Scholarlink

<p>The manufacturing industry is currently facing an urgent need for intelligent transformation, but problems such as long innovation cycles and insufficient decision-making data support constrain entrepreneurial efficiency. To address this, this study proposes an optimization model integrating a multimodal large-scale model and a programmable logic controller (PLC) system. By constructing a dedicated data interface, it achieves cross-modal alignment of text commands, equipment images, and real-time control signals. The innovations of this study include: first, establishing a bidirectional information flow architecture between the multimodal large-scale model and the industrial control system; second, developing a dynamic analysis method for control logic based on natural language; and third, constructing a multi-dimensional evaluation system for quantifying innovation and entrepreneurship efficiency. Experiments conducted on an automotive parts production line show that after deployment, the overall equipment efficiency increased by 7.2%, the average fault diagnosis time was reduced by 43%, and the iteration cycle for new process parameters was reduced from 14 to 5 days. This model can effectively improve the agility and scientific decision-making of manufacturing systems, providing a new path for the digital transformation of the manufacturing industry.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Optimization of manufacturing innovation and entrepreneurship efficiency based on the integration of multimodal large models and PLC systems

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年7月20日

查看全文
  • TingTing Zhang

摘要

The manufacturing industry is currently facing an urgent need for intelligent transformation, but problems such as long innovation cycles and insufficient decision-making data support constrain entrepreneurial efficiency. To address this, this study proposes an optimization model integrating a multimodal large-scale model and a programmable logic controller (PLC) system. By constructing a dedicated data interface, it achieves cross-modal alignment of text commands, equipment images, and real-time control signals. The innovations of this study include: first, establishing a bidirectional information flow architecture between the multimodal large-scale model and the industrial control system; second, developing a dynamic analysis method for control logic based on natural language; and third, constructing a multi-dimensional evaluation system for quantifying innovation and entrepreneurship efficiency. Experiments conducted on an automotive parts production line show that after deployment, the overall equipment efficiency increased by 7.2%, the average fault diagnosis time was reduced by 43%, and the iteration cycle for new process parameters was reduced from 14 to 5 days. This model can effectively improve the agility and scientific decision-making of manufacturing systems, providing a new path for the digital transformation of the manufacturing industry.