Solar Driven Agricultural Greenhouse Integrated with Desalination System; Energy-Water-Food Nexus


 This study presents the effective performance of a sustainable solar driven agricultural greenhouse (GH) self-reliant of energy and irrigation water via desalination. The GH is furnished with infrastructures such as; (i) - an inlet condenser for cool air exchanger and partial water production, (ii) - an internal cavity for crop production (iii) - roof transparent solar distillers (TSD) for solar desalination and partial shading and (iv)- a thermal chimney for natural air ventilation. A mathematical model is developed to predict the performance of the sustainable GH system. A coupled approach of MATLAB/Simulink and computational fluid dynamics (CFD) based on three simulation models were used: solar radiation, thermal energy balance and CFD model. Two parametric studies were carried out. The first one analyzed the effects of different air velocity on the system thermal performance and natural ventilation rate. The second study assessed the effects of different covering material on the transmitted solar radiation. Results from the model shows that 8.5 MJ/m2.day of total solar radiation is transmitted into the GH. The greenhouse air temperature is lowered by 5 °C and humidified by 20%, to satisfy the required conditions necessary for plant growth. Maximum water yield of 11.5 L/m2.day was obtained, aided by the addition of Al-metal net. Additionally, 2.6 kWh/m2.day of power is consumed by the air-cooling condenser. At air velocity of 0.3 m/s, there is a natural tendency of air to flow by draft, due to air temperature difference of up to 4 °C. Furthermore, glass and EVA cover materials transmit 52 and 48% of solar radiation into the GH respectively. The proposed system will enable the parallel production of water and food and enhance economical plant productivity.

Two parametric studies were carried out. The first one analyzed the effects of different 23 air velocity on the system thermal performance and natural ventilation rate. The second 24 study assessed the effects of different covering material on the transmitted solar 25 radiation. Results from the model shows that 8.5 MJ/m 2 .day of total solar radiation is 26 transmitted into the GH. The greenhouse air temperature is lowered by 5°C and 27 humidified by 20%, to satisfy the required conditions necessary for plant growth. 28 Maximum water yield of 11.5 L/m 2 .day was obtained, aided by the addition of Al-metal 29 12 The conceptual configuration of the GH is demonstrated in Fig. 1. Briefly, the GH 174 system consist of an interior glass covered plant cavity, a chilled water condenser at the 175 entry of the GH cavity, a vertical and inclined riser, set of transparent solar distillers 176 placed on the risers, a thermal chimney and a vertical down comer channel. The GH 177 structure was designed facing the south in order to receive maximum solar radiation 178 whiles minimizing heat loss. Geometrical dimension of the GH is giving in Table 1. 179 Ventilation fresh air (a1) at ambient condition enters the GH through an inlet 180 condenser. In the condenser, the air is partially cooled and humidified, depending on 181 the ambient temperature and relative humidity. The cool air (a2) then mixes with the 182 bypassing air and moves through the interior GH cavity (ai), where it gains heat and 183 water vapor through convection and plant transpiration. The heated air then leaves the 184 GH cavity through an outlet condenser (a3), where it gets cooled to saturation condition 185 and condense the water derived from plants transpiration. The cool air (a4) then mixes 186 with the bypassing air and flow through the vertical (a5) and inclined risers (a6), to gain 187 extra heat from solar stills and glass covers. Depending on the seasonal condition, the 188 heated air then either leaves the GH to the atmosphere through a thermal chimney or 189 partially recirculated from the down comer (a7) into the GH cavity to mix with the fresh 190 13 ambient air. In the current paper, the outlet condenser is not active (switched off) and 191 no air recirculation through the down comer. 192 The integrated transparent solar distiller is illustrated in Fig. 2. The solar distillers 193 are oriented to face the south with a cover tilt angle of 30°, which is equal to the solar 194 latitude angle of the target study area [28]. The solar distillers are designed to be 195 transparent to enable solar light reach plant region for photosynthetic process. Further,196 the solar distillers will serve as partial shading thereby reducing the GH cooling load 197 and utilize the excess solar radiation (above plant need) for water desalination. The 198 solar still desalination works based on evaporation and condensation [2]. Solar radiation 199 passes through the glass cover and converts the saline water into vapor. The water vapor 200 flows up and condenses on the inner surface of the inclined cover due to temperature 201 differences. The purely desalinated water is then collected via the distillation trough 202 [33]. Due to the low gained output ratio (GOR) of solar still desalinations, an aluminum 203 (Al) metal net is added to the base of the distillers to enhance productivity by increasing 204 the absorption of solar radiation. A parametric study is performed to determine the 205 effect of net area ratio on the total solar radiation transmitted into the GH cavity.
Where is the incident global solar radiation, is the direct beam radiation. 233 is the diffused radiation.
is the reflected solar radiation. 234 The zenith angle ( ) on an inclined surface of the GH is calculated according to Eqn.4, 235 To calculate the total solar radiation received on the GH and TSS components, the Eqn.

