Condition monitoring of systems and detection of changes in the systems are of significant importance for an automated system, whether it is for production, transport, amusement, or any other application. Although condition monitoring is already widely used in machinery, the need for it is growing, especially as systems become increasingly autonomous and self-contained. One of the toughest tasks concerning embedded condition monitoring is to extract the useful information and conclusions from the often large amount of measured data. The use of self-organising maps, SOMs, for embedded condition monitoring is of interest for the component manufacturer who lacks information about how the component is to be used by the system integrator, or in what applications and load cases. At the same time, there is also a potential interest on the part of the system builders. Although they know how the system is designed and will be used, it is still hard to identify all possible failure modes. A component does not break at all locations or in all functions simultaneously, but rather in one, more stressed, location. Where is this location? Here, the collection of as much data as possible from the system and then processing it with the aid of SOMs allows the system integrators to create a map of the load on the system in its operating conditions. This gives the system integrators a better chance to decide where to improve the system. Automating monitoring and analysis means not only being able to collect prodigious amounts of measured data, but also being able to interpret the data and transform it into useful information, e.g. conclusions about the state of the system. However, as will be argued in this thesis, drawing the conclusions is one thing, being able to interpret the conclusions is another, not least concerning the credibility of the conclusions drawn. This has proven to be particularly true for simple mechanical systems like pneumatics in the manufacturing industry.

As the failure of a bearing could cause cascading breakdowns of the mechanical system and then lead to costly repairs and production delays, bearing condition monitoring has received much attention for decades. One of the primary methods for this purpose is based on the analysis of vibration signal measured by accelerometers because such data are information-rich. The vibration signal collected from a defective bearing is, however, a mixture of several signal components including the fault-generated impulses, interferences from other machine components, and background noise, where fault-induced impulses are further modulated by various low frequency signal contents. The compounded effects of interferences, background noise and the combined modulation effects make it difficult to detect bearing faults. This is further complicated by the nonstationary nature of vibration signals due to speed variations in some cases, such as the bearings in a wind turbine. As such, the main challenges in the vibration-based bearing monitoring are how to address the modulation, noise, interference, and nonstationarity matters. Over the past few decades, considerable research activities have been carried out to deal with the first three issues. Recently, the nonstationarity matter has also attracted strong interests from both industry and academic community. Nevertheless, the existing techniques still have problems (deficiencies) as listed below: (1) The existing enveloping methods for bearing fault feature extraction are often adversely affected by multiple interferences. To eliminate the effect of interferences, the prefiltering is required, which is often parameter-dependent and knowledge-demanding. The selection of proper filter parameters is challenging and even more so in a time-varying environment. (2) Even though filters are properly designed, they are of little use in handling in-band noise and interferences which are also barriers for bearing fault detection, particularly for incipient bearing faults with weak signatures. (3) Conventional approaches for bearing fault detection under constant speed are no longer applicable to the variable speed case because such speed fluctuations may cause “smearing” of the discrete frequencies in the frequency representation. Most current methods for rotating machinery condition monitoring under time-varying speed require signal resampling based on the shaft rotating frequency. For the bearing case, the shaft rotating frequency is, however, often unavailable as it is coupled with the instantaneous fault characteristic frequency (IFCF) by a fault characteristic coefficient (FCC) which cannot be determined without knowing the fault type. Additionally, the effectiveness of resampling-based methods is largely dependent on the accuracy of resampling procedure which, even if reliable, can complicate the entire fault detection process substantially. (4) Time-frequency analysis (TFA) has proved to be a powerful tool in analyzing nonstationary signal and moreover does not require resampling for bearing fault identification. However, the diffusion of time-frequency representation (TFR) along time and frequency axes caused by lack of energy concentration would handicap the application of the TFA. In fact, the reported TFA applications in bearing fault diagnosis are still very limited. To address the first two aforementioned problems, i.e., (1) and (2), for constant speed cases, two morphology-based methods are proposed to extract bearing fault feature without prefiltering. Another two methods are developed to specifically handle the remaining problems for the bearing fault detection under time-varying speed conditions. These methods are itemized as follows: (1) An efficient enveloping method based on signal Fractal Dimension (FD) for bearing fault feature extraction without prefiltering, (2) A signal decomposition technique based on oscillatory behaviors for noise reduction and interferences removal (including in-band ones), (3) A prefiltering-free and resampling-free approach for bearing fault diagnosis under variable speed condition via the joint application of FD-based envelope demodulation and generalized demodulation (GD), and (4) A combined dual-demodulation transform (DDT) and synchrosqueezing approach for TFR energy concentration level enhancement and bearing fault identification. With respect to constant speed cases, the FD-based enveloping method, where a short time Fractal dimension (STFD) transform is proposed, can suppress interferences and highlight the fault-induced impulsive signature by transforming the vibration signal into a STFD representation. Its effectiveness, however, deteriorates with the increased complexity of the interference frequencies, particularly for multiple interferences with high frequencies. As such, the second method, which isolates fault-induced transients from interferences and noise via oscillatory behavior analysis, is then developed to complement the FD-based enveloping approach. Both methods are independent of frequency information and free from prefiltering, hence eliminating the tedious process for filter parameter specification. The in-band vibration interferences can also be suppressed mainly by the second approach. For the nonstationary cases, a prefiltering-free and resampling-free strategy is developed via the joint application of STFD and GD, from which a resampling-free order spectrum can be derived. This order spectrum can effectively reveal not only the existence of a fault but also its location. However, the success of this method relies largely on an effective enveloping technique. To address this matter and at the same time to exploit the advantages of TFA in nonstationary signal analysis, a TFA technique, involving dual demodulations and an iterative process, is developed and innovatively applied to bearing fault identification. The proposed methods have been validated using both simulation and experimental data collected in our lab. The test results have shown that the first two methods can effectively extract fault signatures, remove the interferences (including in-band ones) without prefiltering, and detect fault types from vibration signals for constant speed cases. The last two have shown to be effective in detecting faults and discern fault types from vibration data collected under variable speed conditions without resampling and prefiltering.

