Mohd Zaki Nuawi

Online tool wear prediction system in the turning process using an adaptive neuro-fuzzy inference system

Tool wear is a detrimental factor that affects the quality and tolerance of machined parts. Having an accurate prediction of tool wear is important for machining industries to maintain the machined surface quality and can consequently reduce inspection costs and increase productivity. Online and real-time tool wear prediction is possible due to developments in sensor technology. Recently, various sensors and methods have been proposed for the development of tool wear monitoring systems. In this study, an online tool wear monitoring system was proposed using a strain gauge-type sensor due to its simplicity and low cost. A model, based on the adaptive network-based fuzzy inference system (ANFIS), and a new statistical signal analysis method, the I-kaz method, were used to predict tool wear during a turning process. In order to develop the ANFIS model, the cutting speed, depth of cut, feed rate and I-kaz coefficient from the signals of each turning process were taken as inputs, and the flank wear value for the cutting edge was an output of the model. It was found that the prediction usually accurate if the correlation of coefficients and the average errors were in the range of 0.989-0.995 and 2.30-5.08% respectively for the developed model. The proposed model is efficient and low-cost which can be used in the machining industry for online prediction of the cutting tool wear progression, but the accuracy of the model depends upon the training and testing data.

Online Cutting Tool Wear Monitoring Using I-Kaz Method and New Regression Model

This study presents a new method for detecting the cutting tool wear based on the measured cutting force signals using the regression model and I-kaz method. The detection of tool wear was done automatically using the in-house developed regression model and 3D graphic presentation of I-kaz 3D coefficient during machining process. The machining tests were carried out on a CNC turning machine Colchester Master Tornado T4 in dry cutting condition, and Kistler 9255B dynamometer was used to measure the cutting force signals, which then stored and displayed in the DasyLab software. The progression of the cutting tool flank wear land (VB) was indicated by the amount of the cutting force generated. Later, the I-kaz was used to analyze all the cutting force signals from beginning of the cut until the rejection stage of the cutting tool. Results of the I-Kaz analysis were represented by various characteristic of I-kaz 3D coefficient and 3D graphic presentation. The I-kaz 3D coefficient number decreases as the tool wear increases. This method can be used for real time tool wear monitoring.

On the Need of Kurtosis-Based Technique to Evaluate the Fatigue Life of a Coil Spring

Automotive suspension systems have big potential of mechanical failure due to fatigue as this system was subjected to the variable amplitude loading under service condition. This study was specifically focuses on coil spring, one of the suspension system parts. When vehicle was driven on any road surfaces and hit potholes or bump, this coil spring is then affected by the significant load. This significant load will cause the damage to this component. The objective of this study to predict the coil spring fatigue damage and then relate it with the I-kaz coefficient, results from a new statistical-based method, I-kaz method. The signal captured from this repeated loading was known as fatigue signal. The car was driven on three different roads, highway, country road and damage-surface road. The strain gauges were mounted on the outer surfaces of the coil spring to capture the fatigue load variable amplitude data on the road. The recorded strain was then analysed for fatigue damage and I-kaz coefficient purposes. From the analysis, it was found that the damage road gives the highest fatigue damage rather than the other road and also the fatigue damage was proportionally related to the I-kaz coefficient.

Study of Noise, Vibration and Harshness (NVH) for Malaysian Army (MA) 3-Tonne Trucks

Moving vehicle will produce dynamic excitation and can be categorized in the study of noise, vibration and harshness (NVH). Studies of NVH can resolve issues associated with the use of a vehicle comfort. Exposure to some level of noise and vibration can be classified as a danger to vehicle passengers if they are exposed in a standard eight-hour time period. Therefore NVH aspect is the most important aspect in selecting a military vehicle other than the factor of safety, reliability, fuel consumption and maintenance costs. This study was carried out on Malaysian Army (MA) 3-tonne truck drivers who are exposed to noise and vibration produced by the truck. Vibration exposure at seat, steering wheel and body floor was measured using tri-axial and single axis accelerometers which were connected to a Brüel & Kjær Type 3649 vibration analyser, while noise exposure was measured at the drivers ear position using a DuO smart noise monitor. From the results obtained, both vibration and noise pressure levels increased proportionally with the increasing speed of the truck. In the long-term, will have an impact on the health of the truck drivers.

Correlating strain and acoustic emission signals of metallic component using global signal statistical approach

This study clarifies the fatigue properties using global signal statistical approach during the crack initiation stage for metallic component. Strain loading and acoustic emission (AE) signatures are captured simultaneously in the form of signal waves recorded by strain gauge and AE transducer. An initial study was carried out in the laboratory on medium carbon AISI 1045 steel specimens at constant amplitude cyclic loading condition, which lead to the fatigue failure characteristics. This study was carried out to investigate the relationship between strain and AE signals in order to make sure that AE technique also can be used as a detecting and monitoring crack initiation in metallic specimens. To achieve the goal, three different loads were applied on three specimens to capture the differences of the signals. The specific data acquisition systems were used to collect strain and AE signatures. For the purpose of analysis, the method of root mean square (r.m.s) and the kurtosis were used. The r.m.s value was used to quantify the overall vibrational energy content whereas the kurtosis was then used because of its sensitivity to high amplitude events. Based on these statistical approaches, the correlation patterns between both signals are expected to give a meaningful baseline to predict and monitor crack initiation of a metallic specimen.

