Machine Learning-Based Model for Optimal Operating Conditions of Thermosyphons for Electronic Cooling Applications

2021 ◽  
Author(s):  
John Kim ◽  
Raffaele L. Amalfi
Keyword(s):  
Machine Learning ◽  
Thermal Performance ◽  
Cooling System ◽  
Electronic Cooling ◽  
Operating Conditions ◽  
Complex Nature ◽  
System Control ◽  
Optimal Operating ◽  
Two Phase ◽  
Wide Range

Abstract Two-phase cooling systems based on the thermosyphon operating principle exhibit excellent heat transfer performance, reliability, and flexibility, therefore can be applied to overcome thermal challenges in a wide range of electronic cooling applications and deployment scenarios. However, extremely complex nature of two-phase flow physics involving flow patterns and phase transitions has been the major challenge for technology adoption in industry. This paper demonstrates a machine learning (ML) based model for evaluating the thermal performance and refrigerant mass flow rate, of a thermosyphon cooling system for telecom equipment. Unlike conventional laboratory approach that requires numerous sensors attached to a cooling system to capture their thermal performance, the new model requires a minimum number of sensors to monitor the health of a thermal management solution. Using the proposed model, a system control module can be further developed which could identify optimal operating parameters in real-time under dynamically changing heat load conditions and actively maintain safety and thermal requirements.

Experimental Validation and Design Simulations of a Passive Two-Phase Cooling System for Datacenters

2020 ◽  
Author(s):  
Jackson B. Marcinichen ◽  
John R. Thome ◽  
Raffaele L. Amalfi ◽  
Filippo Cataldo
Keyword(s):  
Thermal Performance ◽  
Experimental Validation ◽  
Cooling System ◽  
Electronics Cooling ◽  
Operating Conditions ◽  
Two Phase ◽  
Data Set ◽  
Pressure Drops ◽  
Wide Range ◽  
Important Challenge

Abstract Thermosyphon cooling systems represent the future of datacenter cooling, and electronics cooling in general, as they provide high thermal performance, reliability and energy efficiency, as well as capture the heat at high temperatures suitable for many heat reuse applications. On the other hand, the design of passive two-phase thermosyphons is extremely challenging because of the complex physics involved in the boiling and condensation processes; in particular, the most important challenge is to accurately predict the flow rate in the thermosyphon and thus the thermal performance. This paper presents an experimental validation to assess the predictive capabilities of JJ Cooling Innovation’s thermosyphon simulator against one independent data set that includes a wide range of operating conditions and system sizes, i.e. thermosyphon data for server-level cooling gathered at Nokia Bell Labs. Comparison between test data and simulated results show good agreement, confirming that the simulator accurately predicts heat transfer performance and pressure drops in each individual component of a thermosyphon cooling system (cold plate, riser, evaporator, downcomer (with no fitting parameters), and eventually a liquid accumulator) coupled with operational characteristics and flow regimes. In addition, the simulator is able to design a single loop thermosyphon (e.g. for cooling a single server’s processor), as shown in this study, but also able to model more complex cooling architectures, where many thermosyphons at server-level and rack-level have to operate in parallel (e.g. for cooling an entire server rack). This task will be performed as future work.

Operational Map and Thermal Performance of a Thermosyphon Cooling System for Compact Servers

2021 ◽  
Author(s):  
Raffaele L. Amalfi ◽  
Cong H. Hoang ◽  
Ryan Enright ◽  
Filippo Cataldo ◽  
Jackson B. Marcinichen ◽  
...  
Keyword(s):  
Heat Flux ◽  
Thermal Performance ◽  
Cooling System ◽  
Heat Removal ◽  
Total Power ◽  
Two Phase ◽  
Maximum Heat ◽  
Wide Range ◽  
Experimental Apparatus ◽  
Secondary Side