Thermal Model 247
To predict the transient surface temperature of the GH cover components, heat air 248 exchange, condensed water from moist air condenser, yield produced from the TSS and 249 (2) Air is assumed transparent for long and short-wave radiation and its absorptivity is 254 For condensed water from the condenser; 270 From the TSD, the produced distillate is calculated as follows; 272 The electrical power is calculated as follows; 274 The input data for the thermal model are the calculated solar radiation and the climate 276 metrological data of temperature, relative humidity and wind speed.

Governing Equations 283
To study the internal climatic condition of the GH cavity, heat transfer processes 284 are solved using the CFD code (ANSYS Fluent). The flow inside the GH was 285 considered as two-dimensional, an-isothermal, fully turbulent and incompressible flow.
Where , and ℎ represent the total fluid energy, thermal conductivity and and 297 species sensible enthalphy respectively. 298 The species transport equation and conservation equation for air and water vapor 299 fraction was activated using the mass fraction of H2O (w/w) described by equation 14 300 [38]. 301 Where 2 represent water vapor added or removed from the air due to condensation 303 or evaporation. The constant 2 is the diffusion coefficient of water vapor into air 304 which is equal to 2.88x 10 -5 ; is the turbulent Schmidt number which is equal to 305 0.7. 306 To account for turbulences, the standard k-ε model with wall function was used [39]. 307 This model solves for turbulent kinetic (k) and the rate of dissipation of energy (ε) in 308 20 unit volume and time. This model has been used in GH CFD simulations with success, 309 providing good agreement with experimental results [40]. The equations for k and ε are 310 giving below, respectively. 311 From the equations, and represent the generated turbulence kinetic energy due 314 to mean velocity gradients and buoyancy respectively, represents the dilation 315 fluctuation contribution in compressible turbulences. The constants are giving below 316 The Darcy law is giving by; 322 Where and 2 represent the permeability and inertial resistance factors of the 324 porous medium, whose values are chosen to represent the crop under consideration. 325

Geometry creation and grid generation 326 21
The 2D computational domain was created using ANSYS Design Modeler 18.2. 327 The geometrical characteristics of the GH were assigned according to what is presented 328 in Table 1. A 2D structured mesh was generated using Ansys Mesh workbench, which 329 consist of approximately 71,000 cells. A structured Cartesian mesh was chosen to limit 330 the numerical diffusion of errors and facilitate calculation convergence.  properties are detailed in Table 3. Thermo-physical properties of other materials like 359 glass, water and ground are giving in