The aim of this thesis is to design and implement two methods of predictive analysis for company Škoda Auto a.s. In the first part I have conducted a literature search on methods of predictive diagnostics. In the next part, with help from thesis consultant in Škoda Auto a.s., the analysis of the assembly line and data blocks from machinery and measuring has been made. Then I designed and programmed data generator based on real data. I created two methods of predictive diagnostics, capable of analyzing input data and deciding about their condition. In the end I tested these two methods and evaluated accuracy of their prediction. Main output of my thesis is two methods of predictive diagnostics, feasible in real world.

This thesis is concerned with condition monitoring and quantitative analysis of synchronous motors. Constantly rising requirements on the reliability of motors develop new methods of predictive diagnostics, which can identify failure conditions in the initial stage. Created mechatronic systems with the implemented failure from pre-prepared PMSM model are tested by unconventional condition monitoring methods. Software solutions of diagnostics and model designs of the mechatronic systems are implemented in MATLAB & Simulink. Part of this work is also a short introduction to the issue with necessary theoretical fundamentals and research of some selected methods of predictive diagnostics.

Wide band-gap semiconductor materials such as silicon carbide (SiC) and gallium nitride (GaN) have been widely investigated these years for their preferred operation at higher switching frequency, higher blocking voltage, higher temperature, with a compacter volume, in comparison with the traditional silicon (Si) devices. SiC MOSFETs have been utilized in photovoltaic systems, wind turbine converters, electric vehicles, solid-state transformers, more electric ships, and airplanes. GaN based transistors have also been adopted in the DC-to-DC converters in data centers, personal computers, AC-to-DC power factor correction converters for the consumer electronic adaptors, and DC-to-AC photovoltaic micro-inverters. The first part of this dissertation is regarding the lifetime modeling and condition monitoring for the SiC MOSFETs. Since SiC-based devices have different failure modes and mechanisms compared with Si counterparts, a comprehensive review will be conducted to develop accurate lifetime prediction, condition monitoring, and lifetime extension strategies. First, a novel comprehensive online updated system-level lifetime modeling approach will be presented. Second, to monitor the SiC MOSFET ageing, the typical degradation indicators of SiC MOSFET gate oxide will be investigated. Third, to measure the junction temperature, the dynamic temperature-sensitive electrical parameters for the medium-voltage SiC devices will be studied. The other part is the topology investigation of these emerging wide band-gap devices. A generalized topology that would leverage the advantages of the wide band-gap devices will be introduced and analyzed in detail. Following it is a new evaluation index for comparing different topologies with the consideration of the semiconductor die information. The topology and its derivatives will be utilized in the subsequent chapters for three applications. First, a 100 kW switched tank converter (STC) will be designed using SiC MOSFETs for transportation power electronic systems. Second, an updated STC topology integrating with the partial-power voltage regulation will be introduced for electric vehicle applications. Third, two novel single-phase resonant multilevel modular boost inverters will be designed based on the voltage-regulated STC. These topologies will be validated through designed prototypes. As a result, the high power density and high efficiency will be realized by combining the well-suited topologies and the advantages of the WBG devices.

