Selçuk Erkaya

Noise and Vibration Analysis of Car Engines using Proposed Neural Network

An experimental design method for noise and vibration analysis of two car engines by feedforward and radial basis neural networks is presented. Two types of car engines are experimentally analyzed by using intelligent data acquisition card with software. Measured vibration and noise parameters of two car engines are used as desired values of the neural networks. The effectiveness of using Radial Basis Neural Network (RBNN) with backpropagation algorithm is demonstrated for predicting the vibrations and noises of two car engines. The robustness of the proposed RBNN predictor to parameters of vibration and noise as well measurement disturbances is investigated. The result of experiments and simulation show that the proposed RBNN is able to adapt effectively under disturbances.

Design of Neural Networks Model for Transmission Angle of a Modified Mechanism

This paper discusses Neural Networks as predictor for analyzing of transmission angle of slider-crank mechanism. There are different types of neural network algorithms obtained by using chain rules. The neural network is a feedforward neural network. On the other hand, the slider-crank mechanism is a modified mechanism by using an additional link between connecting rod and crank pin. Through extensive simulations, these neural network models are shown to be effective for prediction and analyzing of a modified slider-crank mechanism’s transmission angle.

Analysis of the vibration characteristics of an experimental mechanical system using neural networks

This paper presents an investigation on the vibration analysis of a gearing mechanism using neural network predictors. The experimental system is positioned on a working table with changeable legs. The legs have different shapes such as L, H and O shapes, for finding the exact leg profiles for the experimental system. Two types of neural networks are used to predict vibrations of the system for different leg profiles. The results of two approaches indicate that the proposed neural network with Levenberg–Marquardt learning algorithm has a superior performance to predict vibration parameters of the system.

Oils quality and performance analysis of vehicle's engines using radial basis neural networks

Purpose – The purpose of this paper is to improve the application of neural networks on vehicle engine systems for fault detecting and analysing engine oils.Design/methodology/approach – Three types of neural networks are employed to find exact neural network predictor of vehicle engine oil performance and quality. Nevertheless, two oil types are analysed for predicting performance in the engine. These oils are used and unused oils. In experimental work, two accelerometers are located at the bottom of the car engine to measure related vibrations for analysing oil quality of both cases.Findings – The results of both computer simulation and experimental work show that the radial basis neural network predictor gives good performance at adapting different cases.Research limitations/implications – The results of the proposed neural network analyser follow the desired results of the vehicle engines vibration variation. However, this kind of neural network scheme can be used to analyse oil quality of the car in...

Vibration analysis of food industries mixing systems for long life using neural networks

Due to health problems on food industry, it is necessary to control exact mixing rate of some fruit juices. In this study; whole mixing systems with automation is investigated for different flow rates in the pipes. On the other hand, a robust analyzer is designed to predict real time vibrations on the system. Furthermore, from other investigations; neural networks have superior performance to predict such problems. For that reason, three types of neural networks are used to predict vibrations on different points of three tank mixing system. The results are improved that the proposed Radial Basis Neural Network (RBNN) has good performance at adapting vibration problems on mixing system. Finally, this type of neural network will be employed to analyze food industries automation systems.

Determination of Structural Damping and Optimal Vibration Control of an Adhesively-Bonded Double Containment Cantilever Joint

In this study, the loss factors of an adhesively-bonded double containment cantilever joint were determined for different plate and support lengths. The response of the adhesive joint subjected to a transverse excitation force was measured with a contactless eddy-current sensor and the first bending natural frequency was determined using the Fast Fourier Transform method. The loss factor was calculated using the half-power bandwidth method based on the power spectrum of the joint vibration. After an excitation force was applied to the joint, the damped free vibration analysis was carried out using the finite element method and its measured loss factor. The transverse vibration attenuation was actively controlled with different numbers of actuators located on the top surface of the plate. The optimal control of the vibration attenuation was achieved based on a performance index by considering the strain energy, the kinetic energy, the work done on the adhesive joint by the actuators as well as the vibration attenuation time. Genetic Algorithm was implemented to this optimization problem in which the optimal control force histories, the optimal locations and the optimal numbers of the actuators were searched. Eight actuators exhibited the best control force history minimizing the performance index to 3.34 × 10–2. Thus, the attenuation time was reduced from 16 s to 0.15 s and the absolute displacement was decreased from 13.1 mm to 17.15 × 10–3 mm for 0.15 s. In addition, the modal strain energy and kinetic energy were found to be at lowest levels. As the actuator number was increased only a minor decrease in the performance index was observed after four actuators.

