Experimental and numerical investigations on high temperature cast steel based sensible heat storage system

Applied Energy 251 (2019) 113322

Contents lists available at ScienceDirect

Applied Energy journal homepage: www.elsevier.com/locate/apenergy

Experimental and numerical investigations on high temperature cast steel based sensible heat storage system☆

T

K. Vigneshwarana, Gurpreet Singh Sodhib, P. Muthukumara,b, , Anurag Guhac, S. Senthilmurugana,c ⁎

a

Centre for Energy, Indian Institute of Technology Guwahati, Guwahati 781039, India Department of Mechanical Engineering, Indian Institute of Technology Guwahati, Guwahati 781039, India c Department of Chemical Engineering, Indian Institute of Technology Guwahati, Guwahati 781039, India b

HIGHLIGHTS

GRAPHICAL ABSTRACT

a high-temperature cast steel • Designed based thermal energy storage (TES) system.

thermal behaviour of TES • Tested system at different operating temperature ranges.

thermal models for pre• Developed dicting heat storage characteristics of TES system.

1-D dynamic model for TES • Simplified is developed for process optimization. various potential applica• Discussed tions of developed cast steel based TES system.

ARTICLE INFO

ABSTRACT

Keywords: Thermal energy storage Sensible heat Experimental validation Thermal modeling Charging and discharging

The present study focusses on detailed experimental and numerical investigations of a cast steel based sensible heat thermal energy storage system using air as a heat transfer fluid. A dedicated test facility is designed and developed for analysing the performance of the storage system operating in the temperature range of 393 K to 573 K. Three-dimensional (3-D) and one-dimensional (1-D) models are developed for predicting the heat transfer characteristics of the storage system. The developed storage prototype has a shell and tube configuration having 19 passages in the tube side for heat transfer fluid flow. The performance of the storage system during the charging and discharging processes is analysed by varying the operating temperature range and flow velocity of air. The heat transfer characteristics of the system in terms of axial and radial temperature variations are recorded and analysed. Both the experimental and 3-D simulation results show a significant temperature variation in the axial direction than radial direction. The charging and discharging rates are found to be faster at a higher flow velocity of air. The predictions from both 3-D and 1-D models are consistent with the experimental data. The validated 1-D model can be used for real-time monitoring, control, optimization and integration with various storage applications.

The short version of the manuscript was presented at ICAE 2018, Aug 22–25, Hong Kong. This paper is a substantial extension of the short version of the conference paper. ⁎ Corresponding author at: Department of Mechanical Engineering, Indian Institute of Technology Guwahati, Guwahati 781039, Assam, India. E-mail address: [email protected] (P. Muthukumar). ☆

https://doi.org/10.1016/j.apenergy.2019.113322 Received 11 December 2018; Received in revised form 1 May 2019; Accepted 13 May 2019 Available online 24 May 2019 0306-2619/ © 2019 Elsevier Ltd. All rights reserved.

Applied Energy 251 (2019) 113322

K. Vigneshwaran, et al.

Nomenclature

A Cp D Dpcd, i Dpcd, o d hi k L M m n Np Nu P Pr Q Q q R Re T t Uo v V

Subscript

area [m2] specific heat [J/kg-K] diameter of the storage module [m] inner pitch circle diameter [m] outer pitch circle diameter [m] diameter [m] heat transfer coefficient [W/m2-K] thermal conductivity [W/m-K] length [m] molecular weight of air [kg/mol] mass [kg] number of HTF passages number of HTF passages in pitch circle diameter Nusselt number pressure [N/m2] Prandtl number energy [J] volumetric flow rate [m3/s] heat transfer rate [W] universal gas constant [N m/mol-K] Reynolds number temperature [K] time [s] overall heat transfer coefficient [W/m2-K] velocity [m/s] volume [m3]

avg ch e f in ini out s tube

average channel element fluid inlet initial outlet solid HTF passage (tube)

Greek symbols

µ

density [kg/m3] viscosity [N s/m2]

Acronym CSP EXP HTF SHS SIM TES CSTR

concentrated solar power experiment value heat transfer fluid sensible heat storage simulation value thermal energy storage continuous stirred tank reactor

thermochemical reactions, is characterized by very high energy storage density, but at the same time, the materials required in such systems are relatively expensive [8]. In the Sensible Heat Storage (SHS) system, the amount of heat stored is directly proportional to the specific heat, mass and temperature change of the storage material [9]. Apart from these, thermal conductivity and thermal diffusivity are also some of the crucial factors affecting the performance of the SHS systems [10]. The broadly used materials in the SHS are water, thermal oils, solar salts, earth materials, liquid metals and concrete [11,12]. The thermal stability of the SHS materials is one of the main advantages which ensures that they can be used for the high-temperature application. Apart from liquid metals and oils, SHS materials are cheaper and readily available in the market. The only disadvantage of the SHS systems is their low energy density, due to which the system volume is generally high [13]. Few studies have been conducted to explore the use of low-cost sensible heat storage media. Xu et al. [14] studied the degradation and thermal expansion characteristics of a low-cost SHS material (Xceltherm ® 600 oil-drenched in sand). A lumped system model was validated for a 600 MWe Concentrated Solar Power (CSP) plant. Using this material, the net energy storage has increased by 4.5% compared to the concrete. Recently, a multi-tube lab scale TES model was developed and tested to study the performance of different storage materials. It was concluded that the thermal conductivity of the storage material plays a significant role in improving the charging and discharging characteristics [15]. The performance of the TES system is also improved by using high diffusivity materials like graphite and metals as storage media [16]. Meffre et al. [17] have explored the use of Vitrified fly ash produced from municipal solid waste of incinerator as the storage material. This material possess the desired thermal properties and could be used up to 1100 °C. Gravels are one of the low-cost material mainly used in seasonal TES applications such as space heating and domestic hot water generation system [18,19]. Negi et al. [20] conducted a parametric study to investigate the effect of mass flow rate of Heat Transfer Fluid (HTF) on the performance of polyvinylchloride based packed bed SHS system. The optimum flowrate obtained for charging

1. Introduction Renewable energy sources play a crucial role in addressing the current global energy demands. The global energy demand for the year 2017 has significantly increased by 2.1% compared to the previous sixyear rise [1]. Energy from the sun alone can fulfil the global energy demand if it is effectively utilized [2]. Various technologies adopted by the industries for utilizing the solar energy are parabolic trough collectors, linear Fresnel reflectors, heliostat field collectors and parabolic dish collectors. These systems are also fitted with the sun rays tracking system to increase the system efficiency [3]. Even after exploring various solar thermal collector technologies, there exists a larger disparity between energy supply and demand. The energy storage is one of the substantial solutions to correct the mismatch between the demand and the supply. Energy storage can be mainly classified into mechanical, chemical, biological and thermal energy storage [4]. In recent times, Thermal Energy Storage (TES) is gaining importance in several applications such as concentrated solar power stations, space heating, desalination, solar drying, domestic hot water systems, etc. TES system aims at storing the surplus energy and to utilize the same during the peak demand [5]. The integration of TES system with above said solar thermal technologies are well proven using experimental study [6]. TES technology uses a wide variety of materials based on desirable properties like high specific heat, high thermal conductivity, high latent heat, low volumetric expansion and low flammability. The selection of material for cold and hot storage is generally categorized on the basis of three types of TES systems namely: Sensible heat storage, Latent heat storage and Chemical heat storage. In the latent heat storage system, the energy is stored/released during the phase change of storage medium known as Phase Change Material (PCM) at a constant temperature, whereas sensible heat is stored/released without phase change. The time taken for charging/discharging is more in the case of latent heat storage system due to the low thermal conductivity of the PCM, which is the main disadvantage of latent heat storage [7]. Thermochemical heat storage system, which stores heat via reversible 2