Condition monitoring of hydraulic systems has become more available and inexpensive to implement. However, much of the research on this topic has been done on stationary hydraulic systems without the jump to mobile machines. This lack of research on condition monitoring of hydraulic systems on mobile equipment is addressed in this work. The objective of this work is to develop a novel process of implementing an affordable condition monitoring system for axial piston pumps on a mobile machine, a mini excavator in this work. The intent was to find a minimum number of sensors required to accurately predict a faulty pump. First, an expert understanding of the different components on an axial piston pump and how those components interact with one another was discussed. The valve plate was selected as a case study for condition monitoring because valve plates are a critical component that are known for a high percentage of failures in axial piston pumps. Several valve plates with various degrees of natural wear and artificially generated damage were obtained, and an optical profilometer was used to quantify the level of wear and damage. A stationary test-rig was developed to determine if the faulty pumps could be detected under a controlled environment, to test several different machine learning algorithms, and to perform a sensor reduction to find the minimum number of required sensors necessary to detect the faulty pumps. The results from this investigation showed that only the pump outlet pressure, drain pressure, speed, and displacement are sufficient to detect the faulty pump conditions, and the K-Nearest Neighbor (KNN) machine learning algorithms proved to be the least computationally expensive and most accurate algorithms that were investigated. Fault detectability accuracies of 100% were achievable. Next, instrumentation of a mini excavator was shown to begin the next phase of the research, which is to implement a similar process that was done on the stationary test-rig but on a mobile machine. Three duty cycle were developed for the excavator: controlled, digging, and different operator. The controlled duty cycle eliminated the need of an operator and the variability inherent in mobile machines. The digging cycle was a realistic cycle where an operator dug into a lose pile of soil. The different operator cycle is the same as the digging cycle but with another operator. The sensors found to be the most useful were the same as those determined on the stationary test-rig, and the best algorithm was the Fine KNN for both the controlled and digging cycles. The controlled cycle could see fault detectability accuracies of 100%, while the digging cycle only saw accuracies of 93.6%. Finally, a cross-compatibility between a model trained under one cycle and using data from another cycle as an input into the model. This study showed that a model trained under the controlled duty cycle does not give reliable and accurate fault detectability for data run in a digging cycle, below 60% accuracies. This work concluded by recommending a diagnostic function for mobile machines to perform a preprogrammed operation to reliably and accurately detect pump faults.

Conventional gearbox vibration monitoring techniques are based on the assumption that changes in the measured structural response are caused by deterioration in the condition of the gears in the gearbox. However, this assumption is not valid under fluctuating load conditions, since the fluctuating load will amplitude modulate the measured vibration signal and cause the rotational speed of the system to change. In general monitoring of machines subject to fluctuating load conditions is dealt with by considering the constant load conditions on gearboxes or during free rotational tests. The need to monitor the condition of large gearboxes in mineral mining equipment has attracted greater interest in order to improve asset management. An inherent need for signal processing techniques, with the ability to indicate degradation in gear condition, under fluctuating load conditions exist. Such techniques should enable the online monitoring of gearboxes that operate under fluctuating load conditions. A continued flow of up to date information should consequently be available for asset and production management. With this research, a load demodulation normalisation procedure was developed to remove the modulation caused by fluctuating load conditions, which obscures the detection of an incipient gear fault conditions. A rotation domain averaging technique is implemented which combines the ability of computer order tracking and time domain averaging to suppress the spectral smearing effect caused by the fluctuation in speed, as well as to suppress the amplitude of the vibration which is not synchronous with the rotation of the gear shaft. It is demonstrated that the instantaneous angular speed of a gearbox shaft can be utilised to monitor the condition of the gear on the shaft. The instantaneous angular speed response measurement is less susceptible to phase distortion introduced by the transmission path when compared to conventional gearbox casing vibration measurements. A phase domain averaging approach was developed to overcome the phase distortion effect of the transmission path under fluctuating load conditions. The load demodulation normalisation and rotation domain averaging signal processing procedures were applied to both the conventional gearbox casing vibration and instantaneous angular speed measurements prior to the calculation of a smoothed pseudo Wigner-Ville distribution of the data. Statistical parameters such as the energy ratio were calculated from the distribution. These parameters could be monotonically trended under different load conditions to indicate the degradation of gear conditions. / Thesis (PhD (Mechanical Engineering))--University of Pretoria, 2005. / Mechanical and Aeronautical Engineering / unrestricted

