A Data Driven RUL Estimation Framework of Electric Motor Using Deep Electrical Feature Learning from Current Harmonics and Apparent Power

An effective remaining useful life (RUL) estimation method is of great concern in industrial machinery to ensure system reliability and reduce the risk of unexpected failures. Anticipation of an electric motor’s future state can improve the yield of a system and warrant the reuse of the industrial asset. In this paper, we present an effective RUL estimation framework of brushless DC (BLDC) motor using third harmonic analysis and output apparent power monitoring. In this work, the mechanical output of the BLDC motor is monitored through a coupled generator. To emphasize the total power generation, we have analyzed the trend of apparent power, which preserves the characteristics of real power and reactive power in an AC power system. A normalized modal current (NMC) is used to extract the current features from the BLDC motor. Fault characteristics of motor current and generator power are fused using a Kalman filter to estimate the RUL. Degradation patterns for the BLDC motor have been monitored for three different scenarios and for future predictions, an attention layer optimized bidirectional long short-term memory (ABLSTM) neural network model is trained. ABLSTM model’s performance is evaluated based on several metrics and compared with other state-of-the-art deep learning models.

Download Full-text

Design of Landsman Converter Fed BLDC Motor using IFOC and Back EMF Estimation Method

International Journal of Recent Technology and Engineering - 2 ◽

10.35940/ijrte.e3116.018520 ◽

2020 ◽

Vol 8 (5) ◽

pp. 2958-2963

Keyword(s):

High Efficiency ◽

Low Voltage ◽

Low Cost ◽

Estimation Method ◽

Torque Ripple ◽

Bldc Motor ◽

Brushless Dc ◽

Low Voltage Operation ◽

Indirect Field Oriented Control ◽

Paper Sensor

In this paper, sensor-less control of Brushless DC (BLDC) motor drive fed from Landsman Converter (LC) powered from photovoltaic (PV) is designed to improve the performance of the motor. For obtaining the torque ripple minimization and accurate speed control using simplified Indirect Field Oriented Control (IFOC) is applied to the motor with Back-EMF estimation method. It is used for estimation of speed in sensor-less approach and implements to tracks the continuous changes of speed. This estimated speed is used to initiate the rotor position, and hall commutation signals predicted from the rotor angle. The performance of sensor less BLDC motor with low voltage operation which expresses high efficiency at low cost and reduction in torque ripple are verified using MATLAB/Simulink.

Download Full-text

Remaining Useful Life Estimation for Systems with Abrupt Failures

Annual Conference of the PHM Society ◽

10.36001/phmconf.2018.v10i1.590 ◽

2018 ◽

Vol 10 (1) ◽

Author(s):

Wei Huang ◽

Hamed Khorasgani ◽

Chetan Gupta ◽

Ahmed Farahat ◽

Shuai Zheng

Keyword(s):

Short Term Memory ◽

Failure Time ◽

Estimation Method ◽

Remaining Useful Life ◽

Short Term ◽

Mode Detection ◽

Useful Life ◽

Long Short Term Memory ◽

Early Degradation

Data-driven Remaining Useful Life (RUL) estimation for systems with abrupt failures is a very challenging problem. In these systems, the degradation starts close to the failure time and accelerates rapidly. Normal data with no sign of degradation can act as noise in the training step, and prevent RUL estimator model from learning the degradation patterns. This can degrade RUL estimation performance significantly. Therefore, it is critical to identify degradation mode during the training step. Moreover, in the application step, predicting RUL when the system is in normal mode and is not showing any sign of degradation can generate inaccurate estimations, and reduce faith in the model. In this paper, we propose a new RUL estimation method that incorporates an early degradation mode detection step to automatically identify the earliest point of time at which the degradation starts to happen. When the degradation mode is detected, a Long Short Term Memory (LSTM) neural network is applied to predict system RUL. As a case study, we apply the proposed method for RUL estimation in 2018 PHM Data Challenge. The case study demonstrates that our solution achieves more accurate RUL estimation compared to several baseline methods.

