# Artificial Intelligence Techniques for Power System Transient Stability Assessment

^{*}

## Abstract

**:**

## 1. Introduction

## 2. Datasets

#### 2.1. WAMS/PMU Measurement Data

#### 2.2. Simulation-Generated Data

^{™}is often employed for producing these datasets.

#### 2.3. Features Engineering

## 3. AI Model Life-Cycle

**Data processing pipeline example:**Power system transient assessment datasets, which have been generated through simulations of benchmark test cases, usually consist of many thousands of time-domain signals of measured (both electrical and mechanical) quantities for the network elements and machines. These data need to be processed by means of a previously discussed pipeline. Figure 5 graphically presents an example of a data processing pipeline for power system transient assessment data coming from simulations of a benchmark test case [16]. It features statistical processing of systematic simulations data, followed by the train/test dataset split and data scaling process, and concluded with a dimensionality reduction. This particular pipeline assumes that data already consist of features (vectors/tensors of floating point instances), whatever they may be, which will be fed to an ML model. In other words, features engineering is not part of this pipeline and, if undertaken as an independent step, ought to precede it.

**AI cloud infrastructures:**The above-described five steps of the AI model life-cycle are often implemented on a purposefully developed cloud infrastructure, which includes purpose-built tools for every step of the ML development, including labeling, data preparation, feature engineering, statistical bias detection, auto-ML, training, tuning, hosting, explainability, monitoring, and workflows. The user can choose between different cloud AI infrastructure providers, such as: Microsoft Azure (https://azure.microsoft.com/en-us/services/machine-learning/, accessed on 18 November 2021) Machine learning, Amazon AWS SageMaker (https://aws.amazon.com/sagemaker/, accessed on 18 November 2021), Google Vertex AI (https://cloud.google.com/vertex-ai, accessed on 18 November 2021), Huawei MindSpore (https://www.mindspore.cn/en, accessed on 18 November 2021), NVIDIA AI Enterprise (https://www.nvidia.com/en-us/data-center/products/ai-enterprise-suite/, accessed on 18 November 2021) with VMware vSphere, Paperspace (https://www.paperspace.com, accessed on 18 November 2021), and others. All these well-known providers offer fully integrated end-to-end cloud solutions for AI development and hosting. They automate many manual tasks and can even completely eliminate software development from the model building (i.e., auto-ML).

#### 3.1. Machine Learning

#### 3.1.1. Hyperparameter Optimization

**Dealing with a class imbalance problem:**Datasets on which the ML model is being trained, as was already mentioned, have a class imbalance problem, due to the low probability of power system loss from stability events. This imbalance will influence the classifier in favor of the dominant class. In order to remedy this situation, classifiers use class weighting, which puts more weight on samples from the under-represented class. Weights are adjusted inversely proportionally to the class frequencies in the training data. For that, it is important to preserve the level of class imbalance between the training and test datasets, which is why the stratified shuffle split strategy is needed. It should be mentioned that this is different from sample weighting, which can be used in addition to class weighting.

#### 3.1.2. Ensembles

#### 3.1.3. Estimator Performance Metrics

**Machine learning framework example:**The Python (https://www.python.org, accessed on 18 November 2021) programming language is fast becoming a dominant language for data science and artificial intelligence applications. ScikitLearn (https://www.scikit-learn.org, accessed on 18 November 2021), as an open-source Python library, is probably one of the most prominent frameworks for developing traditional machine learning applications. It features a beautifully designed application programming interface (API), which enables building powerful pipelines, diverse ML models, and complex ensembles. It also streamlines many ML-related tasks, such as: data processing, transforming, scaling, splitting, cross-validating estimators, interchanging different metrics and losses, calibrating and evaluating estimators, model performance visualization, and others.

#### 3.2. Deep Learning

**Deep learning framework example:**One of the most prominent open-source deep learning frameworks is TensorFlow 2 (https://www.tensorflow.org, accessed on 18 November 2021) (TF2), developed and maintained by Google. It features a comprehensive and flexible ecosystem of tools and libraries, including a first-class Python API. TF2 offers multiple levels of abstraction, from the high-level Keras API, down to low-level tensor manipulations and gradient computations. It also features efficient data processing structures (designed to handle extremely large datasets) and can be extended with powerful add-on libraries and models, such as TF Probability. Furthermore, TF Extended enables building full production ML pipelines. A model built with TF2 can be trained on CPUs, GPUs, and even TPUs, without any changes. TF2 is also available inside the previously mentioned AI cloud infrastructures.

