Search

Net ecosystem productivity (NEP) is a critical indicator of the CO2 capture capacity of vegetation ecosystems. Based on the land classification and clumping index (CI) datasets, the key parameters of the Carnegie–Ames–Stanford Approach (CASA) model, including fraction of photosynthetic active radiation (FPAR) and maximum light use efficiency $({{\rm{\varepsilon }}}_{\text{max}})$, were optimized. Then, the NEP of China's terrestrial ecosystems was estimated, using the improved CASA coupled with a soil respiration model. Finally, the accuracy of NEP estimation was evaluated by observation data from ChinaFLUX station. The research results indicated the RMSE of the improved NEP estimations decreased from 21.139 to 10.179 (unit: $\text{gC} \cdot \text{m}^{- \text{2}} \cdot \text{month}^{ - \text{1}}$), and the R2 value increased from 0.413 to 0.832, indicating that optimizing the parameters ${{\rm{\varepsilon }}}_{\text{max}}$ and FPAR are both effective methods to improve the model. The spatiotemporal variation of China's NEP was analyzed using the optimized results. The NEP value of China shows a decreasing distribution pattern from southeast to northwest, and the values of different regions are in the order South > North > Qinghai–Tibet > Northwest. The monthly NEP variation in different regions of China is a unimodal curve, reaching the maximum in summer. This study optimized the NEP estimation, which can better characterize the distribution pattern of carbon sinks/sources in China's terrestrial ecosystems and lay a scientific foundation for developing regional carbon neutrality schemes.

The realization of intelligent mining is the only method for realizing high-quality development in the coal industry. As the forefront working link in mine production, achieving automatic roadway tunneling control is key to improving production efficiency, enhancing intelligence, and reducing accident rates at fully mechanized tunneling working faces. Among various detection techniques, machine vision technology stands out with advantages of non-contact measurement, rich information acquisition, and high detection accuracy. The detection and control of tunneling equipment groups based on machine vision has become a research hotspot in the intelligence process of coal mines. This study first introduces the key technologies of a visual detection system, including camera calibration, image preprocessing, feature extraction, visual matching, target segmentation and recognition, visual measurement, and 3D reconstruction. It then elaborates on detection principles, workflows, limitations, precautions, and development status of various vision detection systems in practical application scenarios at tunneling faces, such as tunneling equipments, anchoring systems, transportation systems, and safety auxiliary systems, which significantly improve production safety and efficiency. Finally, considering challenging work conditions and strong interference in mines, the successful adaptation of machine vision to excavation sites relies on addressing technical challenges related to poor environment adaptability, limited imaging field of view, and low intelligence level. Furthermore, according to existing research results and the current technical status, this paper forecasts key technologies that need to be developed in the future for coal mine intelligent equipment systems based on machine vision, including multi-sensor information fusion, equipment group collaborative control, and digital twin-driven remote monitoring.

With the continuous development of information technology, data centers (DCs) consume significant and ever-growing amounts of electrical energy. Renewable energy sources (RESs) can act as clean solutions to meet this requirement without polluting the environment. Each DC serves numerous users for their data service demands, which are regarded as flexible loads. In this paper, the willingness to pay and time sensitivities of DC users are firstly explored, and the user-side demand response is then devised to improve the overall benefits of DC operation. Then, a Stackelberg game between a DC and its users is proposed. The upper-level model aims to maximize the profit of the DC, in which the time-varying pricing of data services is optimized, and the lower-level model addresses user's optimal decisions for using data services while balancing their time and cost requirements. The original bi-level optimization problem is then transformed into a single-level problem using the Karush-Kuhn-Tucker optimality conditions and strong duality theory, which enables the problem to be solved efficiently. Finally, case studies are conducted to demonstrate the feasibility and effectiveness of the proposed method, as well as the effects of the time-varying data service price mechanism on the RES accommodation.

Net ecosystem productivity (NEP) is a key indicator to describe terrestrial ecosystem functions and carbon sinks. The CASA model was improved by optimizing the parameters optimum temperature and maximum light use efficiency (${\varepsilon }_{\max }$), and the NEP value of the European terrestrial ecosystem was calculated by combining the soil respiration model. The results showed that when using vegetation classification data to optimize parameter ${\varepsilon }_{\max }$, the R2 of NEP between estimates and observations increased from 0.252 to 0.403, and the RMSE decreased from 84.557 to 64.466 gC·m−2·month−1. After further optimizing the optimum temperature, R2 increased to 0.428, and the RMSE decreased to 63.720 gC·m−2·month−1. It indicated that the CASA model could be improved by optimizing ${\varepsilon }_{\max }$ as well as optimum temperature, which was a good approach to improve the NEP estimations. Based on this, the NEP spatiotemporal changes in various regions of Europe were analyzed using the optimization results. The NEP values of European terrestrial ecosystem has regional differences, showing a pattern of western region > southern region > central region > eastern region > northern region. The monthly change of NEP in each region is a single peak curve with high in summer and low in winter, and the annual overall value is positive (i.e., it shows a carbon sink). The research can enable us to obtain the carbon source/sink distribution information in Europe more accurately and provide a scientific reference for carbon balance policy formulation in the region.