Applied Energy 251 (2019) 113322

K. Vigneshwaran, et al.

and discharging was 0.0125 kg/s and 0.00796 kg/s, respectively. For a high-temperature application like CSP technologies, different samples of sand have been characterised and were found promising as energy storage materials [21]. Solid SHS materials are preferred over liquid materials due to their greater longevity and negligible degradation over a wider number of thermal cycles [22]. The storage temperature range is decided based on the end use of thermal energy. Gibb et al.[23] have conducted a detailed review on the integration of TES with the real-time applications operating between low to high-temperature. Many researchers have studied the application of SHS system for the following applications: (i) steam generation for power [24], chemical process [25], desalination [26] and cooking [27] and (ii) district heating and water heating [28]. These applications are integrated with either solid or liquid based energy storage mediums. The solid mediums used are concrete, rock and cast iron, whereas, in the liquid medium, PCMs and synthetic oils are most commonly used. In all above-mentioned applications, the total operating cost is significantly influenced by the type of HTF used. The HTF is one of the most essential components for storing and transferring energy within the TES system. The selection of HTF is mainly decided by the operating temperature range, cost, thermo-physical properties and the rate of degradation. Various types of HTF used in the SHS systems are molten salts, air, water, thermic fluid, organic fluids and liquid metals. Dinker et al. [29] examined the use of different HTFs like aqueous suspended graphite, Al2O3 solution and pure water. It was observed that the charging and discharging times decrease significantly as the mass flow rate and inlet temperature of HTF increase. The time taken for charging of SHS material with HTFs like suspended graphite solution and Al2O3 solution has decreased by 14.2% and 21.2%, respectively compared to pure water. For storing sensible heat in the packed concrete bricks, molten salts are more preferred as the HTF [30]. In a packed bed system, the commonly used SHS materials are quartzite rock and sand, whereas air is preferred as the HTF [31]. Many researchers have attempted to study different SHS-HTF combinations, and it was concluded that both the SHS and HTF play a vital role in deciding the system cost. Therefore, for high-temperature applications (above 300 °C), there is a need to explore a good combination of SHS-HTF with desired properties and low operational cost. Presently, Jülich solar tower [32] and Ait-Baha CSP pilot plant [33] are the few plants which are using air as the HTF. The main drawback of using other commercially available HTFs such as synthetic oils and organics is their high cost. Also, their operation is limited to a maximum operating temperature of 400 °C above which their thermal properties deteriorate with continuous operation [9,34]. Molten salts have high freezing temperature [35]. In order to use molten salts, the complete HTF pipelines have to be maintained above their freezing temperature irrespective of the operating condition of the system. Hence, using them for prototype testing requires additional heating source. Though the thermal properties of air such as thermal conductivity and specific heat are poorer than other commercially available HTFs, air has its own advantages such as free availability, high corrosion-resistance [36] and higher operating temperature range of up to 1100 °C. It is suitable to be used for a wide variety of thermal applications. Additionally, air has good flowing characteristics at high temperature due to low dynamic viscosity [37]. It is well known that the optimal design of heat storage system plays a crucial role in achieving low energy storage cost. The idea of optimizing the system design is mostly based on minimizing the charging and discharging times through mathematical models. Various SHS system configurations can be explored and optimized employing validated numerical models instead of building expensive working prototypes [38]. Zanganeh et al. [39] developed a mathematical model of rock-based TES system using air as the HTF and predicted the heat transfer behaviour of the rock in the axial direction. System performances with different HTFs such as oil, air and molten salts were

investigated using a 1-D model. The rate of energy stored and released from the storage model were found to be very low for air as HTF due to its weak thermal properties [40]. Mahmood et al. [41] developed 1-D and 2-D dynamic models, where the storage characteristics of honeycomb ceramic module TES system were validated with the help of an experimental study. It was concluded that the computational time and cost could be reduced by solving the system using 1-D dynamic model while comparing with the commercially available 3-D modeling software’s. 1-D models developed using Dymola software are also extensively used for predicting the performance of the TES system [42]. Recent studies reported on SHS materials have been focused mainly on low thermal conductivity materials like concrete [43], therminol oil, sand [21], fly ash, graphite and gravels. There is a lack of profound experimental studies on TES systems with the SHS materials having high thermal conductivity mainly for high-temperature energy storage applications. Also, the use of naturally available HTFs like air has been limited to packed bed and solid SHS storage technologies [32,40]. Only limited works have been reported using air as the HTF in a shell and tube TES system. Further, most of the reported studies on the SHS are limited to low operating temperatures [44]. Few detailed 3-D numerical studies have been carried out for analysing energy and mass interactions between different components of the TES system [45]. In general, the computational cost and time for simulating the characteristics of the SHS employing 3-D model are high [41]. To overcome this barrier, a real-time monitoring and optimization of the SHS module using 1-D dynamic modeling is very essential [42]. Further, financial aspects, ease in manufacturing, better thermo-physical properties and long life are the other major areas which need further investigation while developing a cost-effective SHS system. The present work is focused on the design and development of a multi-passage high-temperature SHS system with an SHS-HTF combination having a higher heat transfer rate at lower system cost. Atmospheric air has been chosen as the HTF. The use of air highlights its potential as HTF for high-temperature application, which further reduces the overall cost of the system. The experiments are designed based on the inlet process conditions for studying the thermal behaviour of the SHS module and the HTF fluid. To observe the intrinsic spatial behaviour during energy transfer between the SHS material and the air, a 3-D model based on the finite volume method is developed. This model helps in studying the charging and discharging characteristics of the SHS system. In addition, a simplified 1-D dynamic model is developed to study the performance of the SHS module and also to explore the possibility of analysing system behaviour in a relatively shorter time as compared to the 3-D model. Numerical predictions from the developed 3-D and 1-D models are validated with the experiment results. Based on the results obtained, various applications of the SHS system are explored. 2. Description of experimental setup and procedure 2.1. Experimental setup In order to analyse the performance of the TES system in the realtime, an experimental setup has been designed and fabricated at the Indian Institute of Technology Guwahati. The TES system has been designed to have maximum flexibility on the basis of criterion such as topology, orientation, quantity of storage material and placement of temperature sensors for testing various TES prototypes. The schematic representation given in Fig. 1 illustrates the experimental setup of the TES system. The TES system comprised of several major components like control panel, air blower, nichrome element heaters, pitot tube, data acquisition system and thermal energy storage module. Air is chosen as HTF for energy interaction with the storage medium. The major components and their technical specifications are explained below; 3

Applied Energy 251 (2019) 113322

K. Vigneshwaran, et al.

Fig. 1. Schematic of the fully insulated experimental setup of cast steel based TES system with important components.

• A centrifugal blower of capacity 120 m /h, made up of stainless 3

• • •

•

Table 3 SHS material properties [48].

steel is used for supplying air. The blower is powered by a variable frequency driven motor (3 phase induction – 2 hp) to control its speed. Butterfly and spherical ball valves are used to control the air flow direction. Calibrated K- type thermocouples are fitted at various locations for continuous recording of temperature variation inside the TES using a Data Acquisition System with a resolution of ± 0.1 K (Model: Agilent 34970A). A Testo make Pitot tube based differential pressure meter (Model: 512) is used for monitoring the air velocity. An air heater of capacity 48 kW (1 kW per heater element) with nichrome elements is used for supplying air at the required temperature. The power control of the heater is automatic. Selec make temperature controller ensures power on/off control according to the required conditions of inlet air temperature. A flame-retardant glass fibre single core cable supplied by LAPP India (ÖLFLEX® HEAT 650 SC) is used for power supply to the heater elements. The entire system is insulated with 102 mm thick ceramic wool (k = 0.12 W/m K) blanket.

The horizontal cut-section view of the SHS module is shown in Fig. 2a. Cast steel used as the storage material in the present study has Table 1 Geometric configuration of the SHS module. Length, L (mm)

Diameter, D (mm)

n

NP

Dpcd,i (mm)

Dpcd,o (mm)

dtube (mm)

Cast-Steel

740

267

19

1,6,12

110

220

12.7

ks (W/m K)

Cps (J/kg K)

ρs (kg/m3)

Cast steel

40

600

7800

high volumetric heat capacity and thermal conductivity. Even though, the initial cost of the cast steel is comparatively higher than other solidstate storage materials such as cast iron, concrete and magnesia bricks [46], cast steel is selected as the storage material due to its favourable thermal properties and no deformation over time. The dimensions of the storage module are given in Table 1. Nineteen holes (HTF passages/ tubes) of diameter 12.7 mm each are drilled throughout the length of the storage module. Nine K-Type thermocouples are used for measuring the temperature of the SHS module at different axial and radial locations (illustrated in Fig. 2(a) and (b) and described in Table 2). The total volume of the storage module is divided into three sections namely: ‘Section- A’, ‘Section- B’ and ‘Section- C’. Additionally, the air inlet and outlet temperatures of the SHS module are also measured and recorded. The number of HTF passages are decided on the basis of previous studies on the SHS design optimization conducted by the author’s research team [47]. It was observed that though the charging time got reduced by adding more number of tubes, there was a definite threshold after which the charging time does not reduce with further increase in the number of HTF passages. So, SHS module with 19 tubes embedded in concrete as the storage medium was selected for performance evaluation. In the current study, SHS material has been replaced by cast steel (1.0% C) having a higher thermal conductivity and hence, lesser charging and discharging times are expected than the concrete module. The properties of cast steel are presented in Table 3. The arrangement of HTF tubes in the current SHS module is presented in Fig. 2c.

