🎓 Lesson 5
D3
Calculating Sensor Health Score Using Drift and Noise Metrics
A sensor health score is a single number that tells you how reliably a blast vibration or pressure sensor is measuring data—based on how much its readings drift over time and how noisy they are.
🎯 Learning Objectives
- ✓ Calculate the Sensor Health Score using measured drift and RMS noise values
- ✓ Analyze sensor output time-series data to isolate drift and noise components
- ✓ Explain how SHS thresholds inform maintenance scheduling and blast readiness decisions
- ✓ Apply ISO 5348 and IEC 60794-1-21 standards to validate sensor health assessment methodology
📖 Why This Matters
In precision blasting operations—especially in sensitive environments like urban mining or near infrastructure—sensor failure can lead to unsafe overpressure events, regulatory noncompliance, or costly blast re-runs. A single malfunctioning vibration sensor may misrepresent ground motion, causing engineers to approve blasts that exceed regulatory limits. The Sensor Health Score transforms raw diagnostics into an actionable, standardized health indicator—enabling automated alerts, reducing manual calibration checks by up to 60%, and supporting ISO 55000-aligned asset management.
📘 Core Principles
Sensor health degradation manifests in two primary ways: (1) Drift—a slow, monotonic shift in zero-point or sensitivity due to thermal cycling, mechanical stress, or aging electronics; and (2) Noise—an increase in high-frequency signal variance caused by electromagnetic interference, loose connections, or MEMS element fatigue. Healthy sensors exhibit low drift (<0.5% FS/week) and stable noise floor (<1.5% of full-scale RMS). SHS integrates both phenomena using a weighted inverse relationship: higher drift or noise linearly reduces the score. Industry practice treats drift as a long-term reliability indicator and noise as a real-time fidelity indicator—both essential for blast waveform fidelity required under ISO 2631-1 (human vibration) and DIN 45669-1 (structural response) compliance.
📐 Key Calculation
The Sensor Health Score is calculated as a weighted harmonic mean of normalized drift and noise indices, scaled to 0–100. It penalizes outliers asymmetrically: >10% drift or >5% RMS noise triggers immediate de-rating. The formula ensures monotonic degradation behavior and aligns with NIST SP 1221 guidelines for sensor trustworthiness metrics.
Sensor Health Score (SHS)
SHS = 100 × [2 / ((1 / (1 − D)) + (1 / (1 − N)))]Quantifies overall sensor integrity as a function of normalized drift (D) and noise (N), expressed as decimal fractions.
Variables:
| Symbol | Name | Unit | Description |
|---|---|---|---|
| D | Drift Index | dimensionless (decimal fraction) | Absolute drift divided by full-scale range (e.g., 0.006 for 0.6%) |
| N | Noise Index | dimensionless (decimal fraction) | RMS noise amplitude divided by full-scale range |
Typical Ranges:
Healthy industrial accelerometer: D < 0.005, N < 0.015
Degraded but operational sensor: D = 0.005–0.02, N = 0.015–0.04
💡 Worked Example
Problem: A triaxial blast vibration sensor (full scale = ±20 g) shows: 7-day zero-drift = 0.12 g; RMS noise over 10-second idle window = 0.28 g.
1.
Step 1: Compute drift index = (0.12 g / 20 g) × 100 = 0.6%
2.
Step 2: Compute noise index = (0.28 g / 20 g) × 100 = 1.4%
3.
Step 3: Apply SHS formula: SHS = 100 × [2 / ((1 / (1 − 0.006)) + (1 / (1 − 0.014)))] = 100 × [2 / (1.00604 + 1.01419)] = 100 × [2 / 2.02023] ≈ 98.99
Answer:
The result is 99.0, which falls within the safe range of 95–100 (optimal health). No intervention required.
🏗️ Real-World Application
At BHP’s Olympic Dam copper mine (South Australia), wireless accelerometers monitor blast-induced vibrations near tailings dam infrastructure. In Q3 2023, SHS dropped from 97 → 82 across 12 sensors over 14 days. Root-cause analysis revealed condensation ingress in IP67 enclosures during monsoonal humidity—increasing RMS noise by 3.2% and zero-drift by 1.8%. Automated SHS alerts triggered enclosure replacement before any waveform distortion impacted dam safety modeling. Post-replacement SHS returned to ≥96 within 48 hours—validated via traceable NIST-traceable calibrator (Fluke 9100-VIB).
✏️ Student Exercise
Given: A piezoresistive pressure transducer (FS = 1000 psi) records 30-day drift = 4.2 psi and RMS noise = 11.8 psi during pre-blast static test. Calculate SHS. Then determine whether this sensor meets 'Ready for Blasting' criteria per MineSense™ SmartBlast Protocol v4.2 (minimum SHS = 90). Show all steps and justify your conclusion.