Elegant: Leveraging Large Language Models for Enhanced Electrical Engineering Design

The integration of artificial intelligence into engineering workflows has opened new frontiers for design automation and innovation. This paper explores an AI-driven approach to electrical engineering design, leveraging the capabilities of large language models (LLMs) such as GPT-4 to enhance both accuracy and efficiency. We propose a novel framework named ELEGANT (Electrical Engineering design Assisted by Natural-language Transformers), which utilizes LLMs to assist in key stages of electrical design, including schematic generation, component selection, and simulation parameter tuning. Unlike existing approaches, ELEGANT uniquely integrates a multi-stage natural language prompt parsing with domain-specific circuit simulation feedback, enabling adaptive and precise design automation. This leads to significant improvements in design speed and accuracy, while providing an intuitive user experience that bridges human intent with AI-driven execution. By aligning natural language prompts with engineering intent, our method enables intuitive interaction between human designers and AI systems, facilitating faster iteration and reduced design errors. We further introduce prompt engineering techniques and domain-specific adaptations to improve the performance of LLMs in electrical engineering tasks. Experimental results on several design scenarios demonstrate that our AI-assisted approach outperforms traditional methods in terms of design speed and correctness. This work highlights the transformative potential of LLMs in accelerating and augmenting electrical engineering design processes.