We live in a time where electronic devices are present on every step. In relation
with agriculture the so-called precision agriculture has developed. One of the
further division of it is precision beekeeping. It deals with the non-invasive data
collection about bee colonies in a wish to optimize resource usage and increase
honey production. The aim of this master thesis is to contribute precisely to this
area.
In the introductory section we placed the bees within the animal kingdom. We
made a review of the commercial solutions in the field of precision beekeeping.
We have found that the majority were electronic scales. In the research field
we identified articles which on the basis of acoustic features look for a way of
identifying the activities of the bees. The academic project E-Ruche focuses on
collecting data about temperature inside the hive up to ten measuring points. In
a somewhat earlier project called ITAPIC, communication was realized through
wireless communication sensors with the sensor node.
The following chapter presents the basic technologies and the terms used in
this master thesis. Among them are: load cell (strain gauge), protocol I2C,
the concept of IoT, data formatting JSON, REST principle, JavaScript, nonrelational
database MongoDB and Google Material.
Within our central topic, we first describe the production of fourpoint electronic
scales. We developed it with load cell technology. In the manufacturing,
we encountered some problems due to low output voltage and charging limitations.
At the end the measurements were carried with only one load cell. The
temperature and humidity inside and outside the hive was measured by sensors
DHT22. The sensor with the use of the protocol I2C covered measurements at
regular intervals and submitted them to the web server.
In the chapter regarding the web application, we described the programming
of user data structures and the optimization of the sensor data retention. The
application contains a web interface, which is modeled after Google Material. It
allows the identification of the user, a display of sensors and illustrates recent
measurements using a graph.
In the last chapter, we described a test system built in the beginning of August.
The scale was padded under the hive, we placed the temperature and
moisture sensor in a special modified honeycomb and housed it outside the hive.
Measurements were collected for approximately four days. The amount of data
was too small for serious analysis, but we could still observe some short-term
phenomena and their impact on the measurements.
|
