Observed Anomaly AI-Detected Condition
The AI system has recognized early-phase wear patterns consistent with inner race deterioration.
About
In its early phases (Stage 2 or 3), bearing degradation manifests through elevated spectral energy between 2kHz and 6kHz, with AI detecting fault-specific frequencies such as BPFI, BPFO, BSF, and FTF within the envelope analysis.
Anticipated Wear Pattern
Detectable by
<span data-metadata=""><span data-buffer="">Analytical Diagnostic Summary
The presence of characteristic BPFI signatures in the envelope analysis indicates developing wear on the bearing’s inner race.
<span data-metadata=""><span data-metadata=""><span data-buffer="">ROI Impact
Prevents catastrophic bearing seizure
Reduces bogie swap by up to 65%
Optimizes lubrication scheduling
Labor Cost: $16,000–$22,000
Repair Time: 8–12 hours
<span data-metadata=""><span data-buffer="">Bearing Health Assessment – TBU 100 Unit
The AI system has recognized early-phase wear patterns consistent with inner race deterioration.
<span data-metadata=""><span data-buffer="">Fault Type: Inner Race Defect (BPFI) – Staged Progression Analysis
The envelope spectrum analysis performed on the TBU 100 bearing unit has revealed clear BPFI (Ball Pass Frequency of Inner race) harmonics, indicating progressive inner race degradation. Based on the amplitude, harmonic clarity, and sideband behavior, the fault can be characterized in four distinct stages:
Anticipated Wear Pattern
Detectable by
<span data-metadata=""><span data-metadata=""><span data-buffer="">ROI Impact
Prevents axle seizure
Reduces emergency bogie replacement by 60%
Enables early re-lubrication plans
Estimated Labor Cost: $16,000–$22,000
Repair Time: 8–12 hours
<span data-metadata=""><span data-buffer="">Analytical Diagnostic Summary
The detection of strong BPFI harmonics in the envelope spectrum indicates progressive wear on the inner race of the TBU 100 bearing. Based on vibration amplitude and harmonic pattern, the fault corresponds to:
• Stage 1–2: Micro fatigue and localized pitting
• Stage 3: Multiple defect points, increasing amplitude and sidebands
• Stage 4: Severe inner race damage requiring immediate replacement
Current analysis suggests the bearing is between Stage 2 and 3. Preventive maintenance or replacement is recommended to avoid escalation.
<span data-metadata=""><span data-metadata=""><span data-buffer="">Observed Anomaly AI-Detected Condition
Misalignment or mechanical looseness detected in the coupling between the traction motor and gearbox (Unit No. 4).
<span data-metadata="">About
Vibration data from the monitored metro rail vehicle indicates structural irregularities at the coupling interface between the traction motor and gearbox. This issue is often the result of:
• Mechanical looseness at the shaft ends,
• Axial/radial misalignment due to improper installation,
• Or wear and tear in the coupling elements (e.g., flexible elements, bolts, or keys).
The irregularities are evidenced by increased vibration amplitudes and intermittent impact-like signatures observed in the vibration-time graph, especially during acceleration and deceleration phases. These are typical indicators of faulty mechanical transmission across a coupling under load transitions.
<span data-metadata=""><span data-buffer="">Gearbox Coupling Misalignment / Looseness
Detectable by
<span data-metadata=""><span data-metadata=""><span data-buffer="">ROI Impact
• Prevents premature wear in coupling sleeves and bolts
• Reduces torsional stress on motor and gearbox shafts
• Minimizes vibration-induced fatigue in the motor-bogie interface
• Estimated Labor Cost: $23,000–$45,000
• Repair Time: 10–16 hours (includes bogie access, alignment tools, re-centering)
<span data-metadata=""><span data-buffer="">Analytical Diagnostic Summary
1. Vibration Signatures:
• Time-domain data reveals high-energy impacts during acceleration and deceleration.
• Envelope RMS and kurtosis values exceed normal thresholds during these dynamic phases.
2. Cluster Analysis (AI):
• Vibration data was grouped into 3 clusters based on severity.
• This sample was classified under Cluster 3: Anomalous behavior with elevated RMS and non-linear kurtosis—typical of coupling faults.
• Distinct deviation from baseline Cluster 1 (healthy behavior).
3. Mechanical Interpretation:
• The AI and vibration trends point to mechanical looseness or misalignment at the coupling connection.
• Impacts correlate with speed change, suggesting torque shock transmission through the joint.
• Possible contributing factors: bolt fatigue, coupling wear, or improper alignment between motor shaft and gearbox input.
<span data-metadata=""><span data-metadata=""><span data-buffer="">Observed Anomaly AI-Detected Condition
Non-uniform surface wear detected on Wheel No. 2. No bearing defect observed.
<span data-metadata="">About
This diagnostic report was prepared as part of a predictive maintenance program for metro rail vehicles equipped with smart vibration sensors. The objective is to detect early-stage mechanical anomalies in the wheel-rail interface, enabling proactive intervention before safety or reliability is compromised.
