Environmental monitoring of a smart greenhouse powered by a photovoltaic cooling system

Abstract

In hot and arid environments such as prevailing in Madinah city (KSA), many plantations need an appropriate environment climate. So, a solar cooling system inside greenhouses is necessary to get successful crops. For this, environmental monitoring of a smart solar cooling system is provided to reach high crops by supervising in real time the appropriate environment for the smart greenhouse. Recently, the integration of clean energy sources and the Internet of Things (IoT) has a primordial role in the creation of smart greenhouses. The principal objective of this present research is to design a smart greenhouse prototype based on a photovoltaic (PV) system. This allows for powering the different parts of the greenhouse such as DC-air conditioning, fans, water pumps and electronic circuits. The results reveal the potential and the necessity of such smart monitoring solution for remote greenhouses located in arid and hot locations.

KEYWORDS:

• Smart greenhouse
• environmental monitoring
• internet of things
• photovoltaic cooling system

1. Introduction

An adequate environment is necessary inside greenhouses for getting the best corps and avoiding plant diseases due to the harsh climatic environment such as long-time rainfall or high temperature [1]. The greenhouse allows for the protection of plants against the external influence of weather conditions such as sand storms, very hot temperatures and high solar irradiation.

In Saudi Arabia, many scientific research studies were developed to evaluate the performance of greenhouse farming for some specialized products in the last decades [2]. Fertile land and water are the principal parameters in the agriculture sector, and Saudi Arabia prevents importation for its domestic needs by 2050 [3]. The traditional greenhouses were initially used for some plantations which need high temperatures, while the outside temperature is low and not adequate for such types of crops in cold weather. So, many parameters should be controlled inside the greenhouse such as variation in temperatures, high humidity, CO₂ concentration, water evaporation and condensation of water inside the greenhouse. Research work showed that it is important to keep an environment that minimizes energy consumption [4]. Recently, a review of greenhouse climate models has been done [5]. In this review, analysis and identification of different greenhouse models have been presented. It was concluded that each model depends on its objectives and complexity. Thus, greenhouses are used to make suitable environmental conditions for getting a good production of the plantation. The use of solar energy for powering greenhouses in desert climates is one of the interesting applications of renewable energy sources and carries great significance. Accomplishing a favourable environment inside greenhouses under hot climates and summer conditions, such as KSA, is one of the key challenges still inventors and cultivators are still facing [6]. It is well known that crop productivity is related mainly to environmental conditions and specifically to the performance of the greenhouse system. In greenhouse applications, PV systems are used to supply the cooling, ventilation and lighting systems [7]. Thus, PV can contribute to sheltering the required energy of greenhouses in irrigation (pumping systems), heating (night), ventilation (hot climate), lighting (cloudy regions) and other energy requirements [8]. In addition, some research focussed on the use of soil sensing in greenhouses [9]. Soil sensing allows the best climate and environment inside the greenhouse to enhance the efficiency of agricultural production. Many research works have been done in the field of greenhouse climate control problems [10,11]. The use of fuel to run the cooling, heating and lighting systems is not economically practical due to increasing fuel and electricity costs, furthermore, greenhouses are not always installed nearby to the electrical grid, especially in remote areas. For complex geographical topology and high-cost implementation from the grid, electricity generation from PV systems is a suitable solution [12].

A model greenhouse has been elaborated to control four robots using a wireless system communication [13].

Many research works, about using the Internet of Things (IoT), have been done in many applications [14] such as IoT-based smart monitoring for greenhouse [15] and developing adequate environments [16,17]. An automatic system, based on the IoT technique, for monitoring crops inside a greenhouse has been developed [18]. Energy management using a control for a smart greenhouse integrating a microgrid has been investigated [19]. Advanced energy management using model predictive control has been developed to optimize and control the global internal environment and crop growth. An investigation of environmental conditions has been done [20]. The effect of wind speed and humidity was studied on the soiling losses for the PV panel. An IoT-based system to monitor the environmental factors inside a greenhouse has been designed to give information to floral farmers [21]. Another research work has been developed about wireless sensor implementation inside two different tropical greenhouses [22].

