Lazaros Aresti | MSc in Electronics and Technology

Multiobjective Deep Reinforcement Learning Driven Collaborative Optimization of TSV-Based Microchannel and PDN for 3-D ICs

Sun, 01 Feb 2026 00:00:00 +0000

This study introduces a multiobjective deep reinforcement learning (MODRL) framework for the concurrent thermal-hydraulic optimization of the through-silicon-via (TSV) microchannel heat sink (MCHS) embedded in 3-D integrated circuits (3-D ICs) power delivery network (PDN). By exploiting the inherent structural synergy between TSVs and pin-fin MCHS, the proposed method enhances thermal management in high-density 3-D ICs. The framework integrates deep reinforcement learning (RL) with multiobjective optimization and computational fluid dynamics (CFD) simulations, enabling an efficient exploration of the high-dimensional design space to resolve tradeoffs 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 3-D integrated circuits, especially in applications requiring high power density and high reliability.

Smart cooling: Hydrogel-enhanced adaptive jet impingement utilizing through silicon via for integrated microsystems

Wed, 01 Jan 2025 00:00:00 +0000

In three-dimensional integrated circuits, disparities in power usage across various areas result in inconsistent temperature distributions, potentially jeopardizing system stability and dependability. Although microchannels etched on the back of the chip are frequently employed for cooling, standard designs typically offer a fixed cooling capacity, often failing to effectively target specific hotspots. Furthermore, straight microchannels extending throughout the entire setup can lead to excessive cooling in low-power zones and inadequate cooling in areas of high power consumption. This research introduces an innovative adaptive jet impingement cooling structure that merges hydrogel, jet impingement heat sink (JIHS), and through silicon via (TSV) technologies. The structure incorporates vertical channels and leverages the thermally responsive deformation of hydrogel to facilitate adaptive cooling. This configuration enables the cooler to be strategically positioned at hotspots, dynamically modulating microfluidic injection in reaction to temperature variations. Consequently, it alleviates overcooling in low-power zones while ensuring adequate cooling in hotspots, thereby enhancing thermal uniformity. In comparison to conventional jet impingement sinks, the proposed adaptive model enhances temperature uniformity by 12.21%, decreases thermal spreading resistance by 13%, and only slightly raises the maximum total thermal resistance by 3.08%. The maximum pressure drop experiences a minor increase of just 1.28 kPa.