The analysis focuses on Wheel No. 2, using time-domain vibration signals, polar wear mapping, and AI-based clustering to determine the presence and severity of wear. The sensor data was collected during regular operation under standard load and speed conditions.
Advanced signal processing techniques and machine learning models were applied to extract meaningful patterns that indicate localized surface wear. This report is part of a broader fleet-wide diagnostic framework supported by edge computing and intelligent monitoring systems.
<span data-metadata=""><span data-buffer="">Wear surface of the wheel tread, likely due to:
Detectable by
• Irregular braking or skidding
• Track surface interaction
• Minor flat spot progression
• No inner or outer race bearing frequency peaks observed (BPFI, BPFO absent).
<span data-metadata=""><span data-metadata=""><span data-buffer="">ROI Impact
Detects flat spots and polygonal wear
Reduces lathe-based reprofiling and track wear
Estimated Labor Cost: $14,000–$17,000
Repair Time: 5–7 hours
<span data-metadata=""><span data-buffer="">Analytical Diagnostic Summary
Diagnostic Summary
1. Time-Domain Vibration Signal
• Vibration amplitude shows localized increases up to ±7.5 g, clearly above the baseline.
• Peaks occur periodically, especially during rotation, indicating surface irregularities.
• Wear pattern suggests a repeatable impact typical of tread surface damage.
2. Polar Wear Pattern Analysis
• Polar plot reveals asymmetrical peaks at ~60°, 180°, and 300°, confirming non-uniform contact wear.
• These zones correlate with worn segments of the wheel tread interacting with the rail.
3. AI-Based Clustering
• Vibration features (RMS, Crest Factor) were analyzed using unsupervised machine learning (K-Means).
• Data classified into 3 clusters:
• Cluster 1: Normal behavior
• Cluster 2: Progressive wear
• Cluster 3: High-impact response – this sample falls here
• Confirms mechanical degradation without roller bearing defect.
<span data-metadata=""><span data-metadata=""><span data-buffer="">Observed Anomaly AI-Detected Condition
Track surface irregularity / potential rail crack detected based on consistent natural vibration events recorded by multiple axle-mounted sensors.
<span data-metadata="">About
This diagnostic analysis was generated using Delphisonic’s smart vibration sensors integrated with a mesh communication topology. These sensors are mounted on individual wheels and continuously monitor dynamic interactions between the wheel and the rail.
The purpose of this report is to determine the location and likelihood of a rail defect based on naturally induced vibration signals and sensor-to-sensor consensus mechanisms.
<span data-metadata=""><span data-buffer="">Mechanical Interpretation
Detectable by
• The system suggests a possible surface crack or localized defect on the railhead, affecting the wheel-rail contact interface.
• The shape and symmetry of the vibrations rule out wheel or bearing faults.
• Propagation pattern matches the typical impact response of a rail flaw.
<span data-metadata=""><span data-metadata=""><span data-buffer="">ROI Impact
Replaces wayside inspection with onboard AI
Prevents derailment, human patrol reduction
Estimated Labor Cost: $19,000–$25,000 per event
Repair Time: 6–10 hours (localized weld or rail change)
<span data-metadata=""><span data-buffer="">Analytical Diagnostic Summary
1. Time-Domain Analysis:
• All three sensors (Sensor 1 on Wheel 1, Sensor 2 on Wheel 2, Sensor 3 on Wheel 3) recorded synchronized impact-like peaks as the wheels passed over a specific rail segment.
• These peaks occurred at the same longitudinal position, indicating that a physical irregularity is fixed on the rail.
2. Frequency-Domain (FFT) Analysis:
• Localized spike in low-to-mid range frequency content (e.g., 150 Hz, 900 Hz, 1350 Hz) correlated across sensors.
• Harmonic pattern indicates a track-level resonance rather than a rolling component defect.
3. Mesh-Based Confirmation:
• Each sensor transmits its local event data to its neighboring sensors in a real-time mesh topology.
• Upon confirming that all three sensors detected similar anomaly within a precise time window, the system classified this as a verified track anomaly.
• No detection was dependent on trackside infrastructure, purely vehicle-borne intelligence.
<span data-metadata=""><span data-metadata=""><span data-buffer="">Detection Method: AI-based Envelope Analysis + Visual Inspection
Automated diagnostics point to corrosion pits and streaking aligned with known moisture corrosion fault modes. Corrosion-related anomalies were identified by the AI model, indicating early-stage oxidative damage on the bearing surface.
<span data-metadata="">About
This report analyzes bearing damage caused by prolonged exposure to moisture inside the axlebox housing. The failure was detected using AI-supported signal analysis, which identified characteristic corrosion-related signal patterns. Visual confirmation was performed by examining the raceway for rust, pitting, and etching.
<span data-metadata=""><span data-buffer="">Mechanical Interpretation
Detectable by
The AI model recognized corrosion-related anomalies in the envelope spectrum, supported by a moderate rise in vibration RMS and Crest Factor.
These patterns matched a moisture contamination signature, typically associated with insufficient sealing, water ingress, or lubricant breakdown.
