Artificial-Intelligence-Based Design for Circuit Parameters of Power Converters

Parameter design is significant in ensuring a satisfactory holistic\nperformance of power converters. Generally, circuit parameter design for power\nconverters consists of two processes: analysis and deduction process and\noptimization process. The existing approaches for parameter design consist of\ntwo types: traditional approach and computer-aided optimization (CAO) approach.\nIn the traditional approaches, heavy human-dependence is required. Even though\nthe emerging CAO approaches automate the optimization process, they still\nrequire manual analysis and deduction process. To mitigate human-dependence for\nthe sake of high accuracy and easy implementation, an\nartificial-intelligence-based design (AI-D) approach is proposed in this\narticle for the parameter design of power converters. In the proposed AI-D\napproach, to achieve automation in the analysis and deduction process,\nsimulation tools and batch-normalization neural network (BN-NN) are adopted to\nbuild data-driven models for the optimization objectives and design\nconstraints. Besides, to achieve automation in the optimization process,\ngenetic algorithm is used to search for optimal design results. The proposed\nAI-D approach is validated in the circuit parameter design of the synchronous\nbuck converter in the 48 to 12 V accessory-load power supply system in electric\nvehicle. The design case of an efficiency-optimal synchronous buck converter\nwith constraints in volume, voltage ripple, and current ripple is provided. In\nthe end of this article, feasibility and accuracy of the proposed AI-D approach\nhave been validated by hardware experiments.\n