A Convolutional Neural Network Aided Physical Model Improvement for AC Solenoid Valves Diagnosis

Tod, Georges; Mazaev, Ganjour; Eryilmaz, Kerem; Ompusunggu, Agusmian Partogi; Hostens, Erik; Hoecke, Sofie Van

doi:10.1109/phm-paris.2019.00044

Cited by 12 publications

(14 citation statements)

References 6 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The above methodologies using BNNs all focus on purely data-driven prediction performance. In this work, we adopt a hybrid approach by adding salient physical features from a physical model [8] to the input of the BCNN, with the aim to lower RUL errors. Moreover, an important aspect not included in these works is that uncertainty estimations should show good calibration performance.…”

Section: A Deep Learning In Phmmentioning

confidence: 99%

“…Physical models derived from first principles couple the current signal of the SV to dynamical states such as the magnetic flux and the plunger position [8]. When energizing or de-energizing the valve, its mechanical deterioration is visible through this current signal.…”

Section: B Datasetmentioning

confidence: 99%

“…Physical domain knowledge is used to improve the predictive performance of a BCNN trained on the SV's current signals. Two physical forces obtained as features from the current signals are included to the BCNN input, since they contain salient information on the degradation state of the valve [8]. Feature attribution methods [9] are applied to the trained BCNN, highlighting which aspects of its input are important for its predictions.…”

mentioning

confidence: 99%

See 2 more Smart Citations

Bayesian Convolutional Neural Networks for Remaining Useful Life Prognostics of Solenoid Valves With Uncertainty Estimations

Mazaev

Crevecoeur

Hoecke

2021

IEEE Trans. Ind. Inf.

Self Cite

View full text Add to dashboard Cite

Solenoid valves are essential components of industrial systems and therefore widely used. As they suffer from high failure rates in the field, fault prognosis of these assets plays a major role for improving their maintenance and reliability. In this work, Bayesian convolutional neural networks are used to predict the remaining useful life (RUL) of solenoid valves, by training them on the valve's current signatures. Predictive performance is further improved upon by using salient physical features obtained from an electromechanical model as the network's training input. Results show that our designed network architecture produces well-calibrated uncertainty estimations of the RUL predictive distributions, which is an important concern in prognostic decision-making. I. INTRODUCTIONT HE purpose of Prognostics and Health Management (PHM) is to enable optimal maintenance strategies, as to prevent machine failure, extend the lifetime of machines and reduce operational costs. This is achieved by detecting incipient faults, fault isolation, identification of different fault types (fault diagnostics) and fault prognosis. These techniques typically imply the analysis of healthy and/or faulty conditions indicated by process measurements.Two main approaches exist for estimating the remaining useful life (RUL). Physics-based approaches rely on physical domain knowledge, which describe normal operation and physical degradation laws. Data-driven approaches are based on condition monitoring data, which are used for constructing statistical or machine learning models [1]. The term "hybrid approaches" is commonly used for approaches that combine physics-based and data-driven techniques.Deep learning (DL) techniques are an important subcategory of data-driven approaches. Features are learned automatically at multiple levels of feature representations, which allows DL to learn complex relations mapping the input to the output directly. This contrasts DL with feature-based approaches such as decision tree ensembles or Gaussian process regression, which rely on the construction of features [2]. DL is T.

show abstract

Section: A Deep Learning In Phmmentioning

confidence: 99%

Section: B Datasetmentioning

confidence: 99%

mentioning

confidence: 99%

See 1 more Smart Citation

Bayesian Convolutional Neural Networks for Remaining Useful Life Prognostics of Solenoid Valves With Uncertainty Estimations

Mazaev

Crevecoeur

Hoecke

2021

IEEE Trans. Ind. Inf.

