Traditional ventilation control decisions for Chinese Solar Greenhouses (CSGs) rely primarily on farmers’ empirical judgment. With the development of a novel ventilation system (comprising bottom vents, top vents, and back roof vents), farmers lack adequate practical experience in operating such systems. To evaluate the cooling performance of the new ventilation system in CSGs, a Wireless Sensor Network (WSN) system was established using multiple wireless temperature and humidity sensors. The Ordinary Kriging (OK) interpolation method, which was developed in LabVIEW, was employed to visualize and monitor the real-time air temperature distribution under various ventilation opening configurations and vent combinations. Additionally, the cooling amplitude and temperature uniformity were comparatively analyzed. The results indicated that the synergy of multiple vents could enhance the chimney effect, achieving efficient cooling in summer. The cooling amplitudes of the dual-vent combination (bottom vent+top vent) and the three-vent configuration were 7.1°C and 10.4°C, respectively. Increasing the ventilation area improved both the cooling effect and temperature uniformity, with more open vents and a larger top vent area offering additional benefits. The findings of this study can provide a theoretical reference for farmers to optimize greenhouse ventilation management and offer data support for the application of novel ventilation systems.

Temperature is a critical factor influencing crop growth in controlled environment agriculture. Accurate regulation of air temperature within a greenhouse is essential for promoting optimal crop development and enhancing production efficiency. In this study, a Categorical Boosting model based on SMOTETomek mixed sampling method and improved Sparrow Search Algorithm (SMOTETomek-ISSA-CatBoost) was proposed to predict the categories of greenhouse temperature control modes. This study utilized historical temperature control mode data, which had been accumulated by cultivation experts through practical production and demonstrated effective in temperature management. To enhance the model’s performance and achieve real-time, precise temperature regulation in greenhouses, firstly, the SMOTETomek mixed sampling method was utilized to expand the original training set, effectively addressing the issue of data imbalance. Secondly, the Latin Hypercube Sampling (LHS) method, the Cauchy mutation perturbation operator, and a greedy rule were employed to refine the Sparrow Search Algorithm to enhance the global search capability. Ultimately, the improved Sparrow Search Algorithm was employed to optimize the hyper-parameters of CatBoost model to improve its predictive accuracy. Compared with SMOTETomek-CatBoost models optimized by Whale Optimization Algorithm (WOA), Fruit Fly Optimization Algorithm (FOA), Particle Swarm Optimization (PSO), and standard Sparrow Search Algorithm (SSA), the SMOTETomek-ISSA-CatBoost model demonstrated better prediction efficacy, with F1-score and AUC values reaching 0.8147 and 0.9629, respectively. The SMOTETomek-ISSA-CatBoost model exhibited the capability to predict the category of temperature control modes in a greenhouse accurately, thereby providing a decision-making foundation for intelligent management of greenhouse environments.

How to achieve precise and coordinated control of greenhouse microclimate factors under strong coupling and nonlinear conditions remains a key challenge in protected agriculture. To address this issue, this study integrates intelligent control and multi-objective optimization to regulate greenhouse temperature and humidity in a coordinated manner. A mechanistic model of a Venlo-type greenhouse was first developed in Matlab R2022a. Then, three control methods, namely LQR, MPC, and NMPC, were compared, and NMPC showed the best performance. Finally, NSGA-II was introduced to optimize the objective function weights of NMPC, further improving the control results. Compared with NMPC alone, the optimized method reduced the RMSE and MAE by 0.3366 and 0.0812 for temperature control, and by 0.2192 and 0.6853 for humidity control, respectively. The proposed method improves the precision and coordination of greenhouse environmental control and provides support for efficient greenhouse production. Ultimately, this study offers a promising technical paradigm for transitioning traditional greenhouse management towards highly autonomous and sustainable precision agriculture.

