Non-stationary Operating Research Articles

Reconstruction of angular speed variations in the angular domain to diagnose and quantify taper roller bearing outer race fault

Abstract Monitoring of rotating machines is usually performed on vibration signals and associated signal processing tools focus on extracting weak cyclic information induced by faults on mechanical components like bearings. Their sensitivity to operating conditions in speed and torque are of huge interest in recent research activities. An alternative signal consists in measuring Instantaneous Angular Speed (IAS) of the machine which has been empirically demonstrated as a good vector of information. Angularly sampled by nature, this signal investigates other transfer paths from fault to transducer and introduces new challenges in understanding the relationships between operating conditions and effects of faults on the rotating speed of the machine. The main contribution of this paper lies on the introduction of basic signal processing techniques in order to analyze the shape of rotating speed fluctuations introduced by bearing faults. Modifications due to operating conditions and size of the defect are clearly separated in our suggested analysis on IAS. Another point of interest relies on the fact that this tool can be used in stationary or non-stationary operating conditions. All the analysis are given on experimental data performed on a specific test bench with different artificial bearing faults in various operating conditions.

Efficiency at Maximum Power of Dissipative Thermoelectric Generators: A Finite-time Thermodynamic Analysis

In this paper we analyze the thermodynamic efficiency expected for a fully dissipative thermoelectric generator (TEG) operating under stationary conditions at a finite rate. Although a finite-time thermodynamic analysis of TEGs has been aimed at since long time, no complete theory is available yet. The state of the art of theory is reviewed, and a simple expression for the maximum achievable efficiency of TEGs operating under fully irreversible conditions is obtained. This also sets a reference efficiency for forthcoming studies of nonstationary operation of TEGs.

Advanced bearing diagnostics: A comparative study of two powerful approaches

Abstract The last decade has witnessed spectacular advances in vibration-based fault detection of rotating machines and, in particular, rolling element bearings. Nowadays, the related state of the art can be considered mature thanks to a set of powerful signal processing techniques able to denoise and process the vibration signal to detect fault symptoms. Among these techniques, two emerging approaches have specifically captured the interest of the scientific community thanks to their efficiency and robustness. They have also been recommended in the bearing diagnostic tutorial written by professors R. B. Randall & J. Antoni, published in MSSP in 2011. The first approach consists of pre-processing the random part of the vibration signal (after removal of deterministic components) through the minimum entropy deconvolution (MED) method, followed by the spectral kurtosis (SK), before analyzing the spectrum of the signal envelope. The MED enhances the signal impulsivity by deconvolving the system transfer function through an optimization approach that maximizes the kurtosis of the filter output. Then, the SK is applied to conceive the optimal filter that promotes the most informative spectral band before computing the (squared) envelope spectrum. The second approach is based on a cyclostationary modeling of the bearing signal. It applies a bi-variable map— called the spectral coherence— of (i) the cyclic frequency, which describes the cyclic content of modulations, and (ii) the spectral frequency which describes the spectral content of the carrier. When applied to the random part of the signal, this quantity is able to detail the signal in this plane according to the signal-to-noise ratio, thus allowing weak fault components to appear in the distribution. This paper investigates and compares these two approaches on real bearing vibration datasets including run-to-failure tests. The study also addresses the extension of these approaches to the nonstationary operating regime.

An intelligent diagnosis scheme based on generative adversarial learning deep neural networks and its application to planetary gearbox fault pattern recognition

Abstract Planetary gearbox has complex structures and works under various non-stationary operating conditions. The vibration signals of planetary gearbox are complicated and usually polluted by noise and interference. It is difficult to extract effective features of early faults. In addition, there are only a small number of fault samples for planetary gearbox fault diagnosis. All of these increase the difficulty of planetary gearbox fault diagnosis. Aiming at these problems, a novel fault diagnostic method is proposed which combines Generative Adversarial Networks (GAN) and Stacked Denoising Autoencoders (SDAE). The generator of GAN can generate new samples which has similar distribution with original samples from planetary gearbox vibration signals. Then, generated samples are transformed to the discriminator together with original samples which expand the sample size. SDAE is used as the discriminator of GAN which can automatically extract effective fault features from input samples and discriminate their authenticity and fault categories. Through novel adversarial machine learning mechanism, the generator and discriminator are concurrently optimized to enhance the quality of generation samples and the ability of fault mode classification. The experimental results show that the developed SDAE-GAN method for planetary gearbox has good anti-noise ability and achieve better fault diagnosis performance in the case of small samples.

