University of Twente Student Theses
Predicting Soil Moisture for Improved Environmental Sustainability: A Multivariate Time-Series Forecasting Approach using Machine Learning
Hartmans, L.A. (2023) Predicting Soil Moisture for Improved Environmental Sustainability: A Multivariate Time-Series Forecasting Approach using Machine Learning.
|
PDF
524kB |
| Abstract: | Soil moisture plays an essential role in the overall health of newly planted vegetation. Hence, it is crucial to develop precise soil moisture forecasting techniques that facilitate precision irrigation, effectively mitigating the risk of vegetation succumbing to drought. Recent research compared the performance of different machine learning techniques and found it to be an effective forecasting approach. This research was about extending this existing research on machine learning approaches by assessing the performance of a Long Short-Term Memory Recurrent Neural Network using multivariate time-series data of soil moisture sensors which were subject to manual irrigation complemented with meteorological data from weather stations within the Netherlands. The performance was evaluated using MAE, MSE, RMSE and MAPE and showed compelling results, with an average MAPE of 3.41\% for a 1-day horizon, 5.85\% for a 3-day horizon, and 10.09\% for a 7-day horizon. |
| Item Type: | Essay (Bachelor) |
| Faculty: | EEMCS: Electrical Engineering, Mathematics and Computer Science |
| Subject: | 43 environmental science, 54 computer science |
| Programme: | Business & IT BSc (56066) |
| Awards: | Best Presentation Award |
| Link to this item: | https://purl.utwente.nl/essays/95892 |
| Export this item as: | BibTeX EndNote HTML Citation Reference Manager |
Show download statistics for this publication
Show download statistics for this publicationRepository Staff Only: item control page