Multi-Objective Deep Reinforcement Learning Driven Collaborative Optimization of TSV-Based Microchannel and PDN for 3D ICs
Journal
IEEE Transactions on Components Packaging and Manufacturing Technology
Date Issued
October 8, 2025
DOI
10.1109/TCPMT.2025.3618021
Abstract
This study introduces a multi-objective deep reinforcement learning (MODRL) framework for the concurrent thermal-hydraulic optimization of the through-silicon via (TSV) microchannel heat sink (MCHS) embedded in three-dimensional integrated circuits (3D ICs) power delivery network (PDN). By exploiting the inherent structural synergy between TSVs and pin-fin MCHS, proposed method enhances thermal management in high-density 3D ICs. The framework integrates deep reinforcement learning (RL) with multi-objective optimization and computational fluid dynamics (CFD) simulations, enabling an efficient exploration of the high-dimensional design space to resolve trade-offs between thermal efficacy and fluidic resistance. Relative to baseline, the optimized design achieves a reduction in maximum chip temperature of up to 3.3% while concurrently lowering the overall pressure drop by 17.2%. Impedance analysis further validates the design’s superiority, showing that the optimized TSV geometry effectively suppresses high-frequency peak impedance. Compared with standard deep reinforcement learning (SDRL) and genetic algorithm (GA), MODRL converges faster by 57.1% and 62.5% respectively, showing stronger convergence. These results highlight the advantages of the MODRL intelligent optimization framework in design speed and its great potential in driving the development of next generation 3D integrated circuits, especially in applications requiring high power density and high reliability.