Machining Process Monitoring Research Articles

Prognosis of Bearing Acoustic Emission Signals Using Supervised Machine Learning

Acoustic emission (AE) technique can be successfully utilized for condition monitoring of various machining and industrial processes. To keep machines function at optimal levels, fault prognosis model to predict the remaining useful life (RUL) of machine components is required. This model is used to analyze the output signals of a machine whilst in operation and accordingly helps to set an early alarm tool that reduces the untimely replacement of components and the wasteful machine downtime. Recent improvements indicate the drive on the way towards incorporation of prognosis and diagnosis machine learning techniques in future machine health management systems. With this in mind, this work employs three supervised machine learning techniques; support vector machine regression, multilayer artificial neural network model and gaussian process regression, to correlate AE features with corresponding natural wear of slow speed bearings throughout series of laboratory experiments. Analysis of signal parameters such as signal intensity estimator and root mean square was undertaken to discriminate individual types of early damage. It was concluded that neural networks model with back propagation learning algorithm has an advantage over the other models in estimating the RUL for slow speed bearings if the proper network structure is chosen and sufficient data is provided.

Fault Diagnosis of Face Milling Tool using Decision Tree and Sound Signal

The monitoring of machining process can improve the quality of product and economy of production. The monitoring system helps to recognize and monitor the surface roughness, dimensional tolerance and tool condition. In this way, the condition monitoring system provides precise dimensional products, high productivity and enhanced machine tool life. This paper presents the classification of healthy and faulty conditions of the face milling tool using Decision tree (J48 algorithm) technique through machine learning approach. The sound signals of the face milling tool under healthy and faulty conditions are acquired. A set of discrete wavelet features are extracted from the sound signals using discrete wavelet transform (DWT) method. Decision tree technique is used to select prominent features out of all extracted features. The selected features are fed to the same algorithm for classification. Output of the algorithm is used to study and categorize the tool conditions. The decision tree model has provided a good classification accuracy of about 81% for the given sound signals and can be considered for fault diagnosis/condition monitoring. From the experimental results, it is suggested that the proposed method which comprises of decision tree and DWT techniques with sound signals can be recommended for the applications of fault diagnosis of the face milling tool.

Dynamic Bayesian Network-Based Approach by Integrating Sensor Deployment for Machining Process Monitoring

Many condition monitoring systems based on artificial intelligence process models for machining process monitoring have been developed intensively. However, given that machining processes are very complex (i.e., nonlinear and nonstationary), there is still no clear methodology to acquire machining monitoring systems allowing machining processes to be optimized, predicted, or controlled. In this paper, the coupled hidden Markov model, based on dynamic Bayesian networks, is proposed to monitor a machining process by using multi-directional data fusion and to analyze the effect of the sensor layout on the monitoring accuracy. The features extracted by a singular spectrum and wavelet analysis constitute the input information to the system. The technique is tested and validated successfully by using two scenarios: tool wear condition monitoring (initial wear, gradual wear, or accelerated wear) for the milling process and surface roughness accuracy grade prediction (accuracy grade 9, accuracy grade 8, or accuracy grade 7) for the turning process. In the first case, the maximum recognition rate obtained by the single-sensor placement for tool wear is 83%, whereas in the case of the three-sensor placement, the model recognition rate is 89%. In the second application for turning, the maximum recognition rate obtained by the single-sensor and the double-sensor placements for surface roughness accuracy prediction is 77% and 85%, respectively. In the case of the three-sensor placement, the model recognition rate is 89%. The proposed approach can also be integrated into the diagnosis architecture for condition monitoring in other complex machining systems.

Tool path strategy and cutting process monitoring in intelligent machining

Intelligent machining is a current focus in advanced manufacturing technology, and is characterized by high accuracy and efficiency. A central technology of intelligent machining—the cutting process online monitoring and optimization—is urgently needed for mass production. In this research, the cutting process online monitoring and optimization in jet engine impeller machining, cranio-maxillofacial surgery, and hydraulic servo valve deburring are introduced as examples of intelligent machining. Results show that intelligent tool path optimization and cutting process online monitoring are efficient techniques for improving the efficiency, quality, and reliability of machining.

Investigation of KDP crystal surface based on an improved bidimensional empirical mode decomposition method

Abstract This paper proposed a novel spatial frequency analysis method for the investigation of potassium dihydrogen phosphate (KDP) crystal surface based on an improved bidimensional empirical mode decomposition (BEMD) method. Aiming to eliminate end effects of the BEMD method and improve the intrinsic mode functions (IMFs) for the efficient identification of texture features, a denoising process was embedded in the sifting iteration of BEMD method. With removing redundant information in decomposed sub-components of KDP crystal surface, middle spatial frequencies of the cutting and feeding processes were identified. Comparative study with the power spectral density method, two-dimensional wavelet transform (2D-WT), as well as the traditional BEMD method, demonstrated that the method developed in this paper can efficiently extract texture features and reveal gradient development of KDP crystal surface. Furthermore, the proposed method was a self-adaptive data driven technique without prior knowledge, which overcame shortcomings of the 2D-WT model such as the parameters selection. Additionally, the proposed method was a promising tool for the application of online monitoring and optimal control of precision machining process.