2.2. Thermal energy storage module

SHS medium

Material

Table 2 Thermocouple positions in the SHS module. Thermocouple nomenclature

Axial length position from module inlet section of HTF (mm)

Radial distance from module outer periphery (mm)

A_25/A_50/A_75 B_25/B_50/B_75 C_25/C_50/C_75

22/370/718 22/370/718 22/370/718

25/50/75 25/50/75 25/50/75

4

Applied Energy 251 (2019) 113322

K. Vigneshwaran, et al.

The module is designed by fixing a heat storage capacity as 15 MJ. The estimated mass of the storage medium is 309.28 kg. The volume (V) of the SHS module can be calculated as follows [47];

V=

4

(D 2

2 ndtube )L

(1)

where, D and dtube are the diameter of the SHS module and HTF passage, and L is the length of the SHS module.

Fig. 2. (a) Isometric view of the cut sectioned SHS module with thermocouple positions, (b) Picture of the SHS module without insulation and (c) Cross-sectional view of HTF passages in the SHS module. 5

Applied Energy 251 (2019) 113322

K. Vigneshwaran, et al.

The amount of heat energy (Qs) stored and discharged are calculated using following expressions [47];

Tini )

(2)

Ts, avg (t ))

(3)

Qs (t ) =

s Vs Cps (Ts, avg (t )

Qs (t ) =

s Vs Cps (Tini

where,

minimal temperature gradient. Charging and discharging experiments are conducted with an initial temperature difference of 80 K between the module and air. After conducting charging experiment, discharging does not start immediately. The discharging begins only after the volume average temperature of the SHS module reaches a uniform temperature (Tini), and the same can be estimated once all the thermocouples placed in the SHS module reaches a steady value. Therefore, charging and discharging experiments are not conducted in a continuous cyclic mode. The volume average temperature of the module is estimated by dividing them into nine elements (e) as shown in Fig. 4. The temperature of the respective element is measured with the corresponding thermocouple placed in that element. The module average is estimated for the complete system including the SHS volume and the HTF passage (tube) volume. The volume average temperature (Ts,avg) of the module is calculated using Eq. (4);

and Cp are density and specific heat of the SHS material.

2.3. Testing procedure The layout of the experimental setup is shown in Fig. 3. The experiments can be divided into two major processes; (i) charging and (ii) discharging. The procedures for conducting the charging and discharging experiments are as follows; Initially, the fresh HTF (air) is drawn inside the centrifugal blower. The air is allowed to pass through the air heater and heated until the air reaches the temperature set point. The set point in the temperature controller is fixed as per the required inlet temperature specified in Tables 4 and 5. The heated air is supplied to the module through an inlet channel, and later this flow is divided into nineteen passages, as shown in Fig. 2a. As the TES system is not equipped with an automated PID controller, maintaining the bulk fluid at a constant temperature is very difficult. Hence, a slight variation is observed in the inlet temperature of the HTF. The flow velocity is monitored at the inlet channel, immediately before the storage module. The charging/discharging experiment is started once the module reaches the initial temperature (Tini ). The experiment begins with the supply of the HTF at the required temperature inside the SHS module. The SHS module either absorbs or releases the heat energy from or to the HTF. The experiment is stopped when the SHS module reaches a steady state volume average temperature (Ts, avg ). After passing through the SHS module, the air is allowed to pass to the atmosphere through the outlet valve. The bypass from the heater towards valve 2 is opened only if the inlet temperature (Tf , in ) does not reach the required operating temperature. In such a case, the air is recirculated to blower and back to the heater so as to attain the required temperature. Before starting any charging or discharging experiment, the whole module is brought to a uniform volume average temperature with a

9

Ts, avg (t ) =

Tini (K)

Operating range (80 K)

vch (m/ s)

vtube (m/ s)

1 2 3 4 5 6 7 8 9

573

493

493–573

523

443

443–523

473

393

393–473

4.5 3.5 2.5 4.5 3.5 2.5 4.5 3.5 2.5

20 15 11 20 15 11 20 15 11

Tf , in (K)

Tini (K)

Operating range (80 K)

vch (m/ s)

vtube (m/ s)

10 11 12 13 14 15 16 17 18

493

573

573–493

443

523

523–443

393

473

473–393

4.5 3.5 2.5 4.5 3.5 2.5 4.5 3.5 2.5

20 15 11 20 15 11 20 15 11

(4)

The uncertainty in the calculated quantities arise due to the error associated with the various measured quantities (obtained from measurement devices). It can be estimated by differentiating the calculated quantities with respect to directly measured quantity and multiplying them with an error on directly measured quantity [49]. For example, the mass flow rate (m ) is given by Eq. (5), and the corresponding error on the calculated mass flow rate can be mathematically expressed by Eq. (6). (5)

m = Av m m m v+ v A+ A v A

(6)

where, v , A and are the error in the measured quantities such as velocity and calculated quantities such as cross-sectional area of the channel and density of air, respectively. Similarly, the experimental error on other calculated variables such as energy stored/discharged can be calculated. The estimated maximum experimental error on the mass flow rate is ± 2.2% ( ± 0.00109 kg/s), and on energy stored and discharged are ± 4.9% ( ± 0.7 MJ). 3. 3-D model of the SHS system The three-dimensional (3-D) model is useful to understand the spatial behaviour during the energy exchange between the HTF and the SHS module. The 3-D modeling of the system is carried out using the simulation tool Comsol Multiphysics 4.3a [50]. The conjugate heat transfer problem combining the HTF flow and energy exchange is solved by imposing different initial and boundary conditions on the governing equations. The HTF flow and energy interactions between the HTF and the SHS module are governed by the following set of equations [51]; Continuity equation for flow:

Table 5 Experimental conditions for the discharging process. Experiment No.

Vs

2.4. Uncertainty analysis

m= A

Tf , in (K)

Ve Ts, e (t )

To analyse the charging and discharging characteristics of the system, the experiments are conducted for three different HTF velocities and three different operating temperature ranges. All the experimental conditions are described in Tables 4 and 5.

Table 4 Experimental conditions for the charging process. Experiment No.

e=1

.(

f

vf ) = 0

Momentum equation for flow: 6

(7)

Applied Energy 251 (2019) 113322

K. Vigneshwaran, et al.

Fig. 3. Layout of TES system indicating flow directions.

Fig. 4. The pictorial representation of nine elements of the SHS module representing various thermocouples for determining Ts,avg.

D(

f

Dt

vf )

=

P + µf

2v

f

4. The flow of air through the inlet channel is assumed to be perfectly divided into 19 passages inside the SHS module based on the below equation.

(8)

Energy equation for flow:

D ( f Tf )

Cp, f

Dt

= kf

2T f

Table 6 Initial and boundary conditions for the 3-D model.

(9)

For charging and discharging (Initial condition)

Heat conduction for SHS module is given by; s Cp, s

Ts = ks t

2T s

(10)

where, , Cp and k represent density, specific heat and thermal conductivity of the cast steel, respectively. P and v represent the pressure and velocity. Ts and Tf represent the temperature of solid (cast steel) and fluid (air). The numerical analysis is conducted on the basis of the following assumptions:

For SHS

At t = 0; T = Tini, vf = 0

At z = 0;

T (r , , z , t ) z

=0

At z = L ;

T (r , , z , t ) z

=0

T (r , , z , t ) z

At r = D /2; For HTF

1. The model is perfectly insulated, i.e. heat loss from the model to the surrounding is zero. 2. The storage material is isotropic and homogenous. 3. Effect of radiation heat transfer from the system is neglected.

