Effective detection of electricity theft is essential to maintain power system reliability. With the development of smart grids, traditional electricity theft detection technologies have become ineffective to deal with the increasingly complex data on the users’ side. To improve the auditing efficiency of grid enterprises, a new electricity theft detection method based on improved synthetic minority oversampling technique (SMOTE) and improve random forest (RF) method is proposed in this paper. The data of normal and electricity theft users were classified as positive data (PD) and negative data (ND), respectively. In practice, the number of ND was far less than PD, which made the dataset composed of these two types of data become unbalanced. An improved SOMTE based on K-means clustering algorithm (K-SMOTE) was firstly presented to balance the dataset. The cluster center of ND was determined by K-means method. Then, the ND were interpolated by SMOTE on the basis of the cluster center to balance the entire data. Finally, the RF classifier was trained with the balanced dataset, and the optimal number of decision trees in RF was decided according to the convergence of out-of-bag data error (OOB error). Electricity theft behaviors on the user side were detected by the trained RF classifier.

This paper presents an optimized load frequency control (LFC) approach for interconnected power systems with conventional and renewable energy sources. A proportional integral derivative (PID) controller, tuned using particle swarm optimization (PSO) and genetic algorithm (GA) techniques, manages frequency in a thermal two-area tie-line IPS and a one-area multi-source power network. The integral time absolute error (ITAE) serves as the fitness function for optimization. Key contributions include developing sustainable power network models with renewable energy fluctuations, applying PSO and GA for robust PID tuning, and comparing controller performance under varying conditions. The PSO-PID controller shows superior frequency stability, achieving deviations of 59.97 Hz, 59.72 Hz, and 59.99 Hz for single-area multi-source, area 1, and area 2 in the two-area IPS. It outperforms the GA-PID controller in reducing undershoot, overshoot, and settling time, offering a more reliable LFC solution for interconnected power systems.

The deep integration of sensor, computer, and communication networks has dramatically improved the efficiency and operational performance of the electric grid. However, it also exposes the grid to rising threats of cyber-physical attacks. False data injection attack (FDIA) is one of the most representative attacks. To improve the security of grid operation, we propose a Hellinger-distance-based FDIA detection method by tracking the dynamic characteristics of measurement variations at adjacent moments. First, the irrelevant components of measured data are sieved out by empirical modal decomposition. Second, the image transform algorithms are used to deal with the mapping of measurement variations to refine the distribution characteristics. Last, the discrepancies between the probability distributions are derived based on Hellinger distance to determine whether FDIA exists. Concerning state-variable attacks on different nodes, the method is tested using the IEEE 14-bus system. The results indicate that the proposed scheme has high-level detection precision for false data injection attacks.

Voltage fluctuation and power mismatch which are caused by the decentralization, randomness, and intermittence of distributed energy resource along with load variability cause severe impacts on the security and quality of power network operation. Aiming at alleviating this issue, the structure of an AC/DC hybrid microgrid based on solid-state transformer is presented in this paper. A proper control coordination is developed to guarantee stable and reliable operation of the system. An energy management strategy is proposed to coordinate the power flow among the solid-state transformer, AC microgrid, DC microgrid and energy storage and to effectively suppress the fluctuation of the DC bus voltage. A novel adaptive droop control for the energy storage is proposed to prolong the supercapacitor life and achieve an optimum economic benefit to consumers. The droop coefficient is obtained by fuzzy logic controller of which the state of charge of the supercapacitor and unit-time electricity charge are assumed to be the input parameters. Simulation attest the feasibility of the proposed control coordination and energy management strategy.

Electrical power networks consist of numerous energy control zones connected by tie-lines, with the addition of nonconventional sources resulting in considerable variations in tie-line power and frequency. Under these circumstances, a load frequency control (LFC) loop gives constancy and security to interconnected power systems (IPSs) by supplying all consumers with high-quality power at a nominal frequency and tie-line power change. This article proposes employing a proportional–integral–derivative (PID) controller to effectively control the frequency in a one-area multi-source power network comprising thermal, solar, wind, and fuel cells and in a thermal two-area tie-line IPS. The particle swarm optimization (PSO) technique was utilized to tune the PID controller parameters, with the integral time absolute error being utilized as an objective function. The efficacy and stability of the PSO-PID controller methodology were further tested in various scenarios for proposed networks. The frequency fluctuations associated with the one-area multi-source power source and with the two-area tie-line IPS’s area 1 and area 2 frequency variations were 59.98 Hz, 59.81 Hz, and 60 Hz, respectively, and, in all other investigated scenarios, they were less than that of the traditional PID controller. The results clearly show that, in terms of frequency responses, the PSO-PID controller performs better than the conventional PID controller.