| 1. |
- Gehrmann, Sebastian, et al.
(author)
-
GEMv2: Multilingual NLG Benchmarking in a Single Line of Code
- 2022
-
In: Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing: System Demonstrations. - : Association for Computational Linguistics (ACL). ; , s. 266-281
-
Conference paper (peer-reviewed)abstract
- Evaluations in machine learning rarely use the latest metrics, datasets, or human evaluation in favor of remaining compatible with prior work. The compatibility, often facilitated through leaderboards, thus leads to outdated but standardized evaluation practices. We pose that the standardization is taking place in the wrong spot. Evaluation infrastructure should enable researchers to use the latest methods and what should be standardized instead is how to incorporate these new evaluation advances.We introduce GEMv2, the new version of the Generation, Evaluation, and Metrics Benchmark which uses a modular infrastructure for dataset, model, and metric developers to benefit from each other’s work. GEMv2 supports 40 documented datasets in 51 languages, ongoing online evaluation for all datasets, and our interactive tools make it easier to add new datasets to the living benchmark.
|
|
| 2. |
- Hu, Ming-Hsuan, et al.
(author)
-
Experimental analysis of submerged coil and encapsulated slab latent heat storage
- 2022
-
In: Applied Thermal Engineering. - : Elsevier BV. - 1359-4311 .- 1873-5606. ; 209
-
Journal article (peer-reviewed)abstract
- Latent heat thermal energy storage (LHTES) has two primary LHTES packing methods, encapsulation and bulk PCM storage. Since there is still a lack of experimental comparison between these two methods, two types of LHTES units based on separated methods were built for direct comparison. Moreover, the impact of heat transfer fluid (HTF) flowrate on LHTES performance was evaluated. Unit one is a 0.38 m(3) tank containing slab-shaped macro-encapsulated phase change material (PCM); unit two is a 0.54 m(3) tank containing submerged spiral coil heat exchanger (SCHE) in PCM. PCM with a melting temperature of 58-60 ? was charged/discharged between 46 and 72 ?. Parametric studies on constant and time-varying HTF flowrates were conducted to test the impact on the thermal storage performance.& nbsp;The time-varying flowrate control enables the system to supply the needed power at different discharging stages. Moreover, partial charging/discharging demonstrates higher mean thermal power than full charge/discharge rendering this control strategy adequate under specific operating conditions. Finally, the comparison between the slab-encapsulated PCM storage unit and the SCHE based unit shows that the former requires a shorter completion time while the latter has a higher energy storage density.
|
|
| 3. |
- Khan, Mohammad, et al.
(author)
-
Assessment of PECM as an efficient numerical analysis tool for investigating convective heat transfer phenomena during PCM melting
- 2019
-
In: Journal of Energy Storage. - : Elsevier BV. - 2352-152X .- 2352-1538. ; 24, s. 100743-
-
Journal article (peer-reviewed)abstract
- In the framework of this research work, the principle focus is to assess the applicability & reliability of the Phase change Effective Convectivity Model (PECM) as a numerical analysis tool to investigate natural convective heat transfer in single and two-fluid density PCM molten pools. The model is applied in ANSYS FLUENT as User Defined Function (UDF) to predict convective melt pool thermal hydraulics in a volumetrically heated PCM (Phase Change Material) melt pool. As a part of this work, PECM is tested first by a benchmark case against CFD to gain confidence in its applicability as an analysis tool. Two commercial PCMs: RT50 and C58, are introduced in a 3D semicircular vessel slice with their thermo-physical properties as input for modelling. The sidewalls made of quartz glass are used for direct visualization of convective heat transfer phenomena. It is ensured that the conditions of nearly constant density of power deposition over the entire volume of the PCM melt pool throughout the series of simulation cases. The values of characteristic numbers ranged within the following limits with different pool height corresponding modified Rayleigh number Ra=1012-1013 and for Prandtl number Pr=5-7. The selected modelling approach is validated against SIGMA experiment with respect to the angular distribution of heat flux that qualify our model to run in the proceeding calculation using PECM. Following benchmark test results of PECM compared with that of conventional enthalpy porosity method embedded in ANSYS FLUENT, PECM is applied in 1-layer and 2-layer PCM configuration to study in details of the influence of different boundary conditions, internal heat sources (QV) and heat transfer fluid (HTF) cooling condition to quantify the thermal loads. Finally, the comparison is made between two PCM configurations in terms of the quantification of the thermal load to justify PECM as an efficient numerical analysis tool for investigating convective heat transfer phenomena during PCM melting.
|
|
| 4. |
- Konig-Haagen, Andreas, et al.
