Performance Estimation of Water Thermal Energy Storage Tank by Using Simplified Simulation



Similar documents
CFD SIMULATION OF SDHW STORAGE TANK WITH AND WITHOUT HEATER

THERMAL STRATIFICATION IN A HOT WATER TANK ESTABLISHED BY HEAT LOSS FROM THE TANK

Thermal Mass Availability for Cooling Data Centers during Power Shutdown

How To Improve Energy Efficiency Through Raising Inlet Temperatures

Effect of design parameters on temperature rise of windings of dry type electrical transformer

International Telecommunication Union SERIES L: CONSTRUCTION, INSTALLATION AND PROTECTION OF TELECOMMUNICATION CABLES IN PUBLIC NETWORKS

Coupling Forced Convection in Air Gaps with Heat and Moisture Transfer inside Constructions

ENERGY SAVING BY COOPERATIVE OPERATION BETWEEN DISTRICT HEATING AND COOLING PLANT AND BUILDING HVAC SYSTEM

oil liquid water water liquid Answer, Key Homework 2 David McIntyre 1

Impacts of Static Pressure Set Level on the HVAC Energy Consumption and Indoor Conditions

Zhao et al. 2.2 Experimental Results in Winter Season The analysis given below was based on the data collected from Nov to Mar. 15, 2004.

Natural Convection. Buoyancy force

Exergy Analysis of a Water Heat Storage Tank

Eco Pelmet Modelling and Assessment. CFD Based Study. Report Number R1D1. 13 January 2015

Problem Statement In order to satisfy production and storage requirements, small and medium-scale industrial

ANSI/ASHRAE Standard Building Thermal Envelope and Fabric Load Tests

DSRQ - DSRSQ - DSRSQ-THERM

Convection in water (an almost-incompressible fluid)

ENERGY SAVINGS BY CHANGING CONSTANT AIR VOLUME SYSTEMS (CAV) TO VARIABLE AIR VOLUME SYSTEMS (VAV) IN EXISTING OFFICE BUILDINGS.

Using Computational Fluid Dynamics (CFD) for improving cooling system efficiency for Data centers

Dealing with Thermal Issues in Data Center Universal Aisle Containment

EXPERIMENTAL AND CFD ANALYSIS OF A SOLAR BASED COOKING UNIT

Energy Efficient HVAC-system and Building Design

Federation of European Heating, Ventilation and Air-conditioning Associations

International Telecommunication Union SERIES L: CONSTRUCTION, INSTALLATION AND PROTECTION OF TELECOMMUNICATION CABLES IN PUBLIC NETWORKS

SECTION EG. Engineering Guide Air Distribution. Please refer to the Price Engineer s HVAC Handbook for more information on Air Distribution.

Impacts of Perforated Tile Open Areas on Airflow Uniformity and Air Management Performance in a Modular Data Center

CONSERVATION LAWS. See Figures 2 and 1.

AIR STREAMS IN BUILDING FOR BROILERS

Effect of Self-Heat Circulation on VOCs Decomposition in Regenerative Thermal Oxidizer

HEAVY DUTY STORAGE GAS

Saving Heating Costs In Warehouses

Methods for Effective Room Air Distribution. Dan Int-Hout Chief Engineer, Krueger Richardson, Texas

A NODAL MODEL FOR DISPLACEMENT VENTILATION AND CHILLED CEILING SYSTEMS IN OFFICE SPACES

The CEETHERM Data Center Laboratory

Software Development for Cooling Load Estimation by CLTD Method

Experimental Evaluation Of The Frost Formation

THERMAL ANALYSIS OF A DATA CENTRE COOLING SYSTEM UNDER FAULT CONDITIONS

AIR DISTRIBUTION FOR COMFORT AND IAQ

Building HVAC Load Profiling Using EnergyPlus

CFD Grows Up! Martin W. Liddament Ventilation, Energy and Environmental Technology (VEETECH Ltd) What is Computational Fluid Dynamics?

Comparison of two calculation methods used to estimate cooling energy demand and indoor summer temperatures

Air Conditioners Technical Data

DEVELOPMENT OF HIGH SPEED RESPONSE LAMINAR FLOW METER FOR AIR CONDITIONING

Rate of Heating Analysis of Data Centers during Power Shutdown

Chapter 3.4: HVAC & Refrigeration System

Development of Seismic-induced Fire Risk Assessment Method for a Building

CAN DEMAND CONTROLLED VENTILATION REPLACE SPACE HEATING IN OFFICE BUILDINGS WITH LOW HEATING DEMAND?