A statistical based analysis approach on fatigue failure assessment of carbon steel

This paper discussed on an application of the statistical based analysis approach on fatigue failure of carbon steel. A new statistical based method known as Integrated Kurtosis-based Algorithm for Z-filter (I-kaz) and the quadratic mean were used in this fatigue failure assessment. Fatigue tests were performed according to the ASTM E466-96 standard with the attachment of a strain gauge to the tested specimen. Test was performed at a constant loading leading to the collection of strain data throughout the test. A fatigue based data acquisition system was used in this work. All obtained data were analysed using specific software. Fatigue failure behaviour was observed through the strain signal analysis to investigate significant correlation between I-kaz coefficient, root mean square value and the fatigue failure behaviour. It was found that I-kaz coefficient and root mean square value are proportional to life cycle.

Determination of sensor location for cutting tool deflection using finite element method simulation

The aim of this study was to determine the the optimum locations to mount the sensor for measuring the cutting tool deflection during turning process using finite element method simulation. In this study, stress analysis had been conducted using Autodesk Inventor Professional 2010 integrated with ANSYS software. The simulation results were validated using a strain gauge as the sensor for the detection of cutting force signal during the turning of hardened plain carbon steel JIS S45C using carbide tool. Two strain gauges were mounted on the tool holder at two defined locations I and II, at distances of 37 and 47 mm, respectively, from the cutting point. Only one set of cutting parameters was conducted at spindle speed, N = 1000 rpm, feed, f = 0.25 mm/rev and depth of cut, d = 0.80 mm. The turning process was stopped and the insert discarded when the average flank wear, VBm, reached 0.30 mm. The main cutting force, Fy, and the feed force, Fx, for each machining run were measured, collected and analysed at locations I and II. It was found that when strain gauges were placed at a distance of approximately 43 mm from the cutting point, it was the optimum location for sensing the cutting force signals.

Correlation between Whole Body Vibration Exposure and I-KazTM 3D Coefficient for Malaysian Army 3-Tonne Truck Drivers

Exposure to high levels of whole body vibration (WBV) from the seat can present risks to health and safety and cause injuries to drivers. This study presents a study on WBV exposure to Malaysian Army 3-tonne truck drivers using the method defined in International Standard ISO 2631-1:1997. The vehicle that was chosen for this study was based on the highest number of vehicles available within the Malaysian Army services. 3-tonne truck is the most widely used in Malaysian Army services. WBV from the drivers were measured using tri-axial accelerometer and was connected to Brüel & Kjær Vibration Analyser Type 3649, which is capable to produce raw data to calculate the vibration dose value (VDV). The raw data from each axis were analyzed using statistical analysis method, Integrated Kurtosis Algorithm for Z-notch Filter (I-kazTM) 3D to determine the vibration values in relation to varying speed of the vehicle. Based on the results, WBV experienced by the drivers can be presented in VDV and I-kazTM 3D coefficient which the value will increase proportional to the truck speed.

Eddy current braking experiment using brake disc from aluminium series of A16061 and A17075

The electromagnetic braking using eddy current was studied, focused on two series of aluminium as the brake disc which are A16061 and A17075. This paper presents the comparison for both series in a few varied parameters related to eddy current braking such as air-gap, number of turns and brake disc thickness. Optical tachometer has been used along with PULSE analyzer to capture the speed (rpm) and time (s). The findings shows that the smaller the air-gap, the larger of electromagnet turns and the thicker disc thickness is, will generate higher braking torque to stop the rotational motion of disc brake and give great performance for eddy current braking. Thos parameters that been evaluated also addressed a potential on expanding this knowledge to develop an electromagnetic braking system to replace the conventional braking system.

Using the pressure transmission coefficient of a transmitted wave to evaluate some of the mechanical properties of refractory metals

Refractory metals have attracted increasing interest in recent years because of their use in many high-temperature applications. However, the characteristics of these metals calculated using loaded tests (such as tensile strength tests) differ considerably from those calculated using one of the most famous methods in NDT which is called time of flying of the wave (TOF).The present study presents two solutions based on calculating the pressure transmission coefficient (PTC) of the transmitted wave between the test sample and magnesium metal. The first is based on the development of a highly accurate algorithm that lowers the cost by determining the acoustic impedance of the test specimen to calculating mechanical properties. Up to 26 theoretical tests were done (10 of these tests for refractory materials) according to their known mechanical properties to verify the accuracy of the algorithm. The convergence in results ranged from 92% to 99%. The second solution was designed to solve the same problem for specimens with a thickness of less than 1mm. Eight experimental tests were done (five using refractory materials) to verify the accuracy of the second solution, with the convergence in the results ranging from 94% to 97%. The relationships of the Vrms measured from the oscilloscope with the PTC and with the Fourier transform spectrum were derived. The results of this research were closer to the standard mechanical properties for refractory metals compared with several recent acoustic tests.