Neuron soundware extends the life of escalators in the Prague metro

The Prague Public Transport Company is taking a progressive approach to new trends in the field of predictive maintenance, reflecting the increasing expectations of passengers, as well as requirements for efficiency, human resource management, long-term investments, and the digitization of equipment and processes.

Thanks to the results of the technology using artificial intelligence to process sound and vibration signals, in cooperation with the Prague Public Transport Company, we were able to identify opportunities for extending the life of escalators, reducing their downtime, and prioritizing the activities of the maintenance team.

Over a year-long collaboration on escalator monitoring through Neuron soundware solutions has culminated in Over a year-long collaboration on escalator monitoring through Neuron soundware solutions has culminated in controlled stress testing of individual escalator components located in a subway transfer station. The test always aimed to verify the detection of an impending fault. This case study will guide you through the testing and show you the results.

The monitoring of escalators as specific equipment has resulted in the adaptation of technology and procedures that NSW normally uses in the field of industrial enterprises (monitoring pumps, fans, compressors, etc.). This case study also served to build segment expertise in the field of escalators in their entirety, not just individual sub-components.

Simulation results

Thirteen machine parts on the left part of the escalator were selected to simulate failure conditions.  These were the monitored type of escalators Thysen FT-732.

Fig. susceptible parts of the escalator fitted with sensors
Simulation groupSimulation typeFitted part
I. Rollers – chain and stepped– Cracks in the roll bandage 1, 5, 10mm
– Degreasing rolls, 1-3 rolls
II. Tracks – main/reverse, auxiliary/reverse– Ripples on the track segment
– Hole in the track segment
– Permitted track segment couplings
III. Relieving curvesAdjustment change – drop (upper and lower)
IV. Spreading materialForeign object in the ridge plate
V. HandleAdjustment change slack pressure/handle pressure too tight
VI. Guiding rulerAdjustment change – increase the gap of the guide ruler
VII. Step band alignmentChanging the adjustment – tensioning/releasing the tensioner springs
VIII. Driven gearsDraining the oil filling
IX. Worm gearboxDraining the oil filling
Table: list and types of simulations together with the escalator part to be fitted

The change in sound and anomaly detection using a mathematical model of the spreading material on the ridge plate, the loss of oil buildup in the worm gearbox and gear train, the change in the adjustment of the relief curves, and the loosening of the handle proved conclusive. If at least a weak detection occurred, the model upgrade would presumably improve the result significantly (rollers, tracks).

Table: the manifestation of the simulation on the individual components of the escalator

The neural network

This analysis used a neural network (link), which was trained on data from all 21 machines previously fitted with NSW sensors in collaboration with DPP.

Spectograms & Audio

Before the audio signal is processed by the neural network, the audio is converted into spectrograms via STFT (Short Term Fourier Transform). A spectrogram is a representation of the intensity of an audio signal at different frequencies over time.

The y-axis represents the frequency of the signal, the x-axis represents the time, and the coloration rate represents the intensity of the signal – the loudness. (Yellow = highest energy, signal intensity, Dark purple = low energy, low signal intensity).

To illustrate, the top image is the pure sound signal, the middle is a spectrogram representation of that signal, and the bottom shows the frequency distribution for the nominal sound (green curve) and the sound simulation of the damaged machine part (brown curve).

Fig. Example of graphs used for sound analysis and their frequencies

I. Spreading material stuck in the ridge plate

The crest plate is the element of the escalator that is located at the beginning and end of the escalator. It serves as a transition space between the staircase and the escalator board.

Fig. Escalator crest plate

Description of the simulated fault

While the escalator is in operation, the spreader material becomes jammed under the crest plate, causing the stair treads to grind out, and in the worst-case scenario, the teeth may break out of the crest plate. This type of failure can escalate very quickly from the beginning of the problem to its fatality, making it extremely important to implement real-time monitoring and to perform the measurements frequently.

Possible consequences of a malfunction

Spreading material trapped under the ridge plate that reaches further into the escalator can cause costly consequences such as incorrect adjustment of the stair guide ruler and its subsequent total misalignment.

Manifestations when gritting material is present in the ridge plate

The sound was distinctive even to the human ear. There was a clear difference in the recordings compared to normal operation as evidenced by the spectrogram and frequency spectrum.  These manifestations were then reflected by the model that detected the fault.

Fig.: anomaly score, sound wave, and spectrogram with a comparison of the sound before and after the placement of the spreading material in the ridge plate show its clear detection

Conclusion

As can be seen from this fault analysis, the model clearly detected the fault. When compared to normal operation without the fault, with people walking on the escalator, the model was able to distinguish the difference and did not generate false alerts.

II. Operation with lubricant leakage from the gearbox (worm gearbox)

Fig. Worm gearbox and thermal imaging camera recording of an oil leak

Description of the simulated fault

The gearbox is a set of mechanical gears located between the electric motor and the main shaft of the escalator. The purpose of the gearbox is to transfer torque from the electric motor to the main shaft and reduce the rotational speed. The gearbox is filled with oil at the bottom to keep the gearbox running smoothly, which requires a sufficient level of oil in the gearbox.

While the machine was in operation, the oil gradually leaked from the gearbox.

Possible consequences of a fault

If the gearbox leaks oil, the surrounding parts of the escalator can easily be damaged, while the lack of oil has a direct effect on the function of the bearings in the gearbox and the gears. These components can be partially or completely destroyed due to lack of oil and put the gearbox completely out of service.

Manifestations of failure

As a result of the loss of oil in the gearbox, the noise from the gearbox in the engine room gradually increased in proportion to the increasing temperature and oil loss. However, it was only audible when most of the contents of the gearbox had leaked.