The application of IoT in agronomy such as greenhouses has been developed by showing different advantages and inconveniences in a detailed review paper [17]. One of the important applications in renewable energy systems is to supply greenhouses in isolated sites [23]. Also, using the IoT technique by applying solar energy has been illustrated recently [24]. To predict crop diseases inside the greenhouse, identification and classification have been tested using image processing and IoT technique [25]. Also, to classify plant diseases, IoT and a machine learning algorithm have been used [26]. Another study in agricultural applications has been developed using an expert system to control the epidemic disease [27]. IoT technique has also been applied for the identification [28], detection [29] and quantification of tomato plant diseases [30]. The identification and classification of diseases have been elaborated using the acquired images of the plants via the developed web page [31]. In arid and desert regions, greenhouses are useful to make an adequate environment to allow a good production and quality of crops which need regulated temperature and humidity [32].

The principal aim of this research is to design a smart solar cooling greenhouse, based on monitored environmental parameters. A photovoltaic system is installed and used to power the main components of the greenhouses (Air conditioning, fans, water pumps and the electronic board). The IoT technique is used to upload measured data into a Webpage for remote monitoring.

The proposed smart greenhouse prototype will help users to

• collect and analyze data online.

• make adequate environments inside the greenhouse.

• inform remotely about greenhouse behaviours by sending notifications to users.

• The main novelties are given below:

• Realization of an adequate environmental smart monitoring system for greenhouses.

• Setting up a friendly Webpage for data collection and visualization of a greenhouse located in an arid region.

• Using the IoT technique to monitor and supervise the greenhouse remotely.

The different sections of the manuscript are given as follows: The description of the designed system is given in Section 2. In Section 3, the developed monitoring of solar cooling greenhouse is presented. Section 4 presents the results and discussions. Section 5 concludes about this work by giving some perspective.

2. System description and design

A block diagram of the suggested smart greenhouse is presented in Figure 1. It consists mainly of a photovoltaic system (1 kW), a cooling system (24 DC supply), a water pump (24 V), a tank (1 m³), sensors, a monitoring system and a weather station. The monitored parameters are temperature inside the greenhouse (T), soil moisture (Sm) and relative humidity (Rh). Other parameters are also recorded by the weather station such as global solar irradiance (G), ambient temperature (Ta) and Wind speed (Ws).

Figure 1. The basic structure of the proposed smart greenhouse prototype.

2.1. Data environment in the studied location

Madinah City is considered an arid location which is characterized by a big potential for solar energy [33], as indicated in Figure 2, and also characterized by very hot temperatures in the summer months, as shown in Figure 3.

Figure 2. Evolution of global daily solar irradiation (H_G) and corresponding extraterrestrial irradiation (H₀).

Figure 3. The ambient temperature at Madinah city during July 2022.

The data were collected from January 2020 until December 2022. Figure 3 shows that almost ambient temperatures at Madinah City, during July 2022 are in the interval [40–50]°C. Sometimes, during the same period, the ambient temperature reached 53°C. Due to the high temperature, many plantations need an appropriate environment climate. So, a solar cooling system inside greenhouses is necessary to get successful crops.

2.2. PV cooling system

The photovoltaic (PV) system applied in this work for cooling the greenhouse is presented in Figure 4. It consists of solar panels (1 kW), solar charger controllers, batteries bank and a splitter cooling system (24 VDC). The sizing and performance of the installed PV system have been respected by using a maximum power point algorithm [34].

Figure 4. The photovoltaic system applied to power the DC air cooler inside the greenhouse.

The produced power by the PV system allows supplying the Air-conditioning (cooling the greenhouse), the fans (air ventilation inside the greenhouse), the water pump (watering), LED (lighting) and the monitoring system (Data-acquisition equipment).

2.3. The designed prototype

Figure 5 illustrates the different parts inside the greenhouse (window, pumps, sensors, Fan and DC air conditioning).

Figure. 5. Different parts inside the greenhouse (solar regulator, window, pumps, sensors, fan and DC air conditioning) and general view of the greenhouse with solar panels.

2.4. Management process inside the greenhouse

To define the threshold values many attempts have been carried out, these values depend mainly on the region specification, the planted vegetables and the accuracy measurement are as follows:

To measure the temperature and the relative humidity, the DHT11 has been used. It can measure humidity from 20% to 90% and temperature from 0 to 50 degrees Celsius:

• Operating voltage: 3–5 V

• Temperature range: 0–50°C/ ± 2°C

• Humidity range: 20–80%/5%

• Soil moisture sensor: 6%.

