This study investigates the impact of varying side wind velocities and nozzle inclination angles on droplet penetration during plant protection spraying operations, focusing on citrus trees. Experiments were conducted across four wind speed levels (0, 1, 2, 3 m/s) perpendicular to the nozzle direction and seven nozzle inclination levels (0°, 8°, 15°, 23°, 30°, 38°, 45°) to evaluate droplet distribution under different spraying parameters. A baseline condition with 0 m/s wind speed and a 0° nozzle angle served as the control. Utilizing Computational Fluid Dynamics (CFD) and regression analysis techniques in conjunction with field trials, the droplet penetration was analyzed. Results indicate that at constant wind speeds, adjusting the nozzle inclination angle against the direction of the side wind can significantly enhance droplet deposition in the canopy, with a 23° inclination providing the optimal increase in deposition volume, averaging a change of +16.705 μL/cm². Multivariate nonlinear regression analysis revealed that both wind speed and nozzle inclination angle significantly affect the droplet penetration ratio, demonstrating a correlation between these factors, with wind speed exerting a greater impact than nozzle angle. Increasing the nozzle inclination angle at higher wind speeds improves the penetration ratio, with the optimal parameters being a 23° angle and 3 m/s wind speed, showing a 12.6% improvement over the control. The model fitted for the impact of nozzle angle and wind speed on droplet penetration was validated through field experiments, identifying optimal angles for enhancing penetration at wind speeds of 1, 2, and 3 m/s as 8°, 17°, and 25°, respectively. This research provides insights for improving droplet penetration techniques in plant protection operations.

Creating a suitable growing environment is necessary to ensure good plant growth in a plant factory, which requires wireless sensor networks (WSNs) to monitor the environment in real time. However, existing WSN clustered routing methods hardly take into account the network unreliability caused by varying link quality among nodes, resulting in reduced stability and accuracy of environmental monitoring. This study proposes a wireless sensor network system strategy for improving network reliability in large-scale reliable wireless sensor networks suitable for plant factory scenarios. Firstly, a hybrid wireless sensor network was designed and built based on Wi-Fi and ZigBee communication protocols. Secondly, a nonlinear link quality prediction model for plant factory scenarios was developed using a function fitting method, taking into account the interference and attenuation caused by the dense concentration of agricultural facilities and plants in plant factories on the wireless signal propagation. Finally, a network coverage optimization scheme was designed by combining a particle swarm optimization (PSO) algorithm and link quality prediction model, and a reliable cluster routing protocol was designed by combining K-means algorithm. The results indicated that the coefficient of determination (R²) for the prediction model was 0.9962. The impact of agricultural facilities and vegetation on link quality was most significant when the node height was 0.7 m. Under the optimal node deployment, the number of nodes was 33, and the network coverage rate (CR) reached 97.512%. Compared with the traditional clustered routing method, the wireless sensor network designed in this study is more applicable to the field of plant factories; it further enhances data transmission effectiveness and link quality, improves the reliability of the network, and realizes the load balancing of the internal transmission of the network, which in turn ensures the accuracy of environmental monitoring and the stability of the system.

Greenhouse tomato production is highly influenced by environmental variability, where root-zone temperature plays a crucial role in fruit growth, biomass accumulation, and energy-use efficiency, particularly under winter low-temperature conditions. This study aimed to develop and validate a root-zone temperature dynamic regulation strategy for cherry tomato cultivation under multifactor coupled greenhouse conditions. Four root-zone temperature treatments were initially established: control (CK, no temperature regulation), T15 (setpoint of 15℃), T20 (setpoint of 20℃), and T25 (setpoint of 25℃). An IoT-based monitoring system was constructed to continuously collect root-zone temperature, substrate electrical conductivity, substrate moisture, air temperature, air humidity, CO2 concentration, light intensity, and fruit diameter data. With environmental factors as inputs, four machine learning models—Support Vector Regression (SVR), Gaussian Process Regression (GPR), Random Forest Regression (RFR), and Extreme Gradient Boosting (XGBoost)—were developed and evaluated to predict fruit diameter growth. Taking fruit diameter growth as the optimization objective, Particle Swarm Optimization (PSO), Genetic Algorithm (GA), and Simulated Annealing (SA) were employed to establish production-oriented root-zone temperature dynamic regulation strategies under environmental uncertainty. For the field validation, the dynamic temperature strategy was set as a treatment (DT) and compared with CK and constant-temperature treatments regarding to plant growth and energy consuming. The results showed that SVR achieved the best predictive performance for fruit diameter growth, with an R² of 0.8568 and an RMSE of 0.5142. All three optimization algorithms generated highly consistent optimal solutions, confirming the reliability of the scenario-based optimization framework. GA converged rapidly in terms of iteration number under the peak-growth and yield-oriented objectives, SA maintained comparatively balanced computational efficiency across the three objectives, whereas PSO exhibited greater sensitivity to the characteristics of the optimization objective. The yield-oriented strategy, which maximized the expected mean fruit growth response across representative environmental scenarios, was selected for field validation because it was most consistent with the commercial objective of improving population-level fruit growth and overall greenhouse yield. Under fluctuating winter greenhouse conditions, root-zone temperature regulation generally improved plant and fruit growth compared with the unheated control. DT produced the largest observed functional leaf phenotype, followed by T15, while T20 and T25 showed intermediate leaf development and CK exhibited the smallest leaves. During the early fruit expansion stage, DT showed the fastest expansion rate, while DT, T15, and T20 achieved greater fruit diameter increments than CK and T25. All heated treatments substantially increased fruit fresh and dry biomass compared with CK, and the largest numbers were observed in DT. Among the heated treatments, cumulative heating duration decreased in the order of T25 (2038.11 h), T20 (1985.12 h), T15 (1813.10 h), and DT (1605.41 h). Overall, DT achieved the best integrated agronomic and energy-use performance, which validated the developed framework of root-zone temperature dynamic regulation strategy.

