The optimal power flow (OPF) problem can be rapidly and reliably solved by employing responsive online solvers based on neural networks. The dynamic nature of renewable energy generation and the variability of power grid conditions necessitate frequent neural network updates with new data instances. To address this need and reduce the time required for data preparation time, we propose a semisupervised learning framework aided by data augmentation. In this context, ridge regression replaces the traditional solver, facilitating swift prediction of optimal solutions for the given input load demands. In addition, to accelerate the backpropagation during training, we develop novel batch-mean gradient estimation approaches along with a reduced branch set to alleviate the complexity of gradient computation. Numerical simulations demonstrate that our neural network, equipped with the proposed gradient estimators, consistently achieves feasible and near-optimal solutions. These results underline the effectiveness of our approach for practical implementation in real-time OPF applications.

Nonintrusive load monitoring (NILM) identifies the status and power consumption of various household appliances by disaggregating the total power usage signal of an entire house. Efficient and accurate load monitoring facilitates user profile establishment, intelligent household energy management, and peak load shifting. This is beneficial for both the end users and utilities by improving the overall efficiency of a power distribution network. Existing approaches mainly focus on developing an individual model for each appliance. Those approaches typically rely on a large amount of household-labeled data that are hard to collect. In this article, we propose a multi-appliance-task framework with a training-efficient sample augmentation (SA) scheme that boosts the disaggregation performance with limited labeled data. For each appliance, we develop a shared-hierarchical split structure for its regression and classification tasks. In addition, we also propose a 2-D attention mechanism in order to capture spatio-temporal correlations among all appliances. With only one-day training data and limited appliance operation profiles, the proposed SA algorithm can achieve comparable test performance to the case of training with the full dataset. Finally, simulation results show that our proposed approach features a significantly improved performance over many baseline models. The relative errors can be reduced by more than 50% on average.

Non-convex AC optimal power flow (AC-OPF) is a fundamental optimization problem in power system analysis. The computational complexity of conventional solvers is typically high and not suitable for large-scale networks in real-time operation. Hence, deep learning based approaches have gained intensive attention to conduct the time-consuming training process offline. Supervised learning methods may yield a feasible AC-OPF solution with a small optimality gap. However, they often need conventional solvers to generate the training dataset. This paper proposes an end-to-end unsupervised learning based framework for AC-OPF. We develop a deep neural network to output a partial set of decision variables while the remaining variables are recovered by solving AC power flow equations. The fast decoupled power flow solver is adopted to further reduce the computational time. In addition, we propose using a modified augmented Lagrangian function as the training loss. The multipliers are adjusted dynamically based on the degree of constraint violation. Extensive numerical test results corroborate the advantages of our proposed approach over some existing methods.

This paper develops a novel graph convolutional network (GCN) framework for fault location in power distribution networks. The proposed approach integrates multiple measurements at different buses while taking system topology into account. The effectiveness of the GCN model is corroborated by the IEEE 123 bus benchmark system. Simulation results show that the GCN model significantly outperforms other widely-used machine learning schemes with very high fault location accuracy. In addition, the proposed approach is robust to measurement noise and data loss errors. Data visualization results of two competing neural networks are presented to explore the mechanism of GCNs superior performance. A data augmentation procedure is proposed to increase the robustness of the model under various levels of noise and data loss errors. Further experiments show that the model can adapt to topology changes of distribution networks and perform well with a limited number of measured buses.

A systematic framework is put forth in this paper to integrate renewable energy sources (RES), distributed storage units, cooling facilities, as well as dynamic pricing into the workload and energy management tasks for a data center network. To cope with RES uncertainty, the resource allocation task is formulated as a robust optimization problem minimizing the worst-case net cost. The resulting problem is reformulated as a convex program, and then solved in a distributed fashion using the dual decomposition approach. Numerical tests demonstrate the performance gain of the proposed approach over the existing alternative.

A large number of geo-distributed data centers begin to surge in the era of data deluge and information explosion. To meet the growing demand in massive data processing, the infrastructure of future data centers must be energy-efficient and sustainable. Facing this challenge, a systematic framework is put forth in this paper to integrate renewable energy sources (RES), distributed storage units, cooling facilities, as well as dynamic pricing into the workload and energy management tasks of a data center network. To cope with RES uncertainty, the resource allocation task is formulated as a robust optimization problem minimizing the worst-case net cost. Compared with existing stochastic optimization methods, the proposed approach entails a deterministic uncertainty set where generated RES reside, thus can be readily obtained in practice. It is further shown that the problem can be cast as a convex program, and then solved in a distributed fashion using the dual decomposition method. By exploiting the spatio-temporal diversity of local temperature, workload demand, energy prices, and renewable availability, the proposed approach outperforms existing alternatives, as corroborated by extensive numerical tests performed using real data.

