Integrated Prognostics Observer for Condition Monitoring of an Automated Manual Transmission Dry Clutch System

Ramalingam, Sivakumar; Prasad, Hanumath Vv; Regalla, Srinivasa Prakash

doi:10.36001/ijphm.2017.v8i2.2638

Cited by 2 publications

(2 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Analyzing the position data of either the actuator or the pedal is crucial to identifying potential issues with the clutch. For instance, if there's a malfunction like misplacement, sticking, or a sudden shift in the clutch, the sensor can relay this information for inspection by a mechanic [3].…”

Section: Introductionmentioning

confidence: 99%

Deep learning based fault detection of automobile dry clutch system using spectrogram plots

Sai S,

Sridharan,

Chakrapani

et al. 2024

Eng. Res. Express

View full text Add to dashboard Cite

In this research study, a novel approach that leverages the power of transfer learning (a famous deep learning technique) is proposed to diagnose multiple types of clutch faults including, worn release fingers, fractured pressure plates, deteriorated pressure plates, loss of friction material and distorted tangential strips using spectrogram plots. To train and validate the diagnostic system, vibration readings were taken from a specially designed test rig with the help of piezoelectric accelerometer while the clutch system was operated under different load conditions of 0 (no load), 5 and 10 kg This procedure of data collection was then repeated to acquire the vibration data for all of the fault conditions by replacing the good with fault components individually. These vibration signals were further processed and transformed into spectrogram plots that serves as the input data for the deep learning models considered. Fine-tuning techniques were applied on pretrained networks to maximise the prediction accuracy of the models to effectively determine and diagnose faults in the clutch system. For this study 12 pre-trained networks were chosen namely, Xception, InceptionResNet, DenseNet, AlexNet, VGG16, GoogLeNet, VGG19, ResNet101, ResNet50, InceptionV3, MobileNetV2 and ShuffleNet. To optimize the performance of deep learning models, a systematic adjustment of hyperparameters such as the train-test split ratio, learning rate, optimizer and batch size for each network model was carried out. Through careful experimentation and analysis, significant improvements in fault classification accuracy were achieved thereby enhancing the reliability and effectiveness of the diagnostic system. From the results it was noted that 100% classification accuracy was displayed by AlexNet (for the no load condition and the 10kg load condition) and GoogLeNet (for 5 kg load condition) with extremely low computation times.

show abstract

Section: Introductionmentioning

confidence: 99%

Deep learning based fault detection of automobile dry clutch system using spectrogram plots

Sai S,

Sridharan,

Chakrapani

et al. 2024

Eng. Res. Express

View full text Add to dashboard Cite

show abstract

“…Remaining useful lifetime was predicted based on this health index by a weighted moving average. Another data-driven approach for dry clutches was presented in (Ramalingam, Prasad, Regalla, & Srinivasa Prakash, 2017). Lifetime prediction by using measured clutch actuator position was illustrated, but a proof of concept was missing.…”

Section: Introductionmentioning

confidence: 99%

Remaining Lifetime Prediction for Reliability-based Dry Clutch Control

Strommenger¹

2018

PHME_CONF

View full text Add to dashboard Cite

Increasing requirements for reliability of modern powertrainscan be achieved by predictive maintenance and reliabilitybasedcontrol based on lifetime prediction. This contributionpresents lifetime prediction for a dry clutch, being an essentialcomponent of automated manual transmissions. Modelbaseddevelopment of lifetime prediction requires knowledgeof dry clutch wear, which was identified in previous experiments.The derived wear model allows estimation of characteristicwear-dependent values, like friction lining materiallosses and friction coefficient changes. Based on these estimatedvalues the presented lifetime prediction was developedby fusing these estimated values into a health index (HI) describingthe systems healthiness. Furthermore, the remaininguseful lifetime (RUL) becomes predictable from observationsof health index trend using an exponential weightedmoving average. Eventually, the presented lifetime predictionwas implemented and tested on real-time operating hardwaresimilar to common transmission control units. In order tocontrol the system lifetime in normal operation, target trendsfor health index and predicted remaining useful lifetime weredefined. Based on trend deviations, a fuzzy-logic based controlstrategy was realized, which sets the optimization targetfor a reliability-based control. Thus, the optimization targetcan be varied between comfort-optimized or wear-optimizedclutch engagements. Finally, an outline of reliability-basedcontrol concepts is given.

show abstract

Integrated Prognostics Observer for Condition Monitoring of an Automated Manual Transmission Dry Clutch System

Cited by 2 publications

References 13 publications

Deep learning based fault detection of automobile dry clutch system using spectrogram plots

Deep learning based fault detection of automobile dry clutch system using spectrogram plots

Remaining Lifetime Prediction for Reliability-based Dry Clutch Control

Contact Info

Product

Resources

About