Search

Traditional bearing sensory diagnostic include touching and hearing rely on personal experience, and for more complex system are unable to meet the needs of equipment fault diagnosis. The research on bearing fault diagnosis is developing significantly. Bearings are used in rotating machinery and most machinery failures are caused by bearing failures. The fault diagnosis of bearings is an important research area. The core of bearing fault diagnosis is the pattern recognition of fault features. The key of pattern recognition is to develop a reasonable classifier. Intelligent pattern recognition has been developed such as principal components, support vector machine, neural network. In this study, a bearing fault diagnosis based on exploring pattern recognition is proposed. The key to pattern recognition is to design a significant classifier. A number of features from bearing vibration of normal and fault bearing are extracted and processed using principal components of correlation matrix. Plot of principal components shows the visualization of normal and fault bearing and the classifier is chosen subjectively. The principal components exploration will be confirmed using least squares support vector machine. The parameter of support vector machine estimated using heuristic optimization particle swarm optimization. The proposed method can be applied in the detection of faults of bearing. [ABSTRACT FROM AUTHOR]

Milling is one type of cutting process that is most often used in the manufacturing industry. Flank wear is a wear phenomenon on the cutter that usually occurs in milling. This wear can affect the accuracy of the product and the quality of the cutting finish surface and can cause damage to the cutting tool. This paper will discuss how to improve the accuracy of cutting tool wear predictions based on vibration signals obtained from the machining process. The signals from the milling process are decomposed by Variational Mode Decomposition (VMD) method before being applied to the machine learning algorithm. As a comparison level of accuracy, the root mean square error (RMSE) parameter is used to determine the error between the actual value and the predicted value of the trained model. The results of the experiments show that using the VMD decomposition feature can improve the performance of the Support Vector Machine (SVM) algorithm in predicting tool wear when compared to those using the time and frequency domain features or those using only the time domain features [ABSTRACT FROM AUTHOR]

A virtual model of a motorcycle is essential during the design phase. Using a virtual model can significantly reduce development time. Large-scale simulations like road load simulators and virtual tests require accurate virtual modelling. This article explains the Finite element method update and create a complete model of the motorcycle's main structure. A valid virtual model of the motorcycle's main structure was created using FEA updating. The first step was selecting the correct element type, meshing size, and geometry. In addition, modelling comparisons were made to observe the differences of the RBE elements and solid reverse engineering models for engine as the main structure of a motorcycle which have a significant influence on stiffness. Components that have a minor effect on the structure's rigidity were omitted to simplify the modelling process. The ignoring component was done by comparing the effect on the experimental modal analysis result under the observed frequency. The frame body modelling result from the Finite Element Method was validated using the Experimental Modal Analysis. The prototype's natural frequency was determined by impact hammer testing and a digital signal analyzer. Experiment Modal Analysis was used to compare the natural frequency obtained from normal mode FEM simulation to the result of valid FEM models. The main structure of the motorcycle was reverse engineered in CATIA V5R25. The mesh was created in SimDesigner 2017 and analyzed in Nastran 2017 SOL 103 normal mode. The comparison of FEA normal mode analysis and experimental mode analysis reveals a satisfactory result and a valid FEA model of the motorcycle's main structure. [ABSTRACT FROM AUTHOR]

Energy efficiency is one of the main performance parameters in power generation systems along with reliability, availability, maintainability and safety (RAMS). In condition monitoring (CM) activities, energy efficiency indicators (EEI) are used as one of the monitored parameters. The purpose of this study is to determine operating indicator variables which have a negative impact on the EEI, namely: thermal efficiency (η, %), net plant heat rate (NPHR, kcal/kWh) or specific fuel costs (Rp/kWh). The set of EEI data over time is then regressed to predict when the EEI will exceed the set threshold. Furthermore, the results of this EEI regression are used to develop the concept of energy efficiency-based fault diagnostic (EEBFD) using the support vector machine (SVM) classification algorithm. The results of the diagnosis indicate that there are three operating indicators that experience operating anomalies, namely: main steam water spray flow, main air preheater gas out temperature and oxygen contents in flue gas. The use of EEI in the diagnosis process through SVM classification algorithm can be used to detect equipment malfunctions. And finally, this proposed EEBFD can be used to make maintenance decisions. [ABSTRACT FROM AUTHOR]