#### 3.3. Reinforcement Learning

**Reinforcement learning framework example:**One of the most advanced and user-friendly RL environments is the Agents (https://www.tensorflow.org/agents, accessed on 18 November 2021) framework, which is based on the TF2 deep learning ecosystem. The TF-Agents library facilitates designing, implementing, and testing new RL algorithms, by providing well-tested modular components that can be easily modified and extended. This framework may be used to create a simulated environment of the power system for TSA. It already contains many different kinds of agents (including deep Q-learning ones) and provides intuitive ways of designing policies and coding reward functions. Finally, it inherits many of the excellent features of the TF2, on which it stands.

## 4. Challenges and Future Research Opportunities

**Researcher’s resources:**It might be of value to mention here several additional resources at disposal to researchers interested in AI applications to power systems. Datasets for the power system TSA applications can be found on Zenodo (https://www.zenodo.org, accessed on 18 November 2021), which is a premier European general-purpose open data depository, e.g., [85]. General AI and ML research papers are freely available on arXiv (https://arxiv.org/archive/cs, accessed on 18 November 2021), where cs.AI and cs.ML would be the subcategories of most interest. Apart from GitHub, the source code that accompanies published research papers can be found from papers with code (https://paperswithcode.com, accessed on 18 November 2021), which curates ML and AI papers along with their supporting source code repositories. LF AI & Data (https://lfaidata.foundation/, accessed on 18 November 2021) is a growing ecosystem of open-source projects in support of AI development. Hugging Face (https://huggingface.co, accessed on 18 November 2021) can be mentioned as a resource that offers the reuse of many different large pre-trained models. Furthermore, some of the popular pre-trained models for transfer learning can be found directly within the TF2. Finally, a PyTorch (https://pytorch.org, accessed on 18 November 2021) framework, developed by Facebook, can be mentioned here as a capable alternative to TF2. It also offers the Python API and is available inside the previously mentioned AI cloud infrastructures.

## 5. Conclusions

## Author Contributions

## Funding

## Institutional Review Board Statement

## Informed Consent Statement

## Conflicts of Interest

## Abbreviations

AI | Artificial intelligence |

ANN | Artificial neural network |

API | Application programming interface |

AUC | Area under the curve |

AVR | Automatic voltage regulator |

CNN | Convolutional neural network |

CPU | Central processing unit |

DET | Detection error trade-off |

DL | Deep learning |

DNN | Deep neural network |

DQN | Deep Q-network |

GAN | Generative adversarial network |

GPS | Global Positioning System |

GPU | Graphics processing unit |

GRU | Gated recurrent unit |

HYP | Hyperparameter |

LSTM | Long short-term memory |

ML | Machine learning |

MRMR | Maximum–relevance minimum–redundancy |

PID | Proportional integral derivative |

PMU | Phasor measurement unit |

PSS | Power system stabilizer |

RES | Renewable energy sources |

RF | Random forest |

RL | Reinforcement learning |

RNN | Recurrent neural network |

ROC | Receiver operating curve |

SVM | Support vector machine |

TPU | Tensor processing unit |

TSA | Transient stability assessment |

TSI | Transient stability index |

WAMS | Wide area measurement system |

## Appendix A. Selected Research Overview

Dataset | Pipeline | Model | Ref. |
---|---|---|---|

IEEE 39-bus test case | GAN with double generator networks | Multi-layer LSTM network | [69] |

IEEE 39-bus & 68-bus test case | Features engineering (500+ features), PCA | SVM, DT, RF, AdaBoost | [25] |

IEEE 39-bus test case | Feature vectors construction | Autoencoder | [8] |

Guangdong Power Grid system | Spectral quantization, frequency histograms | Deep CNN | [12] |

IEEE 39-bus test case | Representation learning | Stacked GRUs | [65] |

21-bus & 1648-bus system | Features engineering | MRMR ensemble | [50] |

IEEE 39-bus test case | Autoencoder, PCA | SVM | [30] |

IEEE 14-bus test case | Features engineering | Naive Bayes, SVM | [35] |

Central China regional system | Representation learning | Deep belief network | [86] |

IEEE 39-bus test case | Denoising stacked autoencoder | Voting ensemble | [16] |

China regional system | Representation learning | Deep belief network | [75] |

IEEE 39-bus & 300-bus system | Time-domain signals, Adjacency matrix | Recurrent Graph CNN | [19] |

IEEE 39-bus, 68-bus & 140-bus test case | Automatic features construction, features crossing | Extreme gradient boosting (XGBoost) | [46] |

IEEE 39-bus test case | LASSO-based features selection | Stacked Denoising Autoencoder | [29] |

IEEE 39-bus test case | Generative adversarial network | Ensemble | [9] |

IEEE 118-bus & 300-bus system | Gini Index-based feature selection | DT-based ensemble | [41] |

IEEE 39-bus test case | Indirect PCA | ANN | [27] |

IEEE 39-bus & 118-bus system | Bootstrap random features selection | Ensemble of regression trees | [22] |