Experimental study of non-stationary operation of a dual-wavelength passively mode-locked fibre ring laser

In this paper, we report an experimental study of different dynamics taking place in a 20 m long passively mode-locked fibre ring laser in dual-wavelength operation, at 1531 nm and 1558 nm. For different polarization adjustments, self-starting mode locking is obtained, yielding different types of emission: bunches of solitons in quasi-stationary regime, a compact bunch of solitons coexisting with loose bunches of solitons, a noise-like pulse coexisting with bunches of solitons and a noise-like pulse displaying quasi-periodic fluctuations. In each regime, we extract information on the pulse dynamics from measurements of the temporal profile evolution using a 16 GHz real-time oscilloscope and, at the same time, we propose a phase-space diagram representation of the intensity versus the energy of the temporal profile of the pulses; the latter allows evidencing patterns that could not be identified using conventional measurement techniques.

Non-stationary operation of gas pipeline based on selections of travel

A technique for calculating the nonstationary regime for the movement of natural gas along a multi-stranded gas main with allowance for travel selections is proposed. On the total flow of transported gas, gas flows are determined at all sections of the pipeline, taking into account the pressure in the controlled areas. When compiling the calculation formula for Φ non-stationary mode, the formulas for calculating travel selections were used.

Mathematical Models for Investigation of WWER-1000/320 Transients

The paper presents mathematical models of the reactor WWER-1000/320, which are designed to investigate non-stationary operating modes of the reactor. The models in relative state variables include a point model of neutron kinetics with six groups of delayed neutrons and models of thermal processes, gradual heat generation, change in xenon concentration. The effects of reactivity on the movement of control rods and changes in reactor power, the effects of reactivity on changes in fuel and coolant temperatures, effect of change in the concentration of xenon are taken into account. The values of the constant parameters of the models are given for the start of stationary fuel loading.

Statements on wavelet packet energy–entropy signatures and filter influence in fault diagnosis of induction motor in non-stationary operations

Electric motors are components of great importance in mechanical systems and in the majority of the equipment used in industrial plants. The several faults that occur in the induction machines may induce severe consequences in the industrial process. Many of these faults are progressive. In this work, a contribution to the study of signal-processing techniques based on wavelet packet transform for parameter extraction of energy and entropy from vibration signals for the detection of faults in the non-stationary operation (start of the motor) is presented. Together with the wavelet transform, methods of dimensionality reduction such as principal component analysis, linear discriminant analysis, and independent components analysis are used. In addition, the use of an experimental bench shows that the model of extraction and classification proposed present high precision for fault classification.

Optimal operation of a Solar Membrane Distillation pilot plant via Nonlinear Model Predictive Control

Solar Membrane Distillation (SMD) is an under-investigation desalination process suitable for developing self-sufficient small scale applications. The use of solar energy considerably reduces the operating costs, however, its intermittent nature requires a non-stationary optimal operation that can be achieved by means of advanced control strategies. In this paper, a hierarchical control system composed by two layers is used for optimizing the operation of a SMD pilot plant, in terms of thermal efficiency, distillate production and cost savings. The upper layer is formed by a Nonlinear Model Predictive Control (NMPC) scheme that allows us to obtain the optimal operation by optimizing the solar energy use. The lower layer includes a direct control system, in charge of attaining the variable references provided by the upper layer. Simulation and experimental tests are included and commented in order to demonstrate the benefits of the developed control system.