Model Validation 367
To validate this model, the interior GH temperature and relative humidity is firstly 368 compared between simulation and published data of [27] for the ambient condition of 369 21 st June. Fig. 4 shows the simulation and validation data of the average air temperature 370 and relative humidity inside the GH cavity. The trend of the air temperature (Fig. 4a) 371 shows close agreement especially during solar hours, the difference between the 372 simulation and validation values was always less than 1°C. Regarding the relative 373 humidity (Fig. 4b), the agreement was particularly good during the solar hours, the 374 maximum difference between both values occurred after midnight, when the validation 375 values dropped more quickly than those predicted by the CFD model. In all cases, the 376 maximum relative humidity difference was less than 7 %. A good comparison between 377 24 the two values suggest that the simulation of temperature and relative humidity 378 propagation inside the GH interior is successful. Therefore, in view of the validation 379 results of the model, the CFD model proved to be a satisfactory predictive model that 380 could be used to predict the temperature and airflow distributions in GHs. Hence, the 381 CFD can therefore, be a very useful tool in the study of the internal microclimatic 382 conditions of the GH system. 383

Effect of TSD on transmitted solar radiation 384
The key for designing transparent GHs is to get enough solar radiation into the GH. 385 The threshold of solar intensity required for plants photosynthetic process is 8.5 386 MJ/m 2 .day [46,47]. During the summer, excess solar radiation about twice the plant 387 need is available. To utilize the excess solar intensity, this study integrates transparent 388 solar distillers into GH roof. To improve productivity of the distillers, an Al-metal net 389 were added to the base of the solar distillers to increase solar absorptivity and decreases 390 transmissivity into the GH. Fig. 5 shows the effect of the Al-metal net on the transmitted 391 solar radiation and the daily variation of the direct solar radiation reaching the GH. 392 Firstly, we study the effect of the Al-metal net area ratio to the solar still base. In Fig.  393 5a, when the net area ratio is zero (i.e. no metal net added), 14 MJ/m 2 .day of total solar 394 25 radiation is available during the day. By increasing the net-area ratio, the total solar 395 radiation transmitted into the GH decreases, reaching the required threshold of 8.5 396 MJ/m 2 .day at a net area ratio of 0.75. Fig. 5b present hourly variation of the transmitted 397 solar radiation into the GH through each cover, for the maximum radiation day of 21 st 398 June, with glass cover material. The solar intensity varied in a sinusoidal way with total 399 available direct solar radiation of 8.5 MJ/m 2 .day, the maximum value of 250 W/m 2 400 recorded during solar hours. The solar distillers reduce the transmitted solar radiation 401 by ~50%. These solar intensity data were then used to compute the temperature on 402 individual roof and wall of the GH using the standard ASHRAE model. 403