Original data of bogie faults is hardly obtained by a real experimental because a failure bogie system will make the train suffer from serious danger. Therefore a simulation train dynamics often used to simulate a bogie to obtain failure data. A high speed train is a multi body system which includes train body, bogie frameworks and wheels. This paper focused on bearing vibration analysis using data obtained from bearing simulation. A fault simulation platform was adopted as the experimental device. Data collection was conducted to normal state, inner fault, outer fault and rolling fault. Each fault was collected under the rotation of 300 rpm, 10 groups of data were collected of 10 second for.1 second of measurements. It is the objective of bogie fault characterization to develop an accurate diagnosis method. Deep neural network was proposed for fault diagnosis of bogie bearing. Features, time domain and frquency domain, should be extrcted prior to fault diagnosis. Sixteen time and frequency feature domains are used in fault characterization. This paper proposes bearing simulation methodology to explore diagnosis of bearing bogie faults. The simulation is based on mechanical modeling, lifetime of bearing can be determined. Lifetime data with operating conditions are modeled using Weibull regression, and then remaining life of bearing can be predicted. Multivariate analysis is propsed to characterized the bogie bearing faults. Neural networks is proposed to explore the characterization of bogie fault diagnosis and compared to multivariate analysis. [ABSTRACT FROM AUTHOR]

Pronostia FEMTO-ST is an experimental platform enables testing and validating related to bearing prognostic. Pronostia provides vibration acceleration. Bearing is one of the most critical component in rotating machinery, and it's failure may lead to catastrophic accidents. This paper proposes a survival regression to explore the reliability of bearings. The response is the survival time and the predctors are the bearing features extraction. A Cox and Weibull regression are used to evaluate the reliabilty. The training step is used to build a bearing regression model by estimating the regression coefficient. The results show that the proposed method can be used to predict the reability of the bearing. [ABSTRACT FROM AUTHOR]

Bogie bearings performances directly influences the safety of the train. Diagnosis aims to detect the bearing failure, prognosis aims to predict the remaining life of a bearing. This study concerns on bogie bearings prognosis using survival function. Bearing health monitoring providing information of the system state evolved rapidly over the past decade. Root mean square is an efficient indicator of bearing degradation for inspection and maintenance. Numerous degradation models have been developed. However, the failure regression for bearing degradation is not yet well explored. The residual life provides a prediction for future operation has grown as a significant issue for bearing health condition. This article explores the failure regression modeling for bearing degradation. In the industry, bearings operate at different rotation and load conditions. The developed models for bearing residual's life are limited to a single operating condition. Failure regression considers both rotation and load as operating conditions. Root mean square or kurtosis (time domain features) are extracted and used as bearing degradation indicators. Femto technology developed experimental platform that provides time to failure in specific conditions of operating. horizontal and vertical vibrations are measured every 10 s at 25.6 KHz by accelerometers. The time to failure is observed for bearing under three speed and three load conditions. The speed and load are considered as covariates in failure regression. The failure regression yields the bearing reliability, and this conditional reliability is used to calculate the bearing residual life. In the practice, bearing can be operated at different speeds, loads and operating conditions. The effect of internal and external covariates needs to be explored in order to provide better residual life prediction and to optimize to maintenance strategy. Failure regression can be considered as a tool for bearing residual life prediction in presence of internal and external covariates. [ABSTRACT FROM AUTHOR]

One of the methods used to surface hardening of ductile iron is chilled cast iron. Chill as the fast cooling rate in the mold during solidification and chill thickness greatly affects the thickness of the hardness layer. The main material used is ductile iron, and the chill material is SS 304. Casting uses the sand casting method. Before pouring, the chill plate has been inserted onto the surface of the pattern that has been formed in the mold, then the chill plate is preheated at 700°C. Pouring was carried out at a melting temperature of 1400°C, and then cooled with argon and O2 sprays into the mold in solidification conditions at exactly 700°C. The results analyzed were the microstructure, hardness value, and the hardness of the thickness layer. This chill coolant will absorb heat very quickly and the Cr and Ni alloy will diffuse to the specimen surface to stabilize the ferrite and austenite phases in the final solidification. The particles on the hard surface are has ferro carbide M7C3, which is in the form of cementite and martensitic phases so that to categorized as white cast iron structure formed on the surface with the area around 1.5-3mm has a hardness of 61-65HRC. But in the center area is 31-49HRC. [ABSTRACT FROM AUTHOR]