Greenhouse tomato production is highly influenced by environmental variability, where root-zone temperature plays a crucial role in fruit development and resource utilization efficiency, particularly under winter low-temperature stress conditions. This study aimed to develop an optimal root-zone temperature regulation strategy for cherry tomato cultivation in multifactor-coupled greenhouse environments by integrating Internet of Things (IoT)-based monitoring, machine learning predictive modeling, and robust optimization methods. Four root-zone temperature treatments were established: control (CK, no temperature regulation), T15 (setpoint 15°C), T20 (setpoint 20°C), and T25 (setpoint 25°C). An IoT-based intelligent monitoring system was constructed to continuously collect soil temperature, electrical conductivity (EC), moisture, air temperature, humidity, CO2 concentration, light intensity, and dynamic fruit diameter growth data. With environmental factors as inputs, four machine learning models, Support Vector Regression (SVR), Gaussian Process Regression (GPR), Random Forest Regression (RFR), and Extreme Gradient Boosting (XGBoost) were developed and evaluated to predict fruit diameter growth. Taking fruit diameter as the optimization objective, Particle Swarm Optimization (PSO), Genetic Algorithm (GA), and Simulated Annealing (SA) were employed to develop optimal root-zone temperature regulation strategies based on multiple environmental factors. The results indicated that moderate root-zone heating (T15, 15–21°C) significantly enhanced fruit diameter, single-fruit fresh weight, cluster weight, and dry matter accumulation, with T15 increasing single-fruit fresh weight by approximately 41.5% compared to the control while preserving high nutritional quality (soluble sugars, vitamin C, and lycopene), thereby achieving an optimal balance between yield and quality. For fruit diameter prediction, SVR demonstrated the best predictive performance (R2 = 0.86), effectively capturing nonlinear relationships between environmental factors and fruit growth. For the optimal root-zone temperature regulation strategy, PSO and GA demonstrated higher robustness and convergence stability compared to SA, while SA exhibited better computational efficiency. Under the experimental conditions, the optimal root-zone temperature was determined as 17.7°C to maximize growth response, 17.1°C to maximize average response, and 15°C to minimize variance. This study developed an integrated root-zone temperature regulation strategy combining IoT-driven monitoring, SVR-based prediction, and robust optimization methods to guide dynamic root-zone temperature regulation, contributing to the precision and intelligence of greenhouse tomato production.

This study proposed and validated an FPGA-based filtering algorithm designed to enhance the measurement accuracy, data reliability, and computational efficiency of greenhouse monitoring systems. To address the precision and reliability challenges present in greenhouse environmental monitoring and plant growth monitoring systems, three filtering methods—mean filter, median filter, and Kalman filter—were implemented and evaluated. Experimental results demonstrated that all three filters significantly improved measurement accuracy, with the Kalman filter yielding the most notable enhancement. Specifically, the Kalman filter improved measurement accuracy to 0.356%, representing a 14.045-fold increase compared to the baseline, significantly outperforming the mean filter (3.005-fold improvement) and median filter (2.354-fold improvement). Additionally, after Kalman filtering, 91.100% of the measurement errors fell within a probability density, indicating superior robustness and stability. To optimize hardware implementation of the Kalman filter, this study introduced an approach based on systolic array matrix multipliers and parallel pipeline design, achieving a 28.197-fold increase in computational speed. On FPGA, a parallel implementation using single-precision floating-point arithmetic maintained the same computational accuracy (0.356%) as conventional serial double-precision implementations, while delivering significant advantages in processing speed and resource utilization. For energy efficiency monitoring in greenhouses, an LMS adaptive filtering algorithm and its FPGA implementation were also proposed. This method effectively suppressed harmonic distortion in current signals, especially at harmonic frequencies of 150 Hz and 250 Hz, where attenuation reached 99.85% and 99.69%, respectively. The filtered signal closely approximated the ideal fundamental wave in both time and frequency domains, and the current signal error was reduced from 0.182 to 0.010, substantially improving signal reconstruction accuracy. In summary, the proposed FPGA-based Kalman filtering and LMS adaptive filtering schemes significantly improved the accuracy, reliability, and computational performance of greenhouse monitoring systems, demonstrating strong potential for application in greenhouse monitoring and intelligent agriculture.