Self Cite

View full text Add to dashboard Cite

show abstract

“…Although limited research studies on solenoid pump diagnostics exist, several fault diagnostics techniques have been reported for solenoid valves. The authors of [20] presented the method of wavelet de-noising of vibration amplitudes for diagnostics while Georges et al [21] proposed a convolutional neural network approach for feature extraction from current signals for fault classification. On a different note, Juwita and Rosdiazli [22] proposed the use of statistical features extracted from acoustic emission signals for seal leakage detection in the control valve at an early stage.…”

Section: Motivation and Related Work A Vibration Monitoring Andmentioning

confidence: 99%

A Multi-Domain Diagnostics Approach for Solenoid Pumps Based on Discriminative Features

Akpudo

Hur

2020

IEEE Access

View full text Add to dashboard Cite

Accurate condition monitoring of industrial cyber-physical systems/components demands the use of reliable fault detection and isolation (FD&I) methodologies. Meta-heuristic algorithms for feature selection have good exploration capability for optimal discriminative feature selection for fault isolation/classification of which the Binary Particle swarm optimization (BPSO) is superior to its counterparts. This study presents a robust approach for vibration-based failure diagnostics of electromagnetic/solenoid pumps which employ a multi-domain feature extraction procedure (statistical time-domain and frequencydomain features, Mel frequency cepstral coefficients, and continuous wavelet coefficients) for capturing linear and nonlinear properties from the signals. Compared with other filter and wrapper methods for supervised feature selection, a hybrid filter-wrapper (Pearson's correlation-BPSO (ρ-BPSO)) feature selection procedure is proposed for global search of optimal discriminative (uncorrelated) features for fault diagnosis with an RBF-kernel support vector machine (SVM*). Subsequently, a practical case study involving five VSC63A5 solenoid pumps at various operating/fault conditions is presented for validating the performance of the proposed approach. Results show the superior performance of the proposed hybrid filterwrapper approach against filter-based and wrapper-based techniques for discriminative feature selection. Also, the proposed ρ-BPSO-SVM* diagnostics model performance was compared with other standard fault isolation/classification methods.

show abstract

“…Approaches based on signal processing [25] and machine learning [26] have been proposed in the literature to diagnose the state of solenoid valves leading to the development of a sensor to detect anomalies [27] or a method for grouping failures [26]. None of these approaches gives a physical explanation of the failure modes related to the solenoid parameters [28]; Other models based on the movement of the armature and Foucault current [2] and the EMD winding current curve appearing with the movement of the armature [29,30] have been developed for diagnosis, but none of these models does treat electrical and mechanical parameters as a source of failure. Morever, most control approaches are using signals such as the coil current or voltage of solenoid to monitor the parameters; yet, the main problems in such approaches are that the detected signals are prone to interference and difcult to obtain [31].…”

mentioning

confidence: 99%

Monitoring of solenoid parameters based on neural networks and optical fiber squeezer for solenoid valves diagnosis

Zahidi

Amrane

Azami

et al. 2021

IJECE

View full text Add to dashboard Cite

As crucial parts of various engineering systems, solenoid valves (SVs) operated by electromagnetic solenoid (EMS) are of great importance and their failure may lead to cause unexpected casualties. This failure, characterized by a degradation of the performances of the SVs, could be due to a fluctuations in the EMS parameters. These fluctuations are essentially attributed to the changes in the spring constant, coefficient of friction, inductance, and the resistance of the coil. Preventive maintenance by controlling and monitoring these parameters is necessary to avoid eventual failure of these actuators. The authors propose a new methodology for the functional diagnosis of electromagnetic solenoids (EMS) used in hydraulic systems. The proposed method monitors online the electrical and mechanical parameters varying over time by using artificial neural networks algorithm coupled with an optical fiber polarization squeezer based on EMS for polarization scrambling. First, the MATLAB/Simulink model is proposed to analyze the effect of the parameters on the dynamic EMS model. The result of this simulation is used for training the neural network, then a simulation is proposed using the neural net fitting toolbox to determine the solenoid parameters (Resistance of the coil R, stiffness K and coefficient of friction B of the spring) from the coefficients of the transfer function, established from the model step response. Future work will include not only diagnosing failure modes, but also predicting the remaining life based on the results of monitoring.

show abstract

A Convolutional Neural Network Aided Physical Model Improvement for AC Solenoid Valves Diagnosis

Cited by 12 publications

References 6 publications

Bayesian Convolutional Neural Networks for Remaining Useful Life Prognostics of Solenoid Valves With Uncertainty Estimations

Bayesian Convolutional Neural Networks for Remaining Useful Life Prognostics of Solenoid Valves With Uncertainty Estimations

A Multi-Domain Diagnostics Approach for Solenoid Pumps Based on Discriminative Features

Monitoring of solenoid parameters based on neural networks and optical fiber squeezer for solenoid valves diagnosis

Contact Info

Product

Resources

About