Download Full-text

Improving the efficiency of oilfield electric power distribution

Oil and Gas Power Engineering ◽

10.31471/1993-9868-2019-2(32)-79-87 ◽

2019 ◽

pp. 79-87

Author(s):

Yu. F. Yu. F. Romaniuk ◽

О. V. Solomchak ◽

М. V. Hlozhyk

Keyword(s):

Power Distribution ◽

Reactive Power ◽

Power Transmission ◽

Transmission Efficiency ◽

Active Power ◽

Oil Field ◽

Total Power ◽

Simultaneous Increase ◽

Active Load ◽

Step Down

The issues of increasing the efficiency of electricity transmission to consumers with different nature of their load are considered. The dependence of the efficiency of the electric network of the oil field, consisting of a power line and a step-down transformer, on the total load power at various ratios between the active and reactive components of the power is analyzed, and the conditions under which the maximum transmission efficiency can be ensured are determined. It is shown by examples that the power transmission efficiency depends not only on the active load, but also largely on its reactive load. In the presence of a constant reactive load and an increase in active load, the total power increases and the power transmission efficiency decreases. In the low-load mode, the schedule for changing the power transmission efficiency approaches a parabolic form, since the influence of the active load on the amount of active power loss decreases, and their value will mainly depend on reactive load, which remains unchanged. The efficiency reaches its maximum value provided that the active and reactive components of the power are equal. In the case of a different ratio between them, the efficiency decreases. With a simultaneous increase in active and reactive loads and a constant value of the power factor, the power transmission efficiency is significantly reduced due to an increase in losses. With a constant active load and an increase in reactive load, efficiency of power transmission decreases, since with an increase in reactive load, losses of active power increase, while the active power remains unchanged. The second condition, under which the line efficiency will be maximum, is full compensation of reactive power. Therefore, in order to increase the efficiency of power transmission, it is necessary to compensate for the reactive load, which can reduce the loss of electricity and the cost of its payment and improve the quality of electricity. Other methods are also proposed to increase the efficiency of power transmission by regulating the voltage level in the power center, reducing the equivalent resistance of the line wires, optimizing the loading of the transformers of the step-down substations and ensuring the economic modes of their operation.

Download Full-text

Optimization Design on Power Filter of TCR Under Asymmetrical Voltage

The Open Electrical & Electronic Engineering Journal ◽

10.2174/1874129001509010591 ◽

2015 ◽

Vol 9 (1) ◽

pp. 591-599

Author(s):

Ma Wenchuan ◽

Zhitong Li ◽

Chen Daochang ◽

Qi Jiaming ◽

Zhou Qiang ◽

...

Keyword(s):

Optimization Design ◽

Reactive Power ◽

Estimation Method ◽

Reactive Power Compensation ◽

Harmonic Suppression ◽

Matlab Simulation ◽

Steel Mill ◽

Harmonic Current ◽

Third Harmonic ◽

Power Filter

For resolving the problem that power filter cannot work normally because TCR (thyristor controlled reactor) generates extra third harmonic current under asymmetrical voltage, the paper proposes the estimation method of current capacity that TCR generates extra third harmonic current under asymmetrical voltage. Considering extra third harmonic current under asymmetrical voltage, Optimum method based on genetic algorithm is used to design the parameters of power filter. With reactive power compensation and harmonic suppression project of a steel mill as example, the proposed method is simulated by Matlab. Simulation results show optimized power filter can eliminate extra third harmonic current effects under asymmetrical voltage, meet the requirement of reactive power compensation, reduce harmonics current that load injects into system, and guarantee the power filter safe operation under asymmetrical voltage.

Download Full-text

Leveraging Label Information in a Knowledge-Driven Approach for Rolling-Element Bearings Remaining Useful Life Prediction

Energies ◽

10.3390/en14082163 ◽

2021 ◽

Vol 14 (8) ◽

pp. 2163

Author(s):

Tarek Berghout ◽

Mohamed Benbouzid ◽

Leïla-Hayet Mouss

Keyword(s):

Transfer Learning ◽

Short Term Memory ◽

Remaining Useful Life ◽

Accelerated Life Tests ◽

Learning Path ◽

Accelerated Life ◽

Unseen Data ◽

Label Information ◽

Life Tests ◽

Ill Posed

Since bearing deterioration patterns are difficult to collect from real, long lifetime scenarios, data-driven research has been directed towards recovering them by imposing accelerated life tests. Consequently, insufficiently recovered features due to rapid damage propagation seem more likely to lead to poorly generalized learning machines. Knowledge-driven learning comes as a solution by providing prior assumptions from transfer learning. Likewise, the absence of true labels was able to create inconsistency related problems between samples, and teacher-given label behaviors led to more ill-posed predictors. Therefore, in an attempt to overcome the incomplete, unlabeled data drawbacks, a new autoencoder has been designed as an additional source that could correlate inputs and labels by exploiting label information in a completely unsupervised learning scheme. Additionally, its stacked denoising version seems to more robustly be able to recover them for new unseen data. Due to the non-stationary and sequentially driven nature of samples, recovered representations have been fed into a transfer learning, convolutional, long–short-term memory neural network for further meaningful learning representations. The assessment procedures were benchmarked against recent methods under different training datasets. The obtained results led to more efficiency confirming the strength of the new learning path.