Balancing of Planar Mechanisms Having Imperfect Joints Using Neural Network-Genetic Algorithm (NN-GA) Approach

As a result of design, manufacturing and assembly processes or a wear effect, clearances are inevitable at the joints of mechanisms. In this study, dynamic response of mechanism having revolute joints with clearance is investigated. A four-bar mechanism having two revolute joints with clearance is considered as a model mechanism. A neural network was used to model several characteristics of joint clearance. Kinematic and dynamic analyses were achieved using continuous contact mode between journal and bearing. A genetic algorithm was also used to determine the appropriate values of design variables for reducing the additional vibration effects due primarily to the joint clearance. The results show that the optimum adjusting of suitable design variables gives a certain decrease in shaking forces and their moments on the mechanism frame.

Modelling and evaluation of light railway system’s noise using neural predictors

BackgroundNoise is defined as a sound or series of sounds that are considered to be invasive, irritating, objectionable and disruptive to the quality of daily life. Noise is one of the environmental pollutants, and in cities it is usually originated from road traffic, railway traffic, airports, industry etc. The tram is generally considered as environmentally friendly, namely non-polluting and silent. However complaints from residents living along the tramway lines prove that it may sometimes cause annoyance. In this study, a Global Pointing System (GPS) receiver for determining the sampling locations and a frequency based noise measurement system for collecting the noise data are used to analyse the noise level in the city centre. Both environmental (background) and tram noises are measured.ResultsThree types of neural networks are used to predict the noises of the tram and environment. The results of three approaches indicate that the proposed neural network with Radial Basis Function (RBF) has superior performance to predict the noises of the tram and environment.ConclusionsFor making a decision about transportation planning, this network model can help urban planners for evaluating and/or isolating the tram noise in terms of human health.

Minimization of Shaking Force and Moment on a Four-Bar Mechanism Using Genetic Algorithm

In this study, optimal balancing of a 2D articulated mechanism is investigated to minimize the shaking force and moment fluctuations. Balancing of a four-bar mechanism is formulated as an optimization problem. On the other hand, an objective function based on the subcomponents of shaking force and moment is constituted, and design variables consisting of kinematic and dynamic parameters are defined. Genetic algorithm is used to solve the optimization problem under the appropriate constraints. By using commercial simulation software, optimized values of design variables are also tested to evaluate the effectiveness of the proposed optimization process. This work provides a practical method for reducing the shaking force and moment fluctuations. The results show that both the structure of objective function and particularly the selection of weighting factors have a crucial role to obtain the optimum values of design parameters. By adjusting the value of weighting factor according to the relative sensitivity of the related term, there is a certain decrease at the shaking force and moment fluctuations. Moreover, these arrangements also decrease the initiative of mechanism designer on choosing the values of weighting factors.

Effects of metal surface grinding at the porcelain try-in stage of fixed dental prostheses.

PURPOSE This study was to evaluate the effect of grinding of the inner metal surface during the porcelain try-in stage on metal-porcelain bonding considering the maximum temperature and the vibration of samples. MATERIALS AND METHODS Ninety-one square prism-shaped (1 × 1 × 1.5 mm) nickel-chrome cast frameworks 0.3 mm thick were prepared. Porcelain was applied on two opposite outer axial surfaces of the frameworks. The grinding was performed from the opposite axial sides of the inner metal surfaces with a low-speed handpiece with two types of burs (diamond, tungsten-carbide) under three grinding forces (3.5 N, 7 N, 14 N) and at two durations (5 seconds, 10 seconds). The shear bond strength (SBS) test was performed with universal testing machine. Statistical analyzes were performed at 5% significance level. RESULTS The samples subjected to grinding under 3.5 N showed higher SBS values than those exposed to grinding under 7 N and 14 N (P<.05). SBS values of none of the groups differed from those of the control group (P>.05). The types of bur (P=.965) and the duration (P=.679) did not affect the SBS values. On the other hand, type of bur, force applied, and duration of the grinding affected the maximum temperatures of the samples, whereas the maximum vibration was affected only by the type of bur (P<.05). CONCLUSION Grinding the inner metal surface did not affect the metal-porcelain bond strength. Although the grinding affected the maximum temperature and the vibration values of the samples, these did not influence the bonding strength.