Routing Planning of Power Electronic Circuit Based on Q-learning
ID：267
Submission ID：297 View Protection：ATTENDEE
Updated Time：2021-12-03 10:54:50 Hits：402
Poster Presentation
Start Time：2021-12-17 14:45 （Asia/Shanghai）
Duration：5min
Session：[Z] Poster Session » [Z6] Poster Session 6: AI-driven technology
Video
No Permission
Presentation File
Tips: The file permissions under this presentation are only for participants. You have not logged in yet and cannot view it temporarily.
Abstract
PCB design of the main power loop is very important to power electronic circuits. PCB design includes three steps of component layout, routing planning, and detailed routing. Component layout and detailed routing can be easily finished in EDA software, but routing planning must consider the influences of electromagnetic compatibility, parasitic inductance, high dV/dt and dI/dt. It requires much expert knowledge and experience to make the routing planning, which exceeds the capabilities of EDA software. In this paper, a Q-learning-based routing planning method for power electronic circuits is proposed, which set reward functions according to routing rules of circuits, so the routing planning result can meet the specifications of power electronic circuit. According to the routing planning result, technicians can quickly complete the follow-up PCB design work. We use this method to complete the routing planning of a Buck circuit, and the experimental result verifies the applicability of the proposed method.
Keywords
Power Electronics, Q-learning, Routing Planning, Electronic Design Automation(EDA)
Speaker
Comment submit