Artificial neural network based tool condition monitoring in micro mechanical peck drilling using thrust force signals

Micro scale machining process monitoring is one of the key issues in highly precision manufacturing. Monitoring of machining operation not only reduces the need of expert operators but also reduces the chances of unexpected tool breakage which may damage the work piece. In the present study, the tool wear of the micro drill and thrust force have been studied during the peck drilling operation of AISI P20 tool steel workpiece. Variations of tool wear with drilled hole number at different cutting conditions were investigated. Similarly, the variations of thrust force during different steps of peck drilling were investigated with the increasing number of holes at different feed and cutting speed values. Artificial neural network (ANN) model was developed to fuse thrust force, cutting speed, spindle speed and feed parameters to predict the drilled hole number. It has been shown that the error of hole number prediction using a neural network model is less than that using a regression model. The prediction of drilled hole number for new test data using ANN model is also in good agreement to experimentally obtained drilled hole number.

Integration of virtual and on-line machining process control and monitoring

This paper presents a virtually assisted on-line milling process control and monitoring system. A part machining process is simulated to predict the cutting forces, torque, power, chip load and other process states. The simulated machining states are accessed by a real time monitoring system which detects the tool failure and adaptively adjusts the feed by predicting the forces from the feed and spindle drive motor current supplied by CNC. The integration of virtual simulation with real time measurements avoids false tool failure detection and transient overloads of the tools during adaptive control. The system has been implemented on a CNC machining centre for use in production.

Integrated Monitoring System of Production Processes

Abstract Integrated monitoring system for discrete manufacturing processes is presented in the paper. The multilayer hardware and software reference model was developed. Original research are an answer for industry needs of the integration of information flow in production process. Reference model corresponds with proposed data model based on multilayer data tree allowing to describe orders, products, processes and save monitoring data. Elaborated models were implemented in the integrated monitoring system demonstrator developed in the project. It was built on the base of multiagent technology to assure high flexibility and openness on applying intelligent algorithms for data processing. Currently on the base of achieved experience an application integrated monitoring system for real production system is developed. In the article the main problems of monitoring integration are presented, including specificity of discrete production, data processing and future application of Cyber-Physical-Systems. Development of manufacturing systems is based more and more on taking an advantage of applying intelligent solutions into machine and production process control and monitoring. Connection of technical systems, machine tools and manufacturing processes monitoring with advanced information processing seems to be one of the most important areas of near future development. It will play important role in efficient operation and competitiveness of the whole production system. It is also important area of applying in the future Cyber-Physical-Systems that can radically improve functionally of monitoring systems and reduce the cost of its implementation.

Process monitoring and effects of process parameters on responses in turn-milling operations based on SN ratio and ANOVA

Abstract Process monitoring in machining constitutes machine performance and machine condition for performing the desired objective. The optimality and effect of machining parameters on machine performance monitoring in tangential and orthogonal turn-milling processes is been studied. Surface Roughness (Ra) and Surface Hardness (H) has been taken as machine performance responses and Tool Vibrations (VIB) as machine condition monitoring response. Laser Doppler Vibrometer (LDV) is used for online capturing of tool vibrations and is analyzed using VibSoft analyzer for processing Acousto-optic emissions (AOE). Single cut machining on A-axis of CNC Vertical Milling center using HSS end mill cutters is adopted. Process parameters like cutter (tool) speed, feed rate and depth of cut with constant rotation of workpiece on A-axis are chosen while machining Brass material under dry condition. Statistical design of experiments (DOE) based on Taguchi’s Orthogonal Array (OA) is adopted for experimentation and Signal-to-Noise ratio (SN ratio) of the responses is used for finding optimality of process parameters. The influence and contribution of the process parameters on the responses is been studied with help of Analysis of Variance (ANOVA).

Inline Plasma Analysis as Tool for Process Monitoring in Laser Micro Machining for Multi-layer Materials

Laser micro machining is an innovative manufacturing technology with a wide range of processable materials and a high level of flexibility. Especially for processing multi-layer materials it is a tool for surface structuring and subsequent functionalization. The manufacturing of e.g. conductor paths for solar cells or batteries requires specific ablation depths on multi-layer materials in order to reveal conducting layers as functional conductor paths. To achieve an ablation of a specific, layer an in-line process monitoring can be used to warrant a robust manufacturing preventing damage on further surface layers. With regards on the need of material specific ablation parameters, this paper addresses the development of a measurement system based on laser induced breakdown spectroscopy (LIBS) as a tool for real-time process monitoring by in-line plasma analysis. The presented results show the specific plasma emission variation at layer interfaces, which can be used for real-time feedback process control with the goal of minimizing the ablation of further layers.

