Modern electricity grids generate large volumes of high-dimensional time series data through Advanced Metering Infrastructure (AMI). While this data contains valuable operational insights, its scale and complexity pose significant analytical challenges, including computational constraints, domain knowledge gaps, and the need for targeted exploration. We present an integrated toolchain for data summarization and clustering-based analysis that bridges this gap, giving grid operators practical capabilities to extract actionable insights from complex measurements without requiring advanced algorithmic expertise. Our toolchain integrates streaming data processing, efficient exploration techniques, configurable and extensible feature engineering, and pattern identification components. This infrastructure enables computationally efficient high-dimensional data processing while maintaining the analytical depth necessary for operational decision-making. In this article, we describe a work in progress and showcase electricity consumption behavior analysis as one example application. The underlying data processing infrastructure supports various analytical tasks across multiple domains. Poster: https://doi.org/10.5281/zenodo.18740094 Article: https://doi.org/10.5281/zenodo.18740102 Github: https://github.com/rasmusthorsson/CLUE
Magnusson et al. (Mon,) studied this question.