Zgadzam się
Nasza strona zapisuje niewielkie pliki tekstowe, nazywane ciasteczkami (ang. cookies) na Twoim urządzeniu w celu lepszego dostosowania treści oraz dla celów statystycznych. Możesz wyłączyć możliwość ich zapisu, zmieniając ustawienia Twojej przeglądarki. Korzystanie z naszej strony bez zmiany ustawień oznacza zgodę na przechowywanie cookies w Twoim urządzeniu.
This research was financially supported by the Ministry of Science and Higher Education in Poland within the grants FD-20/IS-6/025, FD-20/DN-10/027, FD-20/DN-10/013 and LUT university program Grants for publication.
This study explores the application of materials used in 3D printing to man-
ufacture the housings of non-invasive sensors employed in measurements using a TDR
(Time Domain Reflectometry) meter. The research investigates whether sensors designed
with 3D printing technology can serve as viable alternatives to conventional invasive and
non-invasive sensors. This study focuses on innovative approaches to designing humidity
sensors, utilizing Fused Deposition Modeling (FDM) technology to create housings for
non-invasive sensors compatible with TDR devices. The paper discusses the use of 3D
modeling technology in sensor design, with particular emphasis on materials used in 3D
printing, notably polylactic acid (PLA). Environmental factors, such as moisture in building
materials, are characterized, and the need for dedicated sensor designs is highlighted.
The software utilized in the 3D modeling and printing processes is also described. The
Materials and Methods Section provides a detailed account of the construction process
for the non-invasive sensor housing and the preparation for moisture measurement in
silicate materials using the designed sensor. A prototype sensor was successfully fabricated
through 3D printing. Using the designed sensor, measurements were conducted on silicate
samples soaked in aqueous solutions with water absorption levels ranging from 0% to
10%. Experimental validation involved testing silicate samples with the prototype sensor
to evaluate its effectiveness. The electrical permittivity of the material was calculated,
and the root-mean-square error (RMSE) was determined using classical computational
methods and machine learning techniques. The RMSE obtained using the classical method
was 0.70. The results obtained were further analyzed using machine learning models,
including Gaussian Process Regression (GPR) and Support Vector Machine (SVM). The
GPR model achieved an RMSE of 0.15, while the SVM model yielded an RMSE of 0.25.
These findings confirm the sensor’s effectiveness and its potential for further research and
practical applications.