=0

At z = 0 ;

vf (r , , z, t ) = vf , in ,

At z = L ;

n (kf Tf ) = 0, P (r , , L , t ) = Patm (Atmospheric

pressure)

At r = dtube /2 ;

7

(Adiabatic )

Tf (r , , z, t ) = Tf , in

vf (r , , z, t ) = 0 (no slip)

Applied Energy 251 (2019) 113322

K. Vigneshwaran, et al.

Fig. 5. 3-D model of the SHS system displaying the mesh features.

nAtube vf , in = Ach vch

(11)

module is symmetric to ‘ ’, only half diameter SHS model is simulated. Triangular and free tetrahedral elements are used to generate the mesh of the model. To take care of boundary layer effects, a fine mesh is generated near the walls, whereas, it is coarse in the core of the domains in order to reduce the total number of elements. The cost of handling computation relies upon the total number of elements. Also, the quality of the grid has been chosen as optimum in order to match the real-time experimental data. Therefore, the grid is made independent of the solution quality, so that the results do not vary much after selecting the best possible grid size. The numerical model is solved for different grid sizes by varying the total number of elements as follows; 208922, 327000, 376838 and 567068. The variation in the volume average temperature of the module with respect to time for all grid sizes is plotted in Fig. 6 to find an optimal number of elements. It is observed that the predicted volume average temperature is independent of the grid size for 376838 elements and above. So, in order to solve in lesser computation time, all subsequent simulations are carried with 376838 elements.

where n represents the number of HTF passages, Atube and Ach represent the cross-section area of tube and channel. 3.1. Initial and boundary conditions The initial and boundary conditions used for solving mass, momentum and energy equations are given in Table 6. 3.2. Computational methodology The computational model is shown in Fig. 5. The implicit scheme Backward Differentiation Formula is used for solving Eqs. (7)–(11). A parallel direct solver available in Comsol Multiphysics is used. Instead of using a constant time step, dynamic time stepping is adopted to save the computational time. The initial minimum time step and the maximum automatic time step are 0.001 s and 0.1 s, respectively. Since the

Fig. 6. Volume average temperature of the module for different grid sizes.

8

Applied Energy 251 (2019) 113322

K. Vigneshwaran, et al.

4. 1-D mathematical model and parameter estimation

The HTF passage consists of drilled holes in the storage module. Hence, the conductive resistance due to the tube material is not taken into consideration. So, the overall heat transfer coefficient is estimated only based on the convective heat transfer (hi) resistance existing in the fluid [53].

In general, 3-D models provide flexibility in terms of variation in geometrical design features, but due to high computational time, they cannot be employed for real-time monitoring and optimization. In this respect, 1-D dynamic modeling is advantageous over 3-D modeling. Further, the 1-D model has the following advantages:

The HTF pressure (P) drop is calculated by using the HagenPoiseuille equation [53].

(i) It is possible to input the dynamic fluctuations in the HTF inlet temperature observed during the real-time experiments. This helps in predicting the SHS system behaviour more realistically, which is not possible in case of 3-D modeling. (ii) 1-D simulations are several hundred times faster than 3-D simulations. (iii) It is highly feasible to integrate the developed 1-D model with the real time application to study the dynamic behaviour of the system. (iv) Based on the downstream requirement, system capacity modulation can be varied easily using 1-D modeling.

Pout

d ( dt

f , out )

nVtube

d ( dt

f , out cpf , out Tf , out )

= nAtube (

= nAtube (vf , in

f , in vf , in

f , in Cpf , in Tf , in

f , out vf , out )

=

f , out Cpf , out Tf , out )

ms cps

dt

=q

(20)

Nu kf

hi =

log

( ) Tin Tout

The maximum estimated value of Reynolds number for the flow conditions inside the HTF passage is 5048. Reynolds number (Re) and Prandtl number (Pr) are calculated as [53]; f , avg dtube Vf , avg

Re =

µf , avg

Ts, avg

Ts, avg

(24)

kf

where, µ f represents the dynamic viscosity of the fluid (air). Average velocity of air is given by;

(13)

vf , in + vf , out

(25)

2

The total mass of the storage module (ms) is estimated by;

ms =

4

2 dtube n] L

[D 2

(26)

s

Mass of the SHS material per HTF passage,

Tout )

ms n

(27)

The temperature difference between the module average and module initial temperature is given by;

(15)

T = Ts, avg

(28)

Tini

The heat gained/released by the SHS module is calculated using Eqs. (2) and (3) respectively. Different properties of the HTF are estimated as the function of temperature through the following empirical equations [54].

= 345.57(T

(16)

The temperature difference between HTF outlet temperature and module average temperature is given by;

Tout = Tf , out

(23)

µf , avg Cpf , avg

Pr =

where, n and V represent number and volume of the HTF passages. , v, Cpf and Tf represent density, velocity, specific heat and temperature of the HTF, respectively. Uo and q represent overall heat transfer coefficient and heat transfer rate. The temperature difference between HTF inlet temperature and module average temperature is given by;

Tin = Tf , in

(22)

dtube

ms/ tube = q=

(21)

Nu = aRebPr c

(14)

Uo Atube n ( Tin

(19)

where, P, R and M represent the pressure, universal gas constant and molecular weight of the air respectively. Heat transfer coefficient is estimated from the dimensionless parameter Nusselt number (Nu) which is further expressed as a function of Reynolds number (Re) and Prandtl number (Pr) number in the DittusBoelter equation [53].

(12)

q

Atube vf , avg

4 dtube

PM RT

vf , avg =

vf , out

µf , avg

128

=

The density of air at inlet and outlet of the storage module is estimated using ideal gas law [53],

The volume average temperature of the storage module is estimated by heat balance between the HTF and the storage module.

dTs, avg

Pin L

The Modelica language is widely adopted in the automation industry to develop the real-time monitoring and optimization tool. Therefore, in this work, a simplified 1-D dynamic model for the SHS system is developed and implemented using Modelica language in Dymola software tool [52]. The objectives of the 1-D model are (i) to predict the temperature variation along the length of the SHS, and (ii) to estimate the output power of the HTF and charging/discharging time for monitoring and optimization application. To incorporate the temperature variation along the length, the concept of multiple Continuous Stirred Tank Reactor (CSTR) in series is used by adopting an object-oriented framework of Modelica language. From experimental results, it is observed that the temperature variation in the radial direction is negligible. Therefore, elements 1, 2 and 3 are modelled as ‘CSTR-1’, elements 4, 5, 6 are modelled as ‘CSTR-2’ and elements 7, 8, 9 are modelled as ‘CSTR3’ (Fig. 4). All three CSTR’s are connected in series. The model equations for a single CSTR are as given below; The unsteady state mass and heat balance for HTF is given below;

nVtube

(18)

Uo = hi

2.6884) 1,

150 K

(29)

T

3000 K

11T 2

+ 5.0523 × 10 8T + 4.1130 (30)

µ = 2.5914 × 10

(17)

× 10 9

6

15T 3

1.4346 × 10

Applied Energy 251 (2019) 113322

K. Vigneshwaran, et al.

Table 7 Parameter values in Dittus-Boelter equation. Parameters

Values

a b c

0.023 0.8 0.4

Cp = 1.3864 × 10 × 10

4T

13T 4

6.4747 × 10

10T 3

+ 1.0234 × 10 6T 2

+ 1.0613

k = 1.5797 × 10 17T 5 + 9.46 × 10 14T 4 + 2.2012 × 10 10T 2 2.3758 × 10 7T 2 + 1.7082 × 10 4T 7.488 × 10 3

4.3282 (31) (32)

The model parameters of the heat transfer coefficient are unknown and have to be estimated from experimental data. The Dymola’s design

Fig. 7. Validation of numerical 3-D and 1-D dynamic models with experiment results during (a) charging and (b) discharging processes.

