The role of artificial intelligence in energy aspects of super cold chain of agricultural products

Unofficial AI-generated summary based on the public title and abstract. Not an official translation.

🔬研究者:Provides a framework for applying AI to cold-chain energy optimization, highlighting research gaps in agricultural logistics.

🏢実務担当者:Offers insights into AI-based energy management systems for cold-chain operators to reduce costs and emissions.

🏛政策担当者:Suggests policy support for AI adoption in cold chains to meet national decarbonization targets.

Fruits and vegetables rich in vitamins are important food sources for the quality of human life and health. Cold chain refers to the logistics and transportation method in that food is always under the specified low-temperature environment in all aspects of production, processing, storage, transportation and sales to ensure food quality and reduce transportation loss.1 By studying the internal microscopic mechanism of agricultural products and analyzing the heat and mass transfer process of the whole cold chain process, the super cold chain can minimize the consumption of internal substances in fruits and vegetables to maintain the freshness of agricultural products.2 With the rapid development of the cold chain, the energy consumption and environmental problems brought by it have become the focus of our attention. How to reasonably and effectively save energy loss, and ensure the quality of agricultural products has become a major problem that the cold chain urgently needs to solve. The cold chain logistics with low energy consumption, low emission and high efficiency as the main characteristics has become an inevitable trend in the development of agricultural products preservation. Artificial intelligence (AI) and Internet of Things (IoT) are being revolutionized in several fields because of their powerful monitoring and rapid decision-making capabilities. Essentially, machine learning is an approach to AI where algorithms parse data and learn from it to decide predictions about actual events.3 In addition, deep learning is a method to implement machine learning. As a derivative of artificial neural networks, deep learning can automatically extract large amounts of data for analysis and processing.4 In the agricultural supply chain, many researchers apply AI to the prediction of crop yield and soil properties, but there is a lack of discussion on the application of AI in the energy aspects of the agricultural cold chain. This paper presents the current problems concerning energy utilization in the agricultural cold chain and proposes the necessity of applying AI to assist in solving them. Secondly, it details the methods of AI to solve the cold chain energy problems, dealing with energy system planning and operation from multiple perspectives. Finally, it makes further analysis of the challenges and opportunities of the new generation of the super cold chain, hoping to provide a reference for the sustainable and low-carbon development of agricultural products cold chain logistics. With the development of the deep processing of agricultural products and the improvement of people's living standards, the demand for cold chain logistics has increased significantly. When studying the energy optimization of cold chain logistics of fresh and live agricultural products, on the one hand, we have to consider the transportation cost of room temperature logistics. On the other hand, it should also consider the cost of loss of fresh agricultural products in the process of circulation and the cost of energy consumed by the refrigeration equipment of transport vehicles. The traditional cold chain of agricultural products is limited by its technical conditions as well as the environment, which makes it difficult to achieve the best effect. Figure 1 lists the major energy problems that currently exist in the agricultural cold chain. Studies have shown that carbon emissions in cold chain transport are 30% higher than in ambient logistics,5 and the energy consumption of the refrigeration system in the cold chain process accounts for approximately 80% of the total energy consumption. Therefore, the refrigeration system that provides a low-temperature environment for the cold chain is the primary factor in the energy consumption analysis. At present, there are problems such as backward refrigeration technology, unreasonable ventilation and poor insulation of the refrigerator. Many factors affect the refrigeration effect, such as temperature, humidity, ventilation methods, thermal insulation performance and the breathing heat of agricultural products.6 The inability to guarantee a strict low-temperature environment increases product spoilage and results in a significant waste of energy and greenhouse gas emissions. Secondly, the transportation energy consumption of cold chain logistics is also the major consumer of energy. Because of the lack of reasonable layout and planning for the origin and related distribution routes, the cold chain transportation process consumes more energy. In addition, during the transportation process, the distance is often prolonged or the refrigerated vehicles return empty-loaded, resulting in a rapid increase in energy consumption. According to statistics, the impact of global