Elsevier

Applied Thermal Engineering

Volume 21, Issue 17, December 2001, Pages 1769-1778
Applied Thermal Engineering

An ice thermal storage computer model

https://doi.org/10.1016/S1359-4311(01)00046-1Get rights and content

Abstract

In hot humid countries such as Thailand, air conditioning plant is installed in most commercial and industrial buildings. A conventional air conditioning system, which is normally operated when cooling is required, is the most favored option. Ice thermal storage on a large scale, used to provide a cool reservoir for use in peak periods, is however an attractive financial option for large buildings to supply coolness. There are two means of operating ice thermal storage systems, namely full storage and partial storage.

In this paper, a computer model has been developed in order to compare energy use in conventional air cooling systems and ice thermal storage systems. Under Thailand electricity tariff rates, the results from the simulations show that the full ice thermal storage can save up to 55% of the electricity cost required for cooling per month when compared with the conventional system. It is also found that using full storage option can reduce the total energy consumption by 5% for the selected building.

Introduction

On average, cooling of commercial buildings in Thailand can consume approximately 40–60% of their total electricity usage. The cooling process mostly occurs during peak periods resulting in high national electricity peak demand. The Electricity Generation Authority of Thailand (EGAT) has imposed demandside management programs and also reconstructed electricity tariff rates to persuade users to move their electricity demand to off-peak periods. Time of use (TOU) rate is a new tariff rate that is applied to medium and large scale users to ensure that peak demands which occur in peak periods are charged at a very high rate compared with those in non-peak periods [1].

Ice thermal storage technique is an alternative strategy that businesses can use to avoid the high peak-load demand charge for space cooling. The principle of this technique is that coolness is produced at night in the form of ice and the ice is stored in a well-insulated storage. The coolness is then extracted from the ice and used for daytime cooling requirement whereby daytime peak demand is reduced and shifted.

Ice thermal storage can be categorized into five techniques, namely external melt ice on coil, internal melt ice on coil, ice harvesting, encapsulated ice, and slurry ice. Ice thermal storage system can be operated as full or partial storage depending on design, partial storage meaning in this context that the daytime cooling load is met partially from the storage and partially from the chiller running during the day. Details of ice thermal storage system are explained clearly in the Design Guide for Cool Thermal Storage [2] and ASHRAE Application Handbook [3].

Apart from the economic aspects, ice thermal storage also offers an improvement on air conditioning system performance [3]. It is clear that if ice thermal storage is applied the compressor of the air conditioner will be running at almost constant load, so that transient loss is minimized. Moreover, the system is operated at night where ambient temperature is lower than daytime giving a lower condensing temperature and therefore higher coefficient of performance (COP) of the refrigeration system can be achieved. This is to some extent negated by the lower evaporation temperature recovery to overcome the heat transfer resistance of the ice.

There are a number of research works reporting on benefits of ice thermal storage. However, quantitative benefits due to higher COP at night have not been reported in the open literature surveyed. At higher COP, the refrigeration systems consume lower input energy per unit of cooling. Thus due to the higher COP this technique has a potential of energy saving. In addition, modeling of ice storage and refrigeration systems that are working simultaneously to study the interactive effects is not readily available from the published work.

This paper presents quantitative benefits of using ice thermal storage obtained from the computer model developed. The model is used to simulate a conventional cooling system and full storage and partial storage systems. Energy consumption, energy cost saving and COP of the compressor for each system are described.

Section snippets

Modeling of ice-on-coil thermal storage systems

There are a few models available for simulation of internal melt ice-on-coil storage, for instance those dealt by Silver et al. [4], Jekel et al. [5], Vick et al. [6], Drees and Braun [7], and Neto and Krarti [8]. In all of these models, spiral tubes are divided into a finite number of segments. Heat transfer calculations have been performed for each segment using thermal resistance network technique. The total heat transfer is then aggregated from individual contributions from each segment.

In

Cooling system design options

Based on consideration of the TOU tariff rates, cooling load met by the compressors and COP, a comparison is made between three possible systems providing equal coolness for a typical office building (conditioned area of approximately 4700 m2) in Bangkok, Thailand. The base design is the conventional central cooling systems, which consist of a compressor, an air-cooled condenser, a thermal expansion valve, and an evaporator. The other two systems are partial ice storage systems and full ice

Simulation results

The results of this study were obtained from the model developed by Chaichana [9] using FORTRAN 90 programming language and compiled by Digital Fortran Compiler (for Microsoft Windows NT) version 6.0. The model was tested with Bangkok temperature data and the cooling loads of a specified office building (Fig. 1).

In the simulation, the time step used is 1 min and the results shown in this section are on an hourly basis. The hour between midnight and 1 a.m. is designated as hour one, 1 and 2 a.m.

Energy consumption

The total energy consumption for each technique is comprized of energy consumed by the compressor, liquid pump, and condensing fan and is shown graphically in Fig. 2. For each technique, the total consumption for the three types of days is also presented.

Results from the full storage show that it consumes the least energy of all the techniques studied. It should be noted that for the full storage design, the chiller operates only during the nighttime in which the ambient temperature is lower

Conclusions

A dynamic computer model for ice thermal storage systems has been developed for studying the effects of ambient temperature on energy usage in a refrigeration plant. The results from the simulation show that the model developed can be used to model a range of ice thermal storage systems. The following conclusions can be drawn:

  • The lower ambient temperature during nighttime can reduce the total energy consumption by 5% using full storage as the cooling system for the chosen building.

  • The partial

Acknowledgements

The authors would like to thank the International Technologies Centre (IDTC), the Department of Civil and Environmental Engineering and the Department of Mechanical and Manufacturing Engineering. The sponsorship for this study at the University of Melbourne from the Thai Government is gratefully acknowledged by Chatchawan Chaichana.

References (12)

  • Electricity tariffs, Provincial Electricity Authority (PEA), http://www.pea.or.th/rate_e.htm (accessed 4 February,...
  • C.E. Dorgan, J.S. Elleson, Design Guide for Cool Thermal Storage, USA,...
  • ASHRAE, Thermal Storage, 1995 ASHRAE Handbook: HVAC Applications, SI ed., American Society of Heating, Refrigerating...
  • S.C Silver et al.

    Component models for computer simulation of ice storage systems

    ASHRAE Transactions

    (1989)
  • T.B Jekel et al.

    Modeling of ice-storage tanks

    ASHRAE Transactions

    (1993)
  • B Vick et al.

    Model of an ice-on-pipe brine thermal storage component

    ASHRAE Transactions

    (1996)
There are more references available in the full text version of this article.

Cited by (56)

View all citing articles on Scopus
View full text