Banca de DEFESA: GABRIELA RODRIGUES DA COSTA

Uma banca de DEFESA de MESTRADO foi cadastrada pelo programa.
STUDENT : GABRIELA RODRIGUES DA COSTA
DATE: 26/10/2022
TIME: 14:00
LOCAL: Video conferência
TITLE:

Development and evaluation of a capacitive soil moisture sensor and weighing mini-lysimeter for automation of irrigation systems

KEY WORDS:

Irrigation management, electrical permittivity, capacitance, evapotranspiration, weighing lysimetry, tree species

PAGES: 48
BIG AREA: Ciências Agrárias
AREA: Engenharia Agrícola
SUBÁREA: Engenharia de Água e Solo
SPECIALTY: Irrigação e Drenagem
SUMMARY:

The efficient use of water in agriculture is achieved through the use of irrigation management techniques, which consists of applying water in sufficient quantity to meet the water requirement during the plant development cycle. Monitoring the soil water content and the water requirements of crops in real time through electronic devices allows the optimization of the irrigation system, with application of water in adequate amounts and at the right time. Given this context, in this study were developed and evaluated: a) a low cost capacitive sensor for measuring soil moisture in different soil/substrate materials; and b) a weighing mini-lysimeter (ML) aimed at automating the irrigation system and determining the water requirement of seedlings of tree species in the initial stage of growth. For the capacitive sensor, two types of geometry (parallel and coplanar boards) were evaluated in different types of printed circuit boards and insulating materials. After initial evaluation, the coplanar sensor built in fiberglass plate and insulated with varnish (CPS) showed greater sensitivity in the estimation of soil
moisture, in addition to greater ease of construction. The CPS was evaluated and compared to the commercial sensor (CS) 5TE from Decagon Devices, using 3.0 L pots filled with pure biosolid and 3 soil materials of different textures, which initially presented moisture close to field capacity. The volumetric moisture measurements (θ) obtained on precision balances were compared to the sensor readings, generating models relating θ with period (CPS) and permittivity (CS). Subsequently, estimated θ values were compared to θ values measured by
weighing and evaluated by the Willmoth index (d), coefficient of determination (R²) and performance index (c). The ML was built in wood and capable of receiving a tray of 54 tubes of 280 cm³. The structure is equipped with a load cell with a nominal capacity of 10 kg. Using an Arduino Mega board, the mass variations of the set were recorded on a micro SD card, which through an algorithm allowed the automated management of irrigation. Two MLs were built, making it possible to determine the water requirement of seedlings of Dalbergia Nigra (JB) and
Enterolobium contortisiliquum (OM) in the initial phase of growth. The CPS presented excellent performance (c > 0.93) when estimating the water content in the different conditions evaluated, having better results than the CS in substrate with biosolid (sewage sludge). After calibration, the MLs showed adequate sensitivity, with absolute error and mean square error of less than 1% of the total mass, enabling effective irrigation control. During the growth period, 1.47 L seedling -1 of JB and 1.45 L seedling -1 of OM were applied, which reached 89 days after
emergence, height and collar diameter of, respectively, 39.2 cm and 3.4 mm, and 19.3 cm and 4.6 mm.

BANKING MEMBERS:
Presidente - 1224578 - DANIEL FONSECA DE CARVALHO
Externo ao Programa - 2353141 - LEONARDO DUARTE BATISTA DA SILVA - UFRRJExterno à Instituição - CARLOS RODRIGUES PEREIRA - UFF