A Deep Learning-Based Method for Heat Source Layout Inverse Design

Abstract

Heat source layout design is an effective technique to enhance the thermal performance in the whole system, which has become a vital part in many engineering fields, e.g. satellite layout design and integrated circuit design. Traditionally, the optimal design is obtained by searching the design space with the optimization technique which repeatedly runs the thermal simulation to compare the performance of each scheme. Due to the extremely large computational burden with this method, the optimization is greatly limited. To overcome the challenge, heat source layout inverse design (HSLID) is proposed in this paper to directly generate the layout scheme with the given thermal performance requirement. A novel method for heat source layout inverse design, denoted as SAR-HSLID, is proposed based on the recently deep learning technique, Show Attend and Read (SAR) model. Firstly, regarding the mapping from the required temperature field to layout scheme as an image-to-location task, this paper introduces SAR model, which is good at sequence predicting, to generate the layout scheme. The trained SAR is capable of learning the underlying physics of the design problem, thus can efficiently predict the design under given requirement without any physical simulation. Secondly, to ensure that the designed heat source layout exactly satisfies the input temperature field requirement, based on the layout predicted by SAR, we further utilize a simple but efficient optimization process to conduct few post-processing. Finally, a heat source layout inverse design task in a typical two-dimensional heat conduction problem is investigated to demonstrate the feasibility and effectiveness of the proposed method.

Artificial Intelligence for Satellite Communication: A Review

“…Conventionally, the optimal design is acquired by exploring the design space by repeatedly running the thermal simulation to compare the performance of each scheme [199]- [201]. To avoid the extremely large computational burden of traditional techniques, Sun et al [202] employed an inverse design method in which the layout of heat sources is directly generated from a given expected thermal performance based on a DL model called Show, Attend, and Read [203]. Their developed model was capable of learning the underlying physics of the design problem and thus could efficiently forecast the design of heat sources under a given condition without any performing simulations.…”

Physics-constrained Deep Learning for Robust Inverse ECG Modeling

“…For example, the intracardiac electrical signals are often measured by placing electrodes within the heart via cardiac catheters [33], which are of low spatial resolution, expensive to acquire, and incur discomforts to patients. To cope with such limitations, recent techniques have been developed to incorporate the underlying physics laws into the DNN training [34], [35], which has been engaged in solving various problems such as simulating incompressive fluid flow by Navier-Stokes equation [36], [37], elastodynamic problem [38], and heat source layout optimization [39], [40].…”

Regularization, Bayesian Inference, and Machine Learning Methods for Inverse Problems

“…After mentioning some successful case studies in which the ML tools have been successful [20][21][22][23][24][25][26][27][28][29][30][31][32][33][34][35], we arrive at the main conclusions of this article and the future of the possible interactions between model-based and machine-learning tools. We conclude by mentioning the open problems and challenges in both the classical, model-based and the ML tool.…”