Fig: In the sound recordings, the change in sound during an oil leak is already visible in the early phase of the leak, as confirmed by the frequency spectrum. In addition, the model also reacts at this stage and reports the anomaly in time.

Conclusion

Failure analysis has shown that the model is able to detect an oil leak from the gearbox if it drops below a certain level. An anomaly can be detected even from a small oil loss.

III. Handle pressure

The handle pressure is used to properly tension the handle belt so that it does not slip or fall due to excessive slack and is not overtightened or cause increased wear.

Fig.  Pressure and tension of the handle and sensor fitting

Description of the detected fault

The simulation aimed to simulate the fault and verify the possibility of detection by acoustic monitoring when the handle pressure is reduced or increased above the recommended limit.

While the machine was in operation, the model observed a reduction in the escalator handrail pressure. After maintenance intervention, the model continued to exhibit this anomaly. At this point, it was likely caused by too much pressure on the handle.  After further adjustment, everything returned to normal.

Possible consequences of the fault

Too little or too much pressure in the handrail tensioner will cause insufficient or excessive tension in the handrail, resulting in inadequate function and danger to passengers relying on the handrail for support.

Manifestations of failure

On the spot, there was absolutely no difference audible to the ear. The difference only became apparent when examining the spectrogram and frequency spectrum, where the model detected the difference.

Fig.: Example of a detected anomaly on anomaly score, audio recording, and spectrogram

Conclusion

The analysis of this fault demonstrated the model’s ability to detect both low and over-pressure conditions in the guardrail. This fault cannot be detected by listening alone (i.e. without the model). 

IV. Relief curves

The relief curve is a component of the escalator that helps relieve the load on the chain track in the bends of the escalator by lifting the chain and thus reducing the pressure of the rollers on the track. There are a total of eight lightning curves on the escalator. The simulation was carried out on the upper and lower relief curves on the left side at the top of the escalator.

Fig. Relief curve of the upper sensor fitted

Description of the simulated fault

The simulation aimed to simulate a fault and verify the possibility of detection by acoustic monitoring when the lower or upper part of the lightening curve or the whole lightening curve drops, so that the chain stops floating and starts to move along the chain path, thus overloading and wearing the chain path.

Possible consequences of a malfunction

During the operation of the machine, the model detected a change in the height of the upstroke adjustment. This failure led to chain sag and roller contact with the rail at the escalator bend. On site, the difference in sound was barely noticeable. It manifested itself by moving the chain clacking to a different location on the up stroke. However, the model worked correctly here and detected the fault.

Fig: Despite the not very noticeable difference in the spectrogram at the beginning of the upstroke height change, the model was able to detect the difference from normal operation and highlight the anomaly.

Conclusion

This simulation clearly detected the setup change/fault condition in the waveform of the recorded sound and spectrogram, and the model also showed it on the sensors on the tracks, not just on the sensors located directly on the lightning curve. However, the signal change is more pronounced on the curve.

Benefits of acoustic detection monitoring technology

This case study included four simulations that demonstrated the suitability of remote acoustic monitoring solutions using artificial intelligence data processing technology. The benefits that emerged from the testing are as follows:

  • Maintenance of transport equipment gains a tool for detecting emerging faults on monitored escalator components and the resulting ability to plan service interventions and reduce emergency escalator downtime
  • Transport Facilities Maintenance acquires a system tool for remote control of monitored escalator components in the DPP Transport Facilities Maintenance System
  • Maintenance of transport equipment gains a tool for long-term monitoring of escalator operation
  • Maintenance of transport equipment gains the possibility of storing acoustic recordings of individual monitored escalator parts for comparison by listening to recorded machine sounds over time or after repairs have been carried out

Maintenance of transport equipment can instantly display information about the operation of a specific monitored part of the escalator.

Benefits of the project for the DPP client

The project introduces Transport Facilities Maintenance to the monitoring system and the development of internal processes for working with the escalator monitoring data.

The project contributes to learning about the requirements for installing sensor technology in a metro operating environment.

The project is helping to identify escalator components suitable for the installation of acoustic data acquisition technologies. The information obtained is used to assess and select machine components suitable for monitoring by acoustic detection in the future.

The project helps to outline a way to manage the constraints arising from the shortage of skilled maintenance workers.   

Client comments on Neuron soundware solution

DPP cooperates with Neuron SW in the field of continuous monitoring and early detection of defects on 21 escalators as well as extensive simulation and early detection of real defects. The collaboration is very close across the levels of both entities and given the commitment of both teams, beneficial progress is being made in quick succession.

Our team worked with Neuron soundware to identify key locations for sensor placement on traffic devices in order to extend the interval of preventive checks based on the results of the testing. Other benefits include reduced operational downtime, extended escalator lifetime, and prioritization of maintenance team activities. Specific quantifications of the overall benefits are the subject of an ongoing study.

Comprehensive monitoring covers the operation of the escalator motor and gearbox, the condition of the crest and platform plates, the tension status of the pull chain and the position of the tensioning trolley, the tension status of the moving handrails, and the positioning of the relief curves. For these escalator components, the Neuron SW system reliably detected deviations from the nominal state and alerted the emerging problems in their initial phase.

The technical solution from Neuron soundware has proven its functionality and therefore we plan to continue our activities with Neuron SW in the field of digital monitoring of escalators.

Ing. Petr Vondráček, Head of Transportation System Service, Traffice Route Metro, Dopravní podnik hl. m. Prahy, akciová společnost

Share the article

linked in twitter facebook link

Want to read more?
Go back to the news section

Neuron Soundware
Our Office
NeuronSW SE
Panorama Business Center
Škrétova 490, Vinohrady
120 00, Praha 2
Czech Republic

ID: 07967373
VAT-ID: CZ07967373
sales@neuronsw.com
+420 774 989 993
Get our case studies and technology news into your mailbox.