Yeon-Pun Chang

Applications of fault diagnosis in rotating machinery by using time series analysis with neural network

The common diagnosis method of time series analysis is an autoregressive (AR) method, which is a kind of math model that can be established by time difference and vibration amplitude. As the AR model utilized the math method for fitting the variable, the AR coefficients represent the signal features and can be used to determine fault types. This study proposed the difference values of AR coefficients, which indicated that the AR coefficients of ideal signal for normal machine are deducted from faulty machines. It is convention that the relationship between the difference values of AR coefficients and fault types as trained by using back-propagation neural network (BPNN). The new fault diagnosis method by using the difference of AR coefficients with BPNN was proposed in this study. The diagnosis results were obtained and compared with the three methods, which include the difference of AR coefficients with BPNN, the AR coefficients with BPNN and the distance of AR coefficients method for 23 samples. And the diagnosis results obtained by using the difference of AR coefficients with BPNN were superior to AR coefficients with BPNN and distance of AR coefficients methods.

Size effects of SiO2 nanoparticles as oil additives on tribology of lubricant

Purpose – The purpose of this paper is to investigate the tribological properties of liquid paraffin with SiO2 nanoparticles additive made by a sol‐gel method.Design/methodology/approach – The tribological properties of the SiO2 nanoparticles as an additive in liquid paraffin are measured using a ball‐on‐ring wear tester to determine the optimal additive concentration. The mechanism that wear and friction are reduced is studied using scanning electron microscopy (SEM), energy dispersive spectrometry (EDS), and atomic force microscope (AFM).Findings – Experimental results indicate that the sizes of the synthesized SiO2 nanoparticles are distributed uniformly and that the optimal concentrations of SiO2 nanoparticles in liquid paraffin is associated with better tribological properties than pure paraffin oil, and an anti‐wear (AW) ability that depends on the particle size.Originality/value – It is shown in the paper that by reducing friction and AW, the lubricant prepared by the methods described can prolong ...

Combined effects of thermophoresis and electrophoresis on particle deposition onto a wafer

A theoretical study of aerosol particles responding to thermophoresis and electrophoresis, and the deposition rate of the particles onto an axisymmetric wafer, is proposed. The flow is modeled as a two-dimensional, incompressible and steady-state laminar forced convective flow. The mechanisms of particle deposition by convection, Brownian diffusion, sedimentation, thermophoresis and electrophoresis are coupled. Similarity analysis is used to transform the governing equations for mass, momentum and concentration into a system of ordinary differential equations. Through numerical integration of the equation for particle concentration, concentration profiles and deposition rates are obtained. While the results are compared with the simple-addition method for a one-dimensional stagnant film model, the error on the film model increases with an increase in the temperature gradient, particularly when thermophoresis is the dominating mechanism. Comparing the experimental results from previous works with the calculated results of deposition velocity, we find that the comparison shows a very good agreement.

THE EFFECT OF THERMOPHORESIS ON PARTICLE DEPOSITION FROM A MIXED CONVECTION FLOW ONTO A VERTICAL FLAT PLATE

Abstract A theoretical study of the effect of thermophoresis on aerosol particle deposition onto a vertical flat plate is proposed. The flow was modeled as a two-dimensional, incompressible and steady-state laminar flow driven by a combination of forced convection and natural convection. The particle deposition mechanisms were coupled using convection, Brownian diffusion and thermophoresis. Similarity analysis with the box method and block-elimination was used to determine the velocity and temperature fields. Particle deposition velocities were solved using the finite difference method or the approach of numerical integration. The effects of thermophoresis and convection were predicted to be particularly important for submicron particles moving toward a cold surface or blowing away from a hot surface at a given temperature gradient. The results also showed that the particle deposition rates were strongly influenced by thermophoresis and buoyancy force, particularly for opposing flow and hot wall surfaces.

Integrated “CAE” strategies for the design of machine tool spindle-bearing systems

This paper adopts both static and dynamic analyses to examine the necessary integrated procedures for the design of spindle-bearing systems, with modeling and analysis of these systems being based on the finite element method. The study further addresses the sub-structure procedures and the application of a commercial package dynamic solver. Within this study, the effects of design parameters on static and dynamic performance of spindle-bearing systems are analyzed in order to establish the requirement for design modifications, and the paper proposes a number of examples, along with a set of guidelines for the design of machine tool spindles.

