Artificial intelligence leads the multi-disciplinary transformation of the textile industry

Update:08-02-2018
Summary:

At the moment, artificial intelligence is changing the […]

At the moment, artificial intelligence is changing the various fields of world development, including daily life, industrial development, and thinking. The textile industry is no exception, and the impact of artificial intelligence on the industry deserves special attention.

The development of artificial intelligence has entered a new stage

Artificial intelligence technology has a history of more than 60 years. In recent years, it has attracted great attention in many fields, mainly due to three factors: First, computer capabilities have been greatly enhanced, various chips have been continuously developed, and ultra-high-speed operations for artificial intelligence. And data processing provides the basic conditions; second, the Internet is everywhere, e-commerce, mobile social networks, and the popularity of the Internet of Things, generating massive data as the support of artificial intelligence, this is the most important point; third, the core algorithm has Important breakthroughs, especially deep learning algorithms based on big data, make the model more thorough and intelligent. The deep application of a new generation of artificial intelligence technology will inevitably inject new kinetic energy into China's economic development.

In July 2017, the State Council issued the "New Generation Artificial Intelligence Development Plan", marking the development of artificial intelligence in China has entered a new stage, indicating the direction for the application of artificial intelligence technology in all walks of life. In order to implement the deployment of the development plan, in December 2017, the Ministry of Industry and Information Technology issued the “Three-Year Action Plan for Promoting the Development of a New Generation of Artificial Intelligence Industry (2018-2020)”, with the deep integration of information technology and manufacturing technology as the main line. Focusing on the industrialization and integration of a new generation of artificial intelligence technology, we will promote the deep integration of artificial intelligence and manufacturing. Planning and promoting a new technology at the national level is unprecedented, and its strategic significance is self-evident, and has attracted great attention from all walks of life at home and abroad.

Artificial intelligence is used in various fields of the textile industry

In recent years, artificial intelligence, including machine perception, machine learning, machine thinking and other technologies have been applied locally in some areas of the textile industry, and preliminary progress has been made, forming a certain foundation. In particular, in the past two years, a series of developments and applications have been carried out in the fields of intelligent manufacturing and have received significant attention.

First, smart manufacturing

Artificial intelligence technology promotes the development of new models, new methods and new systems in intelligent manufacturing. It is the core technology of textile intelligent manufacturing and the most widely used field. It involves production process feature extraction, production process optimization, production planning and scheduling, equipment scheduling algorithm, production process optimization control, quality inspection and management, production operation and handling intelligent connection, equipment fault location and diagnosis. A series of gratifying results have been achieved in product feature identification, quality inspection, and production logistics handling.

In the current textile intelligent manufacturing projects, intelligent functions such as knowledge acquisition, knowledge base construction, deep learning, and optimization decision-making are still lacking. It is necessary to focus on improving in the next stage, and improve the level of intelligence on the basis of automation, digitization and networking. Make smart manufacturing a veritable one.

Second, textile analysis and design

In recent years, artificial intelligence technology has been applied to textile fabric design, fabric defect identification and analysis, fabric performance evaluation, cotton miscellaneous classification and rating, pilling grade evaluation, and dye uptake calculation. It can also be used to analyze various properties of textiles, such as gas permeability, crease resistance, abrasion resistance and other indicators. If using machine vision, machine learning and other technologies, the detection and analysis of the defects, chromatic aberration, pattern, and pilling of printed fabrics and color fabrics can solve the technical difficulties that existed for many years and significantly improve the level of recognition and analysis.

The next stage of artificial intelligence technology will introduce textile design system, which has the ability of logical reasoning and decision-making. It is led by computer, relying on knowledge base and independent learning system, combining a large number of design examples, experiences and guidelines, according to the design goals. Reduce the scope of exploration to achieve the desired design.

Third, the trend research

Based on the accumulation of a large amount of data on the current network, artificial intelligence technology has great room for development in the field of textile and apparel fashion trends. If you use computer vision and image processing technology to analyze massive photos, you can quickly retrieve the colors, patterns and styles that users prefer; for example, according to consumer color preferences and consumption habits, you can analyze the colors that are often worn in all ages. Summarize different popular colors and so on. In terms of trend prediction, the application of artificial intelligence technology will be significantly better than the current multiple modeling methods, not only can accurately solve many complex problems, but also become an ideal tool for the next stage of textile and apparel fashion innovation.

Fourth, the expert system

In the "Twelfth Five-Year Plan for Science and Technology Progress of the Textile Industry", it has been proposed to apply artificial intelligence technology, develop and establish an expert system for the textile industry, and build a textile macroeconomic decision support system. At the industry level and at the enterprise level, work on process design, quality management, enterprise diagnosis, and textile knowledge base has been carried out, and some experience has been gained. In the next few years, the machine thinking method will be more widely used. The expert system will be an important part of the application of artificial intelligence technology in the textile industry. It will bring together the knowledge and experience of industry experts, accumulate and update a large amount of information and technology, and solve textile product design, technology, raw materials, Key issues in production and equipment, and effective support for relevant decisions.

