From the perspective of energy providers, accurate short-term load forecasting plays a significant role in the energy generation plan, efficient energy distribution process and electricity price strategy optimisation. However, it is hard to achieve a satisfactory result because the historical data is irregular, non-smooth, non-linear and noisy. To handle these challenges, in this work, we introduce a novel model based on the Transformer network to provide an accurate day-ahead load forecasting service. Our model contains a similar day selection approach involving the LightGBM and k-means algorithms. Compared to the traditional RNN-based model, our proposed model can avoid falling into the local minimum and outperforming the global search. To evaluate the performance of our proposed model, we set up a series of simulation experiments based on the energy consumption data in Australia. The performance of our model has an average MAPE (mean absolute percentage error) of 1.13, where RNN is 4.18, and LSTM is 1.93.

Random transmission delay and packet loss may cause load frequency control (LFC) performance degradation or even instability in interconnected power systems. Traditional robust methods mainly focus on guaranteeing the asymptotic stability with pre-estimate a maximum delay case. As a result, the designed controller cannot satisfy the actual operational requirements completely since the improper network performance estimation. In this paper, based on jump system theory, the state-space model for LFC system with the different transmission delay is established as the jumping decision variable, which describes the possible dynamics of the networked LFC system with the bounded time delay. Secondly, in order to avoid the defect of conservative or optimistic design caused by imprecise pre-set delay, the upper boundary of the transmission delay in the power communication system is calculated by deterministic network theory. Furthermore, the asymptotic stability constraints in bilinear matrix inequality forms are deduced via construction of a Lyapunov function. Moreover, to reduce the peak value, peak time and setting time of frequency deviations caused by power mismatches, an iterative optimization algorithm is proposed to obtain the optimal feedback matrix. Simulation studies with two-area and five-area interconnected power systems show that the proposed method can avoid optimistic or conservative design effectively while the stability and dynamic performance can be ensured simultaneously.

The aim of this paper is to address the current situation where business units in smart grid (SG) environments are decentralized and independent, and there is a conflict between the need for data privacy protection and network security monitoring. To address this issue, we propose a distributed intrusion detection method based on convolutional neural networks–gated recurrent units–federated learning (CNN–GRU–FL). We designed an intrusion detection model and a local training process based on convolutional neural networks–gated recurrent units (CNN–GRU) and enhanced the feature description ability by introducing an attention mechanism. We also propose a new parameter aggregation mechanism to improve the model quality when dealing with differences in data quality and volume. Additionally, a trust-based node selection mechanism was designed to improve the convergence ability of federated learning (FL). Through experiments, it was demonstrated that the proposed method can effectively build a global intrusion detection model among multiple independent entities, and the training accuracy rate, recall rate, and F1 value of CNN–GRU–FL reached 78.79%, 64.15%, and 76.90%, respectively. The improved mechanism improves the performance and efficiency of parameter aggregation when there are differences in data quality.

With the development of digitalisation and intelligence, the power system has been upgraded from the traditional single energy transmission and conversion system to a complex cyber‐physical power system (CPPS) with tightly coupled energy and information flows. The cyber‐physical power system achieves the controllability and observability of the power system through ubiquitous sensing technology, advanced measurement technology, and powerful information processing technology. However, the large number of intelligent electronic device accesses and frequent information interactions in CPPS make it more vulnerable and susceptible to be attacked than any previous single structured system. Viruses and intrusions can attack the CPPS through the cyber subsystem, which in turn can deal a fatal blow to the energy supply physical system. Because of the above problems, malicious attacks against CPPS have been occurring in recent years and have generated a great deal of scholarly attention. This paper precisely focusses on this problem, by profiling the structure of CPPS and the potential threats, conducting an in‐depth analysis of CPPS attack modes from the cyber and physical subsystems, and summarising the three‐level security defence methods for CPPS in detail. Finally, the future technological development prospects of CPPS security research are explicitly addressed, which will provide technical support for building reliable, safe, and robust energy systems.

With the massive deployment of geographically distributed data centers (DCs) and the demand surge for cloud services, their high energy consumption and carbon pollution are increasingly problematic. Therefore, mitigating the harmful effects of the carbon footprint of DCs has become a critical challenge. This paper takes the first step toward analyzing the complementary characteristics of global multi-regional renewable energy sources (RES). There is an opportunity to schedule workloads flexibly and track RES to reduce emissions. Integrating DCs and inter-DC network to form a refined emission model for the trade-off between emission cutting effects from scheduling and carbon costs of workload migration, we propose the spatio-temporal task migration mechanism to pursue low carbon in dual-dimension: This paper shifts intensive workloads to locations sufficient in RES at a coarse scale, and adjust the execution time of the workload in response to real-time RES fluctuations at a fine scale. Thus, the emission overage is shifted and offset by RES in a complementary manner; meanwhile, the acceptance of RES is enhanced. Finally, experiments with real-world data show that our method can optimally coordinate demand with RES and mitigate carbon pollution in geographically distributed DCs, and verify the performance and applicability with various-parameter scenarios.

With the deep coupling between the electrical power systems and cyber systems, there is a increasingly detrimental effect of the communication delays and packet losses on the load frequency control (LFC) performance. Traditional methods mainly focus on guaranteeing the asymptotic stability in maximum delay case but ignore improving the system dynamic performance. The resulting designed controller cannot satisfy the actual operational requirements completely. Hence, this paper investigates a novel decentralized LFC strategy based on switching control theory. First, the networked LFC system is modelled as a series of subsystems with transmission delay as the switching decision variable. Compared with the existing modelling methods, our modelling approach can depict the influences of time delay and packet loss on system dynamics more accurately. Second, a memoryless state-feedback control design scheme is proposed. Design constraints are deduced by constructing a Lyapunov function and then presented in a bilinear matrix inequality form. To improve the system dynamic performance in power mismatch and communication change case, an iterative linear matrix inequality (LMI) algorithm is proposed. Finally, simulation results illustrate that our proposed method can restore the frequency deviation to zero quickly.