10

Applied Energy 251 (2019) 113322

K. Vigneshwaran, et al.

Fig. 8. Variation of module volume average temperature with time during (a) charging and (b) discharging processes.

library is used to minimize the error between the model prediction and the experimentally measured temperature along the length of the module. The decision variables for optimization are ‘a’, ‘b’ and ‘c’ of the Dittus-Boelter equation (see Table 7). The error function is given below [52]; t

Error (t ) =

1 t=1

(Ts, avg (t ))Sim (Ts, avg (t ))Exp

2

+

1

(Tf , out (t ))Sim

The methodology for 1-D dynamic model simulation is described as flowchart given in SUPPLEMENTARY section. 5. Results and discussion In this section, results obtained from 3-D and 1-D mathematical models, and experimental studies are presented. The performance characteristics of the SHS system such as energy stored and discharged during the charging and discharging processes are also discussed in detail. The experiments are performed as per the design of experiments reported in Tables 4 and 5. Numerically predicted energy storage and discharge rates are

2

(Tf , out (t ))Exp (33)

where, Sim and Exp represent the Simulation value and Experiment value, respectively. 11

Applied Energy 251 (2019) 113322

K. Vigneshwaran, et al.

Fig. 9. Variation in temperature at different thermocouple positions of the SHS module during (a) charging and (b) discharging.

validated with the experiment data. Experiment Nos. 7 and 16 are considered for the validation of numerically predicted charging and discharging rates, respectively. In numerical simulation, the initial temperature (Tini) of the SHS module and the HTF inlet temperature (Tf,in) are fixed as 393 K and 473 K during charging and 473 K and 393 K during discharging. For both 1-D and 3-D validation studies, the velocity of the HTF (vtube) is maintained at 20 m/s. Fig. 7a and b illustrates the validation of the developed thermal models with the corresponding experimental data in terms of energy storage and discharge rates. The flowchart given in the SUPPLEMENTARY section provides the various steps involved in the validation of 1-D dynamic modeling. It is observed that predictions from both 1-D and 3-D models match closely with

experimentally estimated values within the error of ± 3.3% for the 1-D model and ± 9% for the 3-D model. The dynamic fluctuation in the air inlet temperature is not possible to include in 3-D model simulation due to limitations in the Comsol Multiphysics 4.3a software tool. Even if it is possible, due to the large computational requirement, the simulation may require huge computational time. Therefore, a higher deviation is observed in 3-D model prediction as compared to the 1-D model. 5.1. Experimental results 5.1.1. Temperature variation during charging and discharging processes The variation of air inlet temperature (Tf , in ), air outlet temperature 12

Applied Energy 251 (2019) 113322

K. Vigneshwaran, et al.

Fig. 10. Energy profile for charging - Experiment Nos. 1/4/7.

(Tf , out ), module volume average temperature (Ts, avg ) with respect to time are shown in Fig. 8. The air inlet temperature has fluctuated within a temperature range of 5–8 K while conducting both charging and discharging experiments due to practical limitations of the temperature controller. During the charging, the storage module is maintained at an initial temperature of 493 K (for instance, Experiment No. 1 is taken for discussion). Tf , in and vf , in are maintained at 573 K and 20 m/s, respectively.

It is observed that the module temperature increases linearly with time till 100 min, whereas later it follows a non-linear profile and reaches an asymptotic value. Approximately, 70% of the total module energy is stored within 40% of the total time (119 min) and remaining 30% of the total energy storage takes 60% of the total time (159 min). Similarly, for discharging, the initial module temperature is maintained at 573 K (for instance, Experiment No. 10 is taken for discussion), whereas Tf , in and vf , in are maintained at 493 K and 20 m/s, respectively. The

Fig. 11. Energy and power profile for discharging - Experiment Nos. 10/13/16. 13

Applied Energy 251 (2019) 113322

K. Vigneshwaran, et al.

Fig. 12. Axial temperature variation in the SHS module during (a) charging and (b) discharging processes.

temperature profile shown in Fig. 8b reflects that the discharging process is found to be similar to charging, but with heat flow from the module to the HTF. The charging and discharging experiments are stopped once the module reach a steady state condition, such that no heat exchange takes place between the air and the SHS module. Therefore, the net heat loss

from the air decreases with a reduced temperature gradient between the module and air. Further, this can be verified from the variation of T (HTF), which tends to zero at the end of the experiment. Thus, the temperature of the air at the outlet tends towards the inlet temperature as shown by Tf , out in Fig. 8(a) and (b). The temperature profiles measured at different thermocouple locations during the charging and the

14

Applied Energy 251 (2019) 113322

K. Vigneshwaran, et al.

Fig. 13. Radial temperature variation in the SHS module during (a) charging and (b) discharging processes.

discharging processes are shown in Fig. 9(a) and (b). ‘Section-A’, ‘Section-B’ and ‘Section-C’ represent temperature data of three thermocouples located in each of these sections (as illustrated in Table 2). It is observed from the temperature data measured at ‘Section-A’ that the heat transfer in this section is faster as compared to ‘Section-B and Section C’. Such an increased heat transfer rate is observed due to the

high-temperature difference between the air and the module at ‘Section-A’. As expected from the direction of heat transfer, the air temperature drops along the length of the HTF passages during the charging process and increases during the discharging process.

15

Applied Energy 251 (2019) 113322

K. Vigneshwaran, et al.

Fig. 14. Influence of HTF velocity on Ts,avg during charging (red line) and discharging (blue line) between the temperature range of 443–523 K.

5.1.2. Energy stored and energy discharge rates Fig. 10 provides details about the amount of energy stored during the charging process. During this process, vf , in is maintained at 20 m/s and Tf , in is varied at three different inlet conditions 393 K, 443 K and 493 K and their respective Tini are 473 K, 523 K and 573 K. As a result, the total energy stored during charging Experiment Nos. 1/4/7 is calculated as 13.14 MJ/11.85 MJ/11.36 MJ, respectively. For all temperature ranges discussed in the experiments, the rate of heat transfer and quantity of energy transfer by the air are higher at lower temperature range as compared to the higher temperature range. This happens primarily because the density of air is relatively high at low temperature. Thus, the mass flow rate and the amount of energy transferred by the air is higher. Hence, better system performance is observed at relatively lower temperatures of the air. While discharging, achieving a constant power output is one of the vital aspects to run any downstream process at steady state. To accomplish a steady state operation, both HTF flowrate and temperature need careful manipulation. Higher flowrates lead to high rate of charging and discharging, but at the same time, the temperature difference between the HTF inlet and outlet will be less, resulting in less utilization of supplied heat. Higher HTF inlet temperature increases both the quality and the quantity of the heat stored. In the context of the present study, increasing the flow rate restricts the rise in HTF temperature due to lesser residence time inside the air heater. On the contrary, reducing the flowrate helps to attain higher HTF inlet temperatures. In the current experimental investigations, flowrate and temperature of the air

were manipulated and accordingly, different experiments were performed. For a fixed air flow rate, the heat transfer rate between the HTF and the SHS module decreases with time due to drop in temperature gradient. Hence, the obtained output power decreases linearly with time, as illustrated in Fig. 11. The peak power discharged at lower operating temperature range is observed to be high due to the high amount of energy discharged. The trend of power (rate of energy discharged) and energy discharged for the temperature range of 473–393 K is almost similar to the range of 523–443 K. The beginning of the discharging is marked by peak power, which gets reduced gradually towards the end. In sensible heat storage system, the peak power output is delivered at the start of the discharging process, and this may be an advantage for balancing the peak load during intermittent operation. The sensible heat storage may not be suitable for steady power output conditions for a longer duration. 5.1.3. Axial temperature variation of the SHS module during charging and discharging processes The variation in temperature along the axial direction of the SHS module with time, during both charging and discharging processes are illustrated in Fig. 12(a) and (b). During the charging process, Tf , in and Tini are fixed as 573 K and 493 K (Experiment No. 1), respectively; and vice-versa in discharging (Experiment No. 10). As shown in Fig. 2 and Table 2, the thermocouples are located in three different axial positions at 22 mm, 378 mm and 718 mm from the inlet of the SHS module. After 60 min of the charging process, approximately 18 K of temperature

16

Applied Energy 251 (2019) 113322

K. Vigneshwaran, et al.

difference exists consecutively between the three sections (A, B, C) in the axial direction. This behaviour is observed until 120 min, after which the temperature difference between consecutive sections reduces. This happens because the heat transfer between the HTF and the SHS module reduces gradually over time. Finally at steady state, as Tf , out approaches close to Ts, avg , the temperature difference between consecutive sections is only 2 K. The trend of axial variation during the discharging process is similar to the charging process. Therefore, significant temperature variations are observed along the length of the SHS module during the initial period of charging and discharging processes, which are found to reduce over time.

the HTF and the SHS module. It should be noted that similar behaviour is observed for other operating temperature ranges, with the only difference in their respective charging and discharging times. The numerical simulations at all three velocities namely; 11 m/s, 15 m/s and 20 m/s are performed. The predicted temperature contours along the module length show that the front end of the module attains high-temperature much faster for charging and vice-versa for discharging. The percentage completion of the charging and discharging processes in the operating temperature range of 493–573 K at different flow velocities is mentioned in Fig. 15. The predicted charging and discharging rates are found to be similar to the experimental results reported in Section 5.1. A decrease in time is observed due to an increase in the heat transfer rate between the air and the SHS module. At t = 13 min, the temperature contours during charging and discharging reveal that the heat transfer starts in a similar fashion at all the velocities. At t = 52 min, the charging and discharging for 20 m/s is only 7% faster than 11 m/s. The maximum variation in the percentage charging (approximately 8.3%) between the highest and lowest velocity is only seen at t = 278 min. Hence, a very less reduction in charging or discharging time is observed with an increase in velocity.