Method of Data storing, collection and aggregation for definition of life-cycle resources of electromechanical equipment

Analysis of technical and technological conditions for the emergence of emergency situations during the operation of electromechanical equipment of enterprises of the mineral and raw materials complex shows that when developing the basis for ensuring safe operation, it is necessary to take into account not only the technical condition, but also the non-stationary operation of the operating conditions of equipment, and the nonstationarity of operational operating parameters of technological processes. Violations of the operation of individual parts of the machine, not detected in time, can lead to severe accidents at work, as well as to unplanned downtime and loss of profits. That is why, the issues of obtaining and processing Big data obtained during the life cycle of electromechanical equipment, for assessing the current state of the electromechanical equipment used, timely diagnostics of emergency and pre-emergency modes of its operation, estimating the residual resource, as well as prediction the technical state on the basis of machine learning are very important. This article is dedicated to developing the special method of data storing, collection and aggregation for definition of life-cycle resources of electromechanical equipment. This method can be used in working with big data and can allow extracting the knowledge from different data types: the plants’ historical data and the factory historical data. The data of the plants contains the information about electromechanical equipment operation and the data of the factory contains the information about a production of electromechanical equipment.

Hybrid Approach to Remaining Useful Life Prediction of Solid Oxide Fuel Cell Stack

Improvement in efficiency and reliability are essential for more intensive deployment and exploitation of solid oxide fuel cell (SOFC) systems in modern society. Having the information about current health, degradation rate, and estimated remaining useful life (RUL) of the stack, at all times during operation, can greatly benefit the control system performance. Such information can be beneficial also for the service providers and end-users as the maintenance needs, including stack replacements, can be planned in advance. The control system can be able to optimize performance through a trade-off between maximal life span and best efficiency of power conversion. The knowledge on health status and its anticipated evolution assures easy and cost-effective maintainability. However, works that address issue of RUL prediction for SOFC are scant. On top of that, most of authors define voltage as health index and predict RUL accordingly. Such a selection of health index becomes inappropriate when SOFC is working under varying load conditions. In such a case stack voltage becomes affected by the changes in operating conditions and it is difficult to correlate drop in voltage with degradation rate. In this paper, we propose a novel hybrid approach to RUL prediction of SOFC systems, which overcomes the limitations of the known approaches and allows for reliable RUL prediction in non-stationary operating conditions. The approach consists of three main modules: (i.) estimation of area specific resistance (ASR) of the stack, (ii.) prediction of its future progress based on collected data, and (iii.) prediction of RUL. The methodology is evaluated on a 6 kW SOFC system. Figure 1

Hybrid Approach to Remaining Useful Life Prediction of Solid Oxide Fuel Cell Stack

Improvement in efficiency and reliability are essential for more intensive deployment and commercial exploitation of solid oxide fuel cell (SOFC) systems. Apart of advancement in fabrication of new materials and stack designs, there emerges a strong need for innovative control strategies capable of balancing maximal stack life and efficiency of power conversion in a trade-off manner. Reliable online estimation of stack health and prediction of the remaining useful life (RUL) play a key role in new generation of SOFC control systems. In most works until today, the authors utilize voltage as a health index and based on that predict the RUL. Unfortunately, such an approach becomes inappropriate when the SOFC is operating under varying load conditions and, in particular, when the SOFC ages. In this paper, we propose a novel hybrid approach to RUL prediction of SOFC systems, which overcomes the limitations of the known approaches and allows for reliable RUL prediction in non-stationary operating conditions. The approach consists of three main parts, executed continuously online: (i) estimation of area specific resistance (ASR) of the stack, (ii) prediction of its future progress based on collected data, and (iii) prediction of RUL. The methodology is evaluated on a 6 kW SOFC system.

