Traffic prediction on road networks is highly challenging due to the complexity of traffic systems and is a crucial task in successful intelligent traffic system applications. Existing approaches mostly capture the static spatial dependency relying on the prior knowledge of the graph structure. However, the spatial dependency can be dynamic, and sometimes the physical structure may not reflect the genuine relationship between roads. To better capture the complex spatial‐temporal dependencies and forecast traffic conditions on road networks, a multi‐step prediction model named Spatial‐Temporal Attention Wavenet (STAWnet) is proposed. Temporal convolution is applied to handle long time sequences, and the dynamic spatial dependencies between different nodes can be captured using the self‐attention network. Different from existing models, STAWnet does not need prior knowledge of the graph by developing a self‐learned node embedding. These components are integrated into an end‐to‐end framework. The experimental results on three public traffic prediction datasets (METR‐LA, PEMS‐BAY, and PEMS07) demonstrate effectiveness. In particular, in the 1 h ahead prediction, STAWnet outperforms state‐of‐the‐art methods with no prior knowledge of the network.

Distributed renewable energy offers an exciting opportunity for sustainable transition and climate change mitigation. However, it is overlooked in most of the conventional tradable green certificates programs. Blockchain shows an advantage of incorporating a galaxy of distributed prosumers in a transparent and low-cost manner. This paper proposes I-Green, a blockchain-based individual green certificates system for promoting voluntary adoption of distributed renewable energy. Combing the features of blockchain technology and the theories of social norm and peer effects, the novel green ratio incentive scheme and proof of generation consensus protocol are designed for I-Green. A blockchain simulator is constructed to evaluate the effectiveness and efficiency of I-Green system. The simulation results present its potential for facilitating widespread adoption of distributed generation, and confirm the feasibility of blockchain as the information communication technology (ICT).

A critical problem in large neural networks is over parameterization with a large number of weight parameters, which limits their use on edge devices due to prohibitive computational power and memory/storage requirements. To make neural networks more practical on edge devices and real-time industrial applications, they need to be compressed in advance. Since edge devices cannot train or access trained networks when internet resources are scarce, the preloading of smaller networks is essential. Various works in the literature have shown that the redundant branches can be pruned strategically in a fully connected network without sacrificing the performance significantly. However, majority of these methodologies need high computational resources to integrate weight training via the back-propagation algorithm during the process of network compression. In this work, we draw attention to the optimization of the network structure for preserving performance despite compression by pruning aggressively. The structure optimization is performed using the simulated annealing algorithm only, without utilizing back-propagation for branch weight training. Being a heuristic-based, non-convex optimization method, simulated annealing provides a globally near-optimal solution to this NP-hard problem for a given percentage of branch pruning. Our simulation results have shown that simulated annealing can significantly reduce the complexity of a fully connected network while maintaining the performance without the help of back-propagation.

Recently, the ever-increasing development of distributed energy resources has created new opportunities for active distribution networks. In particular, peer-to-peer (P2P) energy management systems facilitate the coordination of prosumers for higher efficiency and flexibility. However, in distributed P2P energy management problems, Byzantine attacks may exist and lead to nonconvergence and infeasible solutions, and the computation performance needs to be further improved. Therefore, a Byzantine-resilient distributed P2P energy management approach is developed to minimize the overall cost of prosumers, considering various constraints such as output bounds, power line congestion, and voltage magnitude limits. The proposed distributed optimization method adopts heavy-ball momentum to improve convergence performance. The convergence and optimality of the proposed distributed optimization method are rigorously proven. The effectiveness, superiority, and scalability of the proposed approach are verified in case studies with various types of Byzantine attacks.

Peer-to-peer (P2P) energy trading offers a promising way for prosumers to achieve multi-bilateral trades, further aids the integration of distributed energy resources into distribution networks and facilitates the low-carbon operation of the system. But realizing this potential requires overcoming challenges in model formulation and distributed optimization. This paper presents a novel P2P energy trading framework of carbon-aware prosumers based on carbon intensity analysis, where explicit emission cap constraints are embedded. To alleviate the computational complexity of each prosumer and accelerate the convergence process, an improved distributed algorithm based on the heavy-ball method is proposed, in which the distribution system operator handles the global constraints. Furthermore, a forgetting-based average online learning method and differential privacy mechanism are integrated into the proposed approach to jointly address the uncertainties and privacy issues. Case studies demonstrate that the proposed methodology can mitigate system carbon emissions while ensuring economics, and simultaneously enhances the algorithm’s performance in terms of convergence speed, uncertainty, and privacy.