Experimental study and neural networks prediction on thermal performance assessment of grooved channel air heater

Highlights

• Triangular grooves are used to improve the thermal performance of a solar air heater.

• Different groove parameters are experimentally examined.

• Correlations for Nu and f for this work are proposed.

• ANN model is used to predict the η, Nu and f.

• Optimum architecture of predictive model are reported.

Abstract

This research presents a study on convective heat transfer and friction factor of turbulent airflow in a channel air heater fitted with isosceles triangular grooves only and isosceles triangular grooves combined with transverse baffles. Experiments were conducted within a rectangular channel with aspect ratio, AR=15 and height, H = 20 mm with three different groove depths (d) of 4, 6 and 8 mm, two different groove inclination angles (α) of 30° and 45° with two different triangular directions of down-stream (DG) and up-steam (UG) of air flow. The airflow rate was presented in terms of Reynolds numbers (Re) based on the inlet hydraulic diameter of the channel and was in the range of 6700 to 17,000. The experimental results showed a significant effect on the heat transfer rate and friction factor in the presence of the isosceles triangular grooves only compared to the smooth-wall channel, by about 1.23–1.91 and 1.12–3.76 times, respectively. While, the case study of isosceles triangular grooves (DG) combined with transverse baffles (TB) provided higher heat transfer rate and friction factor as compared to the smooth-wall channel, by about 1.32–2.00 and 1.25–3.83 times, respectively. The maximum thermal performance enhancement of 1.49 was achieved with the use of 30° down-stream isosceles triangular grooves with the depth of 8 mm combined with transverse baffles at Re ≈6700. Correlations for Nusselt number (Nu) and friction factor (f) for this work were also proposed. The neural networks models were designed for predicting the thermal enhancement factor (η) of a channel air heater with combined turbulators by using the Rapid-Miner Studio 9.5 software. The architecture of the model was 4–20–20–1 with Tanh–Tanh activation function and the use of 5–fold cross validation for the training and testing data segmentation together with the linear_sampling type at a number of the epoch of 1700 provided the best performance for the η prediction which had the coefficient of determination (R²) and mean squared error (MSE) of 0.998864 and 1.2 × 10⁻⁵, respectively.

Introduction

A solar air heater (SAH) is a type of heat exchanger that transfers thermal energy from the solar radiation to the air flowing into it. Although solar energy is an energy that can be used without the cost of energy and in unlimited quantities, the design and development of a high-performance air heater for a better heat transfer process between absorber surface and flowing air is essential. Therefore numerous researchers have conducted studies to improve the heat transfer rate in the SAH by a variety of methods. Such methods include: increasing the flow path of the working fluid [1], [2], [3], [4], [5], [6], [7], [8], [9], [10], [11], using thermal storage materials [12], [13], [14], [15], [16], [17], increasing the surface area of the absorber plate or applying heat transfer enhancement techniques [18], [19], [20], [21], [22], [23], [24], [25], [26], [27], [28], [29], [30], [31], [32], [33], [34], [35], [36], [37].

