Explainable Artificial Intelligence (XAI) techniques for energy and power systems: Review, challenges and opportunities

TLDR

Machine learning models are often black boxes, making it difficult for power system experts to trust and justify their decisions, prompting the development of Explainable Artificial Intelligence techniques to improve model transparency. This paper aims to highlight the potential of applying XAI to power system applications. The authors first outline the common challenges of using XAI in power systems, then review and analyze recent works and emerging research trends in the field. The authors hope this review will spark fruitful discussions and encourage further research on this emerging topic.

Abstract

Despite widespread adoption and outstanding performance, machine learning models are considered as "black boxes", since it is very difficult to understand how such models operate in practice. Therefore, in the power systems field, which requires a high level of accountability, it is hard for experts to trust and justify decisions and recommendations made by these models. Meanwhile, in the last couple of years, Explainable Artificial Intelligence (XAI) techniques have been developed to improve the explainability of machine learning models, such that their output can be better understood. In this light, it is the purpose of this paper to highlight the potential of using XAI for power system applications. We first present the common challenges of using XAI in such applications and then review and analyze the recent works on this topic, and the on-going trends in the research community. We hope that this paper will trigger fruitful discussions and encourage further research on this important emerging topic.