5.1.4. Radial temperature variation during charging and discharging processes To understand the thermal behaviour of the system in the radial direction, the temperature variations along all the radial thermocouple locations for Experiment Nos. 1 and 10 are analysed. The thermocouple positions are illustrated at the bottom right corner in Fig. 13(a) and (b). Due to the distribution of HTF passages uniformly in the radial direction, the conduction heat transfer rate is faster in all the sections. Therefore, the temperature gradient between any radial location is negligible at any time interval throughout charging and discharging processes. The same observations are conferred from the 3-D modeling results. This concludes that the heat transfer does not vary significantly in the radial direction as compared to the axial direction.

5.3. Analysis of SHS performance using the 1-D dynamic model The result obtained from the 1-D model is validated by comparing the model prediction with experimentally measured temperature of three CSTR’s and air outlet temperature. Here, the dynamic variation of the air inlet temperature during experiments is given as an input. The experimental CSTR temperature is estimated by calculating the module volume average temperature using the temperature readings measured at different radial locations. In the 1-D model, the notations A, B and C stands for CSTR 1, 2 and 3 respectively. The variations observed in the volume average temperature of different sections of the SHS module are denoted by A (red), B (blue) and C (yellow) as shown in Figs. 16 and 17. The dotted lines and continuous lines represent the variation of Ts,avg for each section obtained through the 1-D simulation data and experiment data, respectively. After obtaining the initial Ts,avg and Tf,out profiles, the software design library follows specific procedures (refer SUPPLEMENTARY section) to reduce the error in the profiles with reference to actual experimental Ts,avg and Tf,out data. After fulfilling this objective, the final simulated values of Ts,avg and Tf,out are plotted in Figs. 16 and 17. During the mathematical model validation, parameters (‘a’, ‘b’ and ‘c’) in the Dittus- Boelter equation are updated and tabulated in Table 8. The values of ‘b’ and ‘c’ are estimated to be 0.8 and 0.4, respectively. The value of ‘a’ is varying in the range of 2.4–2.72 in the case of the charging, and from 2.51–3 in the case of discharging. While charging, the coefficient ‘a’ is always increasing with respect to velocity. But in the case of discharging, coefficient ‘a’ is found to be constant for high HTF temperature. The updated values of parameters ‘a’, ‘b’ and ‘c’ are the final model variables representing the validated condition of the mathematical model with experimental data. The results conclude that the developed 1-D dynamic model can predict both charging and discharging temperature profiles closely and it is feasible to integrate them with real-time applications and processes.

5.1.5. Influence of the HTF velocity During experiments (Experiment Nos. 4/5/6 and 13/14/15), the influence of different HTF velocities on volume averaged temperature (Ts,avg) of the SHS module throughout the charging and discharging processes are described as red and blue colour, respectively in Fig. 14. The velocity of the HTF is varied to 2.5 m/s, 3.5 m/s and 4.5 m/s for both the processes. As expected, a higher velocity of HTF increases the mass flow rate of the HTF which leads to rapid heat transfer from the HTF to the storage module. The increased heat transfer rate ensures that the steady state of Ts,avg at a faster rate. The observations are same in both charging and discharging processes. The rate of charging and discharging increases with HTF velocity and such increment is observed due to higher heat transfer coefficient and mass flow rate. However, based on the downstream and upstream application requirements, temperature and flow rate of HTF have to be well-adjusted. 5.2. Analysis of the SHS performance using a 3-D model The three-dimensional model of the SHS system is useful to visualize the physical behaviour of the module during the energy interactions. Additionally, it is highly useful to study the axial and radial temperature variations in the SHS module. The simulation result of the SHS module for higher operating temperature range is shown in Fig. 15. It is observed that there is a negligible variation of temperature in a radial direction than axial direction. This behaviour has also been verified from experimental observation. This is mainly due to the domination of conduction heat transfer in radial directions as shown in Fig. 15. Whereas along the axial direction, the heat transfer rate drops gradually along the length due to the reduction in temperature gradient between

17

Applied Energy 251 (2019) 113322

K. Vigneshwaran, et al.

Fig. 15. Temperature contours obtained from the 3-D model results during charging and discharging processes at different inlet HTF velocities.

18

Applied Energy 251 (2019) 113322

K. Vigneshwaran, et al.

Fig. 16. 1-D model simulation results during charging (Experiment Nos. 1–9): variation of volume average temperature of 3 CSTR’s representing Section-A, Section-B and Section-C of the storage module.

19

Applied Energy 251 (2019) 113322

K. Vigneshwaran, et al.

Fig. 17. 1-D model simulation results during discharging (Experiment Nos. 10–18): variation of volume average temperature of 3 CSTR’s representing Section-A, Section-B and Section-C of the storage module.

20

Applied Energy 251 (2019) 113322

K. Vigneshwaran, et al.

As compared to other commonly used SHS material like concrete, silica bricks, sand and rocks, cast steel is having high volumetric capacity (∼450 kWht/m3) and thermal conductivity (40 W/mK) [58]. Therefore, cast steel may be a useful replacement in TES integrated power plants for lesser payback periods.

Table 8 Estimation of Dittus-Boelter co-relation parameter 'a' during charging and discharging processes. Charging experiment no.

vf,in (m/ s)

Simulated ‘a’ value

Discharging experiment no.

vf,in (m/ s)

Simulated ‘a’ value

1 2 3 4 5 6 7 8 9

20 15 11 20 15 11 20 15 11

2.72 2.62 2.45 2.6 2.5 2.5 2.5 2.48 2.4

10 11 12 13 14 15 16 17 18

20 15 11 20 15 11 20 15 11

3 3 3 2.95 2.89 2.51 2.73 2.67 2.51

• Cooking

•

6. Projected applications of the SHS The success of any SHS system depends upon the selection of appropriate energy storage material and application. The energy delivered by the TES system during the initial period of the discharging (as shown in Fig. 11) is highly recommended to be used for the application with frequent peak power demands and locations with inadequate solar intensity. In this section, the various energy storage applications of the SHS system reported in the literature are analysed. The current research on the SHS systems at Indian Institute of Technology Guwahati is in the stage of developing a cost-effective SHS material with high durability. The next phase of the current study is to investigate the utilization of the SHS system for different applications such as steam generation, space heating and drying of agricultural products. The future work will focus on the extension of the developed mathematical model for performance monitoring and optimization of the SHS module for different applications. The list of SHS applications and its potential integration with downstream is briefed below.

7. Conclusions The present work is focused on a detailed investigation of hightemperature sensible heat thermal energy storage system. The storage material and heat transfer fluid used for the analysis are cast steel and air. The estimated storage unit cost for cast steel system employed in the present investigation is 18.43 US$/ MJ. In the experiment section, a detailed investigation has been conducted from low temperature to high-temperature charging and discharging processes at different HTF flow velocities. 3-D numerical model and a simplified 1-D dynamic model are developed and simulated using similar conditions as maintained in the experiments. Based on experimental and numerical observations, the following conclusions can be made;

• Solar heating applications: In many colossal communities, there is

•

application: Cooking by using solar thermal energy is having immense scope in the domestic sector. Solar steam generation projects such as one at Brahmakumaris, India [59] have been found to be cost-effective while addressing the cooking needs of small communities and districts, reducing tons of natural gas consumption every year. By integrating the cast steel based SHS system as developed in the current study, it is possible to have 24 hrs in a day steam generation ability. Waste heat recovery: Industries such as paper and pulp mills, casting of the die, steel works, industrial waste incineration, etc. are generating excess heat beyond their usages. Such industries can save a huge amount of energy and reduce CO2 emissions by re-using the waste heat. This waste heat can be recovered using the compact and adequately sized SHS module tested in this study, which is having a simple approach for multiplying the capacity by changing the module scale and it can become functional with different waste heat generators [19]. The recovered heat could be utilized to cope up with the heat loads in different applications in the industry having high peak power demand.

always a demand for peak power just before the office work timings. Once the peak power load is balanced, the load requirement decreases. Mostly, the requirements are thermal management of hot water generation, cooking, space heating and melting the snow on roads and runways during the winters [55]. The project such as “Drake Landing Solar Community” have made a massive impact in energy saving with the proficient utilization of available natural resources [56]. To balance the power requirement in such peak load conditions with high power intensity during discharging, the module tested in this work is perfectly suitable for such energy requirement. Power production: CSP is one of the attractive energy conversion technologies for power production. However, to overcome the intermittent character of solar power, an efficient and cost-effective TES technology has to be integrated. Due to the high durability, high operating temperature range, stress-free erection, high experimental feedback and low cost, the SHS is mostly preferred to be integrated with CSP plants. By integrating the SHS module developed in the current study, it is possible to generate the steam directly at hightemperature. Thus, the generated steam can be used for power production. The SHS medium used in the current study is found to be thermally stable and no degradation has been observed even after over 400 charging/discharging cycles as compared to other composite materials [57].

