Tool Condition Monitoring Methods Research Articles

Online tool condition monitoring for ultrasonic metal welding via sensor fusion and machine learning

In ultrasonic metal welding (UMW), tool wear significantly affects the weld quality and tool maintenance constitutes a substantial part of production cost. Thus, tool condition monitoring (TCM) is crucial for UMW. Despite extensive literature focusing on TCM for other manufacturing processes, limited studies are available on TCM for UMW. Existing TCM methods for UMW require offline high-resolution measurement of tool surface profiles, which leads to undesirable production downtime and delayed decision-making. This paper proposes a completely online TCM system for UMW using sensor fusion and machine learning (ML) techniques. A data acquisition (DAQ) system is designed and implemented to obtain in-situ sensing signals during welding processes. A large feature pool is then extracted from the sensing signals. A subset of features are selected and subsequently used by ML-based classification models. A variety of classification models are trained, validated, and tested using experimental data. The best-performing classification models can achieve close to 100% classification accuracy for both training and test datasets. The proposed TCM system not only provides real-time TCM for UMW but also can support optimal decision-making in tool maintenance. The TCM system can be extended to predict remaining useful life (RUL) of tools and integrated with a controller to adjust welding parameters accordingly.

Condition monitoring of carbide and non-carbide coated tool insert using decision tree and random tree – A statistical learning

Traditional trial and error methods of tool condition monitoring are time-consuming and not reliable, hence an efficient method for tool condition monitoring is necessary. The following work deals to establish a simple and efficient automatic method to detect and monitor the tool condition. The work includes both the carbide and non-carbide tools used for machining a cylindrical mild steel component. The turning process is carried out under various conditions. Different combinations of speed, feed, and depth of cut are used for turning. A tri-axial accelerometer was used for capturing the vibration signals, which are further used for condition monitoring purposes. A statistical framework has been applied for monitoring the tool insert health condition. The raw vibration signals were processed for extracting the statistical properties. The extracted features were used for applying machine learning. Two machine learning classifiers namely the decision tree and the random tree were used for the classification process. The results were discussed and compared in this study.

Sample Augmentation for Intelligent Milling Tool Wear Condition Monitoring Using Numerical Simulation and Generative Adversarial Network

Recent advances in artificial intelligence (AI) technology have led to increasing interest in the development of AI-based tool condition monitoring (TCM) methods. However, achieving good performance using these methods relies heavily on large training samples, which are both expensive and difficult to obtain in practical TCM applications. This article addresses this issue by employing a much smaller training sample composed of a non-exhaustive sampling of experimentally measured cutting force signals in conjunction with a novel data augmentation method that combines numerical simulation with a generative adversarial network (GAN). First, cutting force signal samples not present in the experimental dataset are obtained by numerical simulation using a finite element method simulated based on the Johnson–Cook model. Second, the GAN is employed to synthesize additional samples that are similar to both the simulated samples and the experimentally measured samples. The synthesized samples are combined with the measured and simulated samples to produce an appropriately large dataset necessary for the effective training of an AI classifier. The proposed sample augmentation method is applied in milling TCM experiments, and the classification accuracies obtained with several AI classifiers trained with the augmented dataset were all close to or equal to 100%.

Tool Wear Condition Monitoring by Combining Variational Mode Decomposition and Ensemble Learning.

Most online tool condition monitoring (TCM) methods easily cause machining interference. To solve this problem, we propose a method based on the analysis of the spindle motor current signal of a machine tool. Firstly, cutting experiments under multi-conditions were carried out at a Fanuc vertical machining center, using the Fanuc Servo Guide software to obtain the spindle motor current data of the built-in current sensor of the machine tool, which can not only apply to the actual processing conditions but, also, save costs. Secondly, we propose the variational mode decomposition (VMD) algorithm for feature extraction, which can describe the tool conditions under different cutting conditions due to its excellent performance in processing the nonstationary current signal. In contrast with the popular wavelet packet decomposition (WPD) method, the VMD method was verified as a more effective signal-processing technique according to the experimental results. Thirdly, the most indicative features that relate to the tool condition were fed into the ensemble learning (EL) classifier to establish a nonlinear mapping relationship between the features and the tool wear level. Compared with existing TCM methods based on current sensor signals, the operation process and experimental results show that using the proposed method for the monitoring signal acquisition is suitable for the actual processing conditions, and the established tool wear prediction model has better performance in both accuracy and robustness due to its good generalization capability.

