Research on intelligent monitoring and optimization control system model of greenhouse environment based on deep learning

This paper presents an improved intelligent control model for greenhouse environments, integrating environmental monitoring data with a deep learning approach combining multilayer CNN and LSTM networks with model predictive control (MPC). By optimizing model parameters and utilizing data fusion techniques, the system’s prediction accuracy and response speed are enhanced. Experimental results from a real greenhouse monitoring system show that the improved model significantly outperforms traditional models and other machine learning algorithms, with a mean absolute error (MAE) of 2.342, mean square error (MSE) of 6.749, and a coefficient of determination (R<sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup>) of 0.931, demonstrating its efficiency and reliability in controlling key environmental indicators like temperature, humidity, and CO<inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</inf> concentration.