• In all the experiments, the time taken for charging to reach a steady

•

21

state condition is higher than discharging. Approximately, 70% of the total module energy is stored in 40% of the time and the remaining is stored in 60% of the time. The energy storage and discharge rate are observed to be faster in “Section-A” followed by “Section-B and Section-C” due to the high-temperature gradient between the air and storage module near the inlet section. The temperature profiles of individual thermocouples show that the heat transfer rate is much more uniform in a radial direction than axial direction. Along the axial direction, the heat transfer rate drops with the fall in the air temperature along the length. The energy charged in the SHS module in the temperature ranges of 393–473 K, 443–523 K and 493–573 K are 13.14 MJ, 11.85 MJ and 11.36 MJ, respectively. The discharging profile of the cast steel based SHS system recommends that the peak load demand can be achieved for a system with an interrupted power source. However, the SHS system can be adopted for constant power output cases by manipulating the HTF inlet temperature and flow rate. High velocity of air leads to a higher rate of heat transfer between the air and storage medium. 3-D numerical and 1-D dynamic model results are found to be in close proximity with the experimental data. 3-D simulation confirms

Applied Energy 251 (2019) 113322

K. Vigneshwaran, et al.

•

that the temperature gradient decreases along the length of the storage module. The real-time operation of the SHS system requires real-time monitoring and optimization solution. One can visualize the thermal characteristics of the complete system by coupling the mathematical model with the real-time applications and such solution can be developed using the 1-D dynamic model. The developed 1-D model is able to predict the temperature profile of the SHS module with minimum error in lesser simulation time. The developed SHS system can potentially act as an important utility component in real-time applications. The advantages of using the SHS system are highlighted by integrating with solar heating, power production, cooking and waste heat recovery applications.

[21] Diago M, Iniesta AC, Soum-Glaude A, Calvet N. Characterization of desert sand to be used as a high-temperature thermal energy storage medium in particle solar receiver technology. Appl Energy 2018;216:402–13. https://doi.org/10.1016/j. apenergy.2018.02.106. [22] Alva G, Liu L, Huang X, Fang G. Thermal energy storage materials and systems for solar energy applications. Renew Sustain Energy Rev 2017;68:693–706. https:// doi.org/10.1016/j.rser.2016.10.021. [23] Gibb D, Johnson M, Romaní J, Gasia J, Cabeza LF, Seitz A. Process integration of thermal energy storage systems – Evaluation methodology and case studies. Appl Energy 2018;230:750–60. https://doi.org/10.1016/j.apenergy.2018.09.001. [24] Martins M, Villalobos U, Delclos T, Armstrong P, Bergan PG, Calvet N. New concentrating solar power facility for testing high temperature concrete thermal energy storage. Energy Procedia 2015;75:2144–9. https://doi.org/10.1016/j.egypro.2015. 07.350. [25] Geyer M, Bitterlich W, Werner K. The dual medium storage tank at the IEA/SSPS project in almeria (Spain); Part I: experimental validation of the thermodynamic design model. J Sol Energy Eng 1987;109:192–8. https://doi.org/10.1115/1. 3268205. [26] Gude VG. Energy storage for desalination processes powered by renewable energy and waste heat sources. Appl Energy 2015;137:877–98. https://doi.org/10.1016/j. apenergy.2014.06.061. [27] Okello D, Nydal OJ, Nyeinga K, Banda EJK. Experimental investigation on heat extraction from a rock bed heat storage system for high temperature applications. J Energy South Africa 2016;27(2):30–7. https://doi.org/10.17159/2413-3051/ 2016/v27i2a1339. [28] Gudmundsson O, Thorsen JE, Brand M. The role of district heating in coupling of the future renewable energy sectors. Energy Procedia 2018;149:445–54. https:// doi.org/10.1016/j.egypro.2018.08.209. [29] Dinker A, Agarwal M, Agarwal GD. Experimental thermal performance study of natural graphite as sensible heat storage material using different heat transfer fluids. Exp Heat Transf 2018;31:219–31. https://doi.org/10.1080/08916152.2017. 1405101. [30] John E, Hale M, Selvam P. Concrete as a thermal energy storage medium for thermocline solar energy storage systems. Sol Energy 2013;96:194–204. https:// doi.org/10.1016/j.solener.2013.06.033. [31] Alva G, Lin Y, Fang G. An overview of thermal energy storage systems. Energy 2018;144:341–78. https://doi.org/10.1016/j.energy.2017.12.037. [32] Zunft S, Hänel M, Krüger M, Dreißigacker V, Göhring F, Wahl E. Jülich solar power tower—experimental evaluation of the storage subsystem and performance calculation. J Sol Energy Eng 2011;133. https://doi.org/10.1115/1.4004358. 031019. [33] Good P, Zanganeh G, Ambrosetti G, Barbato MC, Pedretti A, Steinfeld A. Towards a commercial parabolic trough CSP system using air as heat transfer fluid. Energy Procedia 2014;49:381–5. https://doi.org/10.1016/j.egypro.2014.03.041. [34] Modi A, Haglind F. Performance analysis of a Kalina cycle for a central receiver solar thermal power plant with direct steam generation. Appl Therm Eng 2014;65:201–8. https://doi.org/10.1016/j.applthermaleng.2014.01.010. [35] Goods SH, Bradshaw RW. Corrosion of stainless steels and carbon steel by molten mixtures of commercial nitrate salts. J Mater Eng Perform 2004;13:78–87. https:// doi.org/10.1361/10599490417542. [36] Guillot S, Faik A, Rakhmatullin A, Lambert J, Veron E, Echegut P, et al. Corrosion effects between molten salts and thermal storage material for concentrated solar power plants. Appl Energy 2012;94:174–81. https://doi.org/10.1016/j.apenergy. 2011.12.057. [37] Liu M, Belusko M, Steven Tay NH, Bruno F. Impact of the heat transfer fluid in a flat plate phase change thermal storage unit for concentrated solar tower plants. Sol Energy 2014;101:220–31. https://doi.org/10.1016/j.solener.2013.12.030. [38] Sodhi GS, Jaiswal AK, Vigneshwaran K, Muthukumar P. Investigation of charging and discharging characteristics of a horizontal conical shell and tube latent thermal energy storage device. Energy Convers Manag 2019;188:381–97. https://doi.org/ 10.1016/j.enconman.2019.03.022. [39] Zanganeh G, Khanna R, Walser C, Pedretti A, Haselbacher A, Steinfeld A. Experimental and numerical investigation of combined sensible–latent heat for thermal energy storage at 575°C and above. Sol Energy 2015;114:77–90. https:// doi.org/10.1016/j.solener.2015.01.022. [40] Cascetta M, Cau G, Puddu P, Serra F. Numerical investigation of a packed bed thermal energy storage system with different heat transfer fluids. Energy Procedia 2014;45:598–607. https://doi.org/10.1016/j.egypro.2014.01.064. [41] Mahmood M, Traverso A, Traverso AN, Massardo AF, Marsano D, Cravero C. Thermal energy storage for CSP hybrid gas turbine systems: Dynamic modelling and experimental validation. Appl Energy 2018;212:1240–51. https://doi.org/10. 1016/j.apenergy.2017.12.130. [42] Zauner C, Hengstberger F, Mörzinger B, Hofmann R, Walter H. Experimental characterization and simulation of a hybrid sensible-latent heat storage. Appl Energy 2017;189:506–19. https://doi.org/10.1016/j.apenergy.2016.12.079. [43] Laing D, Bahl C, Bauer T, Fiss M, Breidenbach N, Hempel M. High-temperature solid-media thermal energy storage for solar thermal power plants. Proc IEEE 2012;100:516–24. https://doi.org/10.1109/JPROC.2011.2154290. [44] Krönauer A, Lävemann E, Brückner S, Hauer A. Mobile sorption heat storage in industrial waste heat recovery. Energy Procedia 2015;73:272–80. https://doi.org/ 10.1016/j.egypro.2015.07.688. [45] Selvam RP, Castro M. 3D FEM model to improve the heat transfer in concrete for thermal energy storage in solar power generation. In: ASME 2010 4th Int. Conf. Energy Sustain. Vol. 2, ASME; 2010; 699–707. doi:10.1115/ES2010-90078. [46] Gil A, Medrano M, Martorell I, Lázaro A, Dolado P, Zalba B, et al. State of the art on high temperature thermal energy storage for power generation. Part 1—Concepts, materials and modellization. Renew Sustain Energy Rev 2010;14:31–55. https://