(author)
-
Analysis of the discharging process of latent heat thermal energy storage units by means of normalized power parameters
- 2023
-
In: Journal of Energy Storage. - : Elsevier BV. - 2352-152X .- 2352-1538. ; 72
-
Journal article (peer-reviewed)abstract
- Many efforts are being made to mitigate the main disadvantage of most phase change materials - their low thermal conductivities - in order to deliver latent heat energy storage systems (LHESS) with adequate perfor-mance. However, the effect of applied methods is difficult to compare as they are mostly tested for different storage types and sizes and/or different boundary and initial conditions, which hinders rapid progress in the optimization of these approaches. In this work, a previously developed method for comparing the performance of LHESS is applied to experimental results of different storage systems under different conditions and subsequently analyzed and further refined. The main idea of the method is to normalize the power with the volume and a reference temperature difference and compare its mean value plotted over the normalized mean capacity flow of the heat transfer fluid (HTF). This enables the presentation of the results in a compact and easily comparative way. Attention has to be paid when it comes to the choice of the reference temperature difference, the reference volume and the method for calculating the mean value. Two variants of calculating the mean value (time-weighted and energy-weighted) and two variants of reference temperatures for determining the temperature difference to the inlet temperature of the HTF (initial temperature and melting temperature) are applied and discussed in detail. While the method significantly increases the comparability of results, none of the options listed above are without drawbacks. Approaches are shown to reduce or eliminate these drawbacks in the future. The recommendation for comparing different LHESS under different conditions is to use the method described here and clearly state the chosen reference temperature, reference volume and method for calculating the mean value.
|
|
| 5. |
- Xu, Tianhao, et al.
(author)
-
Comparing integrating configurations of latent heat storage in heat pump heating systems
- 2019
-
In: Proceedings of the 25th IIR International Congress of Refrigeration. - : International Institute of Refrigeration. ; , s. 4883-4890
-
Conference paper (peer-reviewed)abstract
- Short-term latent heat thermal energy storage (LHTES) can be integrated in building heating systems for performing load shifting from on- to off-peak hours. In this study, the coupling effects of LHTES integration in heat pump heating systems are investigated for a multi-family building in Stockholm. As the reference heat pump provides heat in two separate heat exchangers, a de-superheater for domestic hot water and a condenser for space heating, two configurations for LHTES integration with the two respective heat exchangers are compared using a quasi-stationary mathematic system model. The performances of the two integrating configurations are also compared with the basic heat pump heating scenario (no heat storage) regarding the total electricity consumption, the seasonal performance factor (SPF), and the electric bills for end-users over a selected heating-season week. The simulation results reveal that, while using the condenser for charging an LHTES unit containing phase change materials melted at 58 oC inevitably increases the total electricity consumption in the heating system relative to the basic scenario, using the desuperheater instead for LHTES charging can reduce the total electricity consumptions and bills of 24 kWh and 173 SEK (17€), respectively, during that specific week.
|
|
| 6. |
- Xu, Tianhao, et al.
(author)
-
Design Aspects of a Latent Heat Storage Unit for Heat Production Shifting at a Cogeneration Plant
- 2019
-
In: SWC2019 Proceedings.
-
Conference paper (peer-reviewed)abstract
- In the present study, a latent heat thermal energy storage (LHTES) unit with two different configurations (a shell-and-tube design using spiral coils as tubes and an encapsulation design using commercial capsules) are investigated and compared over their thermal performance for providing heat storage and recovery. The designed latent heat storage unit is to be implemented at an existing cogeneration plant for heat production shifting purposes. The design procedure involves several aspects of theoretical investigations: the determination of suitable melting point of the employed phase-change material (PCM); the selection of heat exchanger configuration; and the prediction of the units’ transient charging/discharging thermal behavior under operating conditions set by the cogeneration plant. Numerical approaches are used in this study to estimate the heat transfer conditions in PCMs as well as the transient charging/discharging thermal power of the entire unit. The accumulative stored/released energy throughout a charging process of three hours and a discharging process of one hour is also calculated. With the cylindrical containment tank in the same geometry, the spiral coil design exhibits a 52%-higher total heat storage capacity than the encapsulation design, and the simulation results show that it can store a higher amount of heat by 20% after first three hours of charging. For discharging, however, the encapsulation design exhibits a higher completion rate of 96% than the spiral coil design (53%) after first hour of discharging. Besides, the heat recovery capacity with the encapsulation design is 26% higher. The spiral coil design therefore shows advantage in the charging mode in terms of higher storage capacity while the encapsulation design outperforms when discharging due to is higher discharge rate within the given discharge operating duration of one hour.
|
|
| 7. |
- Xu, Tianhao, et al.
