Traditional Thermal Model Research Articles

Thermal Modeling and Analysis of Cloud Data Storage Systems

An explosive increment of data and a variety of data analysis make it indispensable to lower power and cooling costs of cloud datacenters. To address this issue, we investigate the thermal impact of I/O access patterns on data storage systems. Firstly, we conduct some preliminary experiments to study the thermal behavior of a data storage node. The experimental results show that disks have ignorable thermal impacts as processors to outlet temperatures of storage nodes. We raise an approach to model the outlet temperature of a storage node. The thermal models generated by our approach gains a precision error less than 6%. Next, we investigate the thermal impact of data placement strategies on storage systems. We compare the cooling cost of storage systems governed by different data placement schemes. Our study shows that evenly distributing the data leads to highest outlet temperature for the sake of shortest execution time and energy efficiency. According to the energy consumption of various data placement schemes, we propose a thermal-ware energy-efficient data placement strategy. We further show that this work can be extended to analyze the cooling cost of data centers with massive storage capacity. Big data, which is composed of a collection of huge and complex data sets, has been positioned as must have commodity and resource in industry, government, and academia. Processing big data requires a large-scale storage system, which increases both power and cooling costs. In this study, we investigate the thermal behavior of real storage systems and their I/O access patterns, which offer a guideline of building energy-efficient cloud storage systems. The cooling consumption of data centers can be considerably reduced by using an efficient thermal management for storage systems. However, disk is not considered in traditional thermal models for data centers. In this paper, we investigate the thermal impact of hard disks and propose a thermal modeling approach for storage systems. In addition, we estimate the outlet temperature of a storage server by applying the proposed

Thermal Model for Power Converters Based on Thermal Impedance

In this paper, the superposition principle of a heat sink temperature rise is verified based on the mathematical model of a plate-fin heat sink with two mounted heat sources. According to this, the distributed coupling thermal impedance matrix for a heat sink with multiple devices is present, and the equations for calculating the device transient junction temperatures are given. Then methods to extract the heat sink thermal impedance matrix and to measure the Epoxy Molding Compound (EMC) surface temperature of the power Metal Oxide Semiconductor Field Effect Transistor (MOSFET) instead of the junction temperature or device case temperature are proposed. The new thermal impedance model for the power converters in Switched Reluctance Motor (SRM) drivers is implemented in MATLAB/Simulink. The obtained simulation results are validated with experimental results. Compared with the Finite Element Method (FEM) thermal model and the traditional thermal impedance model, the proposed thermal model can provide a high simulation speed with a high accuracy. Finally, the temperature rise distributions of a power converter with two control strategies, the maximum junction temperature rise, the transient temperature rise characteristics, and the thermal coupling effect are discussed.

Modeling greenup date of dominant grass species in the Inner Mongolian Grassland using air temperature and precipitation data

This work was undertaken to examine the combined effect of air temperature and precipitation during late winter and early spring on modeling greenup date of grass species in the Inner Mongolian Grassland. We used the traditional thermal time model and developed two revised thermal time models coupling air temperature and precipitation to simulate greenup date of three dominant grass species at six stations from 1983 to 2009. Results show that climatic controls on greenup date of grass species were location-specific. The revised thermal time models coupling air temperature and precipitation show higher simulation parsimony and efficiency than the traditional thermal time model for five of 11 data sets at Bayartuhushuo, Xilinhot and Xianghuangqi, whereas the traditional thermal time model indicates higher simulation parsimony and efficiency than the revised thermal time models coupling air temperature and precipitation for the other six data sets at E'ergunayouqi, Ewenkeqi and Chaharyouyihouqi. The mean root mean square error of the 11 models is 4.9 days. Moreover, the influence of late winter and early spring precipitation on greenup date seems to be stronger at stations with scarce precipitation than at stations with relatively abundant precipitation. From the mechanism perspectives, accumulated late winter and early spring precipitation may play a more important role as the precondition of forcing temperature than as the supplementary condition of forcing temperature in triggering greenup. Our findings suggest that predicting responses of grass phenology to global climate change should consider both thermal and moisture scenarios in some semiarid and arid areas.

Techniques and Tools for Collaborative Thermal and Mechanical Modeling of 3D ICs

The Industry has been using Computational Fluid Dynamics (CFD) and Finite Element Analysis (FEA) for thermal and mechanical modeling of IC packages for many years. What was already practiced as an engineering-art within packaging organizations for monolithic IC packages has now become more complex due to the need for collaboration across organizational walls in the case of 3D stacking. The holistic solution needed for collaborative engineering of 3D stacking process calls for newer methodologies and information exchange protocols. One particular necessity while building low power System on Chip (SOC) is the methodology needed to ensure package specific thermal feasibility of IC stacking options at physical layout stage of the 3D IC design. A methodology built on the technique to be presented will avert the unintended consequences of stacked IC hot spot alignment during system operation. The presentation will include a validation of the technique by comparison with traditional thermal modeling technique. The ability to fabricate such stacks using an evolving supply chain of foundry-OSAT combination is equally crucial to the viability and reliability of the packaged SOC. A process focused modeling tool for understanding the warpage, stress, and Chip Package Interaction (CPI) implications of assembly strategy choices is valuable for ensuring time to market at minimal technology development costs. The presentation will include a modeling case study with quantitative data of warpage and stress implications of process, material and stacking choices. This presentation will introduce the tools and techniques for collaborative chip stack thermal and assembly process modeling with in-depth discussions on inputs needed, gaps in existing modeling methodologies and output metrics of engineering relevance.

