Search

Developing the deep learning (DL) technique is a promising way to improve renewable energy forecasting accuracy and offset the negative impacts of renewable energy on the power system. However, the application of the DL technique brings novel cyberthreats to the renewable energy forecast, and its cybersecurity has not received enough attention in previous literatures. To fill the gap, the vulnerability of renewable energy forecasting is, among the first, studied in-depth in this article. First, a novel cyberattack named adversarial learning attack (ALA) is proposed. The ALA is achieved by tampering with the meteorological data obtained by online weather forecasts from external application programming interfaces to undermine the renewable energy forecasting performance, which jeopardizes the power system operation. Then, an iterative algorithm is proposed to solve the ALA-based optimization problem. As the DL model is involved as optimization constraints, the optimization problem is nonconvex and NP-hard, which is unable to be solved by traditional approaches. The proposed algorithm utilizes the proximal gradient descent principle and is effective in iteratively exploring the near-optimal solution. At last, the impact of the ALA strategy on the power system operation is assessed, which considers the economic loss incurred and the potential hazards. The feasibility and efficacy of the ALA strategy are validated by conducting comprehensive and extensive experiments on the IEEE 30-bus benchmarks. The simulation results reveal that the ALA is able to impose severe economic losses on the operation and even induces disastrous hazards, such as power system collapse.

In this article, a multiple time-scale voltage regulation is proposed based on the three-level coordinated control approach. This approach aims at solving the voltage issues in different scopes of space-time for active distribution networks. First, to address the global voltage issue of the entire network on a relatively slow-time scale, a multimode switching control is designed based on the Petri-net, which can effectively switch the on-load tap changer to extend its service life. Second, a multiobjective optimization considering the voltage differences with the security boundary and the transmission loss of the entire network is proposed. As such the global voltage issue can be handled by cooperating with the first-level switching control. Third, to solve the local voltage issue on a fast-time scale, a fully distributed optimal control integrating the active/reactive power sharing of all the distributed units is proposed. This control facilitates the control efficiency and active power consumption for renewable energy sources. Finally, the efficiency and effectiveness of the proposed method are validated under different scenarios in case studies.

An aggregation scheme is an effective transactive manner of Distributed Energy Resources (DER) spreading across distribution networks. Distributed approach locally achieves cost minimization of an aggregator and customers. The uncertainties of wholesale market price and rooftop PV output will impact on aggregator's scheduling decision and each customer's cost, while solar energy fluctuation can cause an overvoltage problem in distribution networks. However, the probability distributions of these uncertainties always have errors, even in emerging data-based methods. There is no stochastic method using real data with an out-of-sample guarantee suitable for this distributed approach so far to help an aggregator avoid price risk and manage customers’ energy against solar energy fluctuation. To address these unsolved issues, we propose a data-driven Wasserstein distributionally robust formulation of the aggregator's agent and customer's agent respectively. The Wasserstein metric is employed to construct the Wasserstein ambiguity set. The mathematical models are then reformulated equivalently to convex programming respectively so that the operating model can be solved by the off-the-shelf solver. To improve the efficiency of the distributed solving framework, an alternating optimization procedure (AOP) process is proposed to overcome the issue caused by binary variables in the alternating direction method of multipliers (ADMM). The proposed operation framework is verified on the modified IEEE 33-bus distribution network and realistic single-feeder LV network.

Dear Editor, This letter presents a novel and efficient adversarial robustness verification method for tree-based smart grid dynamic security assessment (DSA). Based on tree algorithms technique, the data-driven smart grid DSA has received significant research interests in recent years. However, the well-trained tree-based DSA models with high accuracy are always vulnerable caused by some physical noises or attacks, which can misclassify the DSA results. Only with the accuracy index is not enough to represent the performance of the tree-based DSA models. To provide formal robustness guarantee and select the trusted tree-based DSA models, this letter proposes an efficient adversarial robustness verification strategy with a sound robust index to quantify the ability of tree-based DSA models against any adversarial attack. Analysis results verifies the proposed strategy can achieve up to ∼564X speedup.