IEEE 39-bus system with wind farm | Yeo-Johnson power transformation, Z-score | LSTM network | [49] |

IEEE 39-bus test case | Features importance analysis with random forest | SVM with recursive features elimination | [33] |

IEEE 39-bus test case | Time sequence normalization | LSTM network | [63] |

Simulator based PMU data | Time-series data arrangement, feature space mapping | Decision trees | [24] |

IEEE 39-bus test case | Features engineering | Neuro-fuzzy ensemble of SVMs | [40] |

Guangdong Power Grid system | Transforming multiplex trajectories into 2-D images | Two-level CNN | [12] |

IEEE 39-bus test case | (fault-on + 2) trajectories in bitmap forms | LeNet-5 based CNN model | [55] |

China 265-bus AC/DC hybrid power grid | Operation mode and disturbance features | Deep belief network | [77] |

IEEE 39-bus test case | Transfer learning pre-trained on ImageNet | CNN–LSTM hybrid model | [64] |

IEEE 39-bus test case | Artificial features, post-contingency observation window | Stacked GRUs | [67] |

## References

- Bellizio, F.; Cremer, J.; Sun, M.; Strbac, G. A causality based feature selection approach for data-driven dynamic security assessment. Electr. Power Syst. Res.
**2021**, 201, 107537. [Google Scholar] [CrossRef] - Oladeji, I.; Zamora, R.; Lie, T.T. An Online Security Prediction and Control Framework for Modern Power Grids. Energies
**2021**, 14, 6639. [Google Scholar] [CrossRef] - Zhang, S.; Zhu, Z.; Li, Y. A Critical Review of Data-Driven Transient Stability Assessment of Power Systems: Principles, Prospects and Challenges. Energies
**2021**, 14, 7238. [Google Scholar] [CrossRef] - Perilla, A.; Papadakis, S.; Rueda Torres, J.L.; van der Meijden, M.; Palensky, P.; Gonzalez-Longatt, F. Transient Stability Performance of Power Systems with High Share of Wind Generators Equipped with Power-Angle Modulation Controllers or Fast Local Voltage Controllers. Energies
**2020**, 13, 4205. [Google Scholar] [CrossRef] - Zhang, Y.; Liu, W.; Wang, F.; Zhang, Y.; Li, Y. Reactive Power Control Method for Enhancing the Transient Stability Total Transfer Capability of Transmission Lines for a System with Large-Scale Renewable Energy Sources. Energies
**2020**, 13, 3154. [Google Scholar] [CrossRef] - Alimi, O.A.; Ouahada, K.; Abu-Mahfouz, A.M. A Review of Machine Learning Approaches to Power System Security and Stability. IEEE Access
**2020**, 8, 113512–113531. [Google Scholar] [CrossRef] - Sabo, A.; Wahab, N.I.A. Rotor Angle Transient Stability Methodologies of Power Systems: A Comparison. In Proceedings of the 2019 IEEE Student Conference on Research and Development, SCOReD 2019, Bandar Seri Iskandar, Malaysia, 15–17 October 2019; pp. 1–6. [Google Scholar] [CrossRef]
- Azarbik, M.; Sarlak, M. Real-time transient stability assessment using stacked auto-encoders. COMPEL-Int. J. Comput. Math. Electr. Electron. Eng.
**2020**, 39, 971–990. [Google Scholar] [CrossRef] - Ren, C.; Xu, Y. A Fully Data-Driven Method Based on Generative Adversarial Networks for Power System Dynamic Security Assessment with Missing Data. IEEE Trans. Power Syst.
**2019**, 34, 5044–5052. [Google Scholar] [CrossRef] - Phadke, A.G.; Bi, T. Phasor measurement units, WAMS, and their applications in protection and control of power systems. J. Mod. Power Syst. Clean Energy
**2018**, 6, 619–629. [Google Scholar] [CrossRef][Green Version] - Sundararajan, A.; Khan, T.; Moghadasi, A.; Sarwat, A. A Survey on Synchrophasor Data Quality and Cybersecurity Challenges, and Evaluation of their Interdependencies. J. Mod. Power Syst. Clean Energy
**2018**, 7, 449–467. [Google Scholar] [CrossRef][Green Version] - Zhu, L.; Hill, D.J.; Lu, C. Hierarchical Deep Learning Machine for Power System Online Transient Stability Prediction. IEEE Trans. Power Syst.
**2020**, 35, 2399–2411. [Google Scholar] [CrossRef] - Zhu, L.; Hill, D.J.; Lu, C. Intelligent Short-Term Voltage Stability Assessment via Spatial Attention Rectified RNN Learning. IEEE Trans. Ind. Inform.
