High-resolution optical inspection system for fast detection and classification of surface defects

A high-resolution automated optical inspection (AOI) system based on parallel computing is developed to achieve fast inspection and classification of surface defects. To perform fast inspection, the AOI apparatus is connected to a central computer which executes image processing instructions in a graphical processing unit. Defect classification is simultaneously implemented with Hu’s moment invariants and back propagation neural (BPN) approach. Experiments on touch panel glass show that using 100 training samples and 1000?cycle iterations in BPN, the accurate classification of surface defects for a 350?×?350 pixels image can be completed in less than 0.1 ms. Moreover, the inspection of a 43?mm?×?229?mm sample that yields an 800 megapixel raw data can be completed remarkably fast in less than 3?s. Thus, the AOI system is capable of performing fast, reliable, and fully integrated inspection and classification equipment for in-line measurements.     

Monitoring and diagnosis of the tapping process for product quality and automated manufacturing

Tapping has been widely used throughout industry, and its proper operation is paramount in ensuring product quality. Therefore, monitoring and diagnosis are needed to detect the tapping process conditions. They are also important for automated manufacturing. In this work, a combination of ten indices of the tapping process was extracted from tapping torque, thrust force, and lateral forces. The Sequential Forward Search (SFS) algorithm has been used to select the best feature sets. Adaptive Neuro-Fuzzy Inference Systems (ANFIS) were used for the monitoring and diagnosis of tapping process. A 3?×?2 ANFIS structure can distinguish normal tapping process from abnormal tapping process with 100 % reliability. The tapping process conditions can be further classified into five categories with over 95 % success rate using a 10?×?2 ANFIS structure for diagnostic purpose. In simple words, monitoring and diagnosis of tapping process can be carried out successfully using SFS and ANFIS.     

Real-time recognition of ball bearing states for the enhancement of precision,quality, efficiency,safety, and automation of manufacturing

Real-time recognition of ball bearing states can enhance precision, quality, efficiency, safety, and automation of manufacturing. Three features, including peak amplitude of the frequency domain, percent power, and peak RMS, have been extracted from the radial acceleration of ball bearings. The sequential forward search algorithm has been applied to select the best vibration features. Adaptive neuro fuzzy inference systems (ANFIS) have been used. A 2?×?2 ANFIS using the π-shaped built-in membership function can distinguish normal bearing states from defective bearings states with 100 % success rate. Furthermore, a 3?×?5 ANFIS can classify ball bearings into six different states with a success rate of over 95 % for diagnostic purpose. In this manner, real-time recognition of ball bearing states for manufacturing can be performed efficiently and effectively.     

The Application of an ANFIS and Grey System Method in Turning Tool-Failure Detection

The cutting process is a major material removal process; hence, it is important to search for ways of detecting tool failure. This paper describes the results of the application of an adaptive-network-based fuzzy inference system (ANFIS) for tool-failure detection in a single-point turning operation. In a turning operation, wear and failure of the tool are usually monitored by measuring cutting force, load current, vibration, acoustic emission (AE) and temperature. The AE signal and cutting force signal provide useful information concerning the tool-failure condition. Therefore, five input parameters of the combined signals (AE signal and cutting force signal) have been used in the ANFIS model to detect the tool state. In this model, we adopted three different types of membership function for analysis for ANFIS training and compared their differences regarding the accuracy rate of the tool-state detection. The result obtained for the successful classification of tool state with respect to only two classes (normal or failure) is very good. The results also indicate that a triangular MF and a generalised bell MF have a better rate of detection. We also applied grey relational analysis to determine the order of influence of the five cutting parameters on tool-state detection.     

A step towards the in-process monitoring for electrochemical microdrilling

The bandsawing as a multi-point cutting operation is the preferred method for cutting off raw materials in industry. Although cutting off with bandsaw is very old process, research efforts are very limited compared to the other cutting process. Appropriate online tool condition monitoring system is essential for sophisticated and automated machine tools to achieve better tool management. Tool wear monitoring models using artificial neural network are developed to predict the tool wear during cutting off the raw materials (American Iron and Steel Institute 1020, 1040 and 4140) by bandsaw. Based on a continuous data acquisition of cutting force signals, it is possible to estimate or to classify certain wear parameters by means of neural networks thanks to reasonably quick data-processing capability. The multi-layered feed forward artificial neural network (ANN) system of a 6?×?9?×?1 structure based on cutting forces was trained using error back-propagation training algorithm to estimate tool wear in bandsawing. The data used for the training and checking of the network were derived from the experiments according to the principles of Taguchi design of experiments planned as L ₂₇. The factors considered as input in the experiment were the feed rate, the cutting speed, the engagement length and material hardness. 3D surface plots are generated using ANN model to study the interaction effects of cutting conditions on sawblade. The analysis shows that cutting length, hardness and cutting speed have significant effect on tooth wear, respectively, while feed rate has less effect. In this study, the details of experimentation and ANN application to predict tooth wear have been presented. The system shows that there is close match between the flank wear estimated and measured directly.     