A distributed secondary control (DSC) strategy that combines $Q$-learning and pinning control is originally proposed to achieve a fully optimal DSC for droop-controlled microgrids (MGs). It takes advantages of cross-fusion of the two algorithms to realize the high efficiency and self-adaptive control in MGs. It has the following advantages. Firstly, it adopts the advantages of reinforcement learning in autonomous learning control and intelligent decision-making, driving the action value of pinning control for feedback adaptive correction. Secondly, only a small part of points selected as pinned points needs to be controlled and pre-learned, hence the actual control problem is transformed into a synchronous tracking problem and the installation number of controllers is further reduced. Thirdly, the pinning matrix can be modified to adapt to plug-and-play operation under the distributed control architecture. Finally, the effectiveness and versatility of the proposed strategy are demonstrated with a typical droop-controlled MG model.

With the development of Mobile Edge Computing (MEC), it has become a key technology to realize the vision of the Internet of Things. In MEC, users can upload tasks to edge nodes for faster processing speed and lower local energy consumption. However, as the mobility of users and the limited resources of the edge nodes, some edge nodes cannot provide high-quality services. In this case, we study service migration strategy in the MEC system to migrate services from the initial nodes to other edge nodes that can provide services to meet the needs of users. By making service migration decision and allocating computation resource, our work minimizes the delay and the energy consumption caused by finishing tasks. Specifically, we set up an efficient service migration model and formulate the service migration problem as a non-linear 0-1 programming problem. To solve this problem, we design a Particle Swarm based Service Migration scheme (PSSM) which includes Queuing Delay Prediction algorithm (QDP), Delay-aware Computation Resource Allocation algorithm (DCRA), and Modified Quantum Particle Swarm algorithm (MQPS). For evaluating the performance of the proposed PSSM, we conduct simulation in a practical scenario. The results demonstrate that our scheme not only can effectively reduce delay and energy consumption, but also improve the processing capability of servers.

Leaf area index (LAI) is a valuable indicator used in vegetation growth monitoring. To optimize the index selection according to the type of remote sensing data and to improve the inversion accuracy of LAI, this article analyzes the influence of different bandwidths on the accuracy of the inversion model based on vegetation indices. First, the simulation dataset is generated by the PROSAIL model, and on this basis, 15 vegetation indices with high correlation coefficients with LAI. Then, by analyzing the sensitivity of these 15 indices to the variations in bandwidth, and to the coefficient of determination (R2) of the LAI inversion model with the variations of bandwidth, the influence of different bandwidths on the accuracy of LAI estimation by each index is determined. The results show that bandwidth is one of the most important factors in determining the accuracy of LAI inversion, and the influence on different vegetation indices can be divided into the following three categories. First, narrowband vegetation index, the accuracy of inversion models built by vegetation indices decreases with the increase of bandwidth, including SR[800,680], OSAVI, MTVI2, SR[752,690], RDVI, NDCI, and NVI. Second, middleband vegetation index, the accuracy first increases and then decreases with the increase of bandwidth, including SR[700,670], Carte5, and SR[675,700]. Third, broadband vegetation index, the accuracy increases with the increase of bandwidth, including SPVI, Carte2, OSAVI2, MTVI1, and NDVI705. The study provides a scientific basis for vegetation index optimization in the process of LAI inversion.

This paper proposes a computation offloading scheme for precedence-constrained tasks in a base station-assisted device-to-device (D2D) scenario for the information-centric Internet of Things (IC-IoT). When specified precedence among subtasks cannot be described as simple sequential or parallel relations in a task, the selection of task execution helper for subtasks offloading becomes complex due to the constraints of latency and resources. We define this type of precedence and aim to minimize the time and financial cost of computation task offloading for the user by optimizing subtask-helper pairs. This problem is modeled as a dynamic generalized multi-resource-constrained assignment problem. The optimal offloading policy is offered by searching minimum weight matchings in a bipartite graph. Computer simulations indicate the effectiveness of the proposed approach compared with the random helper selection and priority-based offloading scheme.

Ultra-dense networks (UDNs), which can provide extremely high throughput and data rates, have been considered as one of the key techniques for the fifth generation mobile networks. However, it may cause severe inter-cell interference and significant energy consumption due to numerous base stations (BSs) being randomly deployed. To mitigate the interference and boost energy efficiency (EE) of the UDN effectively, we propose a cluster-based energy-efficient resource allocation scheme in this paper. The proposed scheme has two stages: clustering stage and resource allocation stage. In clustering stage, we use a modified K-means algorithm in BS-clustering process to dynamically adjust the number of BS-clusters based on the density of BSs. Then, in each BS cluster, we divide user equipments (UEs) into multiple UE-groups with minimum intra-cluster interference. In this way, the complexity of resource allocation can be greatly reduced. While in resource allocation stage, we design a two-step resource blocks assignment algorithm and an iterative energy efficient power allocation algorithm based on a non-cooperative game. Furthermore, we implement simulations under the realistic broadband channel propagation conditions and the simulation results show that the proposed approach can effectively mitigate the interference and improve the EE of UDN.