A Novel Tool (Single-Flute) Condition Monitoring Method for End Milling Process Based on Intelligent Processing of Milling Force Data by Machine Learning Algorithms

Tool condition monitoring is deemed as an essential technology of the intelligent manufacturing. Tool wear which directly affects the tool life makes a negative influence on the quality and dimensional accuracy of the machined surface, even leads to tool breakage, machine downtime, and other severe problems. Therefore, an available tool condition monitoring system is essential for the machining process to guarantee the processing quality and improve the machining efficiency. This paper proposes a new tool condition monitoring method based on the general judgment of cutting force. Milling force from a single-flute is predicted by deducing a theoretical formula based on un-deformed chip thickness. Based on the formula, cutting force samples used for machine learning paradigms are generated through time domain translation and Gaussian distribution. Nonlinear manifold learning methods are applied in the visualization of high dimensional data. Principal component analysis as a practical feature extraction method is used to reduce the large dimensionality of the sample set. The performance of respectively linear kernel, polynomial kernel, radial basis function and sigmoid kernel are self-compared to estimate the classification results via support vector machine. Experiments are carried out on an annealed Ti–6Al–4V alloy to measure the feasibility of this method.

Tool wear condition monitoring based on a two-layer angle kernel extreme learning machine using sound sensor for milling process

Tool condition monitoring (TCM) in numerical control machines plays an essential role in ensuring high manufacturing quality. The TCM process is conducted according to the data obtained from one or more of a variety of sensors, among which acoustic sensors offer numerous practical advantages. However, acoustic sensor data suffer from strong noise, which can severely limit the accuracy of predictions regarding tool condition. The present work addresses this issue by proposing a novel TCM method that employs only a few appropriate feature parameters of acoustic sensor signals in conjunction with a two-layer angle kernel extreme learning machine. The two-layer network structure is applied to enhance the learning of features associated with complex nonlinear data, and two angle kernel functions without hyperparameters are employed to avoid the complications associated with the use of preset hyperparameters in conventional kernel functions. The proposed TCM method is experimentally demonstrated to achieve superior TCM performance relative to other state-of-the-art methods based on sound sensor data.

Research on automatic monitoring method of face milling cutter wear based on dynamic image sequence

Tool wear is an important factor affecting the quality of finished products, productivity, and the normal operation of machine tools, so tool condition monitoring (TCM) has become a research hotpot in the field of intelligent manufacturing. Compared with traditional monitoring methods, vision-based tool condition monitoring methods are more accurate and intuitive. However, the existing visual monitoring method requires manual adjustment of the tool position, and the degree of automation needs to be improved. Therefore, this paper proposes automatic face milling cutter condition monitoring method based on dynamic image sequence. We first acquire the dynamic image sequence of face milling cutter with the spindle rotating, then forward the dynamic image sequence to the image processing module to extract target area. And the images after image processing are propagated to the image selection module to obtain the image to be measured. Finally, forward the selected image to wear value measurement module to obtain the wear value. The presented automatic face milling cutter condition monitoring method is verified on a five-axis milling center. Compared with the direct measurement results of industrial digital microscope, the measurement error of the proposed method is within 4%, which is a reliable and effective online monitoring method for milling cutter wear.

Tool condition monitoring in milling process using multifractal detrended fluctuation analysis and support vector machine

Tool wear will lead to the reduction of surface quality and machining accuracy. Therefore, tool condition monitoring is vital to the improvement of industrial production efficiency and quality. In this paper, first of all, the internal mechanism of the milling process and the performance characteristics of the milling signals were analyzed. It is found that there are different trends between the large and small fluctuations of milling signals in the process of tool wear increasing. This property can be characterized by various parameters of multifractal spectrum to establish the relationship between tool wear and multifractal parameters. By analyzing the changes of multifractal spectrum parameters, the tool wear monitoring can be realized. Then, the multifractal detrended fluctuation analysis (MFDFA) method is used to calculate the mean square error, generalized Hurst exponent, and multifractal spectrum parameters, which are the eigenvectors, and establish its relationship with tool wear. Finally, the tool condition diagnosis is conducted by a support vector machine (SVM). The results show that the tool condition monitoring method of MFDFA combined with SVM is proved to be effective and the multifractal parameters of MFDFA are very sensitive to tool wear.