Due to increasing threats of global warming and climate change concerns, green wireless communications have recently drawn intense attention toward reducing carbon emissions. Aligned with this goal, the present paper deals with dynamic energy management for smart-grid powered coordinated multipoint (CoMP) transmissions. To address the intrinsic variability of renewable energy sources, a novel energy transaction mechanism is introduced for grid-connected base stations that are also equipped with an energy storage unit. Aiming to minimize the expected energy transaction cost while guaranteeing the worst-case users’ quality of service, an infinite-horizon optimization problem is formulated to obtain the optimal downlink transmit beamformers that are robust to channel uncertainties. Capitalizing on the virtual-queue-based relaxation technique and the stochastic dual-subgradient method, an efficient online algorithm is developed yielding a feasible and asymptotically optimal solution. Numerical tests with synthetic and real data corroborate the analytical performance claims and highlight the merits of the novel approach.

Optimal resource management for smart grid powered multi-input multi-output (MIMO) systems is of great importance for future green wireless communications. A novel framework is put forth to account for the stochastic renewable energy sources (RES), dynamic energy prices, as well as random wireless channels. Based on practical models, the resource allocation task is formulated as an optimization problem that aims at maximizing the weighted sum-rate of the MIMO broadcast channels. A two-way transaction mechanism and storage units are introduced to accommodate the RES variability. In addition to system operating constraints, a budget threshold is imposed on the worst-case energy transaction cost due to the possibly adversarial nature. Capitalizing on the uplink-downlink duality and the Lagrangian relaxation-based subgradient method, an efficient algorithm is developed to obtain the optimal strategy. Generalizations to the setups of time-varying channels and ON-OFF transmissions are also discussed. Numerical results are provided to corroborate the merits of the novel approaches.

The paper deals with dynamic energy management for smart-grid powered coordinated multi-point (CoMP) transmissions. Aiming to minimize the expected energy transaction cost while guaranteeing the worst-case users' quality of service (QoS), an infinite-horizon optimization problem is formulated to obtain the optimal downlink transmit beamformers that are robust to channel uncertainties. Capitalizing on the virtual-queue based relaxation technique and the stochastic dual-subgradient method, an efficient online algorithm is developed in this context. Without a-priori knowledge of any statistics of the underlying random processes, it is rigorously established that the proposed algorithm is able to yield a feasible and asymptotically optimal solution.

A framework is introduced to integrate renewable energy sources (RES) and dynamic pricing capabilities of the smart grid into beamforming designs for coordinated multipoint (CoMP) downlink communication systems. To this end, novel models are put forth to account for harvesting, storage of nondispatchable RES, time-varying energy pricing, and stochastic wireless channels. Building on these models, robust energy management and transmit-beamforming designs are developed to minimize the worst-case energy cost subject to the worst-case user QoS guarantees for the CoMP downlink. Leveraging pertinent tools, this task is formulated as a convex problem. A Lagrange dual-based subgradient iteration is then employed to find the desired optimal energy-management strategy and transmit-beamforming vectors. Numerical results are provided to demonstrate the merits of the proposed robust designs.

Advanced data analytics are undoubtedly needed to enable the envisioned smart grid functionalities. Towards that goal, modern statistical learning tools are developed for day-ahead electricity market inference. Congestion patterns are modeled as rank-one components in the matrix of spatio-temporal prices. The new kernel-based predictor is regularized by the square root of the nuclear norm of the sought matrix. Such a regularizer not only promotes low-rank solutions, but it also facilitates a systematic kernel selection methodology. The non-convex optimization problem involved is efficiently driven to a stationary point following a block successive upper bound minimization approach. Numerical tests on real high-dimensional market data corroborate the interpretative merits and the computational efficiency of the novel method.

Beamforming for multi-input multi-output (MIMO) cognitive networks is considered in the presence of channel uncertainty induced by errors in estimating cognitive-to-primary channels. A robust beamforming problem is formulated to optimize an appropriate cognitive radio network-wide performance metric, while enforcing protection of the primary system. In spite of the non-convexity of the resultant optimization problem, a block coordinate ascent algorithm is developed with provable convergence to a stationary point. Enticingly, the novel scheme also lends itself naturally to a distributed implementation. Numerical results are reported to corroborate the analytical findings.