| Title |
Predicting Internal Temperatures of Cultivation Zones in Smart Farming Based on Simple Modeling |
| Authors |
Seung-Hoon Park ; Gwanyong Park |
| DOI |
https://doi.org/10.6110/KJACR.2024.36.11.546 |
| Keywords |
LSTM 모델; 모델링 방법; RC 모델; 시뮬레이션; 스마트 팜 LSTM model; Modeling method; RC model; Simulation; Smart farming |
| Abstract |
Smart farming is a sustainable agriculture against population growth and a sustainable production method to address worldwide food security concerns. However, with their high dependence on fossil fuels, reducing carbon emissions remains a critical challenge in smart farming. To mitigate this issue, it is essential to implement model-based control and smart farming operations to enhance energy efficiency. In this study, modeling methods for a control purpose were tested between a building physics-based model and a neural network-based model using hourly data from five days. For model test and validation, 18 days of data were used all models were evaluated based on their performances in predicting internal temperatures of cultivation zones. Results revealed that the neural network-based model exhibited a better performance than the building physics-based model. However, after identifying supplementary modeling points from the building physics-based model and applying these to both models, the predictive performance of the building physics-based model ultimately surpassed that of the neural network-based model. |