Hassan and Abo-Elfadl [1] presented an experimental study on the effect of using different absorber plate configurations (flat plate, pin fins, corrugated fins, and corrugated-perforated fins) and different inlet air flow percentages (% inlet air flow through lower port/upper port of 100/0, 66.7/33.3, 33.3/66.7 and 0/100) on the performance of double pass SAH. The results revealed that efficiency of SAH increased with an increase in the upper inlet air percentage for all studied configurations. The absorber plate with corrugated-perforated pin fins provided an efficiency higher than the other configurations. Aldabbagh et al. [2] presented a comparative study the on thermal efficiency between single and double pass solar air heaters fitted with steel wire mesh under different mass flow rates of air (0.012–0.038 kg/s). It was found out that the double pass SAH presented a thermal performance higher than the single pass by about 34–45%. El-Sebaii et al. [3] investigated the thermal performance of a double pass SAH fitted with fins and V-corrugated plates. The influence of mass flow rates of air on thermal output, pressure drop and thermohydraulic efficiency were also studied. It was found that the efficiency of V-corrugated plate was higher than a pass-finned plate by around 9–12%. The effect of air flow rate (0.01, 0.023 and 0.03 kg/s) on thermal and thermohydraulic efficiencies of a double pass SAH converging finned wire mesh packed bed was investigated by Singh et al. [4]. It was reported that the maximum thermohydraulic efficiency was exhibited by the SAH of 80% at a mass flow rate of about 0.023 kg/s. Kumar and Rosen [5] designed a photovoltaic/thermal (PV/T) SAH with a double pass configuration. The double pass SAH used in this study were both with and without vertical fins in the lower channel. The study concluded that the SAH with fins on the absorber surface in the lower air channel improved the electrical, thermal and total equivalent thermal efficiencies. Singh [6] conducted an experimental study on the effect of various parameters (amplitude, hydraulic diameter, porosity, wavelength and mass flow rate) on the thermal performance of SAH packed with a porous serpentine wavy wiremesh and, used numerical studies to evaluate the most suitable parameters. Mahmood et al. [7] demonstrated the increased efficiency of the double pass in comparison to single pass for SAH with transverse fins and used wire mesh as an absorber plate. The heat transfer performance of single and double pass SAH with regular glazing and perforated cover (cover with holes) was studied experimentally by Nowzari et al. [8]. They presented the effects of the center-to-center distance between holes (3 and 6 cm) and the different mass flow rates of air (0.011–0.037 kg/s). It was found that, the double pass SAH in both normal glazing and perforated cover showed a thermal efficiency higher than the single pass. The configuration of double pass SAH of literature review above is demonstrated in Table 1. In addition to the aforementioned research, there are many studies [9], [10], [11] which reported that the double pass SAH provided a better thermal efficiency than the single pass SAH.

Another method to improve the heat transfer process of SAH is the use of thermal storage materials. Abuşka et al. [12] presented SAH using phase change material (PCM) as latent heat storage. The effect of honeycomb as an internal fin structure and the air flow rate on the thermal performance of SAH was studied. Kabeel et al. [13] improved the thermal efficiency of SAH with finned absorber plate (FPSAH) by using paraffin wax as PCM. It was reported that applying PCM in the FPSAH increased the efficiency by around 10.8–13.6%. Ghiami and Ghiami [14] presented the use paraffin wax as the latent storage for the solar collector, same as Kabeel et al. [13], although they analyzed and showed the results in terms of the exergy and energy efficiencies of a SAH with baffles. Metallic finned tubes that contained PCM (RT44HC and RT18HC) were used to increase energy storage and the rate of heat transfer in the double pass SAH was investigated by Sajawal et al. [15]. It was revealed that the overall average thermal efficiencies of the SAH with metallic finned tubes was higher than the conventional SAH by about 15–18% depending on the finned tube setting configurations. Dinesh and Bhattacharya [16] numerically studied the energy storage characteristics in the energy storage systems using PCM with metal foam in different geometry. Baig and Ali [17] comparatively studied the efficiency and storage time of a simple flat plate SAH and the SAH equipped with 2 and 4 copper tubes filled with aluminum foam and molten paraffin wax.