Dispersion and tribological properties of liquid paraffin with added aluminum nanoparticles

Purpose – The purpose of this paper is to study the dispersion and tribological properties of liquid paraffin with aluminum nanoparticles as additive, which are prepared by the surface‐modification method using oleic acid (OA).Design/methodology/approach – The dispersion stability of aluminum nanoparticles in liquid paraffin is measured by spectrophotometry, which can be optimization by Taguchi method. The tribological properties are evaluated by using a ball‐on‐ring wear tester.Findings – The results show that few concentrations of aluminum nanoparticles as additives in liquid paraffin have better antiwear and antifriction properties than the pure paraffin oil. Scanning electron microscopy and energy dispersive spectrometer analyses can show that the thin films on the rubbing surfaces can be formed by these aluminum nanoparticles, which not only bear the load but also separate the both interfaces, thus the wear and friction can be reduced.Originality/value – Machine components and mechanism pairs rely on...

Simulation of gas-assisted injection mold-cooling process using line source model approach for gas channel

Abstract Whether it is feasible to perform an integrated simulation for process simulation based on a unified CAE model for gas-assisted injection molding (GAIM) is a great concern. In the present study, numerical algorithms based on the same CAE model used for process simulation regarding filling and packing stages were developed to simulate the cooling phase of GAIM using a cycle-averaged three-dimensional modified boundary element technique similar to that used for conventional injection molding. However, to use the current CAE model for analysis, gas channel was modeled by two-node elements using line source approach. It was found that this new modeling not only affects the mold wall temperature calculation very slightly but also reduces the computer time by 95% as compared with a full gas channel modeling required a lot of triangular elements on gas channel surface. This investigation indicates that it is feasible to achieve an integrated process simulation for GAIM under one CAE model resulting in great computational efficiency for industrial application.

Certainty improvement in diagnosis of multiple faults by using versatile membership functions for fuzzy neural networks

Because the relationship between frequency symptoms and fault causes are different, this study uses fuzzy neural network (FNN) with versatile membership functions to diagnose multiple faults in rotary machinery. According to the frequency symptom values for each fault causes, three kinds of membership functions are used. Besides, the structure of the FNN is large which spend much training time. Thus, when the matrix between frequency symptoms and fault causes can decoupled, the relational matrix decomposed into several sub-matrixes and the structure of the FNN can also divided into several sub-networks. In this study, two above-mention approaches are combined to diagnose multiple faults and compared with neural network (NN), FNN with single/versatile membership functions in two actual cases.

Condensing heat transfer characteristics of aluminum flat tube

Aluminum brazed heat exchangers are used widely in automobile air conditioning systems. Usually, this kind of heat exchanger is made of a flat tube with several independent passages in the cross-section, and formed into a serpentine or a parallel flow geometry. In these heat exchangers, a multitude of corrugated fins with louvers are inserted into the gaps between flat tubes. As a result of their high performance, some companies are starting to use extruded brazed aluminum heat exchangers in heating, ventilating and air conditioning applications. The purpose of this investigation is to provide the basic condensing heat transfer data for the extruded aluminum flat tube. Four different flat tubes and one micro fin tube were employed in this series of experimental studies. HFC-134a and HCFC-22 were the working fluids used. Our results were closest to Shahs correlation in the low quality region.

Estimation of thermal deformation in machine tools using the hybrid autoregressive moving-average - neural network model:

Abstract A hybrid model, which is composed of an autoregressive moving-average (ARMA) filter and a feedforward neural network (FNN), is proposed to increase prediction accuracy and to reduce learning time for the estimation of thermal deformations in a machine tool. The ARMA filter is used to yield state variables which establish the relationship between the present and past states of thermal deformations for the reservation of the influences of past temperatures and deformation. Otherwise, the quantity of FNN inputs is very vast because of the data needed for the non-linear system. These state variables, which are estimated by past measured temperatures and past estimated deformation, serve as inputs of the FNN. The algorithms of this hybrid model are presented and verified by the experimental results; also, the prediction accuracy is compared with the ARMA and FNN independently for the same learning iterations.