Industrial Internet builds critical infrastructure

The development of artificial intelligence is inseparable from the popularity of the Internet. The application of artificial intelligence in the manufacturing industry in the new stage is even more inseparable from the construction of the industrial Internet. The Industrial Internet reflects the deep integration of the new generation of information technology and industrial systems such as the Internet, and is the key information infrastructure for industrial intelligent development.

The textile industry has accumulated a large number of Internet technology applications. Through the construction of industrial Internet, it first builds an interconnected network infrastructure, interconnects decentralized physical production units, and breaks information islands. The key is to use artificial labor on the basis of information sharing. Intelligent, big data and other technologies provide in-depth analysis of production and operation status, business operation status, industry chain coordination and market demand information to form intelligent decision-making and improve the level of industrial Internet intelligence.

In the future, the popularity of the industrial Internet will focus on key industries to create an integrated solution for industrial Internet that is closely integrated with industry characteristics. Such as relying on the industrial Internet platform to carry out data integration applications, to achieve intelligent decision-making and deep optimization of enterprise production and operation management; encourage enterprises to integrate resources through industrial Internet platform, and build a collaborative manufacturing system that effectively organizes design, production and supply chain resources, so as to fully Utilize the resources of the industry expert system to provide a broader development space for the industrial application of artificial intelligence technology.

Big data provides strong support

From the perspective of technology development, the core of artificial intelligence lies in data support. First of all, big data technology can quickly obtain valuable information from various types of massive data through data collection, preprocessing, storage and management, analysis and mining, and provide solid intelligence for artificial intelligence algorithms such as deep learning. Material foundation; the development of artificial intelligence also requires a lot of knowledge and experience, and these knowledge and experience are extracted from big data. On the other hand, artificial intelligence technology has also promoted the advancement of big data technology, and the two complement each other. The breakthrough of technology in either side will promote the technological improvement of the other side.

In recent years, the textile industry has gradually developed the application of big data technology, and has achieved certain results in the field of network marketing. The industry-level big data information service platform is also under construction. It should be noted that the textile industry has a broad base and urgent need for applying big data. The organic integration with the Internet of Things and cloud computing will effectively promote the integration of resources in product design, manufacturing, operation management, logistics and marketing, and marketing. However, it must be noted that the current big data in the textile industry is still based on open big data on the Internet. Despite the lack of open sharing of public data and the slow progress of data participation in value-added development, it has made gratifying progress; more importantly, Industrial big data based on the industrial Internet, which runs through the textile industry design, process, production, management, service and other aspects, is extremely scarce. Even if it is scattered, fragmented, incomplete, and not shared, it is far from satisfactory. The requirements of industry construction expert system and intelligent decision system have become the bottleneck restricting the development of artificial intelligence.

In the next stage, we should make full use of big data technology to mine big data in textile industry, and gradually improve enterprise products, process databases and knowledge bases at the enterprise level, through comprehensive depth perception, real-time extraction processing, rapid calculation analysis and intelligent construction of artificial intelligence. The optimization of the model enables the industrial system to have intelligent functions such as description, diagnosis, prediction, decision-making, control, etc., to achieve enterprise operation optimization and production organization mode change. At the industry level, we must go further and accelerate the construction of big data service platforms in the industry. At the same time, we must target the application needs of key sub-sectors, form big data solutions and services in the vertical field; systematically develop data resource management systems and actively explore data. Open a new model of shared management, play the role of the basic resources of data and the role of innovation engine, and accelerate the multi-domain transformation led by artificial intelligence.

In the next two to three years, the textile industry should seize the historical opportunity, accelerate the application and development of artificial intelligence technology, become a strategic technology that leads the future, and promote the improvement of the level of intelligence in various fields of textiles, which will bring profound changes. Far-reaching significance must not be underestimated. In the process of advancing, it is essential to accurately grasp the characteristics of artificial intelligence technology and closely combine the needs of textile enterprises. It is necessary to fully implement the national planning and action plan, make overall plans, advance from all directions, and adhere to the scientific definition of artificial intelligence. It should not be too generalized, and put any items on the intelligent hat; Intelligent dazzling achievements in face recognition, automatic driving, drones, etc., should focus on the long-term practical application of the textile industry, and can not blindly copy the success path of other industries; Planning, focusing on the future, we must also fully consider the reality of the application of the industry, recognizing that intelligence is not a one-off, can not be swarmed, must be prioritized, step by step, to lay a solid foundation for the industry's long-term intelligent development. Wu Di/Wen, Deputy Director, Industry Department, China Textile Industry Federation