Real-time monitoring of flank wear behavior of ceramic cutting tool in turning hardened steels

Flank wear of an alumina-based ceramic cutting tool was determined in hard turning two workpieces (AISI 4340 and 52100 hardened steels) at three cutting speeds (142, 181, and 264?m/min) to devise a real-time monitoring system. Results of the six turning tests were assessed using Kruskal?CWallis test, regression models, and linear trend analysis. Multiple non-linear regression models that explained variation in flank wear as a function of time (second) had a range of $ R_{\rm{adj}}^2 $ values of 27.7% for the test 4340-142 to 95% for the test 52100-181. Linear trend models revealed that the highest flank wear rate of the ceramic cutting tool belonged to the test 52100-181. Interaction effect of the three cutting speeds and the two workpiece types was determined to account for 82.2% of variation in flank wear (P?<?0.001). The real-time monitoring system designed in this study appeared to be promising in terms of determining and quantifying flank wear behavior of the ceramic cutting tool and optimal hard turning conditions.     

Online tool condition monitoring in turning titanium (grade 5) using acoustic emission: modeling

This paper presents an online prediction of tool wear using acoustic emission (AE) in turning titanium (grade 5) with PVD-coated carbide tools. In the present work, the root mean square value of AE at the chip–tool contact was used to detect the progression of flank wear in carbide tools. In particular, the effect of cutting speed, feed, and depth of cut on tool wear has been investigated. The flank surface of the cutting tools used for machining tests was analyzed using energy-dispersive X-ray spectroscopy technique to determine the nature of wear. A mathematical model for the prediction of AE signal was developed using process parameters such as speed, feed, and depth of cut along with the progressive flank wear. A confirmation test was also conducted in order to verify the correctness of the model. Experimental results have shown that the AE signal in turning titanium alloy can be predicted with a reasonable accuracy within the range of process parameters considered in this study.     

Development and analysis of an online tool condition monitoring and diagnosis system for a milling process and its real-time implementation

This paper addresses the development of an online tool condition monitoring and diagnosis system for a milling process. To establish a tool condition monitoring and diagnosis system, three modeling algorithms–an Adaptive neuro fuzzy inference system (ANFIS), a Back-propagation neural network (BPNN) and a Response surface methodology (RSM)–are considered. In the course of modeling, the measured milling force signals are processed, and critical features such as Root mean square (RMS) values and node energies are extracted. The RMS values are input variables for the models based on ANFIS and RSM, and the node energies are those for the BPNN-based model. The output variable is the confidence value, which indicates the tool condition states–initial, workable and dull. The tool condition states are defined based on the measured flank wear values of the endmills. During training of the models, numerical confidence values are assigned to each tool condition state: 0 for the initial, 0.5 for the workable and 1 for the dull. An experimental validation was conducted for all three models, and it was found that the RSM-based model is best in terms of lowest root mean square error and highest diagnosis accuracy. Finally, the RSM-based model was used to build an online system to monitor and diagnose the tool condition in the milling process in a real-time manner, and its applicability was successfully demonstrated.     

Development of a tool wear observer model for online tool condition monitoring and control in machining nickel-based alloys

Online monitoring and in-process control improves machining quality and efficiency in the drive towards intelligent machining. It is particularly significant in machining difficult-to-machine materials like super alloys. This paper attempts to develop a tool wear observer model for flank wear monitoring in machining nickel-based alloys. The model can be implemented in an online tool wear monitoring system which predicts the actual state of tool wear in real time by measuring the cutting force variations. The correlation between the cutting force components and the flank wear width has been established through experimental studies. It was used in an observer model, which uses control theory to reconstruct the flank wear development from the cutting force signal obtained through online measurements. The monitoring method can be implemented as an outer feedback control loop in an adaptive machining system.     

