This is an open access article under the CC–BY-SA license. I

Download 0,75 Mb. Pdf ko'rish
bet	2/9
Sana	09.12.2023
Hajmi	0,75 Mb.
	#114189

1 2 3 4 5 6 7 8 9

Bog'liq
01#Повторная калибровка модельного емкостного датчика для измерения влажности почвы IoT

This is an open access article under the
CC–BY-SA
license.
I.

I
NTRODUCTION

Many researchers are interested in calibrating sensors to
measure soil moisture content. Soil moisture sensor
calibration is required to be able to develop a low-cost
automatic irrigation system. Considering that the performance
of soil moisture sensors is highly dependent on the physical
and chemical properties of the soil, the development of a
durable low-cost soil moisture monitoring automation system
depend on the quality of the calibrated sensor [1].
A thermo-gravimetric, which technique by drying the soil
sample in an oven, have been developed to measure soil
moisture content. So, dry soil is given a certain volume of
water and then the moisture sensor value is measured.
Because the moisture sensor is a qualitative measurement, the
soil moisture sensor value is expected to change as the water
volume increases. This technique commonly as the regular
reference
because
it
can
provide
more
accurate
measurements. Extended technique of thermos gravimetric be
carried out based on data-driven [2].
Polynomial and linear regression models are frequently
used to calibrate soil moisture sensors data. Ordinary
polynomial regression analysis is often used for curve fitting
while the linear regression model is used for linear fitting.
Both models express R Squared as a relationship
measurement. The order of the polynomial necessary to
explain the dependent variable.
But, computational effort
required [3].
This research aims to provide an application, which
interactively allows retrieving a calibration equation, based
on user defined requirements. This application makes
gravimetric tests easier and accommodates the use of non-
linear models to determine the best model. Interactive web-
based calibration application as has been done previously by

JAIC
e-ISSN: 2548-6861
Re-Calibration of Model-Based Capacitive Sensor for IoT Soil Moisture Measurements
(Iman Setiawan, Mohammad Dahlan Th. Musa, Saskia Amalia Putri)
151
[4] but not using capacitive soil moisture sensor. Our previous
research [5], [6], we calibrated a percentage-based soil
moisture sensor. The percentage is calculated using
observation data when the sensor is in the air and immersed
in water. There are many missing values found in the stored
data. This indicates that the sensor calibration process is not
adaptive, especially if used for a long period of time. So that,
this research aims to visualise data points and the estimated
linear and non-linear model in a web-based application.
Additional features are an interactive web application handles
small sample sizes problem for model selection by using
adjusted R squared. The coefficients model obtained is then
programmed into an Arduino Uno which is connected to a
mobile phone so that an IoT model-based soil moisture sensor
is formed.

Download 0,75 Mb.

1 2 3 4 5 6 7 8 9

Download 0,75 Mb.

Pdf ko'rish