On electronic energy meters diagnosis exploiting components´ susceptibility to failure

Electronic energy meters have been excessively deployed in the last two decades and further dramatic proliferation is expected for smart grid implementation purposes. Due to design cost-quality trade-offs, variability in component reliabilities, and geographically correlated climatic and power quality characteristics, meter´s failure is inevitable. Consequently, efficient failure diagnosis is crucial for repair-yield and design improvement to maximize meters recovery in short term and minimize field returns in the long term. On the pursuit of efficient diagnosis, this paper proposes a generalized diagnostic framework that minimize time-to-diagnose by exploiting component susceptibility to failures (CSF) in prioritizing symptom gatherings and indicted component lists. The CSF metric is a normalized failure rate unique for any on-board component that is empirically obtained and dynamically updated from diagnosis history and new resolved cases. Results shows performance improvement by 40-300% thus highlighting work´s significance.