In this work, tomatoes have been planted and the reference parameters are (normal conditions)

• 20°C < Tref <  28 °C,

• Rh > 45%,

• 25% < Sm <  95%.

The workflow of the greenhouse is summarized in the following steps (control process):

• Step 1: Parameters’ initialization (Tref, Smref and Rhref) and thresholds definition.

• Step 2: Measuring all parameters (T, Rh and Sm).

• Step 3: Compare measured parameters with reference ones for each sensor.

• Step 4: Sending signals to the components (cooling system, fans and valves) by activating the corresponding relay:

  • Water pumps and valves: start watering and irrigation of the plants,

  • Servomotor: open doors for air fresh,

  • Fans: air ventilation,

  • DC-air conditioning,

  • Lighting.

The management process is simple, for example, if the measured temperature is outside of the Tref interval (20°C < Tref < 28°C), a signal is automatically sent to activate the corresponding rely to start the cooling system, to refresh the environment inside the greenhouse. The cooling system can be achieved in the following order:

1. open the doors,

2. turn ON fans,

3. turn ON the air-conditioning.

Regarding the watering and lighting of the greenhouse we follow the same process. Figure 6 shows the flowchart of the management process inside the greenhouse. Measured data are displayed on a digital screen and uploaded on a Webpage (cloud) for remote monitoring using IoT techniques. Thus, IoT techniques are applied principally to control and supervise the system remotely such as lighting, irrigation and cooling. The following pseudo-code is run to set a suitable environment inside the greenhouse:

Step #1: Measure air temperature (T), relative humidity (Rh) and soil moisture (Sm).

Figure 6. Flowchart of the management process inside the greenhouse.

Step #2: Compare the measured T with the reference Tref.

  If T < 20°C and Rh > 45% then open windows or if T > 28°C then open AC (Air Conditioning) and go to step#1.

 Else close the AC and the window, and go to step #1.

 Endif

If Sm < 25% then open the pump, else go to step #1.

Step#3: Display the results.

2.5. Monitoring of the solar cooling greenhouse system

Figure 7 presents the diagram of the developed monitoring system. The management code is implemented into the Arduino Mega 2560 board.

Figure 7. Environmental monitoring of data from the solar greenhouse.

Figure 8 shows the electric connection of different sensors based on the Arduino Mega2560 board. It consists mainly of some sensors (air temperature and relative humidity) and transducers such as relays, LDC displays and electric valves.

Figure 8. Electric connection of different sensors based on Arduino Mega.

The specifications of solar panels and different loads are listed in Table 1.

Table 1. Photovoltaic module and different load specifications.

Solar panels	Specifications
Maximum power/Pmax (W)	300
Maximum power tolerance (%)	+3%
Open circuit voltage (V)	43.2
Short circuit current (A)	8.85
Max power voltage (V)	36
Max power current (A)	8.35
Weight (Kg)	20.5
Dimensions (mm)	1000
Max System voltage (V)	27.00 VDC + - 0.5
DC pump	24 VDC
Hydraulic actuator	Micro-Linear Actuator Motor for the greenhouse (Stroke 100 mm–15 mm/ s-50N)
Relay maxtor	5 v, 10 A
Exhaust fan	Fan 24 VDC

Specification and precision of the used sensors are shown in the Appendix section.

3. Results and discussion

A photo of the designed modern greenhouse is shown in Figure 9. As an example, Table 2 reports measured parameters inside and outside the greenhouse. Where Tin and Rh_in and Sm are the measured parameters inside the greenhouse, while Tout and Rh_out are the measured parameters outside the greenhouse. Thanks to this kind of modern greenhouses, favourable conditions for good crop growth are created. Figure 10. a shows the real screen of the monitoring system (displaying in real time the measured parameters) and Figure 10. b depicts an example of posted data on the web page (Cloud). Thus, users can select historical or actual data for possible performance analysis and supervision of the greenhouse. As shown in Figure 11a, the temperature and relative humidity inside the greenhouse correspond to the adequate environment to plant crops while outside the greenhouse, the temperature is high with low humidity as we can see in Figure 11b,c.