Experiment 3 Pipe Friction

Integrated Solar Radiant Systems

HVAC Systems: Overview

Differential Relations for Fluid Flow. Acceleration field of a fluid. The differential equation of mass conservation

Figure 1 - Unsteady-State Heat Conduction in a One-dimensional Slab

APC APPLICATION NOTE #112

Data Center 2020: Delivering high density in the Data Center; efficiently and reliably

Yijun Gao, Wei Wu, Zongwei Han, Xianting Li *

CFD STUDY OF TEMPERATURE AND SMOKE DISTRIBUTION IN A RAILWAY TUNNEL WITH NATURAL VENTILATION SYSTEM

Exercise (4): Open Channel Flow - Gradually Varied Flow

Direct Fresh Air Free Cooling of Data Centres

CONCEPTUALIZATION OF UTILIZING WATER SOURCE HEAT PUMPS WITH COOL STORAGE ROOFS

International Journal of Latest Research in Science and Technology Volume 4, Issue 2: Page No , March-April 2015

INTERNATIONAL JOURNAL OF MECHANICAL ENGINEERING AND TECHNOLOGY (IJMET)

REINFORCED CONCRETE. Reinforced Concrete Design. A Fundamental Approach - Fifth Edition. Walls are generally used to provide lateral support for:

Liebert Hiross HPW The High Performance Wallmount Cooling Solutions for Telecom Mobile Remote Access Nodes

Application of CFD modelling to the Design of Modern Data Centres

SBi 2013:12. Use of perforated acoustic panels as supply air diffusers in diffuse ceiling ventilation systems

Study on Performance Evaluation of a Split Air Conditioning System Under the Actual Conditions

Improving thermal insulation of concrete sandwich panel buildings

AIRCONDITIONING Cooling Loads Calculations

IPACK DISTRIBUTED LEAKAGE FLOW IN RAISED-FLOOR DATA CENTERS

The natural ombination

COMPARISON OF THE THERMAL ENVIRONMENT IN ROOMS WITH CHILLED BEAM AND RADIANT PANEL SYSTEMS

Data Center Design Guide featuring Water-Side Economizer Solutions. with Dynamic Economizer Cooling

Research on the Air Conditioning Water Heater System

Federation of European Heating, Ventilation and Air-conditioning Associations

Guidelines for energy efficient heating, ventilation and air conditioning (HVAC) systems

When the fluid velocity is zero, called the hydrostatic condition, the pressure variation is due only to the weight of the fluid.

Experimental Evaluation of the Discharge Coefficient of a Centre-Pivot Roof Window

HDA

How To Design A Room Air Conditioning System

Air-water systems for efficient air conditioning

Thermocline Management of Stratified Tanks for Heat Storage

Code of Practice for Building Energy Audit

2013 Code_Saturne User Group Meeting. EDF R&D Chatou, France. 9 th April 2013

Improving Data Center Energy Efficiency Through Environmental Optimization

EFFECT OF OBSTRUCTION NEAR FAN INLET ON FAN HEAT SINK PERFORMANCE

Ravi Kumar Singh*, K. B. Sahu**, Thakur Debasis Mishra***

Optimization of electronic devices placement on printed circuit board

Frost Damage of Roof Tiles in Relatively Warm Areas in Japan

Lecture 6 - Boundary Conditions. Applied Computational Fluid Dynamics

Battery Thermal Management System Design Modeling

Lecture 3 Fluid Dynamics and Balance Equa6ons for Reac6ng Flows

JET ENGINE PERFORMANCE. Charles Robert O Neill. School of Mechanical and Aerospace Engineering. Oklahoma State University. Stillwater, OK 74078

Automated Commissioning for Energy (ACE) Platform for Large Retail Property. Overview

HEAT TRANSFER ENHANCEMENT AND FRICTION FACTOR ANALYSIS IN TUBE USING CONICAL SPRING INSERT

Norwegian Tax Authority - Oslo Norway

Keywords: Heat transfer enhancement; staggered arrangement; Triangular Prism, Reynolds Number. 1. Introduction

Transcription:

