Datasets:

ClarusC64
/

aviation-propulsion-aerodynamics-decoherence-precursor-detection-v0.1

id stringclasses 6 values	flight_phase stringclasses 5 values	baseline_reference stringclasses 5 values	window_minutes stringclasses 6 values	engine_params_trend stringclasses 6 values	aero_state_trend stringclasses 6 values	coupling_anomaly_summary stringclasses 6 values	known_failure_mode stringclasses 6 values	decoherence_onset_time stringclasses 6 values	precursor_pattern_type float64 0.54 0.78	severity_gradient float64 0.4 0.7	failure_likelihood_index float64 0.42 0.8	estimated_time_to_critical_min int64 14 999	primary_decoupling_channels stringclasses 6 values	notes stringclasses 6 values	constraints stringclasses 1 value	gold_checklist stringclasses 1 value
DPD-001	cruise	baseline idx 0.86 env 0.78-0.92 lag IAS 1.5s	0-120	N1 stable 87; EGT rising +35C; FF stable	IAS stable; AoA rising +0.7; trim increasing	EGT rise not matched by IAS response; lag expands to 3.4s	HPT_crack_precursor	t=38m	0.72	0.64	0.76	92	EGT->IAS lag; EGT-AoA divergence	Pattern shows thermal inefficiency before thrust loss	Under 280 words	t+type+sev+lik+ttc+channels
DPD-002	climb	baseline idx 0.74 env 0.62-0.82 lag ALT 2.6s	0-40	N1 oscillation ±2; EGT spikes; FF jitter	IAS droop -8; AoA spikes; pitch corrections	thrust oscillations with nonlinear AoA response; coherence drops below env	compressor_stall_precursor	t=16m	0.78	0.7	0.8	18	N1-AoA oscillatory mismatch; FF jitter->IAS droop	Stall signature appears as oscillatory decoupling	Under 280 words	t+type+sev+lik+ttc+channels
DPD-003	approach	baseline idx 0.66 env 0.52-0.76 lag IAS 2.2s	0-30	N1 commanded increases; EGT rises; FF rises	IAS fails to increase; AoA climbs; sink rate increases	commanded thrust not yielding airspeed; lag expands; control compensation grows	fuel_control_fault_precursor	t=11m	0.66	0.58	0.72	14	N1_cmd->IAS non-response; AoA compensation	Decoupling shows control loop fighting thrust shortfall	Under 280 words	t+type+sev+lik+ttc+channels
DPD-004	holding	baseline idx 0.63 env 0.48-0.74 lag IAS 2.0s	0-90	N1 stable; EGT slowly rising; FF slowly rising	IAS stable; AoA stable; small bank changes	EGT rise without aero effect; coherence slowly erodes; lag to IAS increases to 3.0s	bleed_air_leak_precursor	t=54m	0.6	0.52	0.6	120	EGT drift->no IAS change; lag expansion	Slow precursor with long horizon	Under 280 words	t+type+sev+lik+ttc+channels
DPD-005	cruise	baseline idx 0.86 env 0.78-0.92 lag IAS 1.5s	0-180	N1 stable; EGT stable; FF rising +8%	IAS stable; AoA stable; trim stable	FF rise not matched by thrust response; efficiency drop; correlation N1-IAS weakens	injector_degradation_precursor	t=96m	0.62	0.55	0.58	150	FF->IAS decouple; efficiency loss	Fuel flow creep can be early clue	Under 280 words	t+type+sev+lik+ttc+channels
DPD-006	descent	baseline idx 0.71 env 0.58-0.80 lag ALT 3.2s	0-60	N1 idle; EGT fluctuates; FF irregular	IAS fluctuates; AoA fluctuates; speedbrake stable	idle engine fluctuations correlate poorly with IAS; lag unstable; coherence drops	sensor_egt_fault_precursor	t=22m	0.54	0.4	0.42	999	EGT noise decouples from IAS; channel inconsistency	Shows sensor-driven decoherence not true failure	Under 280 words	t+type+sev+lik+ttc+channels

What this dataset tests

Whether a system can detect early decoherence between propulsion behavior and aerodynamic response.

The signal is relationship drift: lag expansion correlation collapse nonlinear divergence oscillatory mismatch.

Required outputs

decoherence_onset_time
precursor_pattern_type
severity_gradient
failure_likelihood_index
estimated_time_to_critical_min
primary_decoupling_channels

Scoring conventions

onset time is minutes from window start
severity and likelihood range 0 to 1
time to critical is a minute estimate
primary channels name the broken relationships, e.g. EGT->IAS lag

Use case

Layer two of Propulsion–Aerodynamics Coupling Decoherence Signatures.

Supports:

earlier intervention before alarms
separation of sensor-fault vs true precursor
maintenance and diversion decision support

Downloads last month: 6

Number of rows:

Total file size:

9.86 kB