Starting from a new condition a bearing will operate at a particular level of performance. The degradation process occurs as its use progress. Degradation can be regarded as two-state processes, the first one is a normal process and the second one is the degradation process. Identifying the initiation point of degradation is an important issue in power generation insurance. Support vector regression is a linear combination of training patterns depends on Lagrange multipliers which act as forces pushing prediction toward target responses. Only nonzero values are significant in forecasting the regression curve. This paper aims to explore the application of support vector regression in identifying the starting point of bearing degradation. One point outside of the action line is considered as the initial point of the degradation. The bearing vibration obtained from FEMTO website consist of training and testing dataset from three condition of bearing experiments. The training dataset is used to estimate the action line, and the testing dataset used to identify the initial point of degradation. [ABSTRACT FROM AUTHOR]

One way to extend the life time of the gear is minimizing the distortion during the manufacturing process. One of the most important processes in manufacturing to produce gears is heat treatment process. The purpose of this study is to analyze the distortion of the gear after heat treatment process. The material of gear is AISI 1045, and it was designed with the module (m) 1.75, and a number of teeth (z) 29. Gear was heat-treated in the furnace at a temperature of 800°C, holding time of 30 minutes, and then quenched in water. Furthermore, surface hardening process was also performed on gear teeth at a temperature of 820°C and holding time of 35 seconds and the similar procedure of analysis was conducted. The hardness of gear after heat treatment average 63.2 HRC and the teeth surface hardness after gear to induction hardening was 64.9 HRC at the case depth 1 mm. The microstructure of tested gear are martensitic and pearlite. The highest distortion on tooth thickness to upper than 0.063 can cause high precision at the tooth contact is not appropriate. Besides the shrinkage of tooth thickness will also affect to contact angle because the size of gear tolerance was not standardized. [ABSTRACT FROM AUTHOR]

Roller bearing is one of the vital parts of a rotating machine. Bearing failure can result in serious damage of the machine. This paper aims to develop a bearing fault diagnosis method using parameter evaluation technique to improve the diagnosis accuracy. The parameter evaluation technique is used to select five features that are used as predictors in multi-class support vector machine (SVM) classification. The purpose of this feature reduction was to avoid the curse of dimensionality and to increase the accuracy of the diagnosis. The diagnosis process was performed by classification of bearing states using one-against-one method multi-class SVM. Three types of kernel functions i.e., linear, polynomial, and Gaussian RBF were used in the SVM classification. The bearing conditions which is diagnosed in this paper were normal bearing, inner race fault, and outer race fault conditions. As a result, the classification performance of multiclass SVM using five selected features as the parameter have excellent performance in predict the bearing conditions data for all types of kernel functions. [ABSTRACT FROM AUTHOR]

This research studies the relationship between footprint depths and load in the calcaneal area when human standing in an upright posture. Footprint depths are deformation in the calcaneal area obtained from the z-value extraction of the Boolean operation acquired from unloaded foot scanning using 3D scanner and loaded foot using foot plantar scanner. To compare peak loads estimated from footprint depth maximum, force sensing resistor (FSR) sensor is attached over the shoe insole with zero heel height in the calcaneal area. Twenty participants were selected from students of Mechanical Engineering Department Diponegoro University with the average the age and the body weight 19.5 years and 55.27 kg respectively. Results that were relatively accurate was found on the calcaneal loading estimation by footprint depth is presented by curve and data distribution which are in good agreement with the result of the measurement. A significant difference in estimating calcaneal loading is mainly caused by plantar foot position of research subjects which is not perpendicular to foot ankle and hallux. In addition, plantar foot position which bends to front/back/side affects the result of footprint depths. [ABSTRACT FROM AUTHOR]

One way to extend the life time of the gear is minimizing the distortion during the manufacturing process. One of the most important processes in manufacturing to produce gears is heat treatment process. The purpose of this study is to analyze the distortion of the gear after heat treatment process. The material of gear is AISI 1045, and it was designed with the module (m) 1.75, and a number of teeth (z) 29. Gear was heat-treated in the furnace at a temperature of 800°C, holding time of 30 minutes, and then quenched in water. Furthermore, surface hardening process was also performed on gear teeth at a temperature of 820°C and holding time of 35 seconds and the similar procedure of analysis was conducted. The hardness of gear after heat treatment average 63.2 HRC and the teeth surface hardness after gear to induction hardening was 64.9 HRC at the case depth 1 mm. The microstructure of tested gear are martensitic and pearlite. The highest distortion on tooth thickness to upper than 0.063 can cause high precision at the tooth contact is not appropriate. Besides the shrinkage of tooth thickness will also affect to contact angle because the size of gear tolerance was not standardized. [ABSTRACT FROM AUTHOR]