Development of a system for monitoring tool condition using acousto-optic emission signal in face turning—an experimental approach

In automated manufacturing systems, one of the most important issues is accurate detection of the tool conditions under given cutting conditions so that worn tools can be identified and replaced in time. In metal cutting as a result of the cutting motion, the surface of workpiece will be influenced by cutting parameters, cutting force, and vibrations, etc. But the effects of vibrations have been paid less attention. In the present paper, an investigation is presented of a tool condition monitoring system, which consists of a fast Fourier transform preprocessor for generating features from an online acousto-optic emission (AOE) signals to develop a database for appropriate decisions. A fast Fourier transform (FFT) can decompose AOE signals into different frequency bands in the time domain. Present work uses a laser Doppler vibrometer for online data acquisition and a high-speed FFT analyser used to process the AOE signals. The generation of the AOE signals directly in the cutting zone makes them very sensitive to changes in the cutting process due to vibrations. AOE techniques is a relatively recent entry into the field of tool condition monitoring. This method has also been widely used in the field of metal cutting to detect process changes like displacement due to vibration and tool wear, etc. In this research work the results obtained from the analysis of acousto-optic emission sensor employs to predict flank wear in turning of AISI 1040 steel of 150 BHN hardness using Carbide insert and HSS tools. The correlation between the tool wear and AOE parameters is analyzed using the experimental study conducted in 16 H.P. all geared lathe. The encouraging results of the work pave the way for the development of a real-time, low-cost, and reliable tool condition monitoring system. A high degree of correlation is established between the results of the AOE signal and experimental results in identification of tool wear state.     

Studies on the fabrication of tool steel components with micro-features by PIM

Fabrication of structures with fine features has been a challenge and has required sophisticated equipment. In the current work, the use of a cost-effective method, PIM (powder injection molding) was evaluated to fabricate structures with fine features such as sharp edges with dimension in microns. The selected component consisted of micro features of two steps having a cross-section dimension of 0.6×0.07 mm and 0.6×0.06 mm. In order to achieve good mechanical properties coupled with wear resistance, an M2 tool steel material was selected. The mould tip region with two-step micro features was fabricated using into multi-mould pieces to ensure the edge sharpness of the two-step micro features in the mould cavity and to create an air vent to release the entrapped gas at the moulding stage. The moulded samples were first debound and then sintered at different temperatures. The results of actual sintering at 1,244, 1,252, 1,260, and 1,280°C were analyzed, since transition to liquid phase sintering occurs in this region. Visual observation, microstructural analysis and hardness distribution were carried out on the samples sintering at different temperatures. Incomplete sintering occurred at 1,244°C while at 1,252°C distortion of the samples was identified. Sintering at 1,260°C yielded parts with good integrity and strength and a further increase of sintering temperature deteriorated the characteristics. Five samples were chosen randomly from three batches of 100 pieces for a pilot production scale. Their dimensional flatness was measured and the result showed that flatness less than ±0.3% can be achieved under optimized sintering temperature of 1,260°C?±?0.3°C. This means that the M2 tool steel component with micro features can be possibly produced by PIM process in mass production.     

A new approach to investigate tool condition using dummy tool holder and sensor setup

The industrial demand for automated machining systems to enhance process productivity and quality in machining aerospace components requires investigation of tool condition monitoring. The formation of chip and its removal have a remarkable effect on the state of the cutting tool during turning. This work presents a new technique using acoustic emission (AE) to monitor the tool condition by separating the chip formation frequencies from the rest of the signal which comes mostly from tool wear and plastic deformation of the work material. A dummy tool holder and sensor setup have been designed and integrated with the conventional tool holder system to capture the time-domain chip formation signals independently during turning. Several dry turning tests have been conducted at the speed ranging from 120 to 180?m/min, feed rate from 0.20 to 0.50?mm/rev, and depth of cut from 1 to 1.5?mm. The tool insert used was TiN-coated carbide while the work material was high-carbon steel. The signals from the dummy setup clearly differ from the AE signals of the conventional setup. It has been observed that time-domain signal and corresponding frequency response can predict the tool conditions. The rate of tool wear was found to decrease with chip breakage even at higher feed rate. The tool wear and plastic deformation were viewed to decrease with the increased radius of chip curvature and thinner chip thickness even at the highest cutting speed, and these have been verified by measuring tool wear. The chip formation frequency has been found to be within 97.7 to 640?kHz.     

