The intelligent transformation of the manufacturing industry is accompanied by the visualization of industrial information, and the semantic expression of industrial icons plays a crucial role in improving operators' cognitive performance. Taking the production process of a battery production line as the research object, we studied the semantics‐entity relevance of industrial icons based on the concepts of “signifier” and “signified” in metaphorical cognition as well as “topic” and “vehicle.” On this basis, we designed the icons of eight process stations (precleaning, diffusion, coating, screen printing, power testing, etc.) of the battery production line using the metaphor design method. The experimental results show that the icons complying with the law of semantic cognition have high search efficiency. It is concluded that the industrial icons should be designed according to the relevance between functional semantics and entity specific to each production process and that applying such carefully designed industrial icons to industrial information visualization will effectively improve the cognitive performance of operators.