**2021**, 17, 7005–7016. [Google Scholar] [CrossRef] - Tian, Y.; Wang, K.; Oluic, M.; Ghandhari, M.; Xu, J.; Li, G. Construction of Multi-State Transient Stability Boundary Based on Broad Learning. IEEE Trans. Power Syst.
**2021**, 36, 2906–2917. [Google Scholar] [CrossRef] - Moeini, A.; Kamwa, I.; Brunelle, P.; Sybille, G. Open data IEEE test systems implemented in SimPowerSystems for education and research in power grid dynamics and control. In Proceedings of the Universities Power Engineering Conference, Stoke on Trent, UK, 1–4 September 2015; pp. 3–8. [Google Scholar] [CrossRef]
- Sarajcev, P.; Kunac, A.; Petrovic, G.; Despalatovic, M. Power system transient stability assessment using stacked autoencoder and voting ensemble†. Energies
**2021**, 14, 3148. [Google Scholar] [CrossRef] - Azman, S.K.; Isbeih, Y.J.; Moursi, M.S.E.; Elbassioni, K. A Unified Online Deep Learning Prediction Model for Small Signal and Transient Stability. IEEE Trans. Power Syst.
**2020**, 35, 4585–4598. [Google Scholar] [CrossRef] - Zhu, Q.; Chen, J.; Zhu, L.; Shi, D.; Bai, X.; Duan, X.; Liu, Y. A Deep End-to-End Model for Transient Stability Assessment with PMU Data. IEEE Access
**2018**, 6, 65474–65487. [Google Scholar] [CrossRef] - Huang, J.; Guan, L.; Su, Y.; Yao, H.; Guo, M.; Zhong, Z. Recurrent Graph Convolutional Network-Based Multi-Task Transient Stability Assessment Framework in Power System. IEEE Access
**2020**, 8, 93283–93296. [Google Scholar] [CrossRef] - Huang, J.; Guan, L.; Su, Y.; Yao, H.; Guo, M.; Zhong, Z. A topology adaptive high-speed transient stability assessment scheme based on multi-graph attention network with residual structure. Int. J. Electr. Power Energy Syst.
**2021**, 130, 106948. [Google Scholar] [CrossRef] - Li, X.; Yang, Z.; Guo, P.; Cheng, J. An Intelligent Transient Stability Assessment Framework With Continual Learning Ability. IEEE Trans. Ind. Inform.
**2021**, 17, 8131–8141. [Google Scholar] [CrossRef] - Mi, D.; Wang, T.; Gao, M.; Li, C.; Wang, Z. Online transient stability margin prediction of power systems with wind farms using ensemble regression trees. Int. Trans. Electr. Energy Syst.
**2021**, 31, e13057. [Google Scholar] [CrossRef] - Gnanavignesh, R.; Shenoy, U. An improved relaxation based spatial domain decomposition method for parallel transient stability simulation. Int. J. Electr. Power Energy Syst.
**2022**, 135, 107561. [Google Scholar] [CrossRef] - Behdadnia, T.; Yaslan, Y.; Genc, I. A new method of decision tree based transient stability assessment using hybrid simulation for real-time PMU measurements. IET Gener. Transm. Distrib.
**2021**, 15, 678–693. [Google Scholar] [CrossRef] - Dabou, R.T.; Kamwa, I.; Chung, C.Y.; Mugombozi, C.F. Time Series-Analysis Based Engineering of High-Dimensional Wide-Area Stability Indices for Machine Learning. IEEE Access
**2021**, 9, 104927–104939. [Google Scholar] [CrossRef] - Arvizu, C.M.C.; Messina, A.R. Dimensionality Reduction in Transient Simulations: A Diffusion Maps Approach. IEEE Trans. Power Deliv.
**2016**, 31, 2379–2389. [Google Scholar] [CrossRef] - Chen, K. Indirect PCA Dimensionality Reduction Based Machine Learning Algorithms for Power System Transient Stability Assessment. In Proceedings of the 2019 IEEE Innovative Smart Grid Technologies—Asia (ISGT Asia), Chengdu, China, 21–24 May 2019; pp. 4175–4179. [Google Scholar] [CrossRef]
- Hang, F.; Huang, S.; Chen, Y.; Mei, S. Power system transient stability assessment based on dimension reduction and cost-sensitive ensemble learning. In Proceedings of the 2017 IEEE Conference on Energy Internet and Energy System Integration, Beijing, China, 26–28 November 2017; pp. 1–6. [Google Scholar] [CrossRef]
- Su, T.; Liu, Y.; Zhao, J.; Liu, J. Probabilistic Stacked Denoising Autoencoder for Power System Transient Stability Prediction with Wind Farms. IEEE Trans. Power Syst.