Real Time Monitoring of the CNC Process in a Production Environment- the Data Collection & Analysis Phase

For many years there has been considerable research from both an academic and industrial perspective into the monitoring and control of CNC machining processes, and progress has been well documented. It is widely acknowledged within the CIRP community that collection of information into the performance of material cutting processes is a worthwhile research topic, and this is mirrored in the work of the Scientific Technical Committee-Cutting (STC-C) and other Scientific Technical Committees.This work has been continued by the consortium engaged in the REALISM project, an EU-FP7 funded project which is investigating the use of sensor fusion in a real time production environment, to monitor CNC tool wear through the use of three sensor technologies- Force, Acoustic Emission and Vibration. However, the real work of the project consortium will be in the analysis and interpretation of the data from the collated fusion of the deployed sensors- and the intelligent interrogation of this sensor information. The sensor deployment strategy of this project was outlined in a presentation, poster and paper presented at the 2014 CIRP ICME conference and this current paper provides an update on the ongoing work.An overview is provided in this paper of the challenges that have been overcome as part of the REALISM project, and a brief overview of the initial verification trials that were undertaken on the deployment of the sensors. Further results will also be presented that show promising initial sensor data, which shows conditions leading to Catastrophic Tool Failure (CTF).

Compressive sensing and sparse decomposition in precision machining process monitoring: From theory to applications

Precision machining has been claimed to be the backbone to modern industry. It has been widely applied to the key parts’ production in the aerospace, medical and automotive industries. One of the main problems related to precision machining productivity and safety is the machining condition. By utilizing the acquired information from sensory measurements to direct the further actions, the signal processing bridges the gap between human instruction and full automation. Traditional signal acquisition and processing methods are mainly based on Shannon’s Sampling theory, Fourier methods or wavelet analysis. While these techniques meet challenges in precision machining environment, such as machining at high speed, low signal to noise ratio, and high sampling rate. These factors limit their applications especially the online monitoring implementation. The newly developed compressive sensing theory and sparse decomposition techniques provide a possible solution to these problems, while limited studies have been investigated. This paper serves as an introduction to the theory and shows the theory’s potentials in machining condition monitoring by reviewing related literatures and demonstrating case studies from real experiments.

Indirect Model Based Estimation of Cutting Force and Tool Tip Vibrational Behavior in Milling Machines by Sensor Fusion

Real time prediction of cutting tool condition and machined surface finish have been attractive research objectives over the last decades. However, providing practical and reliable solutions is still a demanding task for milling machine tools. One of the most challenging literature goals is to obtain a robust estimation of the cutting forces through indirect sensor measurements since many process and tool related quantities are indirectly linked to cutting forces. Another challenging issue in machining process monitoring and control is prediction of surface finish and quality. As the vibration plays a major role in the surface generation, this can be done by accurate prediction of the vibrational displacements at the tool tip during machining operation. In this paper, a novel model based estimation of cutting force and tool tip acceleration is designed and tested based on data fusion of different sensors measurements. In this context, two sensors (piezoelectric accelerometer and eddy-current displacement both mounted inside the spindle structure) have been utilized to acquire the experimental signals over a wide range of frequencies.In order to predict the above mentioned quantities, an optimal state estimator based on Kalman Filter (KF) is used. The models have been obtained by system identification method based on experimental measurements performed on a machine tool. The model based estimator is fed by real data. The results show that the estimation of the impulse force and tool tip acceleration can be achieved accurately in low and high frequency ranges by assigning different weights to the measurement sensors.

Development of on the machine process monitoring and control strategy in Robot Assisted Polishing

Robot Assisted Polishing (RAP) can be used to polish rotational symmetric and free form components achieving surface roughness down to Sa 10nm. With the aim to enable unmanned robust and cost efficient application of RAP, this paper presents the development of a monitoring and control strategy for automatic detection of process end point as well as on the machine total surface characterization and local defects identification. The approach is based on a multisensory polishing arm allowing measurement of Acoustic Emission and process forces and a scattered light sensor mounted on the machine. The multisensory approach was experimentally validated in polishing with bonded abrasives demonstrating its suitability for process control in RAP.

