Within the realm of agricultural development, the relentless quest for agricultural effectivity amidst the vagaries of local weather change has positioned greenhouse know-how as a linchpin for safe and sustainable meals manufacturing. The exact administration of greenhouse microclimatic situations i.e., the power to precisely predict and keep supreme temperature and relative humidity, is essential for enhancing plant development and well being, optimizing useful resource use, and guaranteeing sustainable agricultural practices. Nevertheless, sustaining optimum microclimatic situations is a major problem because of the dynamic nature of exterior environmental influences.
This examine goals to handle the important want for superior predictive instruments that may improve the management and administration of greenhouse microclimates, thereby supporting sustainable agricultural practices and meals safety. Our analysis introduces a novel integration of constructing transient simulation (TRNSYS) and synthetic neural networks (ANNs) to foretell temperature and relative humidity inside a greenhouse throughout the calendar 12 months, based mostly on exterior atmospheric situations. The TRNSYS mannequin meticulously simulates the greenhouse’s thermal load, incorporating real-world knowledge to make sure a excessive degree of accuracy in describing the ability’s dynamic conduct. Our ANN mannequin, composed of three layers, underwent optimization to establish the best variety of neurons, studying charges, and epochs, selecting a mannequin configuration that minimized prediction errors. The analysis metrics, together with root imply sq. error (RMSE) and imply absolute error (MAE), demonstrated the mannequin’s effectiveness, with an RMSE of 0.3166 °C for temperature and 5.9% for relative humidity, and MAE values of 0.1002° and three.4%, respectively.
These findings underscore the mannequin’s potential as a strong device for greenhouse local weather management, providing substantial advantages when it comes to power effectivity, useful resource optimization, and general sustainability in agriculture. By leveraging detailed dynamic simulations and superior neural community algorithms, this examine contributes considerably to the sector of precision agriculture, presenting a novel method to managing greenhouse environments within the face of adjusting weather conditions.
Ećim-Đurić, O.; Milanović, M.; Dimitrijević-Petrović, A.; Mileusnić, Z.; Dragičević, A.; Miodragović, R. Prediction of Greenhouse Microclimatic Parameters Utilizing Constructing Transient Simulation and Synthetic Neural Networks. Agronomy 2024, 14, 1147. https://doi.org/10.3390/agronomy14061147