**2021**, 36, 3786–3789. [Google Scholar] [CrossRef] - Tang, J.; Sui, H. Power System Transient Stability Assessment Based on Stacked Autoencoders and Support Vector Machine. IOP Conf. Ser. Mater. Sci. Eng.
**2018**, 452, 042117. [Google Scholar] [CrossRef] - Liu, T.; Liu, Y.; Xu, L.; Liu, J.; Mitra, J.; Tian, Y. Non-parametric statistics-based predictor enabling online transient stability assessment. IET Gener. Transm. Distrib.
**2018**, 12, 5761–5769. [Google Scholar] [CrossRef][Green Version] - Kunac, A.; Sarajcev, P. Ensemble Learning Approach to Power System Transient Stability Assessment. In Proceedings of the 2020 5th International Conference on Smart and Sustainable Technologies (SpliTech), Split, Croatia, 23–26 September 2020; pp. 1–6. [Google Scholar] [CrossRef]
- Shao, D.; Xu, Y.; Ma, S.; Jing, Y.; Sun, W.; Zhou, T.; Yang, J. Transient Stability Assessment Method for Power System Based on SVM with Adaptive Parameters Adjustment. Proceedings of 2021 IEEE 4th International Electrical and Energy Conference, CIEEC 2021, Wuhan, China, 28–30 May 2021. [Google Scholar] [CrossRef]
- Hu, W.; Lu, Z.; Wu, S.; Zhang, W.; Dong, Y.; Yu, R.; Liu, B. Real-time transient stability assessment in power system based on improved SVM. J. Mod. Power Syst. Clean Energy
**2019**, 7, 26–37. [Google Scholar] [CrossRef][Green Version] - Pannell, Z.; Ramachandran, B.; Snider, D. Machine learning approach to solving the transient stability assessment problem. In Proceedings of the 2018 IEEE Texas Power and Energy Conference, TPEC 2018, College Station, TX, USA, 8–9 February 2018; pp. 1–6. [Google Scholar] [CrossRef]
- Moulin, L.S.; Alves Da Silva, A.P.; El-Sharkawi, M.A.; Marks, R.J. Support vector machines for transient stability analysis of large-scale power systems. IEEE Trans. Power Syst.
**2004**, 19, 818–825. [Google Scholar] [CrossRef][Green Version] - Bashiri Mosavi, A.; Amiri, A.; Hosseini, H. A Learning Framework for Size and Type Independent Transient Stability Prediction of Power System Using Twin Convolutional Support Vector Machine. IEEE Access
**2018**, 6, 69937–69947. [Google Scholar] [CrossRef] - Wang, B.; Fang, B.; Wang, Y.; Liu, H.; Liu, Y. Power System Transient Stability Assessment Based on Big Data and the Core Vector Machine. IEEE Trans. Smart Grid
**2016**, 7, 2561–2570. [Google Scholar] [CrossRef] - Mohammadi, M.; Gharehpetian, G.B. On-line transient stability assessment of large-scale power systems by using ball vector machines. Energy Convers. Manag.
**2010**, 51, 640–647. [Google Scholar] [CrossRef] - Arefi, M.; Chowdhury, B. Ensemble adaptive neuro fuzzy support vector machine for prediction of transient stability. In Proceedings of the 2017 North American Power Symposium, NAPS 2017, Morgantown, WV, USA, 17–19 September 2017. [Google Scholar] [CrossRef]
- Mukherjee, R.; De, A. Development of an Ensemble Decision Tree-Based Power System Dynamic Security State Predictor. IEEE Syst. J.
**2020**, 14, 3836–3843. [Google Scholar] [CrossRef] - Aghamohammadi, M.R.; Abedi, M. DT based intelligent predictor for out of step condition of generator by using PMU data. Int. J. Electr. Power Energy Syst.
**2018**, 99, 95–106. [Google Scholar] [CrossRef] - Baltas, N.G.; Mazidi, P.; Ma, J.; De Asis Fernandez, F.; Rodriguez, P. A comparative analysis of decision trees, support vector machines and artificial neural networks for on-line transient stability assessment. In Proceedings of the 2018 International Conference on Smart Energy Systems and Technologies, SEST 2018, Seville, Spain, 10–12 September 2018; pp. 1–6. [Google Scholar] [CrossRef]
- Zhang, C.; Li, Y.; Yu, Z.; Tian, F. A weighted random forest approach to improve predictive performance for power system transient stability assessment. In Proceedings of the Asia-Pacific Power and Energy Engineering Conference, APPEEC, Xi’an, China, 25–28 October 2016; pp. 1259–1263. [Google Scholar] [CrossRef]
