In the realm of agrarian progress , the relentless quest for agricultural efficiency amidst the vagaries of mood alteration has place nursery technology as a linchpin for secure and sustainable food production . The accurate direction of greenhouse microclimatic status i.e. , the ability to accurately predict and asseverate ideal temperature and relative humidity , is essential for enhancing plant life emergence and wellness , optimise resource utilization , and guarantee sustainable agricultural practices . However , maintaining optimum microclimatic condition is a significant challenge due to the dynamical nature of outside environmental influences .

This study aims to turn to the decisive want for advanced prognosticative instrument that can enhance the control and management of greenhouse microclimates , thereby supporting sustainable farming practices and food protection . Our research introduces a novel integration of building transeunt simulation ( TRNSYS ) and artificial neuronic meshwork ( ANNs ) to auspicate temperature and relative humidity inside a greenhouse across the calendar yr , based on international atmospherical condition . The TRNSYS model meticulously simulate the greenhouse ’s thermal load , incorporate real - Earth data to insure a gamey level of accuracy in describing the readiness ’s dynamical behavior . Our ANN model , draw up of three layer , underwent optimisation to identify the ideal routine of nerve cell , learn rate , and epochs , settling on a model conformation that minimized prediction error . The evaluation metrics , admit origin mean straight error ( RMSE ) and mean out-and-out erroneous belief ( MAE ) , demonstrated the model ’s potency , with an RMSE of 0.3166 ° speed of light for temperature and 5.9 % for comparative humidness , and MAE values of 0.1002 ° and 3.4 % , respectively .

These findings underline the model ’s potency as a powerful tool for greenhouse climate dominance , offering strong benefits in terms of zip efficiency , resource optimization , and overall sustainability in agriculture . By leveraging detailed dynamical simulations and advanced neuronal web algorithmic rule , this study contributes importantly to the field of preciseness agribusiness , exhibit a novel approach to finagle greenhouse surroundings in the face of change climatic conditions .

Ećim - Đurić , O. ; Milanović , M. ; Dimitrijević - Petrović , A. ; Mileusnić , Z. ; Dragičević , A. ; Miodragović , R. Prediction of Greenhouse Microclimatic Parameters Using Building Transient Simulation and Artificial Neural Networks . Agronomy 2024 , 14 , 1147 . https://doi.org/10.3390/agronomy14061147

Click here to view the full paper

Frontpage photo : © Ian Allenden | Dreamstime