Tomasz Barszcz

Decomposition of Vibration Signals into Deterministic and Nondeterministic Components and its Capabilities of Fault Detection and Identification

Decomposition of Vibration Signals into Deterministic and Nondeterministic Components and its Capabilities of Fault Detection and Identification The paper investigates the possibility of decomposing vibration signals into deterministic and nondeterministic parts, based on the Wold theorem. A short description of the theory of adaptive filters is presented. When an adaptive filter uses the delayed version of the input signal as the reference signal, it is possible to divide the signal into a deterministic (gear and shaft related) part and a nondeterministic (noise and rolling bearings) part. The idea of the self-adaptive filter (in the literature referred to as SANC or ALE) is presented and its most important features are discussed. The flowchart of the Matlab-based SANC algorithm is also presented. In practice, bearing fault signals are in fact nondeterministic components, due to a little jitter in their fundamental period. This phenomenon is illustrated using a simple example. The paper proposes a simulation of a signal containing deterministic and nondeterministic components. The self-adaptive filter is then applied—first to the simulated data. Next, the filter is applied to a real vibration signal from a wind turbine with an outer race fault. The necessity of resampling the real signal is discussed. The signal from an actual source has a more complex structure and contains a significant noise component, which requires additional demodulation of the decomposed signal. For both types of signals the proposed SANC filter shows a very good ability to decompose the signal.

Wind turbines states classification by a fuzzy-ART neural network with a stereographic projection as a signal normalization

In this paper wind turbines operational states classification is considered. The fuzzy-ART neural network is proposed as a classifying system. Applying of stereographic projection as an input signals normalization procedure is introduced. Both theoretical justification is discussed and results of experiments are presented. It turns out that the introduced normalization procedure improves classification results.

Mobile Laser Scanning Systems for Measuring the Clearance Gauge of Railways: State of Play, Testing and Outlook

The paper contains a survey of mobile scanning systems for measuring the railway clearance gauge. The research was completed as part of the project carried out for the PKP (PKP Polish Railway Lines S.A., Warsaw, Poland) in 2011–2013. The authors conducted experiments, including a search for the latest solutions relating to mobile measurement systems that meet the basic requirement. At the very least, these solutions needed to be accurate and have the ability for quick retrieval of data. In the paper, specifications and the characteristics of the component devices of the scanning systems are described. Based on experiments, the authors did some examination of the selected mobile systems to be applied for measuring the clearance gauge. The Riegl (VMX-250) and Z+F (Zoller + Fröhlich) Solution were tested. Additional test measurements were carried out within a 30-kilometer section of the Warsaw-Kraków route. These measurements were designed so as to provide various elements of the railway infrastructure, the track geometry and the installed geodetic control network. This ultimately made it possible to reduce the time for the preparation of geodetic reference measurements for the testing of the accuracy of the selected systems. Reference measurements included the use of the polar method to select profiles perpendicular to the axis of the track. In addition, the coordinates selected were well defined as measuring points of the objects of the infrastructure of the clearance gauge. All of the tested systems meet the accuracy requirements initially established (within the range of 2 cm as required by the PKP). The tested systems have shown their advantages and disadvantages.

Fault detection in rolling element bearings using wavelet-based variance analysis and novelty detection

Fractal signal processing and novelty detection are used for fault detection in rolling element bearings. The former applies the concept of self-similarity based on wavelet variance, and the latter is based on machine learning and utilises artificial neural networks. The method is demonstrated using simulated and experimental vibration data. The work presented involves validation both on laboratory test rig data and industrial wind turbine data. The results show that the method can be used successfully for automated fault detection in ball bearings under real operational conditions.

ART-2 Artificial Neural Networks Applications for Classification of Vibration Signals and Operational States of Wind Turbines for Intelligent Monitoring

In recent years wind energy is the fastest growing branch of the power generation industry. The largest cost for the wind turbine is its maintenance. A common technique to decrease this cost is a remote monitoring based on vibration analysis. Growing number of monitored turbines requires an automated way of support for diagnostic experts. As full fault detection and identification is still a very challenging task, it is necessary to prepare an “early warning” tool, which would focus the attention on cases which are potentially dangerous.