A tool condition monitoring method based on two-layer angle kernel extreme learning machine and binary differential evolution for milling

Tool condition monitoring (TCM) in numerical control (NC) machines plays a significant role in ensuring manufacturing quality. TCM based on multiple sensors can provide more information related to tool condition and is a topic of great interest. A novel TCM method for milling processes based on a two-layer angle kernel extreme learning machine (TAKELM) and binary differential evolution (BDE) was proposed. The TAKELM could enhance inherent feature extraction in microarray data and don’t require manual presetting of the kernel function or optimization of its hyperparameter. The BDE algorithm was applied to search the optimal feature parameter combinations in multi-domain alternative feature parameters to achieve the fewest feature parameters satisfying a very small prediction error. Finally, the performance of the proposed TAKELM- BDE method was verified on two milling TCM experiments (one open-access benchmark TCM data set and one TCM experiment) by comparison to the KELM- BDE, PCC- TAKELM, and mRMR- TAKELM methods. The results indicated that the mean absolute error, mean absolute percentage error, and root mean square error of the proposed method are less than 0.005 on two milling TCM experiments, and outperforms the other three methods.

Tool Wear Condition Monitoring in Milling Process Based on Current Sensors

Accurate tool condition monitoring (TCM) is essential for the development of fully automated milling processes. This is typically accomplished using indirect TCM methods that synthesize the information collected from one or more sensors to estimate tool condition based on machine learning approaches. Among the many sensor types available for conducting TCM, motor current sensors offer numerous advantages, in that they are inexpensive, easily installed, and have no effect on the milling process. Accordingly, this study proposes a new TCM method employing a few appropriate current sensor signal features based on the time, frequency, and time - frequency domains of the signals and an advanced monitoring model based on an improved kernel extreme learning machine (KELM). The selected multi-domain features are strongly correlated with tool wear condition and overcome the loss of useful information related to tool condition when employing a single domain. The improved KELM employs a two-layer network structure and an angle kernel function that includes no hyperparameter, which overcome the drawbacks of KELM in terms of the difficulty of learning the features of complex nonlinear data and avoiding the need for preselecting the kernel function and its hyperparameter. The performance of the proposed method is verified by its application to the benchmark NASA milling dataset and separate TCM experiments in comparison with existing TCM methods. The results indicate that the proposed TCM method achieves excellent monitoring performance using only a few key signal features of current sensors.

Application of spindle power signals in tool condition monitoring based on HHT algorithm

In the present study, we present the application of spindle power signals in tool condition monitoring (TCM) under different cutting conditions based on the Hilbert-Huang transform (HHT) algorithm. We extracted two features from the original collected data using the HHT algorithm to detect the tool wear and conducted six sets of cutting experiments to verify the feasibility of this tool condition monitoring method. The results show that these features are highly correlated with the wear state of cutting tools, regardless of the cutting parameters, workpiece materials, and machining methods. The calculated correlation coefficients between the extracted features and the actual tool wear reach 0.79–0.98. This demonstrates that the HHT algorithm is suitable for extracting features from the spindle power signals to construct the online tool condition monitoring system.

A novel method for tool condition monitoring based on long short-term memory and hidden Markov model hybrid framework in high-speed milling Ti-6Al-4V

High-speed milling of Ti-6Al-4V produces high temperature in the cutting zone resulting in the rapid notching and severe flank wear, which tend to cause the deterioration of the machined surface and increased manufacturing costs. To track the performance degradation of the cutting tool, more and more studies have been conducted to develop effective tool condition monitoring (TCM) systems, hoping to obtain the tool wear status based on monitoring signals. In recent years, machine learning and deep learning methods have developed rapidly and are becoming increasingly popular, and the related research results inspired this article. In this paper, a novel method based on long short-term memory network and hidden Markov model (LSTM-HMM) is presented to track the flank wear and predict the remaining useful life (RUL). The model is divided into three parts: training phase, diagnostic phase, and prognostic phase. During the training phase, both the observed monitoring signal and extracted features that characterize tool wear status are used as the inputs to the training algorithms to estimate the parameters of the global model and local models. Then proceed to the diagnostic phase, the established global model is utilized to identify the current tool wear stage deciding which model to select from the local model base. Finally, the mean wear amount, the remaining useful life, and the associated confidence interval are calculated in the prognostic phase. The proposed hybrid model and several other published methods are applied to the three datasets acquired from milling experiments. The experimental and analytical results indicate that the integrated framework is applicable to track the tool wear and predict its RUL during the high-speed milling Ti-6Al-4V.

Modulation Assisted Drilling of Inconel-718 Ni-based Superalloy

Abstract Modulation assisted drilling (MAD) of a Ni-based superalloy; Inconel-718 has been reported in this paper. Drilling is one of the most critical machining processes during the manufacturing of various products. During MAD, a controlled low-frequency vibration was superimposed in the feed direction, and its effect on tool flank wear was investigated. For comparison, conventional machining was also done under similar machining parameters. A tribo-MAM tool holder (Patented tool-M4 Sciences) was retrofitted on a computer controlled lathe for providing the vibration frequency to the drill bit. Flank wear was directly measured by the tool condition monitoring method. The maximum flank wear was measured with a tool maker’s microscope. The used carbide drills were inspected under the scanning electron microscope (SEM) to establish the probable wear mechanism. The results of the study showed that the tool flank wear (VBmax) got reduced significantly during MAD in comparison with that during the conventional drilling (CD). Among the investigated modulation conditions, a modulation frequency of 183 Hz with an amplitude of 0.0018 mm resulted in the least tool wear. However, at a higher feed rate in CD, a catastrophic failure of drill occurred due to cutting-edge fracture.