GH system operational performance 404
The operational performance of the GH in terms of air temperature, relative 405 humidity, freshwater production and power consumption of the air-cooling condenser 406 are analyzed. The evolution of these factors throughout the day for the examined air 407 velocity are presented. 408 all cases, the GH air temperature is lower than the ambient temperature because of the 411 26 effect of condenser air-cooling, and the solar distillers serving as a partial shading. 412 While the ambient air temperature varies between 24°C to 31°C, the GH (cavity) 413 average air temperature varied from 20°C to 26°C. During this day, as the sun rose 414 around 6:00 AM, GH temperature began to rise until it reached 26°C. As the sun set, 415 around 6:00 PM, the solar radiation intensity and ambient temperature dropped, the GH 416 air temperature also dropped accordingly to a value below 22°C. Despite the change in 417 air velocity, the GH air temperature were similar, satisfying the required condition 418 needed for plant growth. 419 In Fig. 7, the average air relative humidity inside the GH cavity and the ambient 420 for the four examined air velocities are giving. The GH air relative humidity is 15 to 421 20% higher than that of the ambient. This is because the air gets humidified in the 422 condenser and from plant transpiration. The ambient relative humidity varies between 423 45% and 80%, in the GH it varies between 60% and 90%. The relative humidity obeys 424 the sinusoidal trend with its minimum value at midday. Favorably, lower relative 425 humidity at mid-day aid the air-cooling condenser in reducing the high ambient air 426 temperature. The difference of the average air relative humidity among the examined 427 air velocities are not important, since the mechanism of vapor transfer remains the same. In order to assess the efficiency of each examined air flow rate, distributions and 457 profiles for air velocity, temperature and humidity for the mid-day at a flow rate of 0.5 458 m/s are presented. Fig. 10 shows the CFD simulation results for GH air temperature, 459 velocity vectors and relative humidity contours at 12 P.M of 21 st June with glass cover 460 material. At this time, the ambient air temperature is 304 K, the outside relative 461 humidity is 46 % and the wind speed is 0.5 m/s. 462 29 Fig. 10a indicates the temperature distribution inside the GH cavity. The GH air 463 temperature in the plant region is slightly cooler than the ambient air, with an average 464 GH air temperature of 299 K. This is due to the cooling effect created by the inlet 465 condenser. The CFD model also shows the solar distiller region (roof) of the GH as the 466 hottest surface, which intercept majority of the solar radiation. The GH air temperature 467 is uniform for most of the GH cross section, with the left region of the cross section 468 slightly warmer. This is due to the air heating by convection and plant transpiration as 469 it passes through the GH cavity, and transferring the heat to the GH roof, which 470 becomes warmer than other regions in the GH. 471 Fig. 10(b) and (c) show the relative humidity and water mass fraction contours 472 predicted by the CFD model at 12 P.M for the same set of boundary conditions. Fig.  473 10b shows an increase in humidity immediately after the condenser, since the air gets 474 humidified in the condenser unit. As the air flows in the plant region, the air becomes 475 warmer and humidified from plant transpiration. The average air relative humidity in 476 the plant region is 62 %, which is higher than the ambient RH of 45%. The water mass 477 fraction (Fig. 10c) also increases in a stream wise direction in the plant zone, due to 478 water vapor generated inside the plant region. The water mass fraction was slightly 479 30 higher in areas with higher air temperatures (Fig. 10a), since hot air holds more water 480 vapor. 481

Natural ventilation performance 489
This sustainable GH will aim to rely on natural ventilation driven by two 490 mechanisms, namely the wind induced pressure field around the GH and the buoyancy 491 force induced by the warmer and more humid air in the GH riser. Natural ventilation is 492 effective if the outside temperature is low compared to GH riser air temperature, where 493 the difference in density between the inside and outside air causes natural draft [49]. 494 Air temperature and humidity differences between the inside and outside of a GH 495 produce forces that drive flow. The natural tendency for hot and humid air to rise and 496 accumulate towards the upper part of a space leads to stable stratification, and this has 497 a significant influence on the flow patterns within the GH. The determining factor in 498 the form of the vertical stratification is the location of the openings. The warm and 499 humid air will flow out over the upper area of the opening and the cool air will enter 500 through the lower area of the opening. 501 Buoyancy driven ventilation is significant only at low wind speeds. At low wind 502 speed, the buoyancy effect is the main driving force of ventilation in a GH with plant. 503 To evaluate the natural ventilation of this GH, the temperature field on the inclined riser 504 generated by the four different air velocities is shown in Fig.11. During the day, the air 505 temperature at the inclined riser (T_a6) is higher than the ambient air temperature at 506 wind speed of 0.3 and 0.4 m/s. At speed of 0.3 m/s, the air temperature at the inclined 507 riser is 4 °C higher than the ambient. This is due to the absorbed solar radiations by the 508 solar stills and dissipated subsequently as latent and sensible heat inside the GH. During 509 the non-solar hours, the outside air is warmer than the inside GH due to the cooling 510 effect created by the condenser. To achieve natural ventilation at night, the condenser 511 may have to be turned OFF. Hence, natural ventilation could be maintained in the GH 512 especially during the daytime. This conditions also satisfy the air exchange rate of 0.7 513 32 AC/min. 514

