Mechanical anomaly detection on the hourly displacements in mm, data
unchanged: for each target and axis, the largest gap between the median
of the next 24 h and the previous 24 h, kept above 3 times the target's
daily swing (provisional). Jumps within 6 h form one event, so a common
jump is not several anomalies. Grey curves for all targets, colour for
those that jumped; a click opens Explore, Diagnosis.
Note on the idea, first trial and leads in
vues-deplacements/contexte/detection-d-anomalies.md.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Ten screenshots of the monitoring site (Explore, Target Configuration,
System Summary, Measurements), SIM_CCID and IP addresses masked.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Card checked against the case data and images:
- signature: corrected ranges (corr_y plateau up to +1.2, 18 to 126
MAD; error_value of the local group 3 to 7 times the previous day),
the local group moves far more than the other sensors, not alone;
- new criterion 4: at night the references move between themselves
(not R1), sigma0 rises tenfold and corr_x of most targets drops
without their raw position moving, a plateau that lasts to the end
of the measurements; images become criterion 5;
- event now starts on 18.09 at 23:20 (header, summary, symptom, index);
- diagnosis: night then morning in the mechanism; C24 sits in the
local group and its error_value rises from 05:50;
- figure 1 redone with a corr_x panel, captions corrected.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
- docs/displacement/domaine/incident-de-barcelone.md: the AIMsight
regression case is data/CLARIDEN/ (case C-004); what the data show,
the documentation's reading (reference/compensation instability)
against the expert's (real movement of the structure, confirmed by
other instruments, amplitude likely overestimated), open questions.
- nouveau-cas: platform zips date images in UTC, exports cover local
days; keep every image of the range, even at 10 min.
- colonnes-des-deplacements: native cadence is 15 or 10 min.
- CLAUDE.md: four cases in the base, displacement data of NORDEND,
WILDSTRUBEL and CLARIDEN.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Second run on the same question, case in the index: C-004, complete
match (criteria 1 to 4 verified); monitored-object motion with a
camera/reference compensation event, consistent with; consequence:
real movement; action: intervention. The card needed no change.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Card checked section by section against the case data and images:
- header: "end of measurements" (images run to 23:00);
- summary, symptom: duration, most targets, uneven jumps;
- signature: corrected ratios and ranges, remaining references at 0,
jumps only at some switches (setting R2 aside alone moves nothing),
web visible by day too; C1 is a local reference, outside the
compensation;
- diagnosis: mechanism with one or two references set aside, inflated
MAD for several targets, "masked movement" line removed;
- proofs: three captions corrected; index line aligned.
Domain note: pattern_type `ref` is a local reference (station manual).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Copy the station's displacement export and images into the case folder,
with coded sensor names; the station takes the case number instead of
S01, and graphique.py / vignettes.py drop --alias (D-050).
Rewrite C-001 as the expert's diagnosis of this event only: summary,
symptom, verdict, signature with commands, diagnosis, evidence. Drop
exclusions, expected output, unknowns and test history; matching no
longer uses exclusions (D-051). The corr_x jumps at the reference
switches are judged negligible.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Add the anonymiser-cas skill: a growing registry gives stable codes
(S01, R1, C1), a stdlib script rewrites a case and renames its files,
and graphique.py / vignettes.py take --alias to plot with codes.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Two analysis reports delivered to the client (GN-1.472, GN-5.054, May 2026)
go to docs/aimsight/ unchanged. Their diagnoses are summarized in
docs/displacement/domaine/cas-annotes-par-l-expert.md as reference truth
for displacement, without data: not in the diagnostic agent's case base.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>