Accurate sex identification in Procambarus clarkii is essential for genetic breeding and aquaculture management, as it helps optimize population structure, improve reproductive efficiency, and support sustainable aquaculture development. However, manual identification is time-consuming, labor-intensive, and prone to errors, especially when subtle visual differences need to be distinguished. To address this problem, this study proposed SCM-DETR, a sex identification method for Procambarus clarkii based on multi-dimensional feature fusion and enhancement. A high-resolution imaging system was used to acquire two-dimensional images of Procambarus clarkii, and a labeled dataset, the Procambarus clarkii gonad dataset (PGD), was constructed. To improve identification performance, a multi-dimensional semantics and details fusion method (MSDM) was designed to integrate high-level semantic information with fine-grained detail features, thereby enhancing feature representation and localization accuracy. In addition, a channel-spatial focus network (CSFN) was introduced to capture discriminative multidimensional features, including texture and color, for more accurate identification of subtle sex-related differences. Experimental results showed that SCM-DETR-R18 achieved 95.8% mAP@0.50 and 64.6% mAP@0.50-0.95 on the PGD, improving by 1.9 and 1.1 percentage points over the baseline model, respectively. The AP values of female and male gonads reached 93.3% and 96.5%, with gains of 3.3 and 1.9 percentage points, respectively. Moreover, the proposed model had the lowest parameter count (21.11 M) among all compared methods. The results of this study demonstrate that SCM-DETR can effectively improve automated sex identification in Procambarus clarkii and has good potential for intelligent aquaculture applications.

Onions, often referred to as the “Queen of the kitchen”, play a vital role in enhancing dishes with their distinctive flavor, phenols, and flavonoids. Despite their significance, the onion market faces considerable volatility due to seasonal variations and substantial postharvest losses, resulting in significant price fluctuations. These losses can reach up to 40%, primarily due to improper handling, sprouting, rotting, inadequate storage, and limited infrastructure, with microbial spoilage alone accounting for 15%-20% of the total losses. Conventional onion storage methods demand substantial investments but often fail to effectively manage temperature and relative humidity (RH) fluctuations, which are critical for maintaining bulb shelf life. This study focuses on developing an on-farm evaporative cooled onion storage structure (ECOSS) to provide a long-term storage solution for onions. Results demonstrated that the ECOSS maintained internal temperatures between 24.5°C-31.3°C (average 27.1°C) and suitable RH levels throughout the storage period, with minimum values between 63.7%-67.2% and maximum values between 74.1%-75.5%. Physiological loss in weight (PLW), sprouting, and rotting were lowest at the 0.91 m storage height compared to 1.82 m after 5 months of storage. Onion firmness decreased over time, with the lowest firmness observed at 1.82 m (14.68 N). The net profit per ton of onions stored from May to September was INR6021 for the ECOSS, compared to INR2553 for traditional onion storage structure (TOSS). The ECOSS offers a cost-effective and efficient solution for reducing postharvest losses and enhancing onion storage.

To address the persistent challenges of low harvesting efficiency and seasonal labor shortages in Camellia oleifera production, this study proposes a novel canopy vibration harvester driven by a five-bar PRRRP parallel mechanism configured to generate an epitrochoid excitation trajectory. Through analysis of the epitrochoid excitation trajectory and in accordance with parallel mechanism design principles, the PRRRP configuration was selected, considering structural layout, transmission system, motion performance, and singularity distribution. On this basis, a vibration harvesting device driven by the PRRRP mechanism was designed and developed. Kinematic and dynamic analyses, as well as workspace modeling, were conducted, and real-time position acquisition of the driving components was achieved through inverse kinematics solutions. The harvesting device includes dual servo motor-driven linear modules, articulated linkages, and an adjustable excitation frame fitted with excitation rods to engage the tree canopy. Using Camellia oleifera cultivar of Changlin No. 40 as the target crop, the device generated an epitrochoid trajectory with a vibration frequency of 7 Hz and an amplitude of 90 mm, delivering multidirectional excitation to the canopy. By adjusting tree size parameters, the vibration response of branches with different inclination angles was investigated. The results showed that average vibration response accelerations for branches inclined at 0°-30°, 30°-60°, and 60°-90° were 14.04 m/s², 21.88 m/s², and 21.27 m/s², respectively, providing a theoretical basis for branch pruning. Harvesting trials indicated an overall fruit removal rate of 75.10% and an overall bud shedding rate of 11.55%, showing higher fruit removal and lower bud shedding than a traditional canopy vibration device employing linear reciprocating excitation. These results confirm that epitrochoid-based, parallel-mechanism excitation markedly improves fruit detachment efficiency and reduces collateral damage to tree architecture. Our study provides both theoretical insights and practical guidance for the development of next generation mechanized harvesters for Camellia oleifera and other woody fruit crops.

