The profit of greenhouse production is influenced by management activities (e.g., environmental control and plantation scheduling) as well as social conditions (e.g., price fluctuation). In China, the prevailing horticultural facility is the traditional solar greenhouse. The key existing problem is the lack of knowledge of growers, which in turn leads to inefficient management, low production, or unsalable products. To secure effective greenhouse management, the production planning system must account for the crop growing environment, grower's activities, and the market. This paper presents an agricultural cyber-physical-social system (CPSS) serving agricultural production management, with a case study on the solar greenhouse. The system inputs are derived from social and physical sensors, with the former collecting the price of agricultural products in a wholesale market, and the latter collecting the necessary environmental data in the solar greenhouse. Decision support for the cropping plan is provided by the artificial system, computational experiment, and parallel execution-based method, with description intelligence for estimating the crop development and harvest time, prediction intelligence for optimizing the planting time and area according to the expected targets (stable production or maximum gross profit), and prescription intelligence for online system training. The presented system fits the current technical and economic situation of horticulture in China. The application of agricultural CPSS could decrease waste in labor or fertilizer and support sustainable agricultural production.

Accurate prediction of crop development stages (phenolgoy) plays an important role in crop production management. Many biochemical actions of plants change with their development stage, therefore, fertilizing the plants at plants' need can increase the fertilizer use efficiency and decrease the waste. For this purpose, this paper presents a prediction model of crop phenology, describing development stages of crops as a function of the most important environmental factors, temperature and photoperiod. Machine learning is used to calibrate the crop phenology according to the predicted temperature one month later. The results indicated that our method can improve the prediction precision. The resulting model will be incorporated in a crop production management system platform developed for agriculture.

Synthetic data augmentation holds substantial research and application value in scenarios characterized by limited samples and high dimensions. It enhances the analytical ability and efficiency of spectral analysis models. This paper proposes Autoencoder-Combined Boundary Equilibrium Generative Adversarial Networks (AE-BEGAN) as a new method for augmenting synthetic data in scenarios with limited samples and high dimensions, with a specific emphasis on near-infrared (NIR) spectral data. The spectral data first undergoes preprocessing procedures that encompass advanced noise reduction algorithms and techniques for removing abnormal samples, guaranteeing elimination of unwanted disturbances and outliers. Then, the pre-processed data is utilized to train the AE-BEGAN model, which generates augmented synthetic samples. Finally, real NIR spectral data obtained from lubricant samples exhibiting different water contents were employed to validate and test the performance of the model. The experimental results demonstrate that the AE-BEGAN model outperforms other GANs in generating synthetic data of high quality and diversity, as quantified by two evaluation metrics, α-Precision and β-Recall with scores of approximately 0.86 and 0.28, respectively. The application case study confirms that the AE-BEGAN model exhibits the capability to generate derived NIR spectra and expand the number of spectra in scenarios with limited samples and high dimensions.

Accurate determination of moisture content in transformer oil is critical for assessing transformer insulation, planning maintenance, extending life expectancy, and reducing failure rates. However, existing detection methods are often limited by high costs, time inefficiencies, susceptibility to human error, and low accuracy. To overcome these challenges, this study proposes an oil moisture content detection method that integrates a convolutional neural network (CNN) with a broad learning system (BLS) using near-infrared spectroscopy (NIR). This approach aims to achieve efficient, precise, rapid, and environmentally friendly quantitative analysis. The research began with the creation of a NIR spectral database for trace water content in transformer oil, followed by the application of the Savitzky-Golay convolution smoothing method for data preprocessing. The CNN was employed to extract features from the preprocessed spectral data, with the output from the penultimate fully connected layer serving as input for the BLS. The model's performance was further enhanced through training and parameter optimization. Validation and testing were conducted using a real NIR spectral dataset of transformer oil with trace water content. Experimental results show that the proposed method outperforms CNN, BLS, and PLSR models, achieving a Root Mean Square Error of Prediction (RMSEP=0.064) and Prediction Coefficient of Determination (R²=0.926). These findings confirm the method's effectiveness and superiority in detecting moisture content in oil.