Performance Estimation of Water Thermal Energy Storage Tank by Using Simplified Simulation M. Kubota, K. Sagara, T. Yamanaka, H. Kotani Division of Global Architecture, Graduate School of Engineering, Osaka University, -, Yamadaoka, Suita Osaka 55-7, Japan Tel: +--79-75 E-mail: kubota_michinori@arch.eng.osaka-u.ac.jp H. Kitano Division of Architecture, Mie University, 57 Kurimamachiya, Tsu, Mie, Japan S. Ichinose Chubu Electric Power Co., Inc., Toshin Higashi, Nagoya, Aichi, Japan ABSTRACT It is difficult to evaluate the performance of air-conditioning systems with thermal energy storage tanks, and there are many systems without seeing whether they realize the expected performance or not. The objective of this study is to estimate the performance of water thermal energy storage tank of multi-connected complete mixing type. Measured data was compared with results of simulation using a physical model under ideal conditions in order to detect some faults in the thermal energy storage tank. In this paper, measured data was replaced with generated data by simulation considering influence of buoyancy in order to find an index for detecting faults. The detection method was applied to a real thermal energy storage system and the dead water zone which causes the lower performance of the storage tank was detected.. INTRODUCTION Air-conditioning systems have become increasingly more complex with the progress of technology. When the air conditioning systems have faults, it is difficult for system operators to deal with them. This situation can cause the inefficient energy usage and an uncomfortable indoor environment, and can result in the system breaking down. Even if designers aim to realize an appropriate indoor environment without waste of energy, the highest performance of the system will not be achieved without appropriate system operations. The objective of this study is to estimate the performance of water thermal energy storage system with water storage tank of multi-connected complete mixing type by using a simplified simulation under ideal conditions. The authors have been trying to detect the faults such as abnormal water level and insulation damage of tank wall by using simplified simulation (Sagara 997). In water thermal energy storage tank of multi-connected complete mixing type, there can be a large temperature difference between inflow water and tank water. If inflow water is not mixed with tank water, temperature stratification can be generated in the divided tanks, and the dead water zone can appear. In this paper, the detection method of the dead water zone which causes the lower performance of the storage tank is presented. Assuming the simple system consisting of a heat pump, an air-handling unit (AHU) and a water storage tank, the detection method using a physical model for water storage tank was investigated. Then, the detection method was applied to the measured data in a real building and the size of the dead water zone was estimated.

. PHYSICAL MODEL UNDER IDEAL CONDITIONS The type of water thermal storage tank is roughly classified into two types; the multiconnected complete mixing type and the temperature-stratified type. The former type is more popular in Japan because it is easy to construct and have a long history, though storage performance is lower than the latter type. Many buildings in Japan have a space under basement floor which is divided by high tie beams for protection against earthquakes. The divided space has been used for water thermal storage tank of the multi-connected complete mixing type. In this study, the measured temperature data are compared with the simulation result by using the physical model under ideal conditions in order to detect fault conditions in the multi-connected complete mixing type. The assumptions in modeling are as follows.. Water is mixed completely in each divided tank.. Heat gain and loss through tank wall are neglected. 3. Water density and specific heat are constant regardless of water temperature. Under these ideal assumptions, water temperature in the i th tank is calculated with the following equation (see Figure ). θ ρc V d i p i dt = ρc q θ θ ( ) () p in in i The temperature in each divided tank is calculated by a finite difference method in which the calculation step is second, and the input temperature to the lower/higher temperature is assumed to be equal to output temperature from heat pump/air-handling unit. 3. THE DETECTION METHOD OF THE DEAD WATER ZONE The detection method of the dead water zone was investigated by using the simulation result. The thermal storage system for this study is composed of a water thermal storage tank of the multi-connected complete mixing type, a heat pump and an air-handling unit. The storage tank is divided into sub-tanks connected in series and its total volume is m 3. Operating condition of this system is listed in Table. The thermal storage system is shown in Figure schematically, and the plan of thermal energy storage tank is shown in Figure 3. The connecting holes through the tank to the tank are located at lower part near tank bottom in order to generate the dead water zone intentionally. In the thermal storage tank, the ceiling, bottom and outer wall adjacent to soil are insulated, but the partition walls between divided tanks are not insulated. The water to the heat pump is pumped up from the lower and higher temperature, and chilled water is put into the lower temperature as shown in Figure. The water to the AHU is pumped up from the lower temperature and returns to the higher temperature. 3. Measured Data Generated by Simulation Since generated data by simulation considering influence of buoyancy could well simulate the measured data (Kitano and Sagara ), we use simulated data as measurement one to find