Initial service life data towards structural health monitoring of a concrete arch dam

This paper presents a statistical framework to monitor the performance of an operational concrete arch dam using sensory data acquired during its initial service life. One of the major challenges in dealing with a newly constructed dam is to predict its long-term behaviour by forecasting appropriate thresholds using limited data exhibiting nonstationarity. In this paper, a hybrid model is implemented to predict dam responses using environmental—hydrostatic, seasonal, and temperature—as well as age-related variables. The data from multiple sensors are first analyzed using principal component analysis to incorporate overall dam behaviour into a prediction model. The proposed prediction framework is then employed to estimate the residuals and control limits required to calculate thresholds under nonstationary operating conditions during its initial service life. The dam performance is then monitored using statistical control charts and anomalies are detected by comparing the test statistics, square prediction error, and Hotelling T-squared, calculated from the residuals with the preset control limits. The issue of limited data is addressed by updating the model parameters and thresholds periodically, which is aimed at minimizing the false alarm rate. The proposed method is demonstrated using a 130-m-high double-arch concrete dam located in Bulgaria.

A simplified method for the numerical calculation of nonstationary heat transfer through a flat wall

The nonstationary operation modes of power equipment lead to the nonstationary regimes of heat exchange, in particular heat transfer. The transient processes, related to the accumulation of energy in the heat-transmitting surfaces, which manifest themselves in this case, may affect manageability of the work of equipment. Although in many cases, with a proper approximation, the heat exchange surfaces can be represented in the form of a simple variant of an infinite plate, the existing methods and tools for solving the problems on nonstationary heat transfer are built from the positions of universalism, introducing unjustified complications and hampering the generalization of numerical results obtained. We developed a simplified discrete analog to solve the one-dimensional problems on nonstationary heat transfer through an infinite plate. The realized approach allowed us to obtain the analog and results of calculations based on it in the dimensionless form, which substantially facilitates their generalization. A high stability of computational process is demonstrated relative to the selection of a number of nodes in computational grid and calculation step by time. The possibility of using the maximally small computational grids (3 nodes) makes it possible, at the current calculation step by time. To obtain an analytical solution for determining the temperatures at the surfaces of the plate at initial use of boundary conditions of the third kind. As a result, accumulated energy can be defined as a difference in the heat fluxes at the surfaces of the plate. Performing the calculations on the maximally small grids might be useful to solve the inverse problems on thermal conductivity.

Automatic feature extraction of time-series applied to fault severity assessment of helical gearbox in stationary and non-stationary speed operation

Signals captured in rotating machines to obtain the status of their components can be considered as a source of massive information. In current methods based on artificial intelligence to fault severity assessment, features are first generated by advanced signal processing techniques. Then feature selection takes place, often requiring human expertise. This approach, besides time-consuming, is highly dependent on the machinery configuration as in general the results obtained for a mechanical system cannot be reused by other systems. Moreover, the information about time events is often lost along the process, preventing the discovery of faulty state patterns in machines operating under time-varying conditions. In this paper a novel method for automatic feature extraction and estimation of fault severity is proposed to overcome the drawbacks of classical techniques. The proposed method employs a Deep Convolutional Neural Network pre-trained by a Stacked Convolutional Autoencoder. The robustness and accuracy of this new method are validated using a dataset with different severity conditions on failure mode in a helical gearbox, working in both constant and variable speed of operation. The results show that the proposed unsupervised feature extraction method is effective for the estimation of fault severity in helical gearbox, and it has a consistently better performance in comparison with other reported feature extraction methods.

A hybrid health condition monitoring method in milling operations

Cutting chatter has been a very important issue in the milling operations due to its unexpected and uncontrollable characteristics. Developing an effective healthy condition monitoring method is critical to identify cutting chatter exactly. In this paper, a hybrid healthy condition monitoring (HHCM) method, that combines variational mode decomposition (VMD) with genetic algorithm-based back propagation neural network (BPNN) model, is developed for cutting chatter detection and state classification in complex and non-stationary milling operations. First, cutting chatter vibration signal is decomposed into multiple mode components by the VMD. Then, Shannon power spectral entropy is adopted to extract features from decomposed vibration signals. Furthermore, BPNN model is optimized by traditional genetic algorithm to identify and classify machine states in milling operations. Last, impeller milling experiments are conducted and results show that the proposed HHCM method can effectively realize cutting chatter detection and state classification during milling operations.