To address the challenges in labeling long labels on curved surface vegetables, such as wrinkles and label detachment, a cam-elliptical gear (C&E) labeling mechanism that realizes an improved hypocycloid trajectory is proposed. Firstly, the influence of different parameters on the hypocycloid trajectory is studied, and the tricuspid hypocycloid is selected as the labeling trajectory. Secondly, the trajectory and kinematic equations of the C&E labeling mechanism with a tricuspid hypocycloid trajectory are established. Next, the influence of various parameters on the trajectory and kinematics of the C&E labeling mechanism is examined. A set of optimal parameters is obtained through a comparative study, and a 3D model of the C&E labeling mechanism is established and simulated. Finally, a prototype of the C&E labeling mechanism was built and experimented with. The experiment showed that the normal labeling completion rate is 94%, and the prototype’s efficiency is 56.3 pcs/min. The research in this paper provides a theoretical basis for the design and optimization of a vegetable long-label curved-surface labeling mechanism.

To address high cleaning losses caused by high impurity content and uneven distribution during rapeseed threshing, this study proposes a drum pneumatic-assisted threshing and separation method. Initially, by determining the suspension velocity characteristics of various threshed components, auxiliary blades were designed and integrated to construct a non-uniform gradient airflow field, generating a directional pneumatic transport effect. CFD simulations confirmed that this configuration induces a stable axial spiral airflow. Driven dynamically by this airflow, miscellaneous materials are guided to migrate orderly toward the rear of the concave screen, enabling active source-based regulation of the spatial-temporal flow behavior and component distribution. Multi-parameter correlation analysis indicated a significant negative correlation (coefficient = –0.88) between axial airflow and the average impurity ratio. Field experiments demonstrated that at a feed rate of 2.8 kg/s with auxiliary blades, the grain loss rate was reduced to 3.7%, compared with 6.3% without auxiliary blades, corresponding to a 41.3% reduction. This research demonstrates that axial airflow can effectively regulate threshed material at the source, providing a novel approach for high-efficiency and low-loss threshing and cleaning.

Nitrous oxide (N₂O), a long-lived greenhouse gas, is primarily produced in agricultural soils through biological nitrification and denitrification processes. However, the effects of soil salinity on nitrogen transformation processes remain insufficiently understood, hindering the development of effective nitrogen management in salt-affected farmlands. In this study, laboratory incubation experiments combined with a ¹⁵N stable isotope tracing technique were conducted to quantify the effects of salt stress on nitrification and denitrification rates and their contributions to N₂O emissions. The results showed that during the first week of incubation, slight and moderate salinity (NaCl contents of 0.04% and 0.10%; EC_1:5≤1.10 dS/m) enhanced both nitrification and denitrification rates, whereas strong salinity (0.20% NaCl; EC_1:5≥1.42 dS/m) inhibited these processes. In the first week of incubation, nitrification and denitrification contributed approximately 65%-70% and 30%-35% to the total N₂O emissions under 60% water-filled pore space, respectively. These results indicate that nitrification represents the predominant source of N₂O production in saline soils during the first week following fertigation. The findings suggest that nitrogen management practices for inhibiting the nitrification process (e.g., the addition of nitrification inhibitors in companion with fertigation) may mitigate N₂O emissions in saline fields. This provides a scientific basis for optimizing nitrogen management in saline soils and reducing greenhouse gas emissions.

Current Lycium barbarum L. vibration harvesting equipment exhibits low levels of intelligence and precision, often resulting in a trade-off between efficiency and fruit damage. This study proposed a ripe fruit region detection model, YOLO-RFR, specifically for precision vibration harvesting of L. barbarum. First, the ADown downsampling module was introduced to replace part of the conventional convolution layers. Then, the C3k2-AP module, inspired by the asymmetric padding strategy, was designed to replace the C3k2 module. Additionally, the GCHead detection head was constructed using group convolution. Finally, the EMA-Slide Loss function was developed to optimize the classification performance by combining the slide weighting function with Exponential Moving Average (EMA). The experimental results showed that the model achieved precision, recall, and mAP of 93.7%, 92.0%, and 97.0%, respectively, representing improvements of 4.0%, 4.4%, and 2.6% over the baseline. The parameter, floating-point operations (FLOPs), and model size were 1.7 M, 4.1 G, and 3.8 MB, respectively, corresponding to decreases of 34.6%, 34.9%, and 30.9% compared with the baseline. To further validate its practical feasibility, the improved model was deployed on an NVIDIA Jetson AGX Xavier embedded device, achieving an inference speed of 163 fps with TensorRT acceleration. In conclusion, the YOLO-RFR model demonstrated excellent performance in detection accuracy, model lightweighting, and deployment on embedded devices, providing strong technical support for the precision vibration harvesting of L. barbarum.