an index for the detection. This simulation is based on the mixing model for the temperaturestratified thermal storage tank. Here, vertical one-dimensional diffusive and convective heat transfer was assumed. Heat gain/loss through tank wall is taken into account in order to study the effect of heat gain/loss on the fault detecting method. Inflow water from inlet connecting hole is mixed with the tank water in a certain region of the tank. The region is called mixing region in this paper. The depth of the mixing region depends on the inflow conditions and vertical temperature distribution in the tank. Equation () is the governing equation of the mixing model. = + ( ) θi θi θi Φ κ θ θ + Awik hik U i in i () t z z A k = ρc V i i where Φ is inflow rate per unit vertical depth in the mixing region. Both integrated values of Φ and Ω from to L st are equal to inflow rate, q in. Volume of thermal energy storage tank [m 3 ] Number of divided tanks Capacity of heat pump [kw] Flow rate in heat pump Output temperature of heat pump 5 [m 3 /h] [degc] p Input/output temperature difference of heat pump Input temperature limit of heat pump Input/output temperature difference of AHU 5 [degc] [degc] 7 [degc] i- i i+ Vi q in q g q a AHU heat pump g_in soil concrete 5 Lower temperature 7 3 9 3 5 3 i- i i+ 7 Higher temperature q in Figure : Schematic diagram of the divided tank Connection hole located high Connection hole located low 9 Figure 3: Floor plan of thermal storage tank Lower temperature heat [kw] q The multi-connected complete mixing tank Heat transfer in concrete tank wall and ambient soil is assumed to be governed by the one- Table : Operating condition output temperature from AHU 3 output temperature from heat pump cooling load heat pump output : : 3: :: Figure : Input temperature, chiller output and cooling load N Higher temperature Figure : Studied thermal storage system

dimensional unsteady heat conduction equation. Temperature in concrete and soil was calculated with a finite difference method for each wall of divided tank under constant soil temperature at m distance from concrete wall. Generated data are the data in the th date of simulation started initially with deg C uniform temperature in the whole storage tank. The change of heat pump output, cooling load, and output temperature from heat pump and AHU are shown in Figure. In this thermal storage system, heat pump is operated from : to : and from : to :. AHU is operated from : to : as shown in Figure. In this simulation, the temperature measurement point is located at the center of each divided tank. 3. Temperature Change Rate Since the dead water zone cannot store thermal energy, the water volume which is effectively used for thermal energy storage in each divided tank becomes less than designed water volume. The less water volume causes the more rapid temperature response. Therefore, the temperature change rate can be an index of temperature response changed by the dead water zone of each divided tank. The temperature change rate is defined as equation (3). This rate is non-dimensional number, and the effect of each tank volume on temperature response can be neglected. ( j ) j θ θ j + θ i Vi i i = t q θ in AHU ( ) where t j = qint / V, qin qin q j + = + in / and Δθ AHU is the designed input/output temperature difference of AHU. Vertical temperature profiles in the tank to the tank are shown in Figure 5. In this case, chilled water is put into the tank, and it flows through the tank to the tank. Lower temperature water flows along the tank bottom and almost never flows upward in the tank to the tank 5. This is because lower temperature water has larger density than warmer tank water and the connecting holes are located at lower part near tank bottom through the tank to the tank. As a result, the warmer tank water is not mixed with the chilled water and stays at upper part of the tank to the tank 5. The measurement points in the tank to the tank 5 are located within the dead water zone as shown in Figure 5. The temperature change height [m] temperature change rate [-].7.5 -. -. dead water zone height of temperature measurement -. -. dead water zone tank tank tank3 tank tank5 tank temperature[deg C] Figure 5: Vartical temperature profiles in the tank to the tank Ex mea Tmea Ex sim dead water zone simulation result -. -. (3) -.3 -.3 measured data -.3 tank3 Tsim tank tank -. -. -. Figure : Comparison of simulated result and measured data of temperature change rate dead water zone : : : : :