Download Full-text

Convolutional Long Short-Term Memory Autoencoder-Based Feature Learning for Fault Detection in Industrial Processes

IEEE Transactions on Instrumentation and Measurement ◽

10.1109/tim.2020.3039614 ◽

2021 ◽

Vol 70 ◽

pp. 1-15

Author(s):

Jianbo Yu ◽

Xing Liu ◽

Lyujiangnan Ye

Keyword(s):

Fault Detection ◽

Short Term Memory ◽

Feature Learning ◽

Industrial Processes ◽

Short Term ◽

Term Memory ◽

Long Short Term Memory

Download Full-text

Data Independent Acquisition Based Bi-Directional Deep Networks for Biometric ECG Authentication

Applied Sciences ◽

10.3390/app11031125 ◽

2021 ◽

Vol 11 (3) ◽

pp. 1125

Author(s):

Htet Myet Lynn ◽

Pankoo Kim ◽

Sung Bum Pan

Keyword(s):

Short Term Memory ◽

Feature Learning ◽

Biometric Authentication ◽

Qrs Complex ◽

Data Types ◽

Ecg Signals ◽

Data Independent Acquisition ◽

Deep Networks ◽

Classification Tasks ◽

Electrocardiogram Ecg

In this report, the study of non-fiducial based approaches for Electrocardiogram(ECG) biometric authentication is examined, and several excessive techniques are proposed to perform comparative experiments for evaluating the best possible approach for all the classification tasks. Non-fiducial methods are designed to extract the discriminative information of a signal without annotating fiducial points. However, this process requires peak detection to identify a heartbeat signal. Based on recent studies that usually rely on heartbeat segmentation, QRS detection is required, and the process can be complicated for ECG signals for which the QRS complex is absent. Thus, many studies only conduct biometric authentication tasks on ECG signals with QRS complexes, and are hindered by similar limitations. To overcome this issue, we proposed a data-independent acquisition method to facilitate highly generalizable signal processing and feature learning processes. This is achieved by enhancing random segmentation to avoid complicated fiducial feature extraction, along with auto-correlation to eliminate the phase difference due to random segmentation. Subsequently, a bidirectional recurrent neural network (RNN) with long short-term memory (LSTM) deep networks is utilized to automatically learn the features associated with the signal and to perform an authentication task. The experimental results suggest that the proposed data-independent approach using a BLSTM network achieves a relatively high classification accuracy for every dataset relative to the compared techniques. Moreover, it exhibited a significantly higher accuracy rate in experiments using ECG signals without the QRS complex. The results also revealed that data-dependent methods can only perform well for specified data types and amendments of data variations, whereas the presented approach can also be considered for generalization to other quasi-periodical biometric signal-based classification tasks in future studies.

Download Full-text

Using Energy Storage Inverters of Prosumer Installations for Voltage Control in Low-Voltage Distribution Networks

Energies ◽

10.3390/en14041121 ◽

2021 ◽

Vol 14 (4) ◽

pp. 1121

Author(s):

Rozmysław Mieński ◽

Przemysław Urbanek ◽

Irena Wasiak

Keyword(s):

Energy Storage ◽

Control Strategy ◽

Reactive Power ◽

Low Voltage ◽

Storage System ◽

Distribution Networks ◽

Energy Storage System ◽

Voltage Regulation ◽

Active Power ◽

Total Power

The paper includes the analysis of the operation of low-voltage prosumer installation consisting of receivers and electricity sources and equipped with a 3-phase energy storage system. The aim of the storage application is the management of active power within the installation to decrease the total power exchanged with the supplying network and thus reduce energy costs borne by the prosumer. A solution for the effective implementation of the storage system is presented. Apart from the active power management performed according to the prosumer’s needs, the storage inverter provides the ancillary service of voltage regulation in the network according to the requirements of the network operator. A control strategy involving algorithms for voltage regulation without prejudice to the prosumer’s interest is described in the paper. Reactive power is used first as a control signal and if the required voltage effect cannot be reached, then the active power in the controlled phase is additionally changed and the Energy Storage System (ESS) loading is redistributed in phases in such a way that the total active power set by the prosumer program remains unchanged. The efficiency of the control strategy was tested by means of a simulation model in the PSCAD/EMTDC program. The results of the simulations are presented.

Download Full-text