Hutchings, Ian (2023) Low Powered Remote Monitoring for Smart Beehives. [USQ Project]
![]() |
Text (Project – redacted)
Hutchings_I_Hills_Redacted.pdf Download (15MB) |
Abstract
Each year, we find ourselves reaching for a jar of honey to make warm lemon and honey concoctions to comfort our throat during flu season. We also spread the golden goodness on freshy baked bread and cereals; the delicate sweetness reminiscent of life’s simple pleasures.
When we visit the supermarket, we enjoy the year round availability of our favourite fresh fruit and vegetables without the knowledge that behind these products is a sophisticated insect that silently works in the background and is central to our food production. That insect is the honey bee (Apis mellifera). It is a vital part of our food chain and critical to the pollination of a large majority of agriculture crops in Australia.
Over recent years however, the honey bee has been under threat. This threat is a small mite call the Varroa Mite, which humans have yet to find a solution to. Varroa Mite has just entered Australia and is trying to be contained. Varroa Mite is just one of the problems honey bees face. Other problems including Colony Collapse Disorder (CCD) have also seen a major decline in the bee population in the Untinted States of America. With the demand for food driving our agricultural farms to produce crops with the highest yield and in the shortest time frame, the failing health of bees has become a by-product of the demand.
This project has set out to research the possibility of early detection of pest and disease with the use of low powered remote monitoring for smart beehives. This type of technology would benefit commercial beekeepers with remote hive monitoring reducing overheads and the time beekeepers are required to travel to hive sites to conduct inspections.
As part of the project, remote monitoring technology was installed into the beehive so data could be remotely collected for a month before it had to be removed. The data collected was based on the parameters that best indicated hive health attributes found within the literature. These attributes were humidity, temperature, weight, pressure and bee counters, although not all the expected parameters listed had data collected in the experiment time frame.
Data was collected using an Arduino on the 4G Telstra Cat M1 network and stored in the Arduino cloud making it easy to access and interpret with timely updates. The data was then used with machine learning to predict internal hive temperature based on the rest of the parameters.
Modelling was also done with three models to pick out the most accurate results. Using linear regression, random forest and support vector machine models, a website application was built to give the beekeeper an indication of what the internal temperature should be based on the other parameters.
The remote monitoring and hive temperature predictor will provide real time information on remote hives for the beekeeper, improving overall productivity reducing the risk and incidence of disease in hives.
![]() |
Statistics for this ePrint Item |
Item Type: | USQ Project |
---|---|
Item Status: | Live Archive |
Faculty/School / Institute/Centre: | Current – Faculty of Health, Engineering and Sciences - School of Engineering (1 Jan 2022 -) |
Supervisors: | Hills, Catherine |
Qualification: | Bachelor of Engineering (Electrical/Electronic) |
Date Deposited: | 29 Sep 2025 22:49 |
Last Modified: | 29 Sep 2025 22:50 |
Uncontrolled Keywords: | bees; beehives; remote monitoring |
URI: | https://sear.unisq.edu.au/id/eprint/52959 |
Actions (login required)
![]() |
Archive Repository Staff Only |