首页|生成式人工智能的工业应用技术与前景

生成式人工智能的工业应用技术与前景

扫码查看
随着在工业应用中的不断深化,人工智能(AI)逐渐面临场景定制化、数据要求高、动态环境适应性差等问题.以生成式人工智能(AIGC)为代表的通用AI为突破传统AI的瓶颈提供了新思路.为推动AIGC与工业领域的融合创新、抢占下一轮科技革命的技术高地,对AIGC技术及其工业应用展开综述.首先,梳理了国内外AIGC技术的发展现状,总结了当前AIGC工业应用面临的问题和挑战.然后,提出了 AIGC在工业领域应用的技术架构,以及通用大模型集成、通用大模型微调与知识库外挂、预训练工业大模型这三种应用模式.最后,从研发设计、生产制造、经营管理以及运维服务等四个方面的十二个场景作应用展望,以激发AIGC工业应用的新技术、新方向的进一步发展,赋能工业领域形成新质生产力.
Technologies and Prospects for Industrial Applications of Artificial Intelligence Generative Content
With the deepening of artificial intelligence(AI)in industrial applications,which gradually faces problems such as scene customization,high data requirements,and poor adaptability to dynamic environments,etc.General Al represented by artificial intelligence generative content(AIGC)provides a new way of thinking to break through the bottleneck of traditional AI.To promote the integration and innovation of AIGC and industrial fields and seize the technological high ground of the next round of scientific and technological revolution,an overview of AIGC technology and its industrial applications is carried out.Firstly,the development status of AIGC technology at home and abroad is sorted out,and the current problems and challenges facing the industrial application of AIGC are summarized.Then,the technical architecture of AIGC application in the industrial field and three application modes,namely,generalized large model integration,generalized large model fine-tuning and knowledge base outgrowth,and pre-trained industrial large model,are proposed.Finally,twelve scenarios from four aspects of research and development and design,manufacturing,operation and management,and operation and maintenance services are presented to stimulate the further development of new technologies and new directions for the industrial application of AIGC,and to empower the formation of new productivity in the industrial field.

Artificial intelligence generative content(AIGC)Industrial applicationsGeneralized large modelArtificial intelligence(AI)Large model integrationLarge model fine-tuningPre-trained large model

张朋、张洁

展开 >

东华大学人工智能研究院,上海 201620

生成式人工智能 工业应用 通用大模型 人工智能 大模型集成 大模型微调 预训练大模型

2024

自动化仪表
中国仪器仪表学会 上海工业自动化仪表研究院

自动化仪表

CSTPCD
影响因子:0.655
ISSN:1000-0380
年,卷(期):2024.45(8)
  • 14