Performing condition monitoring related tasks on any machinery is an essential element of their rational maintenance. Endeavours to detect an incipient fault within a system serve multiple purposes from increasing the safety of people responsible for operating the machines through decreasing the running and operational costs, allowing time to plan for the inevitable repairs and making sure that the downtime of the machine is kept to an absolute minimum. All these tasks gain extra importance in a case when machines are operated in dangerous conditions putting people's lives in potential jeopardy - for instance in the field of operating a helicopter. The robust assessment of the condition of gearboxes used by helicopters has recently been given an increased attention due to a number of accidents which followed an undetected drive train component failure. The majority of the on-board mounted condition monitoring systems use vibration response signals which are specifically processed to obtain a single number which is representative of a condition of a given monitored drive train component. Those signal processing methods are called Condition Indicators (CIs). There are a number of such CIs which are already in use and they seem to adequately indicate faults in most of the cases. However in a number of instances it has been observed that the most popular parameters like Crest Factor or FM4 failed to dependably reflect the true condition of the gear causing serious accidents, some of which resulted in a number of lives being lost. For this reason the presented research is focused on investigating the limitations of the existing CIs and designing a set of improved CIs. The development process is based on overcoming the drawbacks of thetechniques used in existing CIs combined with the intelligence gathered while analysing the acceleration vibration signals which contained a gear or a bearing fault. Five new CIs are proposed and the details of their design are documented. Both the existing and the proposed CIs are applied on the available, uncorrelated datasets. The results of the comparison show that the newly developed CIs are capable of indicating a gear or a bearing fault in a more robust and dependable fashion. Each proposed CI alone may not be the most robust indicator of the actual condition of the monitored component hence the output from all proposed CIs is combined into a single indication through use of a novel data fusion model. The Combined CI created based on the data fusion model is observed to be more robust compared to each CI alone, hence it may increase the confidence level of the decision making routine and is expected to decrease the number of false alarms. The methods of the existing CIs, the proposed CIs and the data fusion techniques as well as the results of the comparison between the different approaches are present in this thesis.

Analysis of reliability data plays an important role in the maintenance decision making process. The accurate estimation of residual life in components and systems can be a great asset when planning the preventive replacement of components on machines. Artificial intelligence is a field that has rapidly developed over the last twenty years and practical applications have been found in many diverse areas. The use of such methods in the maintenance field have however not yet been fully explored. With the common availability of condition monitoring data, another dimension has been added to the analysis of reliability data. Neural networks allow for explanatory variables to be incorporated into the analysis process. This is expected to improve the quality of predictions when compared to the results achieved through the use of methods that rely solely on failure time data. Neural networks can therefore be seen as an alternative to the various regression models, such as the proportional hazards model, which also incorporate such covariates into the analysis. For the purpose of investigating their applicability to the problem of predicting the residual life of machines and components, neural networks were trained and tested with the data of two different reliability related datasets. The first dataset represents the renewal case where repair leads to complete restoration of the system. A typical maintenance situation was simulated in the laboratory by subjecting a series of similar test pieces to different loading conditions. Measurements were taken at regular intervals during testing with a number of sensors which provided an indication of the test piece’s condition at the time of measurement. The dataset was split into a training set and a test set and a number of neural network variations were trained using the first set. The networks’ ability to generalize was then tested by presenting the data from the test set to each of these networks. The second dataset contained data collected from a group of pumps working in a coal mining environment. This dataset therefore represented an example of the situation encountered with a repaired system. The performance of different neural network variations was subsequently compared through the use of cross-validation. It was proved that in most cases the use of condition monitoring data as network inputs improved the accuracy of the neural networks’ predictions. The average prediction error of the various neural networks under comparison varied between 431 and 841 seconds on the renewal dataset, where test pieces had a characteristic life of 8971 seconds. When optimized the multi-layer perceptron neural networks trained with the Levenberg-Marquardt algorithm and the general regression neural network produced a sum of squares error within 11.1% of each other for the data of the repaired system. This result emphasizes the importance of adjusting parameters, network architecture and training targets for optimal performance The advantage of using neural networks for predicting residual life was clearly illustrated when comparing their performance to the results achieved through the use of the traditional statistical methods. The potential of using neural networks for residual life prediction was therefore illustrated in both cases. / Dissertation (MEng (Mechanical Engineering))--University of Pretoria, 2007. / Mechanical and Aeronautical Engineering / MEng / unrestricted

The focus of this research is in the area of Remote Condition Monitoring (RCM) for use within intermodal transport and logistics industries. For many years the intermodal transport industry has utilised these RCM systems that have in built flaws due to the subsystems they use.

