Agronomy, Год журнала: 2025, Номер 15(5), С. 1180 - 1180

Опубликована: Май 13, 2025

Soil pore water electrical conductivity (EC), as a comprehensive indicator of soil nutrient status, is closely linked to crop growth and development. Accurate prediction EC therefore essential for informed scientific management. This study focuses on greenhouse rose cultivation site in Jiangchuan District, Yuxi City, Yunnan Province, China. Leveraging multi-parameter sensors deployed within the facility, we collected continuous data (temperature, moisture, EC, EC) meteorological (air temperature, humidity, vapor pressure deficit) from January December 2024. We propose hybrid model—PSO–CNN–LSTM–BOA–XGBoost (PCLBX)—that integrates particle swarm optimization (PSO)-enhanced convolutional LSTM (CNN–LSTM) with Bayesian algorithm-tuned XGBoost (BOA–XGBoost). The model utilizes highly correlated environmental variables forecast EC. experimental results demonstrate that PCLBX achieves mean square error (MSE) 0.0016, absolute (MAE) 0.0288, coefficient determination (R2) 0.9778. Compared CNN–LSTM model, MSE MAE are reduced by 0.0001 0.0014, respectively, an R2 increase 0.0015. Against BOA–XGBoost yields reduction 0.0006 0.0061 MAE, alongside 0.0077 improvement R2. Furthermore, relative equal-weight ensemble BOA–XGBoost, shows improved performance, decreased 0.0005, increased 0.0007. These underscore superior predictive capability over individual baselines. By enhancing accuracy robustness prediction, this contributes deeper understanding physicochemical dynamics offers scalable tool intelligent perception forecasting. Importantly, it provides agricultural researchers managers deployable generalizable framework digital, precise, management nutrients protected horticulture systems.

Язык: Английский