IoT based smart irrigation, control, and monitoring system for chilli plants using NodeMCU-ESP8266

ABSTRACT


INTRODUCTION
The most significant problem in some regions of Malaysia is a lack of water cultivation due to insufficient precipitation and an abundance of water waste due to improper water storage arrangements. Farmers plan their crop outputs mainly based on precipitation forecasts [1]. According to risk in Malaysian agriculture, severe climate change is a concern confronting Malaysia's agricultural industry [2]. Most of the water supplied to irrigation systems comes from, for example, natural rainfall and drain water, which is disrupted during the dry season, affecting crop output [3]. Most farmers water their plants manually, but this method is inefficient. Thus, automatic irrigation systems are used to resolve this issue [4]- [6], as they reduce the amount of labour needed to run the plantation and the time required to do so [7]. Dust control, trash management, and mining also use irrigation systems [8]. The internet of things (IoT) also plays a significant role in most industries. IoT is a network of connected computing devices, mechanical and digital machines, objects, animals, or people that can exchange data over a network without requiring human-to-human or human-to-computer interaction. The use of the IoT has increased due to its advantageous features. Improvements in the agricultural industry are significant, as this industry meets one of the primary needs of society and the economy of the country [9]- [11]. The IoT is expected to deliver promising solutions to alter the functions and positions of many current industrial structures, such as transportation schemes and development processes. The IoT is also currently used in Malaysia's agriculture industries, especially in irrigation systems.
Several methods are proposed to gather crop information or sensor data with better irrigation [12], [13]. IoT devices such as Arduino, node micro-controller units (NodeMCU) and sensors can also track farms in real-time and gather crop information remotely. The costs of these systems are manageable; the main challenge would be properly using massive amounts of sensors and other instruments. This technology is used to predict a field's irrigation requirements using ground parameter sensing such as soil moisture, soil temperature, and environmental conditions, along with weather forecast data.
Today, irrigation systems are an important part of ensuring high plant productivity in the agricultural industry. The irrigation systems currently used in Malaysia, especially for plantations, are divided into two categories [14]. The first category is manual irrigation systems by which farmers rely solely on rainwater to treat their crops. This technique is still used by many farmers because it is very cheap. However, it creates many problems because rainwater only falls during certain times and seasons. In addition, farmers cannot systematically control the water system, as it depends only on rainwater and reservoirs. Moreover, manual irrigation systems cannot provide accurate information on the condition of crops in terms of temperature, humidity and soil moisture.
The second category of irrigation systems is based on automated processes. Essentially, current automated irrigation systems in Malaysia fall into two categories: sprinkler irrigation and drip irrigation [15]. Drip irrigation is a water-saving method by which water is supplied by slowly dripping water into the soil or the roots of plants. Meanwhile, in sprinkler irrigation, the water is supplied by sprinkling water around the crop; this type of irrigation is the most commonly used system in Malaysia's agriculture industries [15].
However, both irrigation systems have problems, such as issues related to real-time monitoring, data collection and wasted water. Current sprinkler systems only supply water to the plant without collecting information such as humidity, temperature, and soil moisture from the plants. Furthermore, farmers who wish to manually collect such data would need to examine the soil conditions for each plant, which would require significant time and energy. In addition, water is wasted when there is no control display to notify the farmer when to start and stop an automated irrigation system. This paper describes the development of an irrigation control and monitoring system testbed based on IoT devices and specifically designed for a chilli plantation. The proposed work endeavours to develop a real-time and automated control and monitoring system using relevant crop growth parameters such as temperature, humidity and soil moisture that can be employed to view and interpret growth performance via a user-interface platform.
The testbed of the chilli plant irrigation control and monitoring system was deployed and experimented with at Kampung Paya Rawa, Kedah. Furthermore, the testbed was integrated with planting equipment, including water barrels, pipes and motors. The rest of this paper is organised as follows: section 2 presents the literature review, which investigates previous studies on irrigation and monitoring systems for plantations; the information has been tabulated in a simple form to describe various software packages, databases and other platforms used. Section 3 explains the development of the proposed irrigation and monitoring system and its operation. Section 4 describes the results of the project. All collected data was displayed, including the prototype plan, monitoring display and analysis data related to the objectives of this study. Finally, section 5 provides the conclusions.

LITERATURE REVIEW
Farm automation can change agricultural work from manual and static to smart and dynamic, resulting in higher production with less human supervision. Farmers have started using computers and software systems over the last decade to coordinate their financial data, track their transactions with third parties and control their crops more efficiently [16]. Agriculture is increasingly becoming a data-intensive industry in the era of the internet, in which information plays a key role in people's lives. Furthermore, in this age, farmers need to collect and analyse massive amounts of information from various devices like sensors and machinery to be more effective in generating and communicating relevant information [17]. With the advent of open-source Arduino boards and cheap moisture sensors, it is feasible to build equipment that can detect soil moisture and irrigate the fields or the landscape as required.