Distributed electro-thermal RF-LDMOS model

To develop an accurate RF-LDMOS large signal model,temperature distribution across a multi-finger device was researched.Using the three dimensional image method,temperature distribution across a finger was easily calculated.Furtherly,for multiple fingers,lateral temperature distribution could be obtained under the superposition principle.A thermal matrix was used to describe the model and the thermal parameters were obtained by fitting with the calculation results.Compared with the traditional thermal model,this model can be used to characterize the heat transferring horizontally by using mutual thermal resistances between fingers and the average thermal resistance network is replaced by distributed thermal RC network to model the multi-fingers.The calculation results show that it can present the temperature distribution and predict the average thermal resistance of scaling devices more accurately.

Hybrid Modeling for Soft Sensing of Molten Steel Temperature in LF

Aiming at the limitations of traditional thermal model and intelligent model, a new hybrid model is established for soft sensing of the molten steel temperature in LF. Firstly, a thermal model based on energy conservation is described; and then, an improved intelligent model based on process data is presented by ensemble ELM (extreme learning machine) for predicting the molten steel temperature in LF. Secondly, the self-adaptive data fusion is proposed as a hybrid modeling method to combine the thermal model with the intelligent model. The new hybrid model could complement mutual advantage of two models by combination. It can overcome the shortcoming of parameters obtained on-line hardly in a thermal model and the disadvantage of lacking the analysis of ladle furnace metallurgical process in an intelligent model. The new hybrid model is applied to a 300 t LF in Baoshan Iron and Steel Co Ltd for predicting the molten steel temperature. The experiments demonstrate that the hybrid model has good generalization performance and high accuracy.

A unified model in the pulsed laser ablation process

In this unified model, we introduce the electron-phonon coupling time (tie) and laser pulse width (tp). For long pulses, it can substitute for the traditional thermal conduction model; while for ultrashort pulses, it can substitute for the standard two-temperature model. As an example of the gold target, we get the dependence of the electron and ion temperature evolvement on the time and position by solving the thermal conduction equation using the finite-difference time-domain (FDTD) method. It is in good agreement with experimental data. We obtain the critical temperature of the onset of ablation using the Saha equation and then obtain the theoretical value of the laser ablation threshold when the laser pulse width ranges from nanosecond to femtosecond timescale, which consists well with the experimental data.

On the Luminosity of Young Jupiters

Traditional thermal evolution models of giant planets employ arbitrary initial conditions selected more for computational expediency than physical accuracy. Since the initial conditions are eventually forgotten by the evolving planet, this approach is valid for mature planets, if not young ones. To explore the evolution at young ages of jovian mass planets, we have employed model planets created by one implementation of the core-accretion mechanism as initial conditions for evolutionary calculations. The luminosities and early cooling rates of young planets are highly sensitive to their internal entropies, which depend on the formation mechanism and are highly model dependent. As a result of the accretion shock through which most of the planetary mass is processed, we find lower initial internal entropies than commonly assumed in published evolution tracks. Consequently, young Jovian planets are smaller, cooler, and several to 100 times less luminous than predicted by earlier models. Furthermore, the time interval during which the young Jupiters are fainter than expected depends on the mass of planet. Jupiter mass planets (1MJ) align with the conventional model luminosity in as little at 20 million years, but 10MJ planets can take up to 1 billion years to match commonly cited luminosities, given our implementation of the core-accretion mechanism. If our assumptions, especially including our treatment of the accretion shock, are correct and if extrasolar Jovian planets indeed form with low entropy, then young Jovian planets are substantially fainter at young ages than currently believed. Furthermore, early evolution tracks should be regarded as uncertain for much longer than the commonly quoted 106 yr. These results have important consequences both for detection strategies and for assigning masses to young Jovian planets based on observed luminosities.

Shear yielding of amorphous glassy solids: effect of temperature and strain rate.

We study shear yielding and steady state flow of glassy materials with molecular dynamics simulations of two standard models: amorphous polymers and bidisperse Lennard-Jones glasses. For a fixed strain rate, the maximum shear yield stress and the steady state flow stress in simple shear both drop linearly with increasing temperature. The dependence on strain rate can be described by either a logarithm or a power law added to a constant. In marked contrast to predictions of traditional thermal activation models, the rate dependence is nearly independent of temperature. The relation to more recent models of plastic deformation and glassy rheology is discussed, and the dynamics of particles and stress in small regions is examined in light of these findings.

Review of Preisach type models driven by stochastic inputs as a model for after-effect

The article reviews a novel approach to the modeling of after-effect phenomena in hysteretic systems. The after-effect model is based on Preisach-type models driven by stochastic inputs. Random thermal perturbations which result in the gradual loss of memory in hysteretic systems are modeled by discrete and continuous-time stochastic processes. The important properties of the after-effect model are summarized and compared with the traditional thermal activation-type models and with some known experimental facts.