Passive methods are one of the most extensive heat transfer enhancement techniques due to their convenient operation and the fact that they do not require the support of external energy. One particular passive method technique is the use of roughened surfaces [18,19], by either roughening the base surface (using grooved/dimpled surface) or placing a roughness adjacent to the surface (using ribs/baffles/fin turbulators). Huang et al. [20] carried out numerical analysis and experimental study on the pressure drop characteristics and heat transfer enhancement of pulsatile flow in a grooved channel with five different groove lengths (4, 6, 8, 10 and 12 mm) in the laminar region. Eiamsa-ard and Promvonge [21] investigated numerically the thermal performance in a channel with periodic transverse grooves with various groove-width to channel-height ratios. Predictive accuracy of the turbulence models; the standard k−ε model, the RNG k−ε model, the standard k−ω model and the SST k−ω model, and a comparison to the previous measured data were presented. Heat transfer performances and flow characteristics in channels with the conventional cylindrical grooves and rounded transitions cylindrical grooves were analyzed and compared numerically by Liu et al. [22]. Layek et al. [23] experimented on heat transfer and fluid flow characteristics in a rectangular duct with transverse chamfered rib and grooved roughness. The effect roughness pitch, chamfer angle, relative groove position and relative roughness height were presented. Eiamsa-ard and Promvonge [24] examined the friction characteristics and forced convection heat transfer in a rectangular duct of turbulent air flow. Three different types of rib-groove arrangements and three pitch ratios were examined. The results showed that the duct with rectangular-rib and triangular-groove arrangement provided heat transfer rate and friction factor higher than the others. The effect of relative roughness pitch, relative roughness height and groove position to pitch ratio on the heat transfer and friction factor in the rib-grooved artificially roughened channel was introduced by Jaurker et al. [25]. Skullong et al. [26] presented the thermal enhancement factor of grooved channel fitted with 45° triangular wavy ribs. The effect of different rib-pitch to channel-height ratios and rib arrangements were examined. Mohammed et al. [27] presented a numerical investigation on hydraulic characteristics and thermal enhancement of turbulent nanofluids flow (Al₂O₃, CuO, SiO₂, and ZnO) in a nine different rib–groove channel configuration. Nanoparticles with different volume fractions and different nanoparticle diameters used in three different base fluid (water, engine oil and glycerin) were reported. Al-Shamani et al. [28] presented a study of heat transfer characteristics in a four different trapezoidal rib-groove channel under conditions similar to previous work [27]. Xinyi and Dongsheng [29] experimentally and numerically studied the enhancement of heat transfer of water flow in discontinuous crossed ribs-grooves channel and reported that the ribbed-grooved channel provided a thermal performance higher than the ribbed channel by about 10–13.6%. Bi et al. [30] numerically studied cooling heat transfer and fluid flow in a mini-channel with cylindrical grooves, dimples and low fins. It was found that the dimpled surface showed the highest average heat transfer performance. The effect of a narrow channel with discrete grooved surface on heat transfer enhancement and pressure drop was investigated by Tang et al. [31]. They reported on the effect of turbulence models, groove configurations, longitudinal and transversal groove-pitch ratios, groove-depth ratios and groove-inclination angles. Lorenzini-Gutierrez et al. [32] investigated thermal and hydraulic behavior in channel inserted with periodic block heaters and curved flow deflectors. The effect of different deflector positions, deflector radius and channel height on the heat transfer enhancement were reported. Yang et al. [33] numerically determined the thermal performance of nanofluids in a three-dimensional rectangular rib-grooved channel and presented the optimal parameters by using the response surface methodology and the genetic algorithm method. Liu et al. [34] studied jet impingement heat transfer on the longitudinal and transverse grooves target surfaces aligned inline and staggered pattern with jet hole. Heat transfer on a target surface were measured using the transient liquid crystal technique. Skullong et al. [35,36] also conducted experimental thermal performances in a wavy grooved SAH channel combined with perforated–delta wings [35] and trapezoidal-winglet pairs [36]. It was reported that a wavy grooved SAH with perforated–delta wings and trapezoidal-winglet pairs yielded higher heat transfer performances by around 45.5% and 43% above only grooved SAH, respectively. Thermal performance assessment in chamfered V-grooved channel with punched holes V-ribs were introduced by Promvonge and Skullong [37]. The effect of relative rib-pitches and inclination angles of rib punched holes were presented. It was found that the thermal performance of the channel with combined V-grooves and V-rib was higher than the grooved alone by about 56–77%. The configurations of the artificially roughened channels of the above literature review are demonstrated in Table 2.