Abstract Despite the large amount of research work in condition based maintenance and condition monitoring methods, there is still a need for more reliable and accurate methods. The clear evidence of that need is the continued dependence on time based maintenance, especially for critical applications such as turbomachinery and airplane engines. The lack of accurate condition monitoring systems could lead to not only the unexpected failures as well as the resulting hazards and repair costs, but also a huge waste of material and time because of unnecessary replacement due to false alarms and unnecessary repair and maintenance. Temperature change is a phenomenon that accompanies every dynamic activity in the universe. However, it has not been adequately exploited for mechanical system condition monitoring. The reason is the slow response of current temperature monitoring systems compared to other condition monitoring methods such as vibration analysis. Many references inferred that the change in temperature is not sensible until approaching the end of the monitored component life and even the whole system life (Kurfess, et al., 2006; Randall, 2011; Patrick, et al., March 7-14, 2009). On the other hand, the most commonly used condition monitoring method, i.e., vibration analysis, is not free from pitfalls. Although vibration analysis has shown success in detecting some bearing faults, for other faults like lubrication problems and gradual wear it is much less effective. Also, it does not give a reliable indication of fault severity for many types of bearing faults. The advancement of thermography as a temperature monitoring tool encourages the reconsideration of temperature monitoring for mechanical system fault detection. In addition to the improved accuracy and responsiveness, it has the advantage of non-contact monitoring which eliminates the need for complex sensor mounting and wiring especially for rotating components. Therefore, in current studies the thermography-based monitoring method is often used either as a distinct method or as a complementary tool to vibration analysis in an integrated condition monitoring system. The main objectives of this study are hence to: 1. Define heat sources in the rolling element bearings and overview two of the most famous bearing temperature calculation methods. 2. Setup a bearing test rig that is equipped with both vibration and temperature monitoring systems. 3. Develop a temperature calculation analytical model for rolling element bearing that include both friction calculation and heat transfer models. The friction calculated by the model will be compared to that calculated using the pre-defined empirical methods. The heat transfer model is used for bearing temperature calculation that will be compared to the experimental measurement using different temperature monitoring devices. 4. Propose a new in-band signal enhancement technique, based on the synchronous averaging technique, Autonomous Time Synchronous Averaging (ATSA) that does not need an angular position measuring device. The proposed method, in addition to the Spectral Kurtosis based band selection, will be used to enhance the bearing envelope analysis. 5. Propose a new method for classification of the bearing faults based on the fault severity and the strength of impulsiveness in vibration signals. It will be used for planning different types of tests using both temperature and vibration methods. 6. Develop and experimentally test a new technique to stimulate the bearing temperature transient condition. The technique is supported by the results of finite element modeling and is used for bearing temperature condition monitoring when the bearing is already running at thermal equilibrium condition.

Reliability of emergency Diesel generator systems, or indeed any Diesel engines in a wide range of fields is critical. Traditional maintenance procedures for these engines follow time based or statistical based methods. Due to the wide variety of uses of Diesel engines it is not possible for these forms of maintenance to be as effective as condition based monitoring. Condition based monitoring holds many advantages over traditional maintenance methods. It allows for the earlier detection and diagnosis of a fault and allows for planned maintenance work avoiding costly and unexpected downtime. It also reduces the overall maintenance costs as parts need only be replaced when they are worn or faulty, not based on a time schedule. The ability to unobtrusively monitor the engines also has many advantages in- cluding reduced sensor cost and negating the need to tamper permanently with the engine. Acoustic monitoring has been identified as the most prominent and effective way in which to achieve this goal. As such, extensive experimentation was carried out on both large and small Diesel engines over a wide range of speeds, loads and faults and the data was then analysed. The data was first investigated statistically and then processed using Independent Component Analysis after the statistical re- sults were found to be poor. A program was written for the automatic comparison of the collected data and the results presented in this thesis show that ICA and acoustic emissions have the ability to aid in engine fault detection and diagnosis. The results have shown to be reliable, consistent and able to distinguish when the engine is healthy or faulty.