carbon emissions from cold vehicles accounts for over 40% of the world's greenhouse gases.7 Finally, there is a lack of a complete cold chain energy management system. In cold chain logistics, the energy load is constantly changing with the cold chain link, application equipment, operating time and control conditions. Low organization among the various links of cold chain logistics has led to an increase in ineffective carbon emissions. Therefore, how to reduce the transportation time and energy cost of fresh and live agricultural products have become an urgent problem to be solved. Researchers have proposed many solutions to the above energy problems encountered in the cold chain of agricultural products. But agricultural products vary in composition, size and shape; there is no single approach to solving the cold chain logistics problem for all products. Recently, several techniques have been proposed and developed, such as cloud computing and machine learning, which make the application of AI in the cold chain possible. IoT data collection technology can collect massive data and monitor logistics objects in real-time. Machine learning has efficient data processing capabilities and can optimize energy use. Therefore, it is necessary to make reasonable considerations for the application of AI in energy saving and carbon reduction in the cold chain. The operational control and management of AI in the super cold chain energy of agricultural products can be divided into two aspects. One is the intelligent energy consumption management system involving the energy utilization in the whole process of the cold chain, mainly including intelligent optimization and intelligent coordination. The second is directly related to optimizing the local energy utilization of the cold chain, mainly including intelligent monitoring, intelligent control and intelligent analysis. The perishability and timeliness of fresh agricultural products require higher organizational coordination in all aspects of cold chain logistics. An AI can provide a comprehensive energy-saving solution for the super cold chain. Starting from the overall situation, contemplating the processes of harvesting, processing, transportation, storage, and sales can ensure the low-carbon operation of the super cold chain of agricultural products. Figure 2 represents the energy management architecture of the intelligent cold chain. In the data collection layer, it covers data collection of equipment operating parameters and energy consumption conditions. It also introduced computer vision systems and nondestructive monitoring technologies to monitor quality parameters, such as the freshness and maturity of agricultural products during the cold chain process. Then the collected data are accurately sent to the data center (cloud platform) through wireless sensor networks (WSN), compressed sensing technology and other IoT technologies. Based on the principle of minimizing cost and maximizing efficiency, neural networks and other methods are used to predict and adjust the parameters and efficiency of refrigeration units to reduce the unnecessary energy consumption of equipment. Finally, the analyzed results are converted into instructions and sent to the terminal through the data center to achieve intelligent energy management and full process traceability. Traditional cold chain information collection methods are less efficient, for example, temperature, humidity and other parameters are measured by one-dimensional point detection. However, it is difficult to determine the quality of food in the cold chain process by a large number of sample point detection methods. Thermal imaging, computer vision systems and nondestructive monitoring techniques are future research and development directions for monitoring food quality. Deep learning with feature learning and generalization capabilities can be used for the detection and analysis of complex food matrices.8 Refrigeration system operation, environmental parameters and agriculture quality are all key factors that affect cold chain energy consumption and carbon emissions. As shown in Figure 3, a computer vision system and a deep learning method are combined. First, the computer vision system is used to get the image and simultaneously upload the vehicle's driving status, cargo loading, and other energy consumption and environmental condition parameters. And then the neural network extracts the features and analyzes them, which can accurately distinguish the fruits and vegetables of different varieties and different maturity in the image. The actual energy consumption of the cold chain got from data mining is used to determine the static and dynamic loads of each link in the cold chain by backcasting and forecasting. Finally, the optimal storage and distribution solutions are matched to reduce the cost and energy consumption. With the assistance of emerging technologies such as AI and IoT, it is possible to detect the internal and external quality attributes of products (the presence or absence of diseases, hardness, soluble solids, etc.), and get high-quality agricultural products. Meanwhile, it can dynamically predict the energy utilizat

• openalex https://doi.org/10.1002/er.8031first seen 2026-08-02 17:49:26 · last seen 2026-08-02 17:51:09