WEAR MONITORING IN TURNING OPERATIONS USING VIBRATION AND STRAIN MEASUREMENTS

For the efficient and reliable operation of automated machining processes, the implementation of suitable tool condition monitoring (TCM) strategy is required. Various monitoring systems, utilising sophisticated signal processing techniques, have been widely researched for a number of different processes. Most monitoring systems developed up to date employ force, acoustic emission and vibration, or a combination of these and other techniques with a sensor integration strategy. With this work, the implementation of a monitoring system utilising simultaneous vibration and strain measurements on the tool tip, is investigated for the wear of synthetic diamond tools which are specifically used for the manufacturing of aluminium pistons. Contrary to many of the earlier investigations, this work was conducted in a manufacturing environment, with the associated constraints such as the impracticality of direct measurement of the wear. Data from the manufacturing process was recorded with two piezoelectric strain sensors and an accelerometer, each coupled to a DSPT Siglab analyser. A large number of features indicative of tool wear were automatically extracted from different parts of the original signals. These included features from the time and frequency domains, time-series model coefficients (as features) and features extracted from wavelet packet analysis. A correlation coefficient approach was used to automatically select the best features indicative of the progressive wear of the diamond tools. The self-organising map (SOM) was employed to identify the tool state. The SOM is a type of neural network based on unsupervised learning. A near 100% correct classification of the tool wear data was obtained by training the SOM with two independent data sets, and testing it with a third independent data set.     

Tool condition monitoring based on numerous signal features

This paper presents a tool wear monitoring strategy based on a large number of signal features in the rough turning of Inconel 625. Signal features (SFs) were extracted from time domain signals as well as from frequency domain transforms and their wavelet coefficients (time–frequency domain). All of them were automatically evaluated regarding their relevancy for tool wear monitoring based on a determination coefficient between the feature and its low-pass-filtered course as well as the repeatability. The selected SFs were used for tool wear estimation. The accuracy of this estimation was then used to evaluate the sensor and signal usability.     

Tool crater wear depth modeling in CBN hard turning

Hard turning has been receiving increased attention because it offers many possible benefits over grinding in machining hardened steel. The wear of cubic boron nitride (CBN) tools, which are commonly used in hard turning, is an important issue that needs to be better understood. For hard turning to be a viable replacement technology, the high cost of CBN cutting tools and the cost of down-time for tool changing must be minimized. In addition to progressive flank wear, microchipping and tool breakage (which lead to early tool failure) are prone to occur under aggressive machining conditions due to significant crater wear and weakening of the cutting edge. The objective of this study is to model the CBN tool crater wear depth (K_T) to guide the design of CBN tool geometry and to optimize cutting parameters in finish hard turning. First, the main wear mechanisms (abrasion, adhesion, and diffusion) in hard turning are discussed and the associated wear volume loss models are developed as functions of cutting temperature, stress, and other process information. Then, the crater wear depth is predicted in terms of tool/work material properties and process information. Finally, the proposed model is experimentally validated in finish turning of hardened 52100 bearing steel using a low CBN content tool. The comparison between model predictions and experimental results shows reasonable agreement, and the results suggest that adhesion is the dominant wear mechanism within the range of conditions that were investigated.     

An inexpensive system for classifying tool wear states using pattern recognition

The purpose of this research work is to develop an inexpensive model tool wear sensing system using pattern recognition. Accordingly, the combined output of radial force, feed force and acoustic emission (r.m.s. value) is utilized to model the tool flank wear in a turning operation. The tool wear sensing system consists of two phases: training and classification. The training phase is done off-line and is used to determine the weight coefficients for the linear decision functions using the prototype patterns from the cutting tests. The classification phase is in real time. In the first stage of the classification phase, the minimum distance classifier selects a prototype (conditions already trained) cutting test that is closest to the cutting test to be performed. The linear decision functions of the prototype test selected are used for classifying the incoming signal of the actual cutting test into one of three wear classes.

The success rate of training for various tests varied between 39.57% and 100%. The success rate of classifying signals from actual tests was also encouraging, demonstrating that the proposed methodology can be successfully applied to predict the status of the cutting tool on-line using low budget equipment.     