The induction motor is a highly utilised electrical machine in industry, with the nuclear industry being no exception. A typical nuclear power station usually contains more than 1000 motors, where they are used in safety and non-safety application. The efficient and fault-free operation of this machine is critical to the safe and economical operation of any plant, including nuclear power stations. A comprehensive literature review was conducted that covered the functioning of the induction machine, its common faults and methods of detecting these faults. The Condition Based Maintenance framework was introduced in which condition monitoring of induction machines is an essential component. The main condition monitoring methods were explained with the main focus being on Motor Current Signature Analysis (MCSA) and the various methods associated with it. Three analysis methods were selected for further study, namely, Current Signature Analysis, Instantaneous Power Signature Analysis (IPSA) and Motor Square Current Signature Analysis (MSCSA). Essentially, the methodology used in this dissertation was to study the three common motor faults (bearings, stator and rotor cage) in isolation and compare the results to that of the healthy motor of the same type. The test loads as well as fault severity were varied where possible to investigate its effect on the fault detection scheme. The data was processed using an FFT based algorithm programed in MATLAB. The results of the study of the three spectral analysis techniques showed that no single technique is able to detect motor faults under all tested circumstances. The MCSA technique proved the most capable of the three techniques as it was able to detect faults under most conditions, but generally suffered poor results in inverter driven motor applications. The IPSA and MSCSA techniques performed selectively when compared to MCSA and were relatively successful when detecting the mechanical faults. The fact that the former techniques produce results at unique points in the spectrum would suggest that they are more suitable for verifying results. As part of a comprehensive condition monitoring scheme, as required by a large population of the motors on a nuclear power station, the three techniques presented in this study could readily be incorporated into the Condition Based Maintenance framework where the strengths of each could be exploited.

No description available.

No description available.

This bachelor thesis has been executed with the product development process double diamond. The thesis has been performed at Karlstad University in cooperation with Kongsberg Maritime in Sweden AB. The objective of the thesis has been to develop a condition monitoring system for Kongsberg’s waterjet department. Substantially it has to do with finding a system which could monitor wear and problematic trends  in  the hydraulic oil and oil lubricated bearings. The wear is to be translated to an electrical signal which could be used to alert and visualize the wear to the end user. The literature study was divided into two parts. The first part of the literature study consisted of analyzing and understanding wear and condition monitoring systems. The second part of the literature study instead consisted of analyzing different kind of measurement methods which could be suitable for so-called online measurements. The used references come from scientific articles and documents publicized by various classification societies. The final solution consists of a system of different sensors and measurement methods. Due to the fact that the system is to be  installed on ships, the system needed to be considered with rules from various classification societies. The rules which are relevant for the project was placed in an elimination matrix where concepts that didn’t live up to the rules were scrapped. Further on, the sub concepts were evaluated  and selected  with a relative matrix and a weighted Kesselring matrix, which gave the most suitable sub concepts. The sub concepts were merged into a complete condition monitoring system at concept level. The final system solution could measure solid particles in oil, humidity in oil, oil flow, torque, rotational speed of the shaft and vibrations. The selected sensors also enabled the system to measure shaft power, oil conductivity and oil permittivity which were not included in the task.

Condition monitoring systems are critical for autonomous detection of damage when operating remote wind turbines. These systems continually monitor the turbine’s operating parameters and detect damage before the turbine fails. Although common in utility-scale turbines, these systems are mostly undeveloped in distributed, small-scale turbines due to their high cost and need for specialized equipment. The Cal Poly Wind Power Research Center is developing a low-cost, modular solution known as the LifeLine system. The previous version contained monitoring equipment, but lacked decision-making capabilities. The present work builds on the LifeLine by developing software-based detection of blade damage. Detection is done by monitoring of tower vibrations, rotor speed, and generator power output. First, testing is completed to inform algorithm design: the tower vibrational response is recorded, and blade damage is simulated by adding a mass imbalance to one blade. From these results, several algorithms are developed, and their performance is analyzed in a cross-validation study. The time-series method known as the Nonlinear State Estimation Technique and Sequential Probability Ratio Test (NSET+SPRT) is implemented first. This algorithm is highly successful, with a 93.3% rate of correct damage detection; however, it occasionally raises false alarms during normal operation. A custom-built algorithm known as the Adaptive Fast Fourier Transform (AFFT) is also built; its strength lies in its elimination of false alarms. The final system utilizes a joint monitoring approach, combining the benefits of the NSET+SPRT and AFFT. The final algorithm is successful, correctly categorizing 95.5% of data when operating above 120RPM, and raising no false alarms in normal operation. This version is then implemented for live monitoring on the Cal Poly Wind Turbine, allowing for robust and autonomous detection of blade damage.