A Feature-Enhanced Remote Machining Process Monitoring Method

ABSTRACTMachining process monitoring is essential to guarantee machining quality and reduce production cost, especially for the parts with complex structures and high accuracy requirement. However, effective machining process monitoring is still a challenge in industry. Particularly, a smooth communication between remote control computers and CNC (Computerized Numerical Control) systems is required. To address this requirement, a remote machining process monitoring method based on features and web is proposed. Machining feature model is established to facilitate the communication among the involved modules in the system. A web-based method is adopted to realize remote monitoring and control for multiple machines, as well as to alleviate the computational load of CNC systems. In addition, a feature-based database is constructed as a communication hub, where feature information, monitoring strategies and monitoring results are stored. A prototype system has been developed to validate the feasibility of the ...

Temporal and frequential analysis of the tools wear evolution

The machining process monitoring plays a very significant role in the prevention of non-desired phenome-na, such as excessive wear and the tool rupture. In this article, we use effective techniques based on the analysis of cutting force and vibratory signals measured for various cutting conditions. The proposed methodology uses some signal processing techniques, such as temporal, frequential approaches and the wavelet analysis. The objective of this work is to show, on one hand, the sensitivity of the scalar indicators to the flank wear variation, and on the other hand the determination of the influence of flank wear on the signals produced during a machining process. In this context the wear evolution results allow determining the abrupt changes that can detect the cutting tool damage. DOI: http://dx.doi.org/10.5755/j01.mech.20.2.6933

Current Research and Development Trends of Micro Machine Tool for Micro-cutting

Micro-cutting technology which has a unique advantage in 3D complex parts processing is a hot spot to precision and ultra-precision manufacturing technology research. Micro-cutting machine tool as an important part in micro-cutting technology has presented a miniaturized trend and became an emphasis in current research of micro manufacturing equipment. The research status of micro machine tool for micro-cutting are systematically summarized from the aspects of machine tool bed, micro spindle unit, feed table, micro cutting tools, NC system and monitoring system. Some key technology problems are analyzed such as miniaturized machine tool bed, high speed and high rotation accuracy of micro spindle, high precision and fast response of micro table, structure and performance of micro cutting tool, openness and high precision interpolation of NC system, visual monitoring of machining process, performance of machine tool and so on. Meanwhile, feasible countermeasures toward mentioned problems are presented. The development trend of micro machine tool for micro-cutting is forecasted.

Pattern recognition based surface roughness prediction in turning hybrid metal matrix composite using random forest algorithm

PurposeThe purpose of this paper is surface roughness prediction using pattern recognition for the aluminium hybrid metal matrix composite (HMMC).Design/methodology/approachHybrid composites were manufactured using liquid metallurgy technique. The cast HMMC was machined using an industrial CNC turning centre and the machining vibration signals were acquired using an accelerometer. The acquired signals were processed and used to build a machine learning model for predicting surface finish based on the tool signature.FindingsThe authors established a technique for predicting and monitoring the surface quality during machining using a low cost accelerometer. It is capable of being integrated with the machine controller for online warning of deviations in surface roughness. The system is reconfigurable for any machining condition with a very short training period. The use of this model facilitates online surface roughness monitoring, avoiding the need for costly measuring equipment.Originality/valueThe model developed is innovative and not reported widely to the best of the authors' knowledge. The use of accelerometer‐based surface roughness prediction and control is an innovative approach for automation of machining process monitoring. These can be integrated into any existing machining centre as a standalone system or can be integrated into the CNC controller like Fanuc or Siemens.

Characterization of Cutting Force Induced Surface Shape Variation in Face Milling Using High-Definition Metrology1

High-definition metrology (HDM) systems with fine lateral resolution are capable of capturing the surface shape on a machined part that is beyond the capability of measurement systems employed in manufacturing plants today. Such surface shapes can precisely reflect the impact of cutting processes on surface quality. Understanding the cutting processes and the resultant surface shape is vital to high-precision machining process monitoring and control. This paper presents modeling and experiments of a face milling process to extract surface patterns from measured HDM data and correlate these patterns with cutting force variation. A relationship is established between the instantaneous cutting forces and the observed dominant surface patterns along the feed and circumferential directions for face milling. Potential applications of this relationship in process monitoring, diagnosis, and control are also discussed for face milling. Finally a systematic methodology for characterizing cutting force induced surface variations for a generic machining process is presented by integrating cutting force modeling and HDM measurements.

Position-oriented process monitoring in freeform abrasive machining

Process monitoring is necessary for the identification and avoidance of process disturbances that could cause poor surface integrity at selected machining parameters. In this paper, a position-oriented process monitoring strategy is introduced which enables determination of process characteristics for freeform abrasive machining. Through correlation of internal machine data of position and power during machining with laser displacement measurement, position-orientated maps of power and specific energy can be generated to enable an evaluation of the machining efficiency of the abrasive machining process. Measurement chains are described, and experimental results reveal that the measurement system provides a significant insight into the process by identifying regions of high power, depth of cut, engagement and specific energy on freeform parts.