Abstract This paper advances the state-of-the-art in novel passive two-phase systems for more efficient cooling of datacenters and telecom central offices compared to the traditional air-based cooling solutions (e.g. aisle-based containment systems). The proposed passive two-phase technology uses numerous server-level thermosyphons to dissipate the heat generated by critical components, such as central processing units, accelerators, etc., with the flexibility of selecting the rack-level and room-level cooling elements depending on the deployment scenarios. The main goal of this paper is to experimentally investigate the thermal performance and maximum heat removal capability of a server-level thermosyphon for cooling compact servers. The experimental apparatus, built at Nokia Bell Labs, incorporates a single 7-cm high liquid-cooled thermosyphon that fits within a 2U server (smaller form factors can be achieved by a proper design that would further reduce the thermosyphon’s height). The heat source is represented by a pseudo-chip, composed of six parallel cartridge heaters installed in a copper block that incorporates local temperature measurements and is able of dissipating a total power of ≈ 500 W over a footprint area of 3.5 cm × 3.5 cm (corresponding heat flux of ≈ 41 W/cm2). Steady-state experiments were carried out at various heat loads up to 240 W (corresponding heat flux of ≈ 20 W/cm2), filling ratios and secondary side inlet conditions (coolant temperatures and mass flow rates), using R1234ze(E) and deionized water as the working fluids on the primary and secondary side, respectively. Test results demonstrate high heat transfer performance of the server-level thermosyphon over a wide range of conditions, and operating points are identified and classified into an operational map. Thermosyphon-based cooling systems across multiple length scales can significantly improve operation in terms of lowering energy consumption, allowing for higher hardware density, increased processing speed and reliability.

COVID-19 Outbreak Prediction with Machine Learning

2020 ◽  
Author(s):  
Sina Faizollahzadeh Ardabili ◽  
Amir Mosavi ◽  
Pedram Ghamisi ◽  
Filip Ferdinand ◽  
Annamaria R. Varkonyi-Koczy ◽  
...  
Keyword(s):  
Machine Learning ◽  
Prediction Models ◽  
Fuzzy Inference ◽  
Control Measures ◽  
Future Research ◽  
Complex Nature ◽  
Inference System ◽  
Wide Range ◽  
Standard Models ◽  
High Level

Several outbreak prediction models for COVID-19 are being used by officials around the world to make informed-decisions and enforce relevant control measures. Among the standard models for COVID-19 global pandemic prediction, simple epidemiological and statistical models have received more attention by authorities, and they are popular in the media. Due to a high level of uncertainty and lack of essential data, standard models have shown low accuracy for long-term prediction. Although the literature includes several attempts to address this issue, the essential generalization and robustness abilities of existing models needs to be improved. This paper presents a comparative analysis of machine learning and soft computing models to predict the COVID-19 outbreak as an alternative to SIR and SEIR models. Among a wide range of machine learning models investigated, two models showed promising results (i.e., multi-layered perceptron, MLP, and adaptive network-based fuzzy inference system, ANFIS). Based on the results reported here, and due to the highly complex nature of the COVID-19 outbreak and variation in its behavior from nation-to-nation, this study suggests machine learning as an effective tool to model the outbreak. This paper provides an initial benchmarking to demonstrate the potential of machine learning for future research. Paper further suggests that real novelty in outbreak prediction can be realized through integrating machine learning and SEIR models.

Combustion Tuning for a Gas Turbine Power Plant Using Data-Driven and Machine Learning Approach

2021 ◽  
Vol 143 (3) ◽  
Author(s):  
Suhui Li ◽  
Huaxin Zhu ◽  
Min Zhu ◽  
Gang Zhao ◽  
Xiaofeng Wei
Keyword(s):  
Machine Learning ◽  
Power Plant ◽  
Gas Turbine ◽  
Operating Conditions ◽  
Data Driven ◽  
Ann Model ◽  
Promising Alternative ◽  
Combustion Performance ◽  
Wide Range ◽  
Gas Turbine Power Plant

Abstract Conventional physics-based or experimental-based approaches for gas turbine combustion tuning are time consuming and cost intensive. Recent advances in data analytics provide an alternative method. In this paper, we present a cross-disciplinary study on the combustion tuning of an F-class gas turbine that combines machine learning with physics understanding. An artificial-neural-network-based (ANN) model is developed to predict the combustion performance (outputs), including NOx emissions, combustion dynamics, combustor vibrational acceleration, and turbine exhaust temperature. The inputs of the ANN model are identified by analyzing the key operating variables that impact the combustion performance, such as the pilot and the premixed fuel flow, and the inlet guide vane angle. The ANN model is trained by field data from an F-class gas turbine power plant. The trained model is able to describe the combustion performance at an acceptable accuracy in a wide range of operating conditions. In combination with the genetic algorithm, the model is applied to optimize the combustion performance of the gas turbine. Results demonstrate that the data-driven method offers a promising alternative for combustion tuning at a low cost and fast turn-around.