(author)
-
Experimental and Numerical Investigation of a Latent Heat Thermal Energy Storage Unit with Ellipsoidal Macro-encapsulation
-
Other publication (pop. science, debate, etc.)abstract
- Experimental and Numerical Investigation of a Latent Heat Thermal Energy Storage Unit with Ellipsoidal Macro-encapsulationTianhao Xu, Emma Nyholm Humire, Silvia Trevisan, Monika Ignatowicz, Samer Sawalha, Justin NW Chiu*Department of Energy Technology, KTH Royal Institute of Technology, 100 44, Stockholm, SwedenAbstractIn this paper, a novel type of macro-encapsulated phase change material (PCM)shaped in ellipsoidis investigated on a component scale.The encapsulated PCM is a paraffin-based commercial material (ATP60); differential scanning calorimetry and transient plane source method are used to measure the thermo-physical properties of the PCM. A numerical model and a 0.382 m3latent heat thermal energy storage (LHTES) prototype have been built and experimentally characterized. The temperature measurements indicate that thermocline is retainedin the packed bed region. The charge/discharge can be accelerated by65% with20 K increase intemperature difference between the phase-change temperature and heat transfer fluid (HTF)inlet temperature. Increasing HTF inlet flowrate from 0.15 m3/h (Re=77) to 0.5 m3/h (Re=256) shortenscompletion time of charge by51%. Furthermore, a one-dimensional packed bed using source based enthalpy method was constructed and validatedwithin6.6% errorfor aReynolds numberof 90 to 922. Compared with a conventional cylindrical PCM capsule, the ellipsoidal encapsulation shows60% increase in discharge speed but 23%lower storage capacity. This work demonstrates a combined experimental and numerical characterization approach for a novel ellipsoidal PCM encapsulationgeometry.
|
|
| 8. |
- Xu, Tianhao, et al.
(author)
-
Experimental and numerical investigation of a latent heat thermal energy storage unit with ellipsoidal macro-encapsulation
- 2022
-
In: Energy. - : Elsevier BV. - 0360-5442 .- 1873-6785. ; 238
-
Journal article (peer-reviewed)abstract
- This paper investigates ellipsoid-shaped macro-encapsulated phase change material (PCM) on a component scale. The selected PCM is a paraffin-based commercial material, namely ATP60; differential scanning calorimetry and transient plane source method are used to measure ATP60's thermo-physical properties. A 0.382 m(3) latent heat thermal energy storage (LHTES) component has been built and experimentally characterized. The temperature measurement results indicate that a thermocline was retained in the packed bed region during charging/discharging processes. The experimental characterization shows that increasing the temperature difference between the heat transfer fluid (HTF) inlet temperature and phase-change temperature by 20 K can shorten the completion time of discharge by 65%, and increasing HTF inlet flowrate from 0.15 m(3)/h (Re = 77) to 0.5 m(3)/h (Re = 256) can shorten the completion time of charge by 51%. Furthermore, a one-dimensional packed bed model using source-based enthalpy method was developed and validated by comparison to experimental results, showing discrepancies in the accumulated storage capacity within 6.6% between simulation and experiment when the Reynolds number of the HTF inlet flow ranges between 90 and 922. Compared with a conventional capsule shaped in 69-mm-diameter and 750-mm-long cylinders, the ellipsoidal capsule shows 60% less completion time of discharge but 23% lower storage capacity. Overall, this work demonstrates a combined experimental and numerical characterization approach for applying novel macro-encapsulated PCM geometries for heating-oriented LHTES.
|
|
| 9. |
- Xu, Tianhao, et al.
(author)
-
Experimental investigation on cylindrically macro-encapsulated latent heat storage for space heating applications
- 2019
-
In: Energy Conversion and Management. - : Elsevier. - 0196-8904 .- 1879-2227. ; 182, s. 166-177
-
Journal article (peer-reviewed)abstract
- The integration of latent heat thermal energy storage (LHTES) units with heating systems in buildings is regarded as a promising technology for heating load management; however, so far a limited number of experimental studies have been reported that focus on space heating applications on a representative scale. In this study, we develop and test a 0.38 m3 LHTES unit containing cylindrically macro-encapsulated phase change materials (PCMs) with a melting temperature range of 44–53 °C and with gross mass of 154 kg. The unit has been tested with two tank orientations, horizontal and vertical. In the horizontal orientation tests, parametric studies show that increasing the difference between heat transfer fluid (HTF) supply temperatures and phase-change temperatures of PCMs, as well as increasing HTF flowrates, can both reduce the complete melting/solidification and complete charging/discharging time. Non-linear charging/discharging rates in PCMs are observed. The vertical orientation enables the forming of either a stratified or mixed flow regime in the tank. For charging, the stratified flow provides higher charging rates in PCMs compared to the mixed flow. When discharging the unit with a stratified HTF flow at 35 °C, lower HTF flowrates prolong the discharging time during which the released heat sustains an outlet temperature above 45 °C. Finally, comparisons between horizontal and vertical orientation tests reveal that although the vertical orientation can shorten the charging/discharging time by up to 20% for the entire unit to reach an energy density of 30 kWh/m3, it leads to decrease in PCM thermal capacity by at most 8.2%. The speculated cause of this loss is phase segregation suggested by observed fluid motions in PCM cylinders. This study comprehensively characterizes an LHTES unit providing insights to optimizing its operating strategies considering its coupling with space heating systems.
