LLM-based autonomous agents have recently shown strong capabilities in solving complex industrial design tasks. However, in domains aiming for carbon neutrality and high-performance renewable energy systems, current AI-assisted design automation methods face critical challenges in explainability, scalability, and practical usability. To address these limitations, we introduce PHIA (Physics-Informed Autonomous Agent), an LLM-driven system that automates modulation design for power converters in Power Electronics Systems with minimal human intervention. In contrast to traditional pipeline-based methods, PHIA incorporates an LLM-based planning module that interactively acquires and verifies design requirements via a user-friendly chat interface. This planner collaborates with physics-informed simulation and optimization components to autonomously generate and iteratively refine modulation designs. The interactive interface also supports interpretability by providing textual explanations and visual outputs throughout the design process. Experimental results show that PHIA reduces standard mean absolute error by 63.2% compared to the second-best benchmark and accelerates the overall design process by over 33 times. A user study involving 20 domain experts further confirms PHIA’s superior design efficiency and usability, highlighting its potential to transform industrial design workflows in power electronics.

The growing demand for energy storage systems has highlighted the importance of accurate battery State-of-Charge (SOC) estimation, which directly affects energy availability and operational safety. Existing deep learning-based methods focus on single-cycle charge or discharge estimation, and their performance degrades when applied to continuous multi-cycle scenarios. To address this, a physics-gated neural network, named SocGate, is proposed for multi-cycle battery SOC estimation. The model incorporates physical knowledge through specially designed gate functions, which constrain the estimation within realistic physical boundaries while retaining the flexibility of data-driven modeling. Experiments conducted on a public lithium-ion battery dataset demonstrate that SocGate effectively estimates SOC across sequences involving three discharge cycles, two charge cycles, and two plateau periods. The model achieves a low mean absolute error of 0.46% and root mean squared error of 0.70%.

In industry, Boost converter is widely used in PV systems, microgrids and electric vehicles. In the components of Boost converter, the LC filter plays important roles to ensure the high-performance of Boost converter. To guarantee the economic and reliable operations of Boost converter, the efficiency of Boost converter is a significant aspect to be optimized with respect to the LC filter design. However, the optimization of the efficiency of Boost converter will always lead to the deterioration of the power density, so the trade-off relationships between efficiency and power density need to be considered in the design of Boost converter. In this paper, to consider both the efficiency and power density simultaneously in the design of LC filter, a three-stage design methodology is proposed. Following the design methodology, some design examples are given and verified experimentally.

The symmetrical CLLC-type DAB converter has become more popular as a DC transformer (DCT) in the DC microgrid for its galvanic isolation, high power efficiency and high power density. As a DCT, the CLLC-type DAB converter is expected to regulate the DC voltages on the bus and keep a robust voltage conversion gain (VCG). However, because of the practical values of the inductors and capacitors may fluctuate with varying temperature and operating power, VCG of the CLLC-type DAB converter may deviate from the designed one, negatively affecting power supplies. Therefore, this paper proposes a swarm intelligence aided parameter design approach for the CLLC-type DAB converter to ensure its robust VCG against fluctuating inductors and capacitors. In this design approach, the particle swarm optimization (PSO) algorithm is adopted to facilitate design process with high accuracy and computational speed. Finally, the effectiveness of the proposed design approach has been verified with experiments.