The high penetration of renewable energy sources, coupled with decommissioning of conventional power plants, leads to the reduction of power system inertia. This has negative repercussions on the transient stability of power systems. The purpose of this paper is to review the state-of-the-art regarding the application of artificial intelligence to the power system transient stability assessment, with a focus on different machine, deep, and reinforcement learning techniques. The review covers data generation processes (from measurements and simulations), data processing pipelines (features engineering, splitting strategy, dimensionality reduction), model building and training (including ensembles and hyperparameter optimization techniques), deployment, and management (with monitoring for detecting bias and drift). The review focuses, in particular, on different deep learning models that show promising results on standard benchmark test cases. The final aim of the review is to point out the advantages and disadvantages of different approaches, present current challenges with existing models, and offer a view of the possible future research opportunities.

Increased integration of renewable energy sources brings new challenges to the secure and stable power system operation. Operational challenges emanating from the reduced system inertia, in particular, will have important repercussions on the power system transient stability assessment (TSA). At the same time, a rise of the “big data” in the power system, from the development of wide area monitoring systems, introduces new paradigms for dealing with these challenges. Transient stability concerns are drawing attention of various stakeholders as they can be the leading causes of major outages. The aim of this paper is to address the power system TSA problem from the perspective of data mining and machine learning (ML). A novel 3.8 GB open dataset of time-domain phasor measurements signals is built from dynamic simulations of the IEEE New England 39-bus test case power system. A data processing pipeline is developed for features engineering and statistical post-processing. A complete ML model is proposed for the TSA analysis, built from a denoising stacked autoencoder and a voting ensemble classifier. Ensemble consist of pooling predictions from a support vector machine and a random forest. Results from the classifier application on the test case power system are reported and discussed. The ML application to the TSA problem is promising, since it is able to ingest huge amounts of data while retaining the ability to generalize and support real-time decisions.

In this paper, a customizable low-cost voltage waveform generator based on a real-time desktop PC and embedded data acquisition card synchronized with Coordinated Universal Time (UTC) is presented. A software approach to phase-locked loop synchronization with an external Global Positioning System (GPS) pulse signal is utilized to achieve a time uncertainty of $\pm 1\mathsf{\mu }$s. This avoids expensive hardware modules for synchronization and timing purposes, which are commonly presented in literature. Besides the application for controlling the test platform, our own phasor data concentrator (PDC) application is running concurrently on the host PC. The latter is used for collecting and comparing the syncrophasor data from the test platform against the syncrophasor data measured by phasor measurement units (PMUs) under the test. The paper describes all procedures for generating reference test signals. Numerous case studies were performed, and experimental results for steady-state compliance as well as frequency ramp and phase modulation tests for dynamic compliance are presented in detail. All tests confirm that customizable test platform meets the requirements of IEEE/IEC standards. Compared to other calibrators, the cost as well as the specifications and point-by-point concept of data processing makes the described test platform suitable for performance analysis of PMU algorithms implemented on various development boards.