Abstract This study concentrates on developing and tuning various oil analysis methods to meet the requirements of modern industry and environmental analytics. Oil analysis methods form a vital part of techniques used to monitor the condition of machines and may help to improve the overall equipment effectiveness value of a factory in a significant manner. Worm gears are used in various production machines, and their breakdowns may cause significant production losses. Wearing of these gears is relatively difficult to monitor with vibration analysis. Analysis of two indicator metals, copper and iron, may reveal wearing phenomena of worm gears effectively, and savings can be significant. Effective wear metal analysis requires good tools. ICP-OES with kerosene dilution is widely used in wear metal analysis, but purchasing and using of ICP-OES is expensive. A cheaper FAAS technique with similar pre-treatment of oil samples was tested and it proved to be useful especially in analyzing small amounts of samples. The accuracy of FAAS was sufficient for quantitative work in machine diagnostics and waste oil characterization. Solid debris analyses are useful in oil contamination control as well as in detection of wearing mechanisms. Membrane filtration, optical microscopy, SEM and automatic particle counting were applied in analysis of rolling and gear oils. Particle counting is an effective way to detect oil contamination, but in the studied cases even larger particles than those detected in normal ISO classes would be informative. However, membrane filtration and optical microscopy may reveal the wearing machine element exactly. Additives provide oils with desired properties thus they should be monitored intensively. A FTIR method for quantitative analysis of fatty alcohols and fatty acid esters in machinery oils was developed during this work. It has already been used successfully in quantitative and qualitative analysis of machinery oil samples. Various kinds of oils may be spilled into the soil during use and in accident situations, and they can migrate to groundwater layers. Biodegradation of oils can remove them from the soil or water completely or at least diminish the amount of harmful substances. An automatic, respirometric BOD OxiTop method was used to evaluate the biodegradability of various oils in water and soil media. The biodegradation of certain bio and mineral hydraulic oils was evaluated in groundwater, where bio oils usually biodegraded more effectively than mineral oils. The use of oils in machines weakened especially the biodegradability of bio oils. Biodegradability of bio oils was also studied in standard conditions of OECD 301 F and bio oils usually biodegraded moderately good in these conditions. The biodegradation of forestry chain oils and wood preservative oils was evaluated in forest soils. Linseed oil biodegraded moderately, but certain experimental wood preservatives biodegraded more effectively. Widely used creosote oil biodegraded in a lesser degree. Rapeseed oil-based chain oils biodegraded more effectively than corresponding tall oil.

Machine diagnostics is usually done via conditioned monitoring (CM). This approach analyses certain thresholds or patterns for diagnostic purposes. This approach can be costly and time consuming for industries. A larger downside is the difficulty in generalizing CM to a wider set of machines.There is a new trend of using a Machine learning (ML) approach to diagnose machines states. An ML approach would implement an autonomous system for diagnosing machines. It is highly desirable within industry to replace the manual labor performed when setting up CBM systems. Often the ML algorithms chosen are novelty/anomaly based. It is a popular hypothesis that detecting anomalous measurements from a system is a natural byproduct of a machine in a faulty state.The purpose of this thesis is to help CombiQ with an implementation strategy for a fault detection system. The idea of the fault detection system is to make prediction outcomes for machines within the system. More specifically, the prediction will inform whether a machine is in a faulty state or a normal state. An ML approach will be implemented to predict anomalous measurements that corresponds to a faulty state. The system will have no previous data on the machines. However, data for a machine will be acquired once sensors (designed by CombiQ) have been set up for the machine.The results of the thesis proposes an unsupervised and semi-supervised approach for creating the ML models used for the fault detection system. The unsupervised approach will rely on assumptions when selecting the hyperparameters for the ML. The semi-supervised approach will try to learn the hyperparameters through cross validation and grid search. An experiment was set up check whether three ML algorithms can learn optimal hyperparameter values for predicting rotational unbalance. The algorithm known as OneClassVM showed the best precision results and hence proved more useful for CombiQ’s criterium.

Modeling, fault assessment, and diagnostics of rolling element bearings and induction motors were studied. Dynamic model of rolling element bearings with faults were developed using vector bond graphs. The model incorporates gyroscopic and centrifugal effects, contact deflections and forces, contact slip and separations, and localized faults. Dents and pits on inner race, outer race and balls were modeled through surface profile changes. Experiments with healthy and faulty bearings validated the model. Bearing load zones under various radial loads and clearances were simulated. The model was used to study dynamics of faulty bearings. Effects of type, size and shape of faults on the vibration response and on dynamics of contacts in presence of localized faults were studied. A signal processing algorithm, called feature plot, based on variable window averaging and time feature extraction was proposed for diagnostics of rolling element bearings. Conducting experiments, faults such as dents, pits, and rough surfaces on inner race, balls, and outer race were detected and isolated using the feature plot technique. Time features such as shape factor, skewness, Kurtosis, peak value, crest factor, impulse factor and mean absolute deviation were used in feature plots. Performance of feature plots in bearing fault detection when finite numbers of samples are available was shown. Results suggest that the feature plot technique can detect and isolate localized faults and rough surface defects in rolling element bearings. The proposed diagnostic algorithm has the potential for other applications such as gearbox. A model-based diagnostic framework consisting of modeling, non-linear observability analysis, and parameter tuning was developed for three-phase induction motors. A bond graph model was developed and verified with experiments. Nonlinear observability based on Lie derivatives identified the most observable configuration of sensors and parameters. Continuous-discrete Extended Kalman Filter (EKF) technique was used for parameter tuning to detect stator and rotor faults, bearing friction, and mechanical loads from currents and speed signals. A dynamic process noise technique based on the validation index was implemented for EKF. Complex step Jacobian technique improved computational performance of EKF and observability analysis. Results suggest that motor faults, bearing rotational friction, and mechanical load of induction motors can be detected using model-based diagnostics as long as the configuration of sensors and parameters is observable. / text

The main idea is focused on the diagnostics of a specific hydraulic system, i.e. sensing the physical quantities of the hydraulic circuit with a tank and a centrifugal pump driven by an asynchronous motor. It is a system of pipes connected to the pump, where due to its work it creates a water flow and a pressure increase. In practice, this issue is also addressed in the energy and nuclear industries. Primary circuits in some cases cannot be designed or modified to be able measure locally the pressure value. It is necessary to measure this quantity indirectly - from the motor currents. The main idea of the work is to diagnose the system by an indirect method - specifically to detect the state of the hydraulic circuit (pressure, flow) from the values that we are able to measure and detect damage in advance. In the second part of the thesis is the application of the parts of a specific hydraulic system in the simulation environment MATLAB Simulink. The model of the hydraulic circuit contains mathematical-physical relations that simulate the course of the mentioned experiment. The results of the simulation are compared with the results of the experiment. The model also investigates the simulation of a fault condition, when we supply pressure pulsations to the hydraulic circuit. It is these changes in the hydraulic part that affect the characteristics of the pump and the asynchronous motor, so we are able to diagnose this system.