- Amraee, T.; Ranjbar, S. Transient instability prediction using decision tree technique. IEEE Trans. Power Syst.
**2013**, 28, 3028–3037. [Google Scholar] [CrossRef] - Li, N.; Li, B.; Gao, L. Transient Stability Assessment of Power System Based on XGBoost and Factorization Machine. IEEE Access
**2020**, 8, 28403–28414. [Google Scholar] [CrossRef] - Liu, X.; Min, Y.; Chen, L.; Zhang, X.; Feng, C. Data-driven Transient Stability Assessment Based on Kernel Regression and Distance Metric Learning. J. Mod. Power Syst. Clean Energy
**2021**, 9, 27–36. [Google Scholar] [CrossRef] - Kamari, N.A.M.; Musirin, I.; Othman, M.M. Application of evolutionary programming in the assessment of dynamic stability. In Proceedings of the 2010 4th International Power Engineering and Optimization Conference (PEOCO), Shah Alam, Malaysia, 23–24 June 2010; pp. 43–48. [Google Scholar] [CrossRef]
- Du, Y.; Hu, Z. Power system transient stability assessment based on snapshot ensemble LSTM network. Sustainability
**2021**, 13, 6953. [Google Scholar] [CrossRef] - Li, X.; Zheng, Z.; Ma, Z.; Guo, P.; Shao, K.; Quan, S. Real-time approach for oscillatory stability assessment in large-scale power systems based on MRMR classifier. IET Gener. Transm. Distrib.
**2019**, 13, 4431–4442. [Google Scholar] [CrossRef] - Liu, X.; Zhang, X.; Chen, L.; Xu, F.; Feng, C. Data-driven Transient Stability Assessment Model Considering Network Topology Changes via Mahalanobis Kernel Regression and Ensemble Learning. J. Mod. Power Syst. Clean Energy
**2020**, 8, 1080–1091. [Google Scholar] [CrossRef] - Baltas, G.N.; Perales-González, C.; Mazidi, P.; Fernandez, F.; Rodríguez, P. A Novel Ensemble Approach for Solving the Transient Stability Classification Problem. In Proceedings of the 2018 7th International Conference on Renewable Energy Research and Applications (ICRERA), Paris, France, 14–17 October 2018; pp. 1282–1286. [Google Scholar] [CrossRef]
- Zhou, Y.; Wu, J.; Yu, Z.; Ji, L.; Hao, L. A Hierarchical Method for Transient Stability Prediction of Power Systems Using the Confidence of a SVM-Based Ensemble Classifier. Energies
**2016**, 9, 778. [Google Scholar] [CrossRef][Green Version] - Khodayar, M.; Liu, G.; Wang, J.; Khodayar, M.E. Deep learning in power systems research: A review. CSEE J. Power Energy Syst.
**2021**, 7, 209–220. [Google Scholar] [CrossRef] - Zhu, L.; Hill, D.J.; Lu, C. Data-Driven Fast Transient Stability Assessment Using (Fault-on + 2) Generator Trajectories. In Proceedings of the 2019 IEEE Power Energy Society General Meeting (PESGM), Atlanta, GA, USA, 4–8 August 2019; pp. 1–5. [Google Scholar] [CrossRef]
- Hou, J.; Xie, C.; Wang, T.; Yu, Z.; Lü, Y.; Dai, H. Power system transient stability assessment based on voltage phasor and convolution neural network. In Proceedings of the 2nd IEEE International Conference on Energy Internet, ICEI 2018, Beijing, China, 21–25 May 2018; pp. 247–251. [Google Scholar] [CrossRef]
- Zhou, Y.; Zhao, W.; Guo, Q.; Sun, H.; Hao, L. Transient Stability Assessment of Power Systems Using Cost-sensitive Deep Learning Approach. In Proceedings of the 2nd IEEE Conference on Energy Internet and Energy System Integration, EI2 2018, Beijing, China, 20–22 October 2018; pp. 8–13. [Google Scholar] [CrossRef]
- Shinde, B.; Wang, S.; Dehghanian, P.; Babakmehr, M. Real-Time Detection of Critical Generators in Power Systems: A Deep Learning HCP Approach. In Proceedings of the 2020 IEEE Texas Power and Energy Conference (TPEC), College Station, TX, USA, 6–7 February 2020; pp. 1–6. [Google Scholar] [CrossRef]
- Yu, J.J.Q.; Hill, D.J.; Lam, A.Y.S.; Gu, J.; Li, V.O.K. Intelligent Time-Adaptive Transient Stability Assessment System. IEEE Trans. Power Syst.