Figure 9. A real photo of the designed greenhouse including a weather station, PV panels, monitoring system and cooling system at Madinah City, KSA.

Figure 10. (a) Monitored parameters by the monitoring system (real screen) (b) posted parameters on the web page (curves).

Figure 11. (a) Evolution of air temperature and relative humidity inside and outside the greenhouse (b). Temperatures versus time (c). Relative humidity versus time.

Table 2. Measured parameters inside and outside the greenhouse, including solar radiation.

Time	Tout (°C)	Tin (°C)	Rhout (%)	Rhin (%)	Sm (%)	G (W/m^2)
08:00	24.1	21.2	36	52	95	150
09:00	27.3	23	34	50	93	310
10:00	31.4	26.2	30	46	90	535
11:00	32	27	28	44	88	750
12:00	34.5	31	24	40	85	931
13:00	36.8	32.6	28	38	70	987
14:00	37.4	30.2	24	34	75	950
15:00	37	32	26	44	77	852
16:00	24.1	31	30	46	80	751
17:00	27.3	29.4	28	48	84	430

Based on the collected parameters, users can check and control remotely the state of the greenhouse, which reduce significantly the efforts as well the maintenance cost. Obtained results indicate clearly the good functionality of the designed smart greenhouse. It should be pointed out that the cooling system is very necessary to keep appropriate environmental conditions for the planted tomatoes in this region, which is characterized by its hot summer climate.

The system started working from April 2021to 27th March 2023. The data collected in the manuscript are from March to May 2022.

The proposed greenhouse could be generalized for large-scale applications. It should be noted that in this prototype we have used a PV array of 600 Wp to supply the greenhouse (size = 3 m by 2.5 m), which is large enough. So, for a large-scale greenhouse, we should first calculate the total consumption, including the consumption of each component and sensor in Wh and especially the capacity of the DC air conditioner. So, the main change is to increase the output power of the PV system which should be adequate to supply all loads inside the greenhouse. In addition, the system voltage could be 24 VDC or 48 VDC because it depends on the power consumption in watts for all loads used.

4. Conclusion and perspectives

In this paper a smart greenhouse powered by a photovoltaic cooling system was designed and verified experimentally. It has been shown that the DC-cooling system contributes well to keeping an adequate environment inside the greenhouse, which significantly increases the crops’ productivity. By using the IoT technique and a friendly webpage users can monitor and check their system easily with less effort. The designed greenhouse uses a free source of electricity (photovoltaic array) which helps to reduce electricity bills and CO₂ emissions. However, for a large-scale greenhouse the cost will be increased and the main components that increase the cost are the DC air conditioners and the photovoltaic generator. The designed monitoring system, including sensors and actuators, does not cost.

We believe that this work will play a very important role in contributing to modern greenhouses. The proposed prototype could be generalized for large-scale farming applications. As perspectives, plant disease detection and diagnosis procedure will be considered and integrated into the monitoring system.

Acknowledgements

The researchers extend their sincere gratitude to the Deanship of Scientific Research at the Islamic University of Madinah for the support provided to the Post-Publishing Program.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Research data policy and data availability statements

The authors confirm that the data supporting the findings of this study are available within the article.

Additional information

Funding

This work was supported by the Deanship of Scientific Research at the Islamic University of Madinah for the support provided to the Post-Publishing Program Post-Publishing Program.

Appendix

Table A1. Specification and precision of the used sensors.

Sensor	Specification and precision
Temperature and humidity (DHT11)	• Operating voltage: 3–5 V • Temperature range: 0–50°C / ± 2°C • Humidity range: 20–80%/5%
Soil moisture	Range: 0–45% volumetric water content in soil (capable of 0–100% VWC with alternate calibration) Accuracy: ± 4% typical 13-bit resolution (using SensorDAQ):0.05% 12-bit resolution (using LabPro, LabQuest, LabQuest Mini, Go!Link, or EasyLink): 0.1% 10-bit resolution (using CBL 2):0.4% Power: 3 mA @ 5VDC Operating temperature: –40°C to +60°C Dimensions: 8.9 cm × 1.8 cm × 0.7 cm (active sensor length 5 cm) Stored calibration: Slope: 108%/ volt Intercept: –42%