Predictive Maintenance of Distributed Energy Sensors through Cloud Telemetry: A Reproducibility-Centred Evaluation

Abstract

Distributed energy-monitoring networks collect large amounts of telemetry, yet seasonal load changes can be mistaken for equipment faults. To address this, the study will distinguish actionable sensor degradation from benign demand fluctuations using time-aware predictive maintenance rules. The research design centres on a prospective-style replay of multiseason sensor traces with independently logged fault and repair events. The choice of evaluation units and exclusion criteria is informed by the broader methodological record [3–7]. The analysis will contrast fixed engineering thresholds, statistical change detection and a guarded machine-learning classifier. The reproducibility extension records dataset versions, model settings, exclusion rules and provenance before any held-out evaluation. Related enterprise-storage studies inform the assessment of operational reliability [1] and the deployment assumptions adopted for system-level comparison [2]. Assessment will focus on alerts per sensor-month, detection lead time, missed faults and unnecessary maintenance decisions. The supporting evidence base motivates the chosen comparator conditions and robustness criteria [8–12]. Critical sensitivity checks address sensor dropout, firmware changes, class imbalance and network interruption during peak loads. Further systems and modelling literature supplies context for the proposed sensitivity analyses [13–17]. An independent replication exercise would repeat the main analysis from the documented artefacts and quantify disagreements. The expected output is a reusable evaluation framework with transparent uncertainty and reporting rules, informed by wider reproducibility literature [18–23].