Acknowledgement The authors gratefully acknowledge the Department of Science and Technology (DST), Government of India, for their financial support (Project No: DST/TMD/SERI/D12(C)). Appendix A. Supplementary material Supplementary data to this article can be found online at https:// doi.org/10.1016/j.apenergy.2019.113322. References [1] International Energy Agency. Global energy & CO 2 status report; 2018. [2] Kalogirou SA. Chapter three – solar energy collectors. Sol Energy Eng 2009:121–217. 10.1016/B978-0-12-374501-9.00003-0. [3] Sukhatme SP, Nayak JK. Solar energy: principles of thermal collection and storage. 3rd ed. New Delhi: Tata McGraw-Hill; 2008. [4] Dincer I, Rosen MA. Energetic, environmental and economic aspects of thermal energy storage systems for cooling capacity. Appl Therm Eng 2001;21:1105–17. https://doi.org/10.1016/S1359-4311(00)00102-2. [5] IEA. Technology roadmap: Concentrating solar power. OECD/IEA Publishing; 2010. [6] Pelay U, Luo L, Fan Y, Stitou D, Rood M. Thermal energy storage systems for concentrated solar power plants. Renew Sustain Energy Rev 2017;79:82–100. https://doi.org/10.1016/j.rser.2017.03.139. [7] Ling TC, Poon CS. Use of phase change materials for thermal energy storage in concrete: An overview. Constr Build Mater 2013;46:55–62. https://doi.org/10. 1016/j.conbuildmat.2013.04.031. [8] Pardo P, Deydier A, Anxionnaz-Minvielle Z, Rougé S, Cabassud M, Cognet P. A review on high temperature thermochemical heat energy storage. Renew Sustain Energy Rev 2014;32:591–610. https://doi.org/10.1016/j.rser.2013.12.014. [9] Medrano M, Gil A, Martorell I, Potau X, Cabeza LF. State of the art on high-temperature thermal energy storage for power generation. Part 2—Case studies. Renew Sustain Energy Rev 2010;14:56–72. https://doi.org/10.1016/j.rser.2009.07.036. [10] Ushak S, Gutierrez A, Flores E, Galleguillos H, Grageda M. Development of thermal energy storage materials from waste-process Salts. Energy Procedia 2014;57:627–32. https://doi.org/10.1016/j.egypro.2014.10.217. [11] Bauer T, Pfleger N, Breidenbach N, Eck M, Laing D, Kaesche S. Material aspects of Solar Salt for sensible heat storage. Appl Energy 2013;111:1114–9. https://doi.org/ 10.1016/j.apenergy.2013.04.072. [12] Vignarooban K, Xu X, Arvay A, Hsu K, Kannan AM. Heat transfer fluids for concentrating solar power systems – A review. Appl Energy 2015;146:383–96. https:// doi.org/10.1016/j.apenergy.2015.01.125. [13] Zhang H, Baeyens J, Cáceres G, Degrève J, Lv Y. Thermal energy storage: Recent developments and practical aspects. Prog Energy Combust Sci 2016;53:1–40. https://doi.org/10.1016/j.pecs.2015.10.003. [14] Xu B, Han J, Kumar A, Li P, Yang Y. Thermal storage using sand saturated by thermal-conductive fluid and comparison with the use of concrete. J Energy Storage 2017;13:85–95. https://doi.org/10.1016/j.est.2017.06.010. [15] Rao CRC, Niyas H, Muthukumar P. Performance tests on lab–scale sensible heat storage prototypes. Appl Therm Eng 2018;129:953–67. https://doi.org/10.1016/j. applthermaleng.2017.10.085. [16] Yalsin Gogus. Energy storage systems - Volume I. EOLSS Publications; 2009. [17] Meffre A, Xavier Py B, Veron E, Patrick Echegut B. High-temperature sensible heatbased thermal energy storage materials made of vitrified MSWI fly ashes. Waste and Biomass Valorization 2015;6(6):1003–14. https://doi.org/10.1007/s12649-0159409-9. [18] Hyman Lucas. Sustainable thermal storage systems planning design and operations. McGraw-Hill Education; 2011. [19] Miró L, Gasia J, Cabeza LF. Thermal energy storage (TES) for industrial waste heat (IWH) recovery: A review. Appl Energy 2016;179:284–301. https://doi.org/10. 1016/j.apenergy.2016.06.147. [20] Negi P, Singh S, Sharma K, Dixit J, Kumar S. Experimental investigation of thermal energy storage system for hot air applications. Singapore: Springer; 2018. p. 101–14. doi:10.1007/978-981-10-7122-5_12.

22

Applied Energy 251 (2019) 113322

K. Vigneshwaran, et al. doi.org/10.1016/j.rser.2009.07.035. [47] Prasad L, Muthukumar P. Design and optimization of lab-scale sensible heat storage prototype for solar thermal power plant application. Sol Energy 2013;97:217–29. https://doi.org/10.1016/j.solener.2013.08.022. [48] Pilkington Solar International GmbH PSI. Survey of thermal storage for parabolic trough power plants; period of performance: September 13, 1999 - June 12, 2000. Golden, CO (United States): National Renewable Energy Laboratory, 2000. [49] Kobe KA. Applied mathematics in chemical engineering (Mickley, Harold S., Sherwood, Thomas K., and Reed, Charles E.). J Chem Educ 1957;34:A552. https:// doi.org/10.1021/ed034pA552.1. [50] COMSOL multiphysics. www.comsol.com; 2017. [51] Niyas H, Prasad L, Muthukumar P. Performance investigation of high-temperature sensible heat thermal energy storage system during charging and discharging cycles. Clean Technol Environ Policy 2015;17:501–13. https://doi.org/10.1007/ s10098-014-0807-7. [52] Dassault systems. www.dymola.com; 2018. [53] Incropera FP, Lavine AS, Bergman TL, DeWitt DP. Fundamentals of heat and mass transfer. Wiley; 2007. [54] Zografos AI, Martin WA, Sunderland JE. Equations of properties as a function of

[55]

[56] [57] [58] [59]

23

temperature for seven fluids. Comput Methods Appl Mech Eng 1987;61:177–87. https://doi.org/10.1016/0045-7825(87)90003-X. Epting J, Müller MH, Genske D, Huggenberger P. Relating groundwater heat-potential to city-scale heat-demand: A theoretical consideration for urban groundwater resource management. Appl Energy 2018;228:1499–505. https://doi.org/10. 1016/j.apenergy.2018.06.154. Mesquita L, Mcclenahan D, Thornton J, Carriere J, Wong B. Drake landing solar community: 10 years of operation, 2017. doi:10.18086/swc.2017.06.09. Laing D, Bahl C, Bauer T, Lehmann D, Steinmann WD. Thermal energy storage for direct steam generation. Sol Energy 2011;85:627–33. https://doi.org/10.1016/j. solener.2010.08.015. Abhishek K, Mishra P, Mehta K, Ray A, Hadia N. India one solar thermal power plant : A case study, Bhubaneswar, Odisha.: International Congress on Renewable Energy (IOCRE-2013); 2013. CSH India. Steam cooking system for 6000 people by using Scheffler Dishes at BrahmaKumaris, Gurgaon. UNDP-GEF Concentrated Solar Thermal Heat (CSH) Project, MNRE, New Delhi, India: Government of India (http://www.cshindia.in/ images/Cooking/BramhaKumarisGurgaon.pdf); 2001.