Artificial neural networks (ANN) methodology has been widely used for heat transfer analysis of thermal process [38], [39], [40], [41], [42], [43], [44], [45], [46], [47], [48]. Esfe [39] designed and proposed ANN to predict relative pressure drop and relative Nusselt number of Ag/water nanofluids in a double tube heat exchanger. The result showed high accuracy regression coefficients (R²) for the relative Nusselt number of 99.76% and relative pressure drop of 99.54%. Beigzadeh and Rahimi [40] developed ANN models for predicting the heat transfer (Nu) and friction factor (f) in helically coiled tubes. The predicted results for tested data were compared with those obtained from experiments which showed a high performance of (R²), about 99.88% and 99.95% for Nu and f, respectively. Verma et al. [41] proposed ANN models for predicting heat transfer coefficient (U), Nusselt number (Nu) and Reynolds number (Re) of double pipe heat exchanger with corrugated/non-corrugated inner tubes. The R² for Re, U and Nu were 0.99999, 0.999997 and 0.999993, respectively. Maddah et al. [42] showed a neural network for predicting thermal performance of heat pipe heat exchanger using CuO/water nanofluid. The results presented the network model with an accuracy to predict the heat transfer coefficient of 0.9938. Ghritlahre and Prasad [43] used ANN technique to predict the heat transfer of different roughened absorber plates of SAH and compared it with experimental data. The values of coefficient of determination (R²) were found to be 0.99532 and 0.99791 for training and testing stage, respectively. ANN with different training algorithms were used to analyze the heat transfer and pressure drop of pulsating nanofluids (TiO₂/water) in a spirally coiled tube under magnetic field by Naphon et al. [44]. It was found that the Levenberg-Marquardt backpropagation (LMB) algorithm was the optimal model.

From previous studies, it is evident that the use of double pass solar air heaters can increase heat transfer and thermal efficiency due to the increase of the flow path and the residence time of the working fluid within the SAH. In addition, fins/corrugated plates/wire mesh [1], [2], [3], [4], [5], [6], [7] are applied in the double pass solar air heaters to help generate turbulent air flow inside the air heater, leading to a better heat transfer process. The PCM are contained under the absorber plate of the SAH providing higher thermal efficiency. The thermal energy received by the PCM is stored in terms of latent heat by transition from solid to liquid phase, and heat is released in changing phase by conversion from liquid to solid phase. The fin turbulator is still applied to improve the heat transfer together with the PCM in the SAH [13,14]. Finally, several rough surface configurations (grooved surface only and grooved surface with turbulators) at various parameters have been used in order to enhance the thermal performance in channel heat exchangers or SAH. The grooved surface and turbulators create re-circulating flow or turbulent air flow which results in the destruction of thermal boundary layer near the surface, a phenomenon which can help further increase heat transfer. In addition, it has been found that there are applications of ANN methodology with different algorithms, activation functions, numbers of hidden layer or numbers of neuron in hidden layer for analyzing the thermal process. The network provides high precision for prediction with adaptability for learning, stability and flexibility in operations and response to complicated and large scale issues [46,48]. The ANN network has no restrictions on input variables, which can be both numerical data and non-numerical data.

Therefore, the expectance of this work is to examine the effect of a roughened surface on thermal performance assessment of channel solar air heater together with the development of the neuron network for prediction. The aims of the present work are as follows:

• To experimentally study the influence of direction, inclination angles and relative depth of the novel isosceles triangular grooves on convective heat transfer, friction factor and thermal performance enhancement factor. The isosceles triangular grooves is a new design of groove turbulator developed from rectangular grooves [20,21,24,26,27,32,33,35] together with triangular grooves [[23], [24], [25],27] as presented in Fig. 1. The walls on both sides of the rectangular grooves are vertically straight, while the triangular groove walls are sloped on both sides. Therefore, the isosceles triangular groove of this research is designed to have a vertical straight wall on one side and a slope on the other to create a different fluid flow behavior.

• Comparatively study the effect of isosceles triangular grooves only with a combination of grooves and transverse baffles on convective heat transfer, friction factor and thermal performance enhancement factor. According to previous studies [26,35,36], it is apparent that the combination of winglets/ribs turbulator together with grooved surfaces provide higher heat transfer and thermal performance than using a grooved surface only, even if it increases the friction factor. Therefore, the idea of combined turbulator between the baffles and grooves are presented after deliberating the influence of some parameters of the novel isosceles triangular grooves to enhancing the heat transfer process.

• Use of the artificial neural network (ANN) model with different algorithms to predict the thermal enhancement factor, convective heat transfer and friction factor of different channel geometries based on the experimental data.