelligence has been gradually applied to the production field of the textile industry, which has brought a significant impact on the production mode, development mode and industrial ecology of the textile industry. Accelerating the development of artificial intelligence in the production field of the textile industry is an important way to promote high-quality development of the textile industry. This paper believes that the key to promoting the application of AI is to do a good job in the application of "3+1" mode (data, computing power, scene algorithm+talent).

　　 1、 Development status of artificial intelligence in the production field of textile industry

　　 The application of artificial intelligence in the production field of the textile industry is still in its infancy. Because AI system needs a lot of data, computing power, algorithms and other resources to support, at this stage, only a few enterprises in the textile industry are trying, and most enterprises are still focusing on automation, digital, and networking transformation. The application in the production field is mainly in design, process optimization, quality inspection, intelligent logistics, product digitization, etc., which needs to be continuously improved and promoted.

　　（1） Application of artificial intelligence in cotton textile industry

　　 The application of AI in the production field of cotton textile industry is still in the process of trial, mainly used in equipment such as different fiber sorting machine, cloth defect detection, and online product quality monitoring, automatic production scheduling, intelligent cotton blending, energy efficiency management, intelligent logistics and other systems. According to the feedback of enterprises with outstanding AI application in the industry, the most advanced production line of the enterprise currently has the functions of full process automation, full process quality monitoring management, equipment health management, energy consumption management, and optimal operation of workshops, which has increased the production efficiency by 38% compared with the conventional production line, increased the energy utilization rate by 21%, saved 80% of labor, and employed about 10 people for 10000 ingots.

　　（2） Application of artificial intelligence in printing and dyeing industry

　　 There are few application scenarios of AI in the production field of printing and dyeing industry, and only a few enterprises try to use intelligent cloth inspection system, AI pattern and style design. The intelligent cloth inspection system is used for the inspection of finished printed and dyed fabrics. The machine vision and AI technology are integrated and applied, and the AI depth learning technology is used to automatically generate the defect detection model to realize the intelligent detection of printed and dyed fabric defects. However, at present, this technology is not mature enough, and the application effect still needs to be further optimized. It has not yet been widely promoted in the industry. AI pattern and style design is mainly used for pattern design and fabric development of printed fabrics. It can imitate the designer's creative idea and quickly present the design works, which greatly promotes the rapid development of fabric pattern design and meets the consumer's demand for personalized and customized consumption.

　　（3） Application of artificial intelligence in garment industry

　　 The application of AI in the production field of the clothing industry has gradually deepened. In the past, traditional clothing manufacturing mainly focused on batch production and traditional management mode, which has a large room for improvement in meeting the diverse market channels and changing consumer needs. Intelligence has changed the production efficiency of the clothing industry and improved the production capacity of rapid response, The application of generative AI can realize the efficiency improvement and deep collaboration of the whole process of order style design, automatic pattern processing, automatic pre production planning, and production process scheduling, and strive to meet consumers' constantly updated and iterative consumption needs more efficiently, more accurately and quickly. According to the prediction of the consulting company, by 2026, more than 80% of technology products will integrate some form of AI technology. In the field of intelligent garment manufacturing, generative AI technology has been comprehensively and deeply involved in all aspects of garment production, making design and development, production and supply chain management, precision marketing, sustainable development and other fields more automated and intelligent. By 2035, driven by generative AI, the digital rate of manufacturing industry will exceed 85%. By 2055, China's generative AI technology will basically realize the digital transformation of all industries, and the digital rate will reach 100%.

　　 2、 Typical scenarios of AI application in textile industry production

　　（1） Digitization of cotton blending

　　 Cotton blending is a very important process for cotton spinning enterprises. Due to the diversity and diversity of raw cotton in performance, cotton blending needs to understand the raw cotton inventory, physical properties, quality continuity and stability, and the amount of calculation is large, which is prone to errors. Cotton blending is facing more and more complex problems. Most enterprises rely on the experience of cotton matchers to complete, which is difficult to achieve efficient and accurate cotton matching. It directly affects the cost of cotton used by enterprises. AI+Digital Cotton Blending The use of digital technology system cotton blending can greatly improve work efficiency, reduce the quality fluctuations caused by human experience, effectively reduce the fluctuation of average grade difference before and after batch reception, and ensure the stability of product quality while reducing the cost of cotton blending. According to enterprise data, the cotton grade of the same kind of products can be reduced by about 0.5 grade on average, and the average grade difference before and after cotton blending succession can be reduced by 0.1 grade after automatic cotton blending. However, we should pay attention to the need to establish data models for various factors affecting product quality fluctuations. In addition to the cotton blending model, we also need to establish the implementation of process standards and tracking records, textile accessories consumption data tracking, temperature and humidity environment information tracking, employee status tracking, etc. To truly achieve intelligent cotton blending, we need a long time of research and accumulation.