|
|
| 10. |
- Xu, Tianhao, 1992-
(author)
-
Integrating Latent Heat Storage into Residential Heating Systems : A study from material and component characterization to system analysis
- 2021
-
Doctoral thesis (other academic/artistic)abstract
- Latent heat thermal energy storage (LHTES) systems can be coupled with heat pump (HP) systems to realize heat load shifting on demand side. With phase change material (PCM), well designed LHTES components exhibit high storage energy density and thus have large potentials to be integrated in residence where a compact energy storage solution is needed. However, real installations of LHTES-HP integrated systems are still rare nowadays; feasibility of this technology in achieving technically sound and economically viable load shifting operations should be demonstrated and understood by stakeholders to promote its implementation. Therefore, this thesis presents an exemplary feasibility study for three selected off-the-shelf macro-encapsulated PCM products, encompassing in-depth experimental and numerical modelling investigations on three levels—material, component, and system. The feasibility is studied with a specific scenario where the macro-encapsulated LHTES systems are designed to integrate with HP-based heating systems in common Swedish residential buildings. On the material level, three commercial PCMs (C48, C58, and ATP60) are selected by the operating temperature levels in typical HP-based space heating systems. Differential Scanning Calorimetry and Temperature-History method are employed to measure PCM enthalpy-temperature profiles; Transient Plane Source method is used to measure the thermal conductivity of ATP60. C58 based on sodium acetate trihydrate is prioritized for in-depth feasibility analyses because of its highest volumetric heat storage capacity.On the component level, three full-scale macro-encapsulated LHTES components (Component 1: cylindrical encapsulation of C48; Component 2: cylindrical encapsulation of C58; Component 3: ellipsoidal encapsulation of ATP60) are developed for integration in single-family houses to achieve full peak load shifting. A test rig is built for characterizing the three components under possible operational conditions in practical systems. The heat transfer enhancing effects from increasing the temperature difference between heat transfer fluid (HTF) inlet temperature and phase-change temperature as well as from increasing the HTF inlet flowrate are quantified. Performance indicators, such as completion time of charge/discharge, energy storage density, and capacity enhancement factor, are evaluated at different operating temperature ranges. Overall, Component 2 is feasible in delivering around 90% of storage capacity (the capacity loss is due to phase separation). However, storage design and control improvements are still needed for realizing full peak load shifting over a three-hour discharging process. For Component 2, an improved storage solution with a reduced capsule diameter and time-increasing HTF flowrate profiles is developed through numerical simulation using an experimentally-validated two-dimensional heat transfer model. Furthermore, a one-dimensional model is developed and validated for simulating storage thermal output of Components 2 and 3. On the system level, a numerical model is developed to calculate electricity input to the LHTES-HP integrated systems for technical, economic, and environmental load shifting evaluation. Three new integration layouts are developed to charge scaled-up Component 2 with a de-superheater and/or a booster heat pump cycle. The new layouts can improve the weekly heating performance factor by 22%–26%, compared with a conventional layout using the condenser for charging. Savings in operational expenses can justify a capital expense of 25,000 Swedish Krona (about 2,500 €) for the LHTES system with a 15-yr operation. Although this justifiable capital expense is lower than the storage component cost alone estimated with the cost of Component 2, it is anticipated that similar LHTES solutions may gain more economic feasibilities with larger peak-valley electricity price differences foreseeable in the future. Through presentation of the multi-level feasibility evaluation, this thesis identifies key design and operational issues which might be neglected in single-level investigations. Furthermore, the thesis develops two new LHTES-HP integrated solutions with improved storage design/control strategies and enhanced system coupling methods from the existing solutions. This provides application-oriented insight for design and operation of the load-shift based LHTES installations in residential buildings, potentially contributing to decarbonisation of the increasingly electrified heating sector.
|
|