scholarly journals Cooling Performance of Arrays of Vibrating Cantilevers

2009 ◽  
Vol 131 (11) ◽  
Author(s):  
Mark Kimber ◽  
Suresh V. Garimella
Keyword(s):  
Heat Transfer ◽  
Thermal Performance ◽  
Vibration Amplitude ◽  
Past Research ◽  
Resonant Mode ◽  
Operating Conditions ◽  
Infrared Thermal Imaging ◽  
Practical Applications ◽  
Wide Range ◽  
Piezoelectric Fans

Piezoelectric fans are vibrating cantilevers actuated by a piezoelectric material and can provide heat transfer enhancement while consuming little power. Past research has focused on feasibility and performance characterization of a single fan, while arrays of such fans, which have important practical applications, have not been widely studied. This paper investigates the heat transfer achieved using arrays of cantilevers vibrating in their first resonant mode. This is accomplished by determining the local convection coefficients due to the two piezoelectric fans mounted near a constant heat flux surface using infrared thermal imaging. The heat transfer performance is quantified over a wide range of operating conditions, including vibration amplitude (7.5–10 mm), distance from heat source (0.01–2 times the fan amplitude), and pitch between fans (0.5–4 times the amplitude). The convection patterns observed are strongly dependent on the fan pitch, with the behavior resembling a single fan for small fan pitch and two isolated fans at a large pitch. The area-averaged thermal performance of the fan array is superior to that of a single fan, and correlations are developed to describe this enhancement in terms of the governing parameters. The best thermal performance is obtained when the fan pitch is 1.5 times its vibration amplitude.

scholarly journals Current Advances in Ejector Modeling, Experimentation and Applications for Refrigeration and Heat Pumps. Part 2: Two-Phase Ejectors

Inventions ◽  
2019 ◽  
Vol 4 (1) ◽  
pp. 16 ◽  
Author(s):  
Zine Aidoun ◽  
Khaled Ameur ◽  
Mehdi Falsafioon ◽  
Messaoud Badache
Keyword(s):  
Heat Pumps ◽  
Operating Conditions ◽  
Industrial Processes ◽  
Experimental Investigations ◽  
Mixing Enhancement ◽  
Data Generation ◽  
Two Phase ◽  
Wide Range ◽  
Potential Applications ◽  
And Performance

Two-phase ejectors play a major role as refrigerant expansion devices in vapor compression systems and can find potential applications in many other industrial processes. As a result, they have become a focus of attention for the last few decades from the scientific community, not only for the expansion work recovery in a wide range of refrigeration and heat pump cycles but also in industrial processes as entrainment and mixing enhancement agents. This review provides relevant findings and trends, characterizing the design, operation and performance of the two-phase ejector as a component. Effects of geometry, operating conditions and the main developments in terms of theoretical and experimental approaches, rating methods and applications are discussed in detail. Ejector expansion refrigeration cycles (EERC) as well as the related theoretical and experimental research are reported. New and other relevant cycle combinations proposed in the recent literature are organized under theoretical and experimental headings by refrigerant types and/or by chronology whenever appropriate and systematically commented. This review brings out the fact that theoretical ejector and cycle studies outnumber experimental investigations and data generation. More emerging numerical studies of two-phase ejectors are a positive step, which has to be further supported by more validation work.

scholarly journals Climate- and Technology-Specific PUE and WUE Predictions for U.S. Data Centers using a Physics-Based Approach

2021 ◽  
Author(s):  
Nuoa Lei ◽  
Eric Masanet
Keyword(s):  
Water Use ◽  
Data Centers ◽  
Cooling System ◽  
Operating Conditions ◽  
Sensitivity Analyses ◽  
Real World Data ◽  
Climate Zone ◽  
System Type ◽  
Tower Equipment ◽  
Wide Range

