Intelligent identification of multiple power quality (PQ) disturbances is very useful for pollution control of power systems. In this study, we propose a novel detection framework for complex PQ disturbances based on multi-fusion convolutional neural network (MFCNN). Our contributions focus on automatic extraction and fusion of features from multiple sources. First, an information fusion structure is introduced in which the time domain and frequency domain information of the PQ disturbance signal are used as inputs. Additionally, the one-dimensional composite convolution is proposed to improve the diversity of network features based on the standard convolution and dilated convolution. Then, to speed up the training and prevent overfitting, batch normalization is used to adjust the distribution of features. Second, we use several visualization methods to resolve the internal mode of MFCNN, and demonstrate the working mechanism of the proposed method. Finally, we conducted various experiments to verify the effectiveness of the MFCNN. Compared with the handcrafted feature design methods and the general convolutional neural network models, the simulation under different noises and hardware platform-based experiments verify the effectiveness of noise immunity, higher training speed, and better accuracy of the method.

This study explores the impact of high-photovoltaic (PV) penetration on the inter-area oscillation modes of large-scale power grids. A series of dynamic models with various PV penetration levels are developed based on a detailed model representing the U.S. Eastern Interconnection (EI). Transient simulations are performed to investigate the change of inter-area oscillation modes with PV penetration. The impact of PV control strategies and parameter settings on inter-area oscillations is studied. This paper finds that as PV increases, the damping of the dominant oscillation mode decreases monotonically. It is also observed that the mode shape varies with the PV control strategy and new oscillation modes may emerge under inappropriate parameter settings in PV plant controls.

With the development and wide deployment of measurement equipment, data can be automatically measured and visualized for situation awareness in power systems. However, the cyber security of power systems is also threated by data spoofing attacks. This letter proposed a measurement data source authentication (MDSA) algorithm based on feature extraction techniques including ensemble empirical mode decomposition (EEMD) and fast Fourier transform (FFT), and machine learning for real-time measurement data classification. Compared with previous work, the proposed algorithm can achieve higher accuracy of MDSA using a shorter window of data from closely located synchrophasor measurement sensors.

Power system frequency disturbances are caused by various generation and transmission events including generator trips, load disconnections, line trips, etc. Accurate detections of the events are crucial to bulk power system situation awareness and event investigation. This paper utilizes the recent advances of deep learning to build a convolutional neural network model to detect events in an accurate yet straightforward manner. In this paper, the rate of change of frequency and the relative angle shift are converted to images as the inputs of the proposed model. Finally, this paper uses two convolutional neural networks and classifier fusion to achieve the detection result. Compared with the conventional event detection algorithm and the frequency only deep learning model, the proposed model improves the detection accuracy by over 48%. As a promising tool for bulk power system situation awareness, the proposed model requires a short decision time, which is suitable for practical scenarios.

The distribution-level electric network frequency (ENF) extracted from an electric power signal is a promising forensic tool for multimedia recording authentication. Local characteristics in ENF signals recorded in different locations act as environmental signatures, which can be potentially used as a fingerprint for location identification. In this paper, a reference database is established for distribution-level ENF using FNET/GridEye system. An ENF identification method that combines a wavelet-based signature extraction and feedforward artificial neural network-based machine learning is presented to identify the location of unsourced ENF signals without relying on the availability of concurrent signals. Experiments are performed to validate the effectiveness of the proposed method using ambient frequency measurements at multiple geographic scales. Identification accuracy is presented, and the factors that affect identification performance are discussed.

The increase of renewable penetration in power grids calls for loads to participate in frequency regulation to avoid under-frequency load shedding after large resource contingencies. The motivation of this paper is to fulfill a fast residential load control for frequency stability enhancement. For this purpose, a novel mobile distribution-level phasor measurement unit (MDPMU) is developed for disturbance event detection, power mismatch estimation and fast load control. First, the systematic design of the MDPMU introduced. The proposed MDPMU has the major advantages of low cost and high frequency measurement accuracy. Second, the distribution-level measurement is utilized for the rate of change of frequency (ROCOF) calculation. Upon detection of a power system frequency event, power mismatch is estimated at the early stage of the event. Details of a robust approach to calculate ROCOF and determine the event starting point are presented. Practical issues including the measurement reporting rate and estimation time are also considered. Last, with the proposed control system, residential load responses can be controlled coordinately based on load availability and compensation prices offered by customers, providing adaptive frequency regulation service after large resource contingencies. The accuracy of frequency measurement and power mismatch estimation are evaluated via experimental analysis. The frequency stability improvement by load control is validated via PSS/E simulation.