**2018**, 33, 1049–1058. [Google Scholar] [CrossRef][Green Version] - Du, Y.; Hu, Z.; Wang, F. A hierarchical power system transient stability assessment method considering sample imbalance. Energy Rep.
**2021**, 7, 224–232. [Google Scholar] [CrossRef] - Mahato, N.; Dong, J.; Song, C.; Chen, Z.; Wang, N.; Ma, H.; Gong, G. Electric Power System Transient Stability Assessment Based on Bi-LSTM Attention Mechanism. In Proceedings of the 2021 6th Asia Conference on Power and Electrical Engineering, ACPEE 2021, Chongqing, China, 8–11 April 2021; pp. 777–782. [Google Scholar] [CrossRef]
- Li, B.; Wu, J.; Hao, L.; Shao, M.; Zhang, R.; Zhao, W. Anti-Jitter and Refined Power System Transient Stability Assessment Based on Long-Short Term Memory Network. IEEE Access
**2020**, 8, 35231–35244. [Google Scholar] [CrossRef] - Yang, H.; Niu, K.; Xu, D.; Xu, S. Analysis of power system transient stability characteristics with the application of massive transient stability simulation data. Energy Rep.
**2021**, 7, 111–117. [Google Scholar] [CrossRef] - Gong, G.; Mahato, N.K.; He, H.; Wang, H.; Jin, Y.; Han, Y. Transient Stability Assessment of Electric Power System based on Voltage Phasor and CNN-LSTM. In Proceedings of the 2020 IEEE/IAS Industrial and Commercial Power System Asia (I CPS Asia), Weihai, China, 13–15 July 2020; pp. 443–448. [Google Scholar] [CrossRef]
- Pan, F.; Li, J.; Tan, B.; Zeng, C.; Jiang, X.; Liu, L.; Yang, J. Stacked-GRU based power system transient stability assessment method. Algorithms
**2018**, 11, 121. [Google Scholar] [CrossRef][Green Version] - Yu, J.J.; Hill, D.J.; Lam, A.Y. Delay aware transient stability assessment with synchrophasor recovery and prediction framework. Neurocomputing
**2018**, 322, 187–194. [Google Scholar] [CrossRef] - Zhou, Y.; Yang, Y. Transient Stability Assessment using Gated Recurrent Unit. In Proceedings of the 2020 5th Asia Conference on Power and Electrical Engineering (ACPEE), Chengdu, China, 4–7 June 2020; pp. 560–564. [Google Scholar] [CrossRef]
- Li, J.; Yang, H.; Yan, L.; Li, Z.; Liu, D.; Xia, Y. Data Augment Using Deep Convolutional Generative Adversarial Networks for Transient Stability Assessment of Power Systems. In Proceedings of the 2020 39th Chinese Control Conference (CCC), Shenyang, China, 27–29 July 2020; pp. 6135–6140. [Google Scholar] [CrossRef]
- Yang, D.; Ji, M.; Zhou, B.; Bu, S.; Hu, B. Transient Stability Assessment of Power System Based on DGL-GAN. Dianwang Jishu/Power Syst. Technol.
**2021**, 45, 2934–2944. [Google Scholar] [CrossRef] - Ren, J.; Chen, J.; Li, B.; Zhao, M.; Shi, H.; You, H. A Method for Power System Transient Stability Assessment Based on Transfer Learning. In Proceedings of the 2020 IEEE Power Energy Society General Meeting (PESGM), Montreal, QC, Canada, 2-6 August 2020; pp. 1–5. [Google Scholar] [CrossRef]
- Tan, B.; Yang, J.; Tang, Y.; Jiang, S.; Xie, P.; Yuan, W. A Deep Imbalanced Learning Framework for Transient Stability Assessment of Power System. IEEE Access
**2019**, 7, 81759–81769. [Google Scholar] [CrossRef] - Karami, A.; Esmaili, S. Transient stability assessment of power systems described with detailed models using neural networks. Int. J. Electr. Power Energy Syst.
**2013**, 45, 279–292. [Google Scholar] [CrossRef] - Yousefian, R.; Kamalasadan, S. A Review of Neural Network Based Machine Learning Approaches for Rotor Angle Stability Control. arXiv
**2017**, arXiv:1701.01214. [Google Scholar] - Yan, R.; Geng, G.; Jiang, Q.; Li, Y. Fast Transient Stability Batch Assessment Using Cascaded Convolutional Neural Networks. IEEE Trans. Power Syst.
**2019**, 34, 2802–2813. [Google Scholar] [CrossRef] - Wu, S.; Zheng, L.; Hu, W.; Yu, R.; Liu, B. Improved Deep Belief Network and Model Interpretation Method for Power System Transient Stability Assessment. J. Mod. Power Syst. Clean Energy
**2020**, 8, 27–37. [Google Scholar] [CrossRef] - Xia, D.; Liu, Y.; Qu, K.; Hu, W.; Ma, K. An intelligent assessment for transient stability of power systems incorporating grid topology characteristics. In Proceedings of the 2021 6th Asia Conference on Power and Electrical Engineering, ACPEE 2021, Chongqing, China, 8–11 April 2021; pp. 1543–1548. [Google Scholar] [CrossRef]
- Zhou, Z.; Ma, S.; Bu, G.; Jing, Y.; Wang, G. Transient Stability Assessment of Large-scale AC/DC Hybrid Power Grid Based on Separation Feature and Deep Belief Networks. In Proceedings of the 2021 Power System and Green Energy Conference (PSGEC), Shanghai, China, 20–22 August 2021; pp. 25–29. [Google Scholar] [CrossRef]