Abstract The onsite water use of data centers (DCs) is becoming an increasingly important consideration within the policy and energy analysis communities, but has heretofore been difficult to quantify in macro-level DC energy models due to lack of reported water usage effectiveness (WUE) values by DC operators. This work addresses this important knowledge gap by presenting thermodynamically-compatible power usage effectiveness (PUE) and WUE values for a wide range of U.S. DC archetypes and climate zones, using a physics-based model that is validated with real-world data. Results enable energy analysts to more accurately analyze the onsite energy and water use of DCs by size class, cooling system type, and climate zone under many different operating conditions including operational setpoints. Sensitivity analyses further identify the variables leading to best-achievable PUE and WUE values by climate zone and cooling system type—including operational set points, use of free cooling, and cooling tower equipment and operational factors—which can support DC water- and energy-efficiency policy initiatives. The consistent PUE and WUE values may also be used in future work to quantify the indirect water use of DCs occurring in electrical power generating systems.

scholarly journals COVID-19 Outbreak Prediction with Machine Learning

2020 ◽  
Author(s):  
Sina Faizollahzadeh Ardabili ◽  
Amir Mosavi ◽  
Pedram Ghamisi ◽  
Filip Ferdinand ◽  
Annamaria R. Varkonyi-Koczy ◽  
...  
Keyword(s):  
Machine Learning ◽  
Prediction Models ◽  
Fuzzy Inference ◽  
Control Measures ◽  
Future Research ◽  
Complex Nature ◽  
Inference System ◽  
Wide Range ◽  
Standard Models ◽  
High Level

Several outbreak prediction models for COVID-19 are being used by officials around the world to make informed-decisions and enforce relevant control measures. Among the standard models for COVID-19 global pandemic prediction, simple epidemiological and statistical models have received more attention by authorities, and they are popular in the media. Due to a high level of uncertainty and lack of essential data, standard models have shown low accuracy for long-term prediction. Although the literature includes several attempts to address this issue, the essential generalization and robustness abilities of existing models needs to be improved. This paper presents a comparative analysis of machine learning and soft computing models to predict the COVID-19 outbreak as an alternative to SIR and SEIR models. Among a wide range of machine learning models investigated, two models showed promising results (i.e., multi-layered perceptron, MLP, and adaptive network-based fuzzy inference system, ANFIS). Based on the results reported here, and due to the highly complex nature of the COVID-19 outbreak and variation in its behavior from nation-to-nation, this study suggests machine learning as an effective tool to model the outbreak. This paper provides an initial benchmarking to demonstrate the potential of machine learning for future research. Paper further suggests that real novelty in outbreak prediction can be realized through integrating machine learning and SEIR models.

scholarly journals Program for determination of zones of gold concentration on sluices of washing devices

2021 ◽  
Vol 895 (1) ◽  
pp. 012002
Author(s):  
V S Alekseev ◽  
R S Seryi
Keyword(s):  
High Performance ◽  
Two Phase Flow ◽  
Experimental Studies ◽  
Operating Conditions ◽  
Phase Flow ◽  
Technological Parameters ◽  
Two Phase ◽  
Alluvial Gold ◽  
Wide Range ◽  
Host Rocks

Abstract Currently sluice washing devices are the most common in alluvial gold mining. Their use provides a sufficiently high performance, relatively low power consumption, and acceptable recovery of valuable components. The theoretical provisions of traditional hydraulics make it possible to determine all the main parameters of the movement of particles of rocks and gold in the pulp, however, in real operating conditions of the sluice box, their actual values will differ greatly from the calculated ones, especially if there are solid fractions in the pulp with a particle size of more than 20 mm. This is explained by significant fluctuations in the values of the surface, average and bottom velocities of the two-phase flow, vertical pulsation velocity in conditions of constrained movement of the different fractional composition of rocks. The article presents the results of experimental studies to identify the dependence of the distance traveled by an individual gold particle and host rocks in a two-phase flow through a sluice, the bottom of which is lined with trapping coatings, on the design and technological parameters of the flushing device. The mathematical model for determining this distance formed the basis of the Gold Enriching program. The program allows, in a wide range of initial data, to determine the zones of concentration of gold of a certain size at the sluice boxes.

Sign in / Sign up

Export Citation Format

Share Document