　　（2） Spinning collaborative manufacturing

　　 Faced with the problems of low efficiency of manual inspection, passive yarn quality control, extensive management, difficulty in recruitment, lack of support platform and other problems of spinning enterprises, based on the characteristics of small batch and multiple varieties of yarn mills, and diversified customer needs, Wuxi Internet of Things Innovation Center developed the spinning industry Internet based collaborative manufacturing management innovation platform to solve the pain points of the spinning industry +The "photoelectric" sensor can monitor the spindle movement status in real time, identify the broken yarn and other problems, automatically trigger the stop feeding action when the roving stops feeding, reduce the waste of roving and the winding problem of the leather roller, quickly guide the car stopping work, improve the patrol efficiency, collect the whole process equipment data, deploy the big data platform, and achieve all-round monitoring and data analysis. The digital empowerment of spinning contributes to lean production, transparent management, digital decision-making, and cost reduction and quality improvement for enterprises.

　　（3） Intelligent spinning detection

　　 Faced with the problems of long detection time, low detection efficiency, easy fatigue and difficult employment in spinning manual detection, Fujian Hengshen Group solved the problem of product appearance detection through the appearance visual detection system, improved the product qualification rate and efficiency, and used machine vision equipment to replace human eyes to complete detection, measurement and judgment, so as to achieve product wool, snag, mesh, molding defects, oil stains Integrated appearance inspection for paper tube damage; The detection time has increased from 5 minutes for manual detection of a vehicle of 48 spindle wafers to 2.5 seconds for each spindle, and only 2 minutes for the whole vehicle to complete the detection. The efficiency has increased by 2.5 times, and the accuracy has increased from 96% to 99% manually.

　　（4） Defect detection

　　 In the face of the problems of difficult recruitment, high rate of missing inspection, and waste of raw materials in weaving enterprises, especially the problems of frequent replacement of warp knitting lace patterns, multiple types of defects, and elastic fabrics, Donglong Knitting applies AI defect detection technology to quickly and efficiently detect lace defects, effectively improving production efficiency. Through technical transformation of the production machine (adding camera imaging and management system, image acquisition and pre-processing system), deployment of 5G network, development of fabric algorithm, anomaly recognition and other models, and use of AI training and other cloud services, the weaving process detection, gray fabric defect detection, printed and dyed fabric defect detection, finished fabric defect detection, etc. are realized.

　　 AI+intelligent flaw detection has great advantages in terms of detection effect and labor cost, which can effectively improve the quality qualification rate and product competitiveness, reduce the waste of raw materials and repair rate, and save labor costs. At the same time, Donglong Knitting is located in Changle District, Fuzhou, with many textile mills and many manufacturers engaged in lace making. The application of flaw detection technology is highly replicable, with low investment costs, and has the function of rapid replication and promotion for other enterprises.

　　（5） Intelligent printing and dyeing

　　 In the face of the common pain points of printing and dyeing enterprises in the production process, such as low operating efficiency, high energy consumption and extensive management, and in combination with the current market trend of small batch and multi variety, Hangzhou Tianfu Company successfully developed a printing and dyeing intelligent manufacturing system. The system comprehensively covers the whole process from cloth dyeing, yarn dyeing, cotton dyeing to garment dyeing, and then printing and weaving and dyeing integration, realizing the comprehensive digital management of the production process.

　　 Through the construction of an integrated ERP management system, the ability of the enterprise in planning and scheduling, quality control, equipment management and other core businesses has been strengthened, and the optimization of business processes and real-time information sharing have been realized. The MES system is established to realize the real-time monitoring and fine management of key equipment such as the shaping machine. Collect energy consumption data of equipment such as dye vats, and accurately measure and assess energy consumption. Take targeted measures to reduce energy waste.

　　（6） Intelligent garment production

　　 As one of the industrial sectors with the characteristics of high degree of marketization, large proportion of small and medium-sized enterprises and labor-intensive, the clothing industry is faced with challenges such as high comprehensive costs, intensified market competition, and digital and intelligent transformation. AI technology is participating in the whole process of clothing design, production, supply chain and sales, which can be roughly divided into five areas: digital human separation, R&D technology management, planning and production scheduling management, equipment IOT, intelligent production management, and visual quality inspection management decision-making.

