Straggler nodes are a major bottleneck in large-scale distributed training, degrading efficiency and stability. However, current solutions, including In-Network Aggregation (INA), lack the adaptability to effectively manage these stragglers in dynamic environments. This paper proposes Straggler Dynamic Management (SDM), an adaptive method for large-scale distributed training that performs dynamic straggler management by coordinating the data and control planes to achieve accurate, time-based detection and efficient mitigation via a performanceaware redundancy strategy and semi-asynchronous aggregation. SDM manages stragglers through a coordinated architecture that decouples the data and control planes for efficient detection and response. It leverages the data plane to estimate each node's remaining completion time, ensuring accurate and low-overhead straggler identification. The control plane then mitigates their impact using two key strategies: a performance-aware redundancy scheme to reduce waiting delays, and a semi-asynchronous aggregation mechanism that dynamically adjusts synchronization to alleviate gradient staleness and improve model convergence. We implement and deploy SDM on a real-world hardware testbed and evaluate its performance under various straggler scenarios. Experimental results demonstrate that SDM significantly improves training efficiency and convergence stability in the presence of straggler nodes, particularly when multiple stragglers occur simultaneously, exhibiting greater robustness and adaptability than existing methods.

With economic development and the increasing electricity demand, distribution network operation has become indispensable for maintaining the power system reliability. However, fault recovery planning for distribution network still faces challenges such as human error, redundant workflows, and duplicated work. Large language models (LLMs), which have exceptional semantic understanding and automated generation capabilities, have recently attracted more and more attention. In this paper, we propose a novel conversational agent based on the mainstream LLMs for fault recovery plan generation. Besides, we introduce a novel tool-learning method that integrates various functionalities, encompassing topology querying, power flow calculations, and formatted text generation. Experiments demonstrate that the fault recovery plan generation agent can effectively leverage the integrated tools, achieving an average success rate of 99.25% in tool invocation.

Dynamic state estimation (DSE) of synchronous machines is essential to real-time monitoring, protection, and control of power systems. DSE can be significantly affected by bad data due to outliers, cyber attack and model uncertainties. This paper proposes a new robust adaptive fading (AF) unscented Kalman filter (UKF) for DSE, which utilizes the AF-UKF to minimize possible scale mismatches in the state and measurement noise covariance matrices of the KF to mitigate these uncertainties. A robust extension of the AF-UKF based on robust statistics is also developed to effectively detect and suppress bad data at each KF update. The proposed method was evaluated and compared with conventional algorithms on the Northeastern Power Coordinating Council 48-machine 140-bus system. Results showed that the proposed decentralized DSE algorithm yields more accurate and reliable performance than conventional methods under bad-data and noise covariance mismatches.

The stable skeletal structure formed by the interlocking mechanism of the spatial movement of aggregate particles during compaction is the mechanism by which asphalt pavements are compacted and shaped. A major problem facing conventional compaction monitoring is that lacking the ability to monitor particle movement and a real-time evaluation method for pavement compaction, which can easily lead to problems such as uneven compaction and over-compaction. In addition, vehicle speed is an important parameter for analysing the dynamic response of a pavement. Current vehicle speed monitoring relies on complex field equipment and the collected vehicle speed parameters are difficult to match with other sensors for fusion analysis. In this paper, a method for monitoring the compaction quality of asphalt pavements and a method for collecting vehicle speed is proposed based on intelligent aggregate and aggregates interaction mechanisms. The experimental results show that with the continuous action of the compaction machinery, the compaction state of the asphalt pavement can be reliably captured and monitored by studying the spatio-temporal movement pattern of the intelligent aggregates. Meanwhile, the analysis of the time nodes of intelligent aggregate attitude change under the action of vehicle load allows accurate acquisition of vehicle speed parameters. Therefore, the method proposed in this study can improve the accuracy of asphalt pavement compaction state control as well as the accurate acquisition of vehicle speed. It can provide a reference for the later analysis of the dynamic response of pavement loads and the intelligent compaction of pavements.

Multiple channels in Wireless Sensor Networks (WSNs) are often exploited to support parallel transmission and to reduce interference. However, the extra overhead posed by the multi-channel usage coordination dramatically challenges the energy-constrained WSNs. In this paper, we propose a Regret Matching based Channel Assignment algorithm (RMCA) to address this challenge, in which each sensor node updates its choice of channels according to the historical record of these channels’ performance to reduce interference. The advantage of RMCA is that it is highly distributed and requires very limited information exchange among sensor nodes. It is proved that RMCA converges almost surely to the set of correlated equilibrium. Moreover, RMCA can adapt the channel assignment among sensor nodes to the time-variant flows and network topology. Simulations show that RMCA achieves better network performance in terms of both delivery ratio and packet latency than CONTROL, MMSNand randomized CSMA. In addition, real hardware experiments are conducted to demonstrate that RMCA is easy to be implemented and performs better.

This paper investigates the feasibility of using a discriminate pricing scheme to offset the inconvenience that is experienced by an energy user (EU) in trading its energy with an energy controller in smart grid. The main objective is to encourage EUs with small distributed energy resources (DERs), or with high sensitivity to their inconvenience, to take part in the energy trading via providing incentive to them with relatively higher payment at the same time as reducing the total cost to the energy controller. The proposed scheme is modeled through a two-stage Stackelberg game that describes the energy trading between a shared facility authority (SFA) and EUs in a smart community. A suitable cost function is proposed for the SFA to leverage the generation of discriminate pricing according to the inconvenience experienced by each EU. It is shown that the game has a unique sub-game perfect equilibrium (SPE), under the certain condition at which the SFA's total cost is minimized, and that each EU receives its best utility according to its associated inconvenience for the given price. A backward induction technique is used to derive a closed form expression for the price function at SPE, and thus the dependency of price on an EU's different decision parameters is explained for the studied system. Numerical examples are provided to show the beneficial properties of the proposed scheme.

In this paper, we study power allocation in cognitive sensor networks where cognitive users (cognitive enabled sensor nodes) opportunistically share a common spectrum with primary users (licensed devices). We define an energy efficiency-oriented utility function as a new metric to evaluate power allocation. Consider that sensor nodes are self-interested to maximize their own utility, we formulate the energy efficient power allocation problem as a non-cooperative game. We firstly prove that there exist Nash equilibriums in the proposed game. Secondly, we prove that the power allocation game is a supermodular game with some conditions. Finally, we use best response algorithm to identify the Nash equilibrium. Simulations are conducted to demonstrate that the proposed power allocation strategy can achieve satisfactory performance in terms of energy efficiency, convergence speed and fairness in cognitive sensor networks.