Here is the abstract you requested from the Thermal_2018 technical program page. This is the original abstract submitted by the author. Any changes to the technical content of the final manuscript published by IMAPS or the presentation that is given during the event is done by the author, not IMAPS.
|Embeddable Heat Transfer Devices based on Self-Organizing Thermodynamic Complex System and Artificial Intelligence Algorithm|
|Keywords: Embedded thermal management, Self-organizing crticality, artificial intelligence algorithm|
|Securing sufficient cooling is always a concern for electronic system, and critically important for applications like high density computing. Other than rising demand for more power from processors, multiplex needs for compactness, lightweight, low noise level, and energy efficiency also make appropriate cooling a great challenge in nowadays electronic system design. Self-Organizing Thermodynamic System (SOTS) is a new type of heat transfer device developed by Coolanyp. The fundamental structure of SOTS is a network formed by connecting microchannels (Dh≤1mm). A two-phase fluid is enclosed in the network, and the network topology defines the fluid flow paths. When heat is added to the network at some joints (nodes of network), and dissipated at some others, two-phase fluid flow will be stimulated and transfer heat between nodal counterparts of the network. Research conducted by Coolanyp disclosed that network patterns define the interaction of two-phase fluid flow in individual channels, and the collective thermal behavior of the two- phase fluid flow displays Self- Organizing Criticality, which represents a dynamic steady state for both mass and heat transfer inside the network. Coolanyp incorporates this phenomenon to build heat transfer devices by embedding 3D micro-channel network into solid base, and further develop thermal solutions which are embeddable to various geometrically complex structures and complicated heat transfer scenarios. Different from other two-phase heat transfer devices, such as heat pipe and vapor chamber, there is no capillary limits for SOTS, and gravitational impact is essentially negligible. And more importantly, SOTS is compatible with a variety of working fluids and solid materials. It was found that network pattern plays a critical role in determining SOTS’ heat transfer performance; therefore, Coolanyp has developed several types of patterns suitable for different heat transfer modes, such as remote cooling or on-site cooling. However, SOTS’ network is a high dimensional space with enormous design freedoms that regular mathematical approaches don’t quite fit the modeling needs. Algorithm based on artificial intelligence is therefore incorporated in analyzing and optimizing SOTS networks. From the conducted investigations thus far, SOTS demonstrates superior heat transfer performance (>10x of capsulated graphite or conductive plate embedded with heat pipes; >>100W/cm2 heat flux), and great flexibility in geometry, material, and operation conditions. The details of development and experimental results will be presented.|