Rozhin Eskandarpour

Rozhin Eskandarpour, Amin Khodaei, Ali Arab

In this paper, in an attempt to improve power grid resilience, a machine learning model is proposed to predictively estimate the component states in response to extreme events. The proposed model is based on a multi-dimensional Support Vector Machine (SVM) considering the associated resilience index, i.e., the infrastructure quality level and the time duration that each component can withstand the event, as well as predicted path and intensity of the upcoming extreme event. The outcome of the proposed model is the classified component state data to two categories of outage and operational, which can be further used to schedule system resources in a predictive manner with the objective of maximizing its resilience. The proposed model is validated using Ä-fold cross-validation and model benchmarking techniques. The performance of the model is tested through numerical simulations and based on a well-defined and commonly-used performance measure.

Rozhin Eskandarpour, Amin Khodaei

Predicting power system component outages in response to an imminent hurricane plays a major role in preevent planning and post-event recovery of the power system. An exact prediction of components states, however, is a challenging task and cannot be easily performed. In this paper, a Support Vector Machine (SVM) based method is proposed to help estimate the components states in response to anticipated path and intensity of an imminent hurricane. Components states are categorized into three classes of damaged, operational, and uncertain. The damaged components along with the components in uncertain class are then considered in multiple contingency scenarios of a proposed Event-driven Security-Constrained Unit Commitment (E-SCUC), which considers the simultaneous outage of multiple components under an N-m-u reliability criterion. Experimental results on the IEEE 118-bus test system show the merits and the effectiveness of the proposed SVM classifier and the E-SCUC model in improving power system resilience in response to extreme events.

Rozhin Eskandarpour, Amin Khodaei, Ali Arab

A machine learning model is proposed in this paper to help estimate potential nodal load curtailment in response to an extreme event. This is performed through identifying which grid components will fail as a result of an extreme event, and consequently, which parts of the power system will encounter a supply interruption. The proposed model to predict component outages is based on a Support Vector Machine (SVM) model. This model considers the category and the path of historical hurricanes, as the selected extreme event in this paper, and accordingly trains the SVM. Once trained, the model is capable of classifying the grid components into two categories of outage and operational in response to imminent hurricanes. The obtained component outages are then integrated into a load curtailment minimization model to estimate the nodal load curtailments. The merits and the effectiveness of the proposed models are demonstrated using the standard IEEE 30-bus system based on various hurricane path/intensity scenarios.

Rozhin Eskandarpour, Amin Khodaei, A. Paaso et al.

In this paper, an artificial intelligence based grid hardening model is proposed with the objective of improving power grid resilience in response to extreme weather events. At first, a machine learning model is proposed to predict the component states (either operational or outage) in response to the extreme event. Then, these predictions are fed into a hardening model, which determines strategic locations for placement of distributed generation (DG) units. In contrast to existing literature in hardening and resilience enhancement, this paper co-optimizes grid economic and resilience objectives by considering the intricate dependencies of the two. The numerical simulations on the standard IEEE 118-bus test system illustrate the merits and applicability of the proposed hardening model. The results indicate that the proposed hardening model through decentralized and distributed local energy resources can produce a more robust solution that can protect the system significantly against multiple component outages due to an extreme event.