One of Bosch Rexroth’s newest developments is the ABPAC power unit, which is both modular and configurable. The modular design of the ABPAC is enhanced by a selfcontained Condition Monitoring System (CMS), which can also be used to retrofit existing designs. This dissertation shows how Industry 4.0-Technology provides special advantages for the diverse user profiles. Today, Hydraulic Power Units have either scheduled intervals for preventive maintenance or are repaired in case of component failures. Preventive maintenance concepts, until now, did not fully utilize the entire life expectancy of the components, causing higher maintenance costs and prolonged downtimes. Risk of unscheduled downtime forces the customer to stock an array of spare parts leading to higher inventory costs or in the event a spare is not readily available, the customer may encounter long delivery times and extended downtime. Bearing this in mind, we’ve conceived the idea of a self-contained intelligent Condition Monitoring System including a predictive maintenance concept, which is explained in the following.

In this work, an automated statistical approach for the condition monitoring of a fluid power system based on a process sensor network is presented. In a multistep process, raw sensor data are processed by feature extraction, selection and dimensional reduction and finally mapped to discriminant functions which allow the detection and quantification of fault conditions. Experimentally obtained training data are used to evaluate the impact of temperature and different aeration levels of the hydraulic fluid on the detection of pump leakage and a degraded directional valve switching behavior. Furthermore, a robust detection of the loading state of the installed filter element and an estimation of the particle contamination level is proposed based on the same analysis concept.

This dissertation proposes hybrid models for (i) diagnosis and (ii) remaining useful life estimation of a single fatigue crack in a low-pressure turbine blade. The proposed hybrid methods consist of physics-based methods and data-driven methods. In this dissertation, blade tip timing is used to measure the relative tip displacement of a rotor blade. The natural frequency of the blade is determined by detecting the critical speeds of the blade using a newly derived least squares spectral analysis method. The method shares its origin from the Lomb-Scargle periodogram and can detect resonance frequencies in the blade’s displacement while the rotor is in operation. A Campbell diagram is then used to convert the critical speed into a natural frequency. Two kinds of shaft transients are considered, a run-up run-down crossing the same critical speed, is used to test the new method. This dissertation shows that the relative displacement of the blade tip is comparable to those simulated from an analytical single degree of freedom model. It is also shown that the newly proposed resonance detection method estimates the natural frequency of the blade to a high degree of accuracy when compared to the measurements from a modal impact hammer test. The natural frequency obtained from the real time measurement is then used in a pre-constructed hybrid diagnostics model. The diagnostics model provides a probability density function estimation of the surface crack length given the measured natural frequency. A Gaussian Process Regression model is trained on data collected during experiments and finite element simulations of a fatigue crack in the blade. The final part of this dissertation is a sequential inference model for improving the estimation of the crack length and the prediction of the crack growth. The suggested model uses an unscented Kalman filter that improves estimations of the crack length and the rate of crack growth from Paris’ Law coefficients. The model is updated each time a diagnosis is performed on the blade. The RUL of the blade is then determined from an integration of Paris’s Law given the uncertainty estimates of the current damage in the blade. The result of the algorithm is an estimation of the remaining number of cycles to failure. The algorithm is shown to improve the overall estimation of the RUL; however, it is suggested that future work looks at the convergence rate of the method. / Dissertation (MEng)--University of Pretoria, 2019. / Eskom Power Plant Engineering Institute (EPPEI) / Mechanical and Aeronautical Engineering / MEng / Unrestricted