Finish Machining of Nickel-Base Inconel 718 Alloy with Coated Carbide Tool under Conventional and High-Pressure Coolant Supplies

Single-point turning of Inconel 718 alloy with commercially available Physical Vapour Deposition (PVD)-coated carbide tools under conventional and high-pressure coolant supplies up to 20.3 MPa was carried out. Tool life, surface roughness (Ra), tool wear, and component forces were recorded and analyzed. The test results show that acceptable surface finish and improved tool life can be achieved when machining Inconel 718 with high coolant pressures. The highest improvement in tool life (349%) was achieved when machining with 11 MPa coolant supply pressure at higher speed conditions of 60 m · min^?1. Machining with coolant pressures in excess of 11 MPa at cutting speeds up to 40 m · min^?1 lowered tool life more than when machining under conventional coolant flow at a feed rate of 0.1 mm · rev^?1. This suggests that there is a critical coolant pressure under which the cutting tools performed better under high-pressure coolant supplies.

Cutting forces increased with increasing cutting speed due probably to reactive forces introduced by the high-pressure coolant jet. Tool wear/wear rate increased gradually with prolonged machining with high coolant pressures due to improved coolant access to the cutting interface, hence lowering cutting temperature. Nose wear was the dominant tool failure mode when machining with coated carbide tools due probably to a reduction in the chip-tool and tool-workpiece contact length/area.     

TOOL LIFE PREDICTION USING RANDOM WALK BAYESIAN UPDATING

According to the Taylor tool life equation, tool life reduces with increasing cutting speed. The influence of additional factors can also be incorporated. However, tool wear is generally considered a stochastic process with uncertainty in the model constants. In this work, Bayesian inference is applied to predict tool life for milling/turning operations using the random walk/surface methods. For milling, Bayesian inference using a random walk approach is applied to the well-known Taylor tool life model. Tool wear tests are performed using an uncoated carbide tool and AISI 1018 steel work material. Test results are used to update the probability distribution of tool life. The updated beliefs are then applied to predict tool life using a probability distribution. For turning, both cutting speed and feed are considered. Bayesian updating is performed using the random surface technique. Turning tests are completed using a coated carbide tool and forged AISI 4137 chrome alloy steel. The test results are applied to update the probability distribution of tool life and the updated beliefs are used to predict tool life. While this work uses the Taylor model, by following the procedures described here, the technique can be applied to other tool life models as well.     

Tribological characterisation of epoxy–graphene–liquid filler composite coatings on steel under base oil external lubrication

In our earlier study, epoxy-based composites with graphene (10?wt-%) and in situ liquid fillers (base oil SN150 or perfluoropolyether at 10?wt-%) were found to provide low friction and highly wear durable as thin coatings on the steel substrate in dry interfacial state. In this present work, we have tested this composite in the presence of an external lubricant (base oil SN150). The lowest coefficient of friction was recorded as 0.04 and the lowest specific wear rate was measured as 9.8?×?10^?7?mm^3?Nm^?1 for the composites without any failure of the coating up to 200,000 sliding cycles. It is shown that such polymeric coatings can be an excellent boundary film in both dry and lubricated conditions for various bearings.     

3D characterisation of tool wear whilst diamond turning silicon

Nanometrically smooth infrared silicon optics can be manufactured by the diamond turning process. Due to its relatively low density, silicon is an ideal optical material for weight sensitive infrared (IR) applications. However, rapid diamond tool edge degradation and the effect on the achieved surface have prevented significant exploitation. With the aim of developing a process model to optimise the diamond turning of silicon optics, a series of experimental trials were devised using two ultra-precision diamond turning machines. Single crystal silicon specimens (1 1 1) were repeatedly machined using diamond tools of the same specification until the onset of surface brittle fracture. Two cutting fluids were tested. The cutting forces were monitored and the wear morphology of the tool edge was studied by scanning electron microscopy (SEM).The most significant result showed the performance of one particular tool was consistently superior when compared with other diamond tools of the same specification. This remarkable tool performance resulted in doubling the cutting distance exhibited by the other diamond tools. Another significant result was associated with coolant type. In all cases, tool life was prolonged by as much as 300% by using a specific fluid type.Further testing led to the development of a novel method for assessing the progression of diamond tool wear. In this technique, the diamond tools gradual recession profile is measured by performing a series of plunging cuts. Tool shape changes used in conjunction with flank wear SEM measurements enable the calculation of the volumetric tool wear rate.