Covering material parametric study 515
GH sustainability depends largely on cover material, as it influences the GH 516 microclimate and protect plants from adverse weather conditions. The evolution of 517 basic factors such as transmitted solar radiation, interior air temperature and water 518 production for each examined cover material are presented. Fig. 12a present the 519 instantaneous total solar radiation transmitted into the GH cavity for the examined 520 cover materials for the maximum radiation day of 21 st June. The highest transmitted 521 solar radiation is achieved with the glass and EVA cover, with 52 and 48 % solar 522 radiation transmitted into the GH respectively. The PVC cover allows lower 523 transmittance (~ 30 %) due to its high absorptivity, which may become less in winter 524 season. The air temperature and relative humidity differences inside the GH cavity for 525 the studied cover materials are not significant (data not shown), because the main 526 mechanism of heat transfer is from convection of the entering air stream and the effect 527 of air-cooling condenser. For water production, Fig. 12b shows the accumulated water 528 production for the examined cover materials. Less water is expected to be produced 529 from the transparent solar distillers for PVC and P.E covers due to high material 530 33 absorptivity and low transmitted solar radiation. Although sufficient amount of fresh 531 water is expected to be produce from the air-cooling condenser unit. 532

Conclusions 533
In this study, a newly developed solar driven agricultural greenhouse for air cooling, 534 and solar energy application for irrigation water production is presented. The potentials 535 of using an integrated TSD-GH system was investigated, combining water production 536 and agricultural farming in the same area. The potential harmony between solar 537 radiation availability and the demand for irrigation water is illustrated. A new coupled 538 approach of MATLAB/Simulink and CFD modelling methodology was developed to 539 predict the performance of the sustainable GH system. This approach takes advantage 540 of the strength of each tool to study, with accuracy, the full functionality and 541 requirement needed to operate a sustainable GH integrated with solar desalination 542 system. The calculations are implemented for the external environmental condition of 543 21 st June, representing the maximum radiation day of Borg-El-Arab, Egypt. 544 In general, the excess solar radiation was utilized, allowing 8.5 MJ/m 2 .day of solar 545 intensity threshold for plant photosynthetic process. An AL-metal net was added to trap 546 the excess solar radiation for enhanced desalination water production. A parametric 547 34 study was carried out to investigate the effect of different air flow velocity and covering 548 materials. Increasing the air velocity by 0.1 m/s increases the water production by 0.75 549 L/m 2 .day, and power consumption by 0.43 kWh/m 2 .day. At wind speed of 0.3 m/s, air 550 temperature at the inclined riser is 4 °C higher than the ambient condition. At this 551 condition, there is a tendency of the air to flow by natural draft, hence reducing the fan 552 power. At air velocity of 0.5 m/s (base line), the average air temperature and relative 553 humidity in the GH cavity were 26 °C and 62 % respectively, at 12:00 of the maximum 554 radiation day. For different covering materials, the highest transmitted solar radiation 555 is achieved with glass and EVA cover, with 52 and 48 % solar radiation transmitted into 556 the GH respectively. 557 In general, to overcome the MENA-GCC drought, this GH is designed to reach a 558 water autonomous irrigation situation by investigating the potential of providing 559 conventional and nonconventional water sources for a water efficient GH. This self-560 sustainable GH will allow sufficient cooling to enable crop production in the extreme 561 summer climate of the Mediterranean regions. The proposed system will also allow 562 parallel production of food and water and enhance plant productivity. The desalination 563 components may perhaps enable regional progress in GH production into arid and 564 35 MENA-GCC regions. In the next step of this research, a comprehensive economic 565 analysis and development of an experimental model are proposed. The economic 566 analysis will include the capital and operating cost of the integrated system. The major 567 capital cost will include building construction, cooling system, transparent solar 568 distillers, irrigation system, water storage tank, fan ventilation systems, lightening and 569 other machinery and equipment. Operational cost will consist of labor cost, material 570 input, marketing, fertilizer and repair and maintenance cost. The experimental 571 validation will be built along the shores of the Mediterranean Sea in Borg Al-Arab 572 Egypt, using cost effective and sustainable materials. 573 Declarations 574

Availability of data and materials 575
All data generated or analyzed during this study are available from the corresponding 576 author on reasonable request. 577