　　 Through integrated generative AI, accurate and flexible analysis and decision-making functions can be played in the process of production, operation and management, improving the ability of enterprises to understand and use big data, and promoting the intellectualization of management to executive levels at all stages. In addition, generative AI has also changed the traditional workflow that relies on unstructured data, opened up the information barrier of the department, made the information transmission more efficient and transparent, and ensured the smoothness of operation. Query and broadcast production and operation data in real time, provide management with accurate and rapid decision-making, realize data visualization, and improve decision-making efficiency and system friendly interaction experience.

　　 3、 Problems and key factors in the application of artificial intelligence in the production field of textile industry

　　（1） Problems in the application of artificial intelligence in the production field of textile industry

　　 In recent years, although AI has achieved initial application in the production field of the textile industry, bringing unprecedented development opportunities and challenges to the textile industry, it still faces many challenges and problems in further improving the application of AI in the production field.

　　 At the enterprise level, most enterprises are still in the stage of digital transformation, and the foundation of digital management and intelligent equipment is relatively weak. At the same time, due to the shortage of talents, financial difficulties and other difficulties, there is still a big gap to achieve the application of artificial intelligence. At the industry level, the data is scattered and reserved within the textile industry, lacking effective cross-border data sharing and accumulation, a set of data standards and annotation protocols widely recognized for key data in the industry production field, and a systematic and large-scale data collection mechanism, which limits the training effect and accuracy of AI models and further application of AI in the industry.

　　（2） Key factors for the development of artificial intelligence in the production field of textile industry

　　 Based on the challenges and problems faced by AI development, we should combine the characteristics of the industry and grasp the key factors of AI development to promote and apply AI in the industry. The enterprise data acquisition, integration and processing capabilities should be improved to consolidate the foundation of AI application; It should identify typical application scenarios in the AI production field, combine the innovative algorithm of specific demand and logic of textile production and manufacturing, cooperate with high-quality resources in the industry, and jointly promote; Cloud computing resource sharing should be used to optimize the use of computing power, explore computing power sharing models, and jointly bear costs; We should jointly cultivate talents who understand AI and textile business and provide channels for talent cultivation. At the same time, the application and development of digital twin technology continue to promote the application of AI in the textile industry. Intelligent factories based on digital twins are gradually trying to be applied, playing an important role in the collection, planning and scheduling of all production factors, production management, etc. Therefore, we should focus on the model that data is the foundation, algorithm is the key, computing power is the guarantee, and talent is the support. Starting from the essence of enterprise capability and efficiency, we should continue to promote and apply in the industry.

　　 4、 Suggestions on accelerating the application of artificial intelligence in the production field of textile industry

　　 In recent years, the application effect of artificial intelligence in the textile industry has begun to show. We should continue to maintain this good momentum of development, and continue to promote in typical application scenarios, key technologies, personnel training and other aspects.

　　（1） Deepen typical application scenarios in the production field of the textile industry

　　 Accelerate the research and application of production optimization, intelligent detection, supply chain management, intelligent warehousing and logistics, product tracking, intelligent marketing and other advanced technical solutions in the production field of the textile industry, cultivate a batch of textile intelligent factories integrating intelligent design, production and management, replicate and promote them in the industry on a large scale, and play a leading role in the demonstration of AI benchmarking enterprises.

　　（2） Accelerate breakthroughs in a number of key core technologies

　　 Rely on leading enterprises in the textile industry, quality service providers and scientific research institutes to promote research on key common technologies of artificial intelligence and research and development of intelligent components, equipment and systems. Develop the application of intelligent cotton blending, intelligent dyeing and other solutions applicable to the textile industry, accelerate the breakthrough of a number of core technologies such as intelligent control and optimization, data acquisition and analysis, fault diagnosis and maintenance, and consolidate the basis of artificial intelligence hardware and software in the textile industry.

　　（3） Strengthen the cultivation of compound talents

　　 Give full play to the role of industry associations, integrate resources from production, learning, research and use, and promote the construction of the artificial intelligence "compound talent" training system in the textile industry. Establish on job learning channels and build enterprise talent echelon with independent development ability. Actively introduce and give play to the role of relevant industry alliances, scientific research institutions and service providers, and organize various special trainings.