Feedback on the Surveillance 8 challenge: Vibration-based diagnosis of a Safran aircraft engine

This paper presents the content and outcomes of the Safran contest organized during the International Conference Surveillance 8, October 20–21, 2015, at the Roanne Institute of Technology, France. The contest dealt with the diagnosis of a civil aircraft engine based on vibration data measured in a transient operating mode and provided by Safran. Based on two independent exercises, the contest offered the possibility to benchmark current diagnostic methods on real data supplemented with several challenges. Outcomes of seven competing teams are reported and discussed. The object of the paper is twofold. It first aims at giving a picture of the current state-of-the-art in vibration-based diagnosis of rolling-element bearings in nonstationary operating conditions. Second, it aims at providing the scientific community with a benchmark and some baseline solutions. In this respect, the data used in the contest are made available as supplementary material.

Angular-based modeling of induction motors for monitoring

Understanding the occurrence of bearing defects in electrical current signals using Motor Current Signal Analysis (MCSA) requires the implementation of numerical models. In this paper, an electro-magnetic-mechanical model is proposed to describe the dynamic behavior of a squirrel cage induction motor coupled to a rotating shaft supported by elastic foundations. The aim of this research work is to gain understanding of the interaction between multiphysics subsystems, mainly in faulty cases, to decipher the transfer path from the defect to its manifestation in stator currents. A new method of writing dynamic equations for simplified simulations of an induction motor is developed using an angular approach. In addition to its capacity to extend the modeling to non-stationary operating conditions, the model proposed highlights the angular periodicity of the rotating motor’s geometry. The electromagnetic field of the motor is redistributed periodically when a geometric defect occurs on a rotating part of the global system. In this case, the electromagnetic torque of the induction motor may present angularly-periodic variations. After having presented the electromagnetic-mechanical coupling methodology, the influence of torque variations is investigated and the importance of the angle-time function is highlighted.

Development of the dynamic tensor model for traffic management in a telecommunication network with the support of different classes of service

A model of traffic management, based on the tensor generalization of nonlinear differential equations of network state was proposed. This made it possible to obtain coordinated solution of the problems of routing and link resource allocation between the flows of different classes. The solution makes it possible to provide for the Quality of Service by the parameters of the average packet transmission rate and the packet delay, including the nonstationary operating mode of the network routers interfaces. The coordinate system of interpolar paths and internal node pairs was used in the tensor geometrization of the network structure. This made it possible to provide for the solution, in which the QoS requirements were satisfied not only along the basic paths, but also on the entire set of calculated routes. The problem of traffic management in the telecommunication network was set as the problem of optimization. In this case, it was necessary to calculate many sets of control variables of two types, responsible for flow routing and for determining the order of the link resource allocation in the network in the course of minimization of the quadratic criterion of optimality. The selection of the criterion of quadratic form was determined by the fact that its use contributes to obtaining solutions of the balanced use of available network resources. On the other hand, the consideration of the state dynamics of network interfaces over time made it possible to provide meeting the requirements on the packets transmission rate and the average delay not in a specific moment of time, but throughout the entire routing table update timer.

Dynamic analysis of gearbox behaviour in milling process: Non-stationary operations

Dynamic behaviour of gearbox transmitting power to milling tool in a milling process application is considered in this paper. The gearbox is subjected to variable cutting torque induced by the milling operation. This variability will lead to fluctuating speed. The dynamic behaviour of the system should be investigated according to this non-stationary operating condition. To achieve this objective, a simple model of motor-gearbox-milling tool system is developed. The influence of variable cutting torque on driving motor speed is taken into account. The gearmesh function which describes the meshing phenomena is modified according to this specific operating condition. Simulations of the dynamic behaviour of the gearbox show a simultaneous frequency and amplitude modulation. These modulations are highlighted using time frequency representation which is necessary to implement so that it is possible to achieve a correct diagnosis of the transmission. Simulations in presence of tooth crack defect confirmed the necessity to implement time frequency techniques to characterize time varying systems.