Machine Learning Approaches for Smart Meter Condition Monitoring in Metropolitan Environments

Smart Meters deployed in Advanced Metering Infrastructure (AMI) generate vast operational data that can be leveraged for condition monitoring in densely populated urban environments. This paper presents a comprehensive machine learning approach to detect Smart Meter irregularities, addressing critical limitations of traditional rule-based expert systems through advanced data analytics. Using systematic feature engineering of multidimensional time-series data from over $\mathbf{1 0, 0 0 0}$ Smart Meters across Hong Kong’s metropolitan network, this study developed a binary classification framework utilizing Gradient Boosting that achieved $93.8 \%$ precision on holdout testing, while demonstrated $100 \%$ detection rate for critical faults at operational threshold 0.7, thus considerably outperforming the existing expert rules system’s $67.7 \%$ precision. This research addresses the inherent challenge of highly imbalanced datasets where faulty meters represent significantly less than $1 \%$ of the population, providing a scalable solution for proactive Smart Meter asset management in metropolitan environments with over 2 million deployed units.

• Publication year

  2025

• Venue

  2025 International Conference on Applied Artificial Intelligence, Data Engineering and Sciences (ICAIDES)

• Publication date

  2025-12-11

• Fields of study

  Not labeled

• Identifiers
  DOI 10.1109/ICAIDES67265.2025.11404074
• External record

  Open on Semantic Scholar

• Source metadata

  Semantic Scholar

• No claims are published for this paper.

• No concepts are published for this paper.

• Resampling approaches to handle class imbalance: a review from a data perspective

  2025influential reference
• Two Statistical Approaches to Justify the Use of the Logistic Function in Binary Logistic Regression

  2023cited by this paper
• A review of ensemble learning and data augmentation models for class imbalanced problems: combination, implementation and evaluation

  2023cited by this paper
• A Novel Time-Series Transformation and Machine-Learning- Based Method for NTL Fraud Detection in Utility Companies

  2022cited by this paper
• Performance Comparison of Imputation Methods in Building Energy Data Sets

  2020cited by this paper
• k-Nearest Neighbour Classifiers - A Tutorial

  2020cited by this paper
• Machine Learning Algorithms for Binary Classification of Liver Disease

  2020cited by this paper
• 100% Classification Accuracy Considered Harmful: The Normalized Information Transfer Factor Explains the Accuracy Paradox

  2014cited by this paper
• Gradient boosting machines, a tutorial

  2013cited by this paper
• 基于Naive Bayes Classifiers的航空影像纹理分类

  2006cited by this paper
• Induction of decision trees

  2004cited by this paper
• Random Forests

  2001cited by this paper

• No citing papers are available for this paper.