- Zeng, L.; Yao, W.; Ai, X.; Huang, Y.; Wen, J. Double Q-learning Based Identification of Weak Lines in Power Grid Considering Transient Stability Constraints. Zhongguo Dianji Gongcheng Xuebao
**2020**, 40, 2429–2440. [Google Scholar] [CrossRef] - Cui, W.; Zhang, B. Lyapunov-Regularized Reinforcement Learning for Power System Transient Stability. arXiv
**2021**, arXiv:2103.03869. [Google Scholar] [CrossRef] - Pereslytskikh, O.; Suvorova, V.; Belyaev, A. Synthesis of Centralized Adaptive Excitation Control System by Reinforcement Learning Approach. In Proceedings of the 2021 IEEE Conference of Russian Young Researchers in Electrical and Electronic Engineering, ElConRus 2021, St. Petersburg, Russia, 26–29 January 2021; pp. 1514–1519. [Google Scholar] [CrossRef]
- Zhang, K.; Xu, P.; Zhang, J. Explainable AI in Deep Reinforcement Learning Models: A SHAP Method Applied in Power System Emergency Control. In Proceedings of the 2020 IEEE 4th Conference on Energy Internet and Energy System Integration (EI2), Wuhan, China, 30 October–1 November 2020; pp. 711–716. [Google Scholar] [CrossRef]
- Liu, H.; Zhang, S.; Sun, L.; Zhang, C.; Peng, X. Transient Stability Analysis of the Two-Area With AC/DC Paralleled Interconnected Power System in Different Operation. Front. Energy Res.
**2021**, 9, 703. [Google Scholar] [CrossRef] - Abedin, T.; Lipu, M.S.H.; Hannan, M.A.; Ker, P.J.; Rahman, S.A.; Yaw, C.T.; Tiong, S.K.; Muttaqi, K.M. Dynamic Modeling of HVDC for Power System Stability Assessment: A Review, Issues, and Recommendations. Energies
**2021**, 14, 4829. [Google Scholar] [CrossRef] - Zhou, Y.; Xu, T.; Ye, L.; Liu, M.; Chen, X.; Yang, Y.; Guo, Q.; Sun, H. Transient Rotor Angle and Voltage Stability Discrimination Based on Deep Convolutional Neural Network with Multiple Inputs. In Proceedings of the 2021 IEEE 4th International Electrical and Energy Conference, CIEEC 2021, Wuhan, China, 28–30 May 2021. [Google Scholar] [CrossRef]
- Sarajcev, P.; Kunac, A.; Petrovic, G.; Despalatovic, M. Power System Transient Stability Assessment Simulations Dataset—IEEE New England 39-Bus Test Case. Zenodo
**2021**. [Google Scholar] [CrossRef] - Zheng, L.; Hu, W.; Zhou, Y.; Min, Y.; Xu, X.; Wang, C.; Yu, R. Deep belief network based nonlinear representation learning for transient stability assessment. In Proceedings of the IEEE Power and Energy Society General Meeting, Chicago, IL, USA, 16–20 July 2018; pp. 1–5. [Google Scholar] [CrossRef]

**Figure 4.**Typical AI model life-cycle, which consists of data preparation, model building, training, deployment, and management.

Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations. |

© 2022 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).

## Share and Cite

**MDPI and ACS Style**

Sarajcev, P.; Kunac, A.; Petrovic, G.; Despalatovic, M. Artificial Intelligence Techniques for Power System Transient Stability Assessment. *Energies* **2022**, *15*, 507.
https://doi.org/10.3390/en15020507

**AMA Style**

Sarajcev P, Kunac A, Petrovic G, Despalatovic M. Artificial Intelligence Techniques for Power System Transient Stability Assessment. *Energies*. 2022; 15(2):507.
https://doi.org/10.3390/en15020507

**Chicago/Turabian Style**

Sarajcev, Petar, Antonijo Kunac, Goran Petrovic, and Marin Despalatovic. 2022. "Artificial Intelligence Techniques for Power System Transient Stability Assessment" *Energies* 15, no. 2: 507.
https://doi.org/10.3390/en15020507