Grain storage plays a crucial role in safeguarding food security and maintaining market stability, and it has therefore attracted growing attention from both academia and industry. The primary objective of storage technologies is to minimize post-harvest losses caused by pests, mold, and mechanical damage. However, conventional storage management methods, which rely heavily on manual labor, are often inefficient and costly. With the rapid advancement of artificial intelligence (AI), various approaches, such as convolutional neural network (CNN)-based models, Transformer-based frameworks, and emerging Mamba architectures, have been introduced into the field of grain storage. This paper presents a comprehensive review of artificial intelligence methodologies applied across multiple stages of the grain storage process. From four complementary perspectives, including application significance, existing AI techniques, comparative analysis, and future development trends, the review systematically summarizes current progress in pest monitoring, unsound kernel detection, and intelligent aeration. It critically examines their respective advantages and limitations, while outlining key challenges and future research directions. The review aims to offer a global perspective on the integration of AI technologies in grain storage and to foster interdisciplinary collaboration toward the development of intelligent, efficient, and sustainable storage systems.

Multistage robust unit commitment (MRUC) is an important decision-making problem in power system operations. The affine policy facilitates problem-solving, but it compromises flexibility. This letter proposes a partially affine policy for MRUC problem with fast-ramping units; this policy imposes affine relations to coupling variables only and leaves the remaining variables to be optimized in the real-time dispatch. As a result, the real-time flexibility of fast-ramping units is retained. By adopting this approach, MRUC with a partially affine policy becomes a special two-stage adaptive robust optimization problem. Numerical tests verify that the proposed partially affine policy significantly reduces the conservativeness compared with affine policy, improving the dispatch economy and flexibility.

Electricity prices have complex features, such as high frequency, multiple seasonality, and nonlinearity. These factors will make the prediction of electricity prices difficult. However, accurate electricity price prediction is important for energy producers and consumers to develop bidding strategies. To improve the accuracy of prediction by using each algorithms' advantages, this paper proposes a hybrid model that uses the Empirical Mode Decomposition (EMD), Autoregressive Integrated Moving Average (ARIMA), and Temporal Convolutional Network (TCN). EMD is used to decompose the electricity prices into low and high frequency components. Low frequency components are forecasted by the ARIMA model and the high frequency series are predicted by the TCN model. Experimental results using the realistic electricity price data from Pennsylvania-New Jersey-Maryland (PJM) electricity markets show that the proposed method has a higher prediction accuracy than other single methods and hybrid methods.