rate of the simulation result under ideal conditions is compared with that of the measured data generated by a simulation in Figure. In the tank 3, temperature change rate of the measured data keeps approximately zero because the dead water zone spreads to the middle of the tank where measurement point is located. In the tank and the tank, temperature change rate of the measured data is larger than the simulation result, and the of extreme value is earlier than the simulation result. The appearance of the dead water zone enlarges the extreme value and quickens the of the extreme value. 3.3 Difference and Time Lag of Extreme Value Figure 7 shows the difference of extreme value (see Figure, Ex mea - Ex sim ) and the lag of extreme value (see Figure, T mea - T sim ) in all divided tanks under three conditions; (a) the dead water zone are enlarged to the middle part of tank from beginning to end, (b) the dead water zone are scaled down from the middle part to upper part of tanks, (c) the dead water zone are appearing only in the upper part of tanks. In each condition, the difference of extreme value takes large positive value and the lag of extreme value takes large negative value in the tank located at the downstream side of the tank including the dead water zone. In addition, these values are proportional to the size of the dead water zone because the larger the dead water zone causes the more rapid temperature response. The dead water zone is appearing in divided tanks located at the upstream side of the tank having the maximum difference of the extreme value as shown in Figure 7. Thus it was shown that the difference of extreme value could be used as an index to detect the dead water zone in this study.. APPLICATION TO MEASURED DATA IN REAL BUILDING The detection method shown above was applied to the measured data in a real building. The thermal storage system of this building is composed of a water thermal storage tank of the multi-connected complete mixing type, a heat pump, an air-handling unit (AHU) and fan coil units (FCU). The storage tank is divided into sub-tanks connected in series and its total volume is 39 m 3. Operating condition of this system is listed in Table. The thermal storage system is shown in Figure schematically. The plan of the thermal energy storage tank and temperature measurement points are shown in Figure 9. In this storage tank, the temperature measurement points are located at the center of the each divided tank except the tank and the tank.. Detection in Charging Mode The data measured from : on 5th to : on th of Aug. were used for the detection of the dead water zone. Input/output temperature of heat pump and flow rate through heat pump in charging mode are shown in Figure. lag of extreme value[h]... condition (a). condition (b). condition (c).... lag of extreme value... - - - -. -. -. - - - - difference of extreme value - -. - -. -. difference of extreme value[-] lag of extreme value[h] Figure 7: Difference and lag of extreme value in each divided tank difference of extreme value[-] lag of extreme value[h] difference of extreme value[-]

The difference of extreme value in each divided tank is shown in Figure. As shown in Figure, the difference of extreme value has the maximum value in the tank 3. According to the results shown in previous chapter, it was expected that the dead water zone was appearing in the tank. In order to estimate the size of the dead water zone in the tank in the simulation, the dead water zone ratio of the tank is changed, and its results is compared with measured temperature data. As a result, the temperature difference between the measured data and the simulation result becomes the minimum when the dead water zone ratio in the tank is set as.5, and the temperature profiles in the whole tank are shown in Figure. The temperature profiles without the dead water zone are shown in Figure 3. As shown Figure and Figure 3, the simulation result changing the dead water zone ratio is similar to the measurement data if compared with the simulation result without the dead water zone. The measured vertical temperature profile shows that the dead water zone was appearing about the upper half of the tank (see Figure ).. Detection in Discharging Mode The data measured from 7:3 to : on th of Aug. were used for detection in discharging mode. Input/output temperature of heat pump, flow rate through heat pump and flow rate from AHU and FCU in discharging mode are shown in Figure. difference of extreme value [-] Volume of thermal energy storage tank 39 [m 3 ] Number of divided tanks Capacity of heat pump 7 [kw] Output temperature of heat pump Input/Output temperature difference of heat pump 7 [degc] 5 [degc]..5..3.. Table : Operating condition [mm] 7 7 Higher temperature 35 A Lower temperature 3 5 measurement point 7 A' 9 :Connection hole located high :Connection hole located low 3 5 7 9 3 7 Figure : Difference of extreme value in charging mode 7 3 Figure 9: Plan of the thermal energy storage tank Lower temperature Figure : Thermal energy storage system temperature[deg C] FCU AHU heat pump q The multi-connected complete mixing tank 7 Heat pump is variable water volume input temperature 9 to heat pump 3 flow rate in heat pump 7 output temperature 9 from heat pump 7 3 : : : : : : : : Figure : Input/output temperature of heat pump, and flow rate in heat pump N Higher temperature 3 5 7 9 3 7 measured simulated : : 3: 3: tank position [m 3 ] 3 39 : 5: : : : 5: : : Figure : Temperature profiles in the whole tank (dead water zone ratio of the tank is set as.5) flow rate[m 3 /h]

