Filling in the Blanks: Applying Data Imputation in incomplete Water Metering Data

TL;DR

This paper evaluates various data imputation techniques to address missing data in smart water meter readings, demonstrating improved accuracy for water management applications like leak detection and maintenance.

Contribution

It applies and compares recent imputation methods on real-world water metering data, highlighting their effectiveness in enhancing water network monitoring.

Findings

Imputation improves data quality for water management tasks.

Transformers and Recurrent Neural Networks outperform traditional methods.

Enhanced data quality aids in leak detection and predictive maintenance.

Abstract

In this work, we explore the application of recent data imputation techniques to enhance monitoring and management of water distribution networks using smart water meters, based on data derived from a real-world IoT water grid monitoring deployment. Despite the detailed data produced by such meters, data gaps due to technical issues can significantly impact operational decisions and efficiency. Our results, by comparing various imputation methods, such as k-Nearest Neighbors, MissForest, Transformers, and Recurrent Neural Networks, indicate that effective data imputation can substantially enhance the quality of the insights derived from water consumption data as we study their effect on accuracy and reliability of water metering data to provide solutions in applications like leak detection and predictive maintenance scheduling.