Visual Confirmation
• Vertical rust lines observed on the raceway surface
• No rotational pitting or brinelling, indicating static exposure to moisture
• Evidence of etching due to water-oil interaction during idle periods.
Root Cause
• Water ingress through compromised seals or condensation during shutdowns
• Oil degradation leading to loss of protective lubrication film
• Lack of movement allowing localized oxidation (“static corrosion”)
<span data-metadata=""><span data-metadata=""><span data-buffer="">ROI Impact
Prevents misdiagnosed early replacements
Detects slow surface erosion under humidity
Estimated Labor Cost: $15,000–$18,000
Repair Time: 6–8 hours
<span data-metadata=""><span data-metadata=""><span data-buffer="">The system registered a progressive thermal escalation in the axlebox, automatically categorized as a severe condition.
Vehicle Type: Metro / Railcar
Component: Axlebox Bearing Assembly
Failure Mode: Overheating (Hot Box Condition)
Detection Method: Thermal Monitoring + AI Trend Analysis
<span data-metadata="">About
This diagnostic report evaluates a temperature anomaly detected in the axlebox unit. Using onboard thermal sensors combined with AI-driven anomaly detection algorithms, the system tracked an abnormal temperature escalation far beyond operational thresholds.
This condition is typically associated with:
• Bearing lubrication failure
• Excessive friction due to internal damage
• Contamination or misalignment within the bearing
• Potential fire hazard if left unresolved
<span data-metadata=""><span data-metadata=""><span data-buffer="">Analytical Diagnostic Summary
Detectable by
Thermal Trend Summary
• Initial temperature: +45°C
• Escalation to +150°C within 60 minutes
• Thresholds breached:
• Warning: 80°C
• Alarm: 110°C
• Critical: 130°C
• AI flagged anomaly after sustained breach above alarm level for >5 minutes
AI-Based Assessment
The AI engine classified this event into the “Critical Risk” cluster, consistent with historic bearing seizure or internal friction events.
Correlated with known lubrication breakdown and seal failure patterns in the learning model.
Visual Confirmation
Field inspection confirmed thermal glow at the axlebox surface (see attached photo), a classic symptom of a hot box.
No flames were observed, but discoloration and grease evaporation were present.
<span data-metadata=""><span data-metadata=""><span data-buffer="">ROI Impact
Detects bearing overheating before damage
Prevents fire, axle burn, and passenger delay
Estimated Labor Cost: $70,000–$130,000
Repair Time: 10–14 hours
System: Water Pump Drive Unit
Components Involved: Electric Motor – Coupling – Centrifugal Pump
Failure Mode: Mechanical Misalignment / Looseness
<span data-metadata=""><span data-metadata=""><span data-buffer="">Identified Fault
Abnormal vibration levels and phase shift patterns detected at the coupling interface between the motor and the pump shaft, consistent with angular and/or parallel misalignment.
<span data-metadata="">About
This diagnostic report evaluates mechanical performance of the pump-motor coupling assembly. Using triaxial vibration sensors mounted near the coupling and bearing housings, signal anomalies were detected in both horizontal and axial directions.
Detectable by
<span data-metadata=""><span data-metadata=""><span data-buffer="">ROI Impact
- Prevents bearing wear, coupling failure
- Reduces vibration energy losses
- Estimated Labor Cost: $6,000–$9,000
- Repair Time: 6–8 hours
<span data-metadata=""><span data-metadata=""><span data-buffer="">Observed Anomaly AI-Detected Condition
System: Gearbox Assembly
Failure Mode: Surface Wear / Pitting / Profile Distortion
<span data-metadata="">About
This report summarizes fault data collected from a gearbox unit exhibiting signs of gear tooth degradation. Using vibration analysis (FFT) and AI-based pattern classification, a clear deterioration trend was identified in the high-speed gear stage.
AI Classification Result
• Clustered in “Severe Tooth Wear” category
• Compared with training data:
• High kurtosis
• Multi-harmonic signature
• Sideband presence typical of pitting or misalignment-induced wear
<span data-metadata=""><span data-metadata=""><span data-buffer="">Identified Fault
Detectable by
Abnormal frequency patterns and high kurtosis levels detected in the vibration signal indicate progressive gear tooth surface damage, likely involving pitting and profile wear.
<span data-metadata=""><span data-metadata=""><span data-buffer="">Analytical Diagnostic Summary
• Harmonics observed at 1×, 2×, and 3× RPM with visible sidebands
• Sidebands suggest modulation due to tooth mesh irregularity
• Elevated RMS vibration levels in horizontal axis
• High kurtosis indicates impulsive contact between worn surfaces
<span data-metadata=""><span data-metadata=""><span data-buffer="">ROI Impact
- Prevents total gear mesh failure
- Early warning saves full gearbox rebuild
- Estimated Labor Cost: $48,000–$74,000
- Repair Time: 8–12 hours
Solutions
Trust Center
Solutions
Trust Center
Solutions
Trust Center
© Copyright 2025, All Rights Reserved by Delphisonic