3055
A remote greenhouse global system for mobile-short message services (GSM-SMS) measurement and control system based on a personal computer were developed by Vimal and Shivaprakasha [18]. The developed system applied a database system linked to the base station. A microcontroller, GSM module, sensors and actuators were used to construct the base station. The central station receives and sent messages through a GSM module during practical service. The criterion values of the parameters to be measured at each base station were determined by the parameters of the central station and then by the parameters of the base station, such as air, temperature and the humidity of the air.
Previous work has primarily focused on assessments in the field of remote monitoring while controlling the technology used and determining its potential benefits. A revolutionary remote-controlled embedded irrigation device based on GSM/Bluetooth was proposed in [19]. Depending on the temperature and humidity read by the sensors, the device determined the irrigation time of the crop and irrigated the field automatically when unattended. The information was shared via SMS on the GSM network between the far end and the planned device. A Bluetooth module, which removed SMS charges when the user was within a few meters of the specified device, was also attached to the main microcontroller chip. The device informed users via SMS on the GSM network or Bluetooth of several conditions, such as electricity status, dry running engine, increased temperature, soil water, and smoke content.
In a previous study, an automated rain gun irrigation system based on a microcontroller was used such that irrigation occurred only when there was a high demand for water, thereby saving a significant quantity of water [20]. Noerhayati et al. [21] argued about the design of plant watering equipment with automatic pumps for drip irrigation on dry land. This study aimed to activate automatic watering equipment on dry land, determine the effective and efficient manner and identify more specific performance in the use of electrical power.
Moreover, Saputra [22] explained soil moisture monitoring through Telegram messages based on the results of watering the plants. The system performance started by detecting plant soil moisture, watering the plants and sending messages about the soil moisture conditions of the plant to the owner's cellphone. Then, Ren et al. [23] described a gardening smart system for web-based orchid plant care. Their prototype was intended to help control plant care systems orchids at the individual level (personal users). Lastly, a horticultural plant growth control and monitoring system in smart gardens has been researched in [24]. This research aimed to create a system capable of controlling the temperature, monitoring the moisture content in the soil and fertilising plants regularly while also providing system reports provided via an easy-to-use interface (namely, smartphones). The synthesis matrix in Table 1 summarises previous studies.

SYSTEM DESIGN AND IMPLEMENTATION
The architectural system design of the chilli plant irrigation control and monitoring system is shown in Figure 1(a). A NodeMCU ESP8266 was connected with a DHT11 humidity and temperature sensor and an FC-28 soil moisture sensor (referred to as a sensor box). ESP8266 [25], developed by espressif systems, is a low-cost open-source IoT platform that runs on an ESP12 RF module and other necessary components such as an antenna, universal serial bus (USB) connector and analogue and digital input/output pins. In this work, the chilli irrigation control and monitoring system was developed using Arduino IDE since it is easy to use. The system testbed was connected to the water pump (see Figure 1(b)) through a solid-state relay of FOTEK SSR-25 DA that applied a photocoupler as the switching motor. Thus, the plant irrigation system could be controlled via NodeMCU, which was connected wirelessly via Wi-Fi or a 4G network. A mobile application and a desktop application were developed using Blynk and Thinger.io (see Figures 2(a) and 2(b)), respectively. Thus, users could monitor the temperature, humidity, and soil moisture of the chilli plantation and then control its irrigation system accordingly. Gauges were used to show the relative humidity, soil moisture, and air temperature. The super chart was also used to plot humidity levels versus temperature readings. When the temperature rises, humidity falls, and vice versa. Switches in the form of on/off button gadgets were used to activate and deactivate the pump. In addition, a notification widget detailing the current soil moisture level was developed. Table 2 categorises the soil condition of the chilli plantation based on the humidity, temperature and soil condition readings. The chilli cultivation system comprising water pumps, 900-gallon water tanks, filters, valves, and poly tubes was set up on 0.5 hectares located in Kampung Paya Rawa, Kedah. However, the chilli plant monitoring and control system testbed was deployed in an area of 10×10 metres, as shown in Figure 3(a). Crop management was facilitated by spacing rows four feet apart and polybags two feet apart in rows. Each crop bag contained micro-tubes containing nutritional solutions. Figure 3(b) shows the location of the sensor box within the chilli plantation area.

RESULTS OF THE CONTROL AND MONITORING SYSTEM
A desktop application for the chilli plant irrigation control and monitoring system was developed using Thinger.io. The interface comprises temperature and humidity readings received from the sensors. The received data were represented in doughnut-shaped gauges with values ranging from 1-60 ℃ depicting instant temperatures (Figure 4(a)). Meanwhile, humidity readings were represented by a time series chart (Figure 4(b)).  Table 3 presents the hourly temperature and humidity readings transmitted by the DHT11 sensor over 12 hours. Temperature changes in the chilli crop influenced the humidity value of the crop area. When the temperature rose, the humidity decreased and vice versa. Variations in temperature in the chilli crop area affected changes in the crop region's relative humidity. When the temperature increased, the relative  Table 4 shows the percentage of soil moisture changed to derive its condition status based on the rules set in Table 2.
For instance, Figure 5 depicts the notification statuses that appeared in the mobile application developed using Thinger.io. In Figure 5(a), the moisture sensor value of 29.11 provided a notification when the water content was low, while in Figure 5(b), the notification indicated that the soil was humid. These indicators assisted users in monitoring their chilli plantation and controlling the irrigation system. In addition to the Thinger.io application, a mobile application was developed using Blynk, a platform used to manage device provisioning and sensor data visualisation.

CONCLUSION
This paper described the construction and assessment of a system designed to regulate and monitor the watering of chilli plants. The chilli plantation in Kedah, Malaysia, served as a testbed for incorporating the system's sensors to track environmental variables such as temperature, humidity, and soil moisture. Additionally, plantation tools and equipment such as water pumps, water barrels, and motors were integrated into the system so that they could be used to manage the watering of the plants. The sensor data were collected and processed by a NodeMCU microcontroller, which also operated as the node's central processing unit. Mobile and desktop apps were built utilising Blynk and Thinger.io to facilitate irrigation system monitoring and analysis. Thus, an IoT based smart irrigation, control and monitoring system for chilli plants using NodeMCU-ESP8266 has been discussed to help farmers on their daily routine. In order to 3059 enhance existing setup, integrating additional sensors into the system box and incorporating power supply technologies such as a solar supply system enabled the sensor box to run continuously without requiring rechargeable batteries per future recommendations. Long-ranged network modules are worth considering when the goal is to cover a large area to replace the existing limited coverage of Wi-Fi setup.