Observer Verification Feature: MAC Matrix for Confident Modal Validation
A MAC Verification Matrix (Modal Assurance Criterion) lets you verify mode shapes identified with different Operational Modal Analysis (OMA) methods. Each cell scores the correlation between one mode in set A and one in set B on a 0–1 scale: strong, near-diagonal values confirm correct mode pairing; off-diagonal or low values flag spurious poles, noise, or geometry mismatches.
Observer’s Verification tab provides an integrated workspace to compare, validate, and cross-check modal parameters estimated with multiple OMA methods. After running your analyses, you can use the MAC Matrix together with frequency and damping checks to validate models before SHM tracking or report export.
The goal is simple but vital: confirm the reliability of identified frequencies, damping ratios, and mode shapes across methods, so you can confidently proceed to continuous monitoring.
Why Verification Matters
Real-world OMA can surface peaks or poles that represent noise or numerical artifacts rather than physical modes. Verification distinguishes true structural behavior from spurious results, ensures consistency across methods, and builds a stable baseline for SHM so you avoid false alarms and unstable trends.
- Noise vs. true modes: Separate physical modes from artifacts and poor SNR outcomes.
- Cross-method consistency: True modes recur across methods with similar frequencies and compatible shapes.
- Reliability for monitoring: Track only verified modes to reduce nuisance alerts and increase diagnostic sensitivity.
Calculated Methods Panel
Observer aggregates all identified modes from applied OMA methods (e.g., EFDD, FDD, FSDD, SSI, pLSCF) and lists for each method:
- Picked Frequency — the frequency chosen during analysis,
- Calculated Frequency — a refined value from curve fitting or stabilization,
- Damping — the estimated damping ratio (if available for that method).
Use this panel to shortlist candidate modes for detailed comparison.
Mode Shape Comparison (3D View)
Compare mode shapes A/B from different methods directly in 3D to inspect pattern, scale, and orientation agreement. The viewer supports rotation and zoom for local/global checks, optional animation to reveal phase and nodal lines, and quick side-by-side inspection to make subtle differences obvious.
MAC Matrix (Modal Assurance Criterion)
The MAC Matrix provides a quantitative, method-agnostic comparison between two sets of mode shapes. Each cell reports the correlation between a mode from Set A and a mode from Set B (0 to 1).
- Y-axis: Modes from one of the compared methods.
- X-axis: Modes from the second method.
Use it to match modes between methods (e.g., EFDD vs. FSDD) and confirm both identify the same physical mode.
Rule of thumb: MAC ≥ 0.90 is strong evidence of consistent mode shapes; 0.70–0.90 requires engineering judgment; < 0.70 often indicates spurious poles, poor SNR, or geometry/DOF mismatches.
Suggested Verification Workflow
- Run multiple OMA methods (FDD, EFDD, FSDD, SSI, pLSCF) on the same data.
- Use the Calculated Methods Panel to shortlist candidate modes near consistent frequencies.
- Compare mode shapes in the 3D view for qualitative agreement.
- Use the MAC Matrix to confirm pairings; aim for MAC ≥ 0.90 for final selections.
- Revisit method settings (windowing, model order, stabilization thresholds) if conflicts appear.
- Lock in verified modes and export to Voyager or your structural model.
Practical Tips for Better MAC & Shape Agreement
- Ensure consistent geometry mapping/DOF ordering across methods.
- Use adequate duration and bandwidth to improve SNR and stabilize damping estimates.
- Balance model orders and stabilization criteria to limit spurious poles.
- When in doubt, trust the shape: prioritize MAC-verified modes over isolated frequency matches.
Conclusion: From Verification to Reliable Condition Monitoring
Observer’s verification environment combines visual (3D shapes) and numerical (MAC) checks so you can confidently select modes for tracking. Verified modes strengthen baselines, reduce nuisance alarms, and heighten sensitivity to real structural change—making your SHM both robust and actionable.