Wind Turbine Main Bearing Diagnosis - A Proposal of Data Processing and Decision Making Procedure under Non Stationary Load Condition

Condition Monitoring of bearings used in Wind Turbines (WT) is an important issue. In general, bearings diagnostics is well-recognized field; however it is not the case for machines working under non-stationary load. An additional difficulty is that the Main Bearing (MB) discussed here, it is used to support low speed shaft, so dynamic response of MB is not clear as for a high-speed shaft. In the case of varying load/speed a vibration signal acquired from bearings is affected by operation and makes the diagnosis difficult. These difficulties come from the variation of diagnostic features caused mostly by load/speed variation, low energy of sought features and high noise levels. In the paper a novel diagnostic approach is proposed for main rotor bearings used in wind turbines. From a commercial diagnostic system two kind of information have been acquired: peak-to-peak vibration acceleration and generator power that is related to the operating conditions. The received data cover the period of several months, when the bearing has been replaced due to its failure and the new one has been installed. Due to serious variability of the mentioned data, a decision-making regarding the condition of bearings is pretty difficult. Application of classical statistical pattern recognition for data from the period A (bad condition) and the period B (after replacement, good condition) is not sufficient because the probability density functions of features overlap each other (pdf of peak-to-peak feature for bad and good conditions). Proposed approach is based on an idea proposed earlier for planetary gearboxes, i.e. to analyse data for bad/good conditions in two-dimensional space, feature - load. It is shown that the final data presentation is a good basis to the very successful classification of data (i.e. recognition of damaged and undamaged bearings).

Bearings Fault Detection in Gas Compressor in Presence of High Level of Non-Gaussian Impulsive Noise

The paper deals with the local damage detection in rolling element bearings in presence of a high level non-Gaussian noise. In many theoretical signal processing papers and engineering application related to damage detection, a simple model of the vibration is assumed. Basically it consists of signal of interest (SOI) and some unwanted (deterministic and/or random) components masking SOI in acquired observation. So, damage detection problem has to concern filtering, decomposition or extraction issue. Unfortunately, in the most of the industrial systems mentioned unwanted sources are in fact not Gaussian, so many of de-noising techniques cannot be applied directly or at least might give unexpected results. In this paper an industrial example will be discussed and novel approach based on advanced cyclostationary-based technique will be proposed. In the paper disturbances include periodic impacts in reciprocating compressor on an oil rig. Existing classical detection techniques (statistics in time domain, analysis of envelope spectrum, time-frequency decompositions) are insufficient to perform the task due to high power of disturbance contribution in comparison to damage signature. In the proposed technique, the Spectral Coherence Density Map (SCDM) is estimated first. Next step is related to analysis of SCDM contents and selection of informative part. If informative and non-informative components lay in separate frequency regions, such selection should fix the problem immediately

Application of Artificial Neural Network for Damage Detection in Planetary Gearbox of Wind Turbine

In the monitoring process of wind turbines the utmost attention should be given to gearboxes. This conclusion is derived from numerous summary papers. They reveal that, on the one hand, gearboxes are one of the most fault susceptible elements in the drive-train and, on the other, the most expensive to replace. Although state-of-the-art CMS can usually provide advanced signal processing tools for extraction of diagnostic information, there are still many installations, where the diagnosis is based simply on the averaged wideband features like root-mean-square (RMS) or peak-peak (PP). Furthermore, for machinery working in highly changing operational conditions, like wind turbines, those estimators are strongly fluctuating, and this fluctuation is not linearly correlated to operation parameters. Thus, the sudden increase of a particular feature does not necessarily have to indicate the development of fault. To overcome this obstacle, it is proposed to detect a fault development with Artificial Neural Network (ANN) and further observation of linear regression parameters calculated on the estimation error between healthy and unknown condition. The proposed reasoning is presented on the real life example of ring gear fault in wind turbine’s planetary gearbox.

Hybrid System of ART and RBF Neural Networks for Classification of Vibration Signals and Operational States of Wind Turbines

In recent years wind energy has been the fastest growing branch of the power generation industry. Maintenance of the wind turbine generates its the largest cost. A remote monitoring is a common method to reduce this cost. Growing number of monitored turbines requires an automatized way of support for diagnostic experts. Early fault detection and identification is still a very challenging task. A tool, which can alert an engineer about potentially dangerous cases, is required to work in real-time. The goal of this paper is to show an efficient system to online classification of operational states of the wind turbines and to detecting their early fault cases. The proposed system was designed as a hybrid of ART-2 and RBF networks. It had been proved before that the ART-type ANNs can successfully recognize operational states of a wind turbine during the diagnostic process. There are some difficulties, however, when classification is done in real-time. The disadvantages of using a classic ART-2 network are pointed and it is explained why the RBF unit of the hybrid system is needed to have a proper classification of turbine operational states.

Automatic and Full-Band Demodulation for Fault Detection—Validation on a Wind Turbine Test Rig

This paper proposes three algorithms for automatic diagnosis of mechanical system. First of all, an angular resampling with speed measurement correction is introduced. Secondly, a method for the association of detected spectral patterns with the characteristic frequencies of the investigated system is presented. This approach takes into consideration the slippage phenomenon of rolling element bearings. Thirdly, a full-band sideband demodulation method is proposed. It features with multi-rate filtering and offers new health indicators. All methods are applied on the real-world signals of a wind turbine test rig for diagnosis a bearing fault. The comparison of results shows the advantages of the proposed algorithms over well-known health indicators.