The difference of extreme value in each divided tank is shown in Figure. As shown in Figure, the difference of extreme value has the maximum value in the tank, and it was expected that the dead water zone was appearing in the tank 7. The temperature profiles in the whole tank are shown in Figure 7. This figure shows that the simulation result is close to the measured data. The connection holes from the tank to the tank are located at the lower part near tank bottom and the higher part near water surface alternately as shown in Figure 9, and the tank water in each tank is expected to be mixed enough. Figure shows the vertical temperature profiles and the dead water zone was not appearing in the tank, 9, and. However, the temperature data of the tank 7 was measured only at the center of tank and the appearance of the dead water zone in the tank 7 cannot be confirmed. In order to figure out the temperature distribution in the storage tank, CFD analysis was Figure : Input/Output temperature of heat pump, and flow rate in heat pump temperature[deg C] height [mm] 3 5 3 5 7 9 3 7.5 3.5.5 9.5 7.5 3 5 7 9 3 7 5.5 5 5 3 39 tank position[m 3 ] Figure 7: Temperature profiles in the whole tank in discharging mode output temperature from heat pump tank connecting hole tank7 connecting hole tank tank position [m 3 ] 3 39 flow rate from AHU and FCU input temperature to heat pump flow rate of heat pump measured simulated : 3: : 5: : : Figure 3: Temperature profiles in the whole tank (without dead water zone) 5 7:3 9:3 :3 3:3 :3 7:3 9:3 measured 7:3 9: : : 7: : : 3: : 5: : : 3 5 5 simulated 7:3 9: : : 7: : flow rate[m 3 /h] height [mm] height [mm] : : 5: : 3: : 9 37 9 37 9 37 tank tank9 tank Figure : Vertical temperature profiles measured in the tank, 9, and difference of extreme value [-]..... -. 3 5 7 9 3 7 Figure : Difference of extreme value in discharging mode :3 :3 7:3 3:3 9:3 connecting hole 3 Figure 9: Temperature distribution at A-A section at : when the temperature change rate of the tank has the maximum value tank tank9 tank Figure : Vertical temperature profiles measured in the tank, 9, and

conducted. Figure 9 shows the temperature distribution at section A-A' (see Figure 9) at : when the temperature change rate of the tank has the maximum value. As shown in Figure 9, it was found that the warmer inflow water from the tank to the tank 7 flows up near water surface and it is not mixed enough in tank 7. Consequently, the dead water zone is appearing near the bottom of the tank 7. In addition, the warmer water in the upper part of the tank 7 flows into the upper zone of the tank. Temperature stratification is formed in the tank, and the temperature change rate of the tank is found to become large when the temperature stratification passes across the measurement point at the center of tank. In this case, the simulation result agreed with the measured data, and tank water was expected to be mixed enough. However, from the result of CFD analysis, the dead water zone was estimated to be appearing in the tank7 and the appearance of the dead water zone was detected by using the difference of extreme value, which shows that the difference of extreme value is useful as the detection index of the dead water zone. 5. CONCLUSION In this paper, the detecting method of the dead water zone using simplified simulation under ideal conditions was investigated. Aiming to confirm the temperature response, the temperature change rate was considered to be an index of temperature responses which is changed by the dead water zone of each divided tank. The appearance of the dead water zone enlarges the extreme value and quickens the of the extreme value. In this study, the difference of extreme value of the temperature change rate between the simulation results and the measured data was proposed to be a detection index of the dead water zone. The detection index was applied to measured data in a real building and the appearance of the dead water zone was detected and its size was estimated. REFERENCES Sagara K. (997). Fault detection in thermal storage tank of multi-connected complete mixing tank. Proc. MEGASTOCK '97, pp.7-. Kitano H. and Sagara K. (). Study on mixing of temperature-stratified chilled-water thermal storage tank under unsteady input condition. Proc. TERRASTOCK, pp.33-3. NOMENCLATURE h : heat gain through tank wall [W/m ] q : flow rate in thermal energy storage tank [m 3 /s] θ i temperature of i th divided tank [deg C] θ in temperature into a divided tank [deg C] θ g_in temperature into a heat pump [deg C] Δθ a designed input/output temperature difference of AHU [deg C] U velocity in horizontal cross section of tank [m/s] Φ : distribution of inflow rate in a divided tank [(m 3 /s)/m] Ω : distribution of outflow rate of a divided tank [(m 3 /s)/m] κ : thermal diffusivity of water in thermal storage tank [m /s] A : horizontal sectional area of tank [m ] Aw : area of tank wall [m ] L st : water depth [m] l in : position of inflow [m] l m : depth of mixing area [m] V : volume of a divided tank [m 3 ] c p : specific heat of water [J/kgK] ρ : water density [kg/m 3 ] SUBSCRIPT i : j : step k : wall number in the divided tank