Research on fault early warning and the diagnosis of machine tools based on energy fault tree analysis

At present, the status of the monitoring of computerized numerical control machine tools is still in the initial stage, and there is no uniform standard for fault early warning and diagnosis of machine tools. A standardized and comprehensive machine tool condition monitoring method based on energy fault tree analysis is proposed. The power, current and temperature inside the numerical control system and the signal of the external vibration sensor are collectively referred to as energy. A new energy fault tree is used to identify the location of the fault. At the same time, the client of computerized numerical control system internal information collection and analysis is developed. The characteristic frequency of each key component is analysed by theoretical mechanics, and an overall monitoring method is proposed to improve the monitoring efficiency. Finally, the reliability of the monitoring method is verified by experiments. The surface roughness of the parts is reduced by approximately 50%, which lays the foundation for the application of the engineering.

Optimisation of tool replacement time in the machining process based on tool condition monitoring using the stochastic approach

ABSTRACTThe efficient cutting tool replacement policy can decrease the machining costs and increase the productivity. This study focuses on the stochastic approach for tool life modelling in the milling process. In this study by considering the costs of tool condition monitoring methods a hybrid policy was developed based on the reliability function for optimising the tool replacement time. The proposed policy covers seven functional modes assuming discrete and continuous modes. The policy in this paper can provide the better solutions for determining the discrete and continuous time interval of the tool condition monitoring and also tool replacement time in the milling process. The proposed policy is defined in the non-convex space, thus this policy is optimised by a particle swarm optimisation (PSO) algorithm. The mentioned policy was applied in an experimental process using a CNC milling machine to assess its performance. Finally, the effect of different costs is demonstrated using the sensitivity analysis and the outcome of the optimised policies under various conditions are compared together. The results show that the proposed policy can optimise the tool replacement time due to its flexibility in covering the different functional modes efficiently.

A Multisensor Fusion Method for Tool Condition Monitoring in Milling

Tool fault diagnosis in numerical control (NC) machines plays a significant role in ensuring manufacturing quality. Tool condition monitoring (TCM) based on multisensors can provide more information related to tool condition, but it can also increase the risk that effective information is overwhelmed by redundant information. Thus, the method of obtaining the most effective feature information from multisensor signals is currently a hot topic. However, most of the current feature selection methods take into account the correlation between the feature parameters and the tool state and do not analyze the influence of feature parameters on prediction accuracy. In this paper, a multisensor global feature extraction method for TCM in the milling process is researched. Several statistical parameters in the time, frequency, and time–frequency (Wavelet packet transform) domains of multiple sensors are selected as an alternative parameter set. The monitoring model is executed by a Kernel-based extreme learning Machine (KELM), and a modified genetic algorithm (GA) is applied in order to search the optimal parameter combinations in a two-objective optimization model to achieve the highest prediction precision. The experimental results show that the proposed method outperforms the Pearson’s correlation coefficient (PCC) based, minimal redundancy and maximal relevance (mRMR) based, and Principal component analysis (PCA)-based feature selection methods.

Review of tool condition monitoring methods in milling processes

Accurate tool condition monitoring (TCM) is essential for the development of fully automated milling processes. However, while considerable research has been conducted in industrial and academic settings, the complexity of milling processes continues to complicate the implementation of TCM. This paper presents a review of the state-of-the-art methods employed for conducting TCM in milling processes. The review includes three key components: (1) sensors, (2) feature extraction, and (3) monitoring models for the categorization of cutting tool states in the decision-making process. In addition, the primary strengths and weaknesses of current practices are presented for these three components. Finally, this paper concludes with a list of recommendations for future research.

Tool wear monitoring and replacement for tubesheet drilling

Indirect Tool Condition Monitoring (TCM) methods have shown significant potential to automatically detect worn tools without intervention in the machining process. This paper presents the development of a non-intrusive and online TCM system for large diameter indexable insert drills. The TCM system developed used two cutting force-related signals of a horizontal boring machine, namely the spindle motor current and the axial feed motor current, and features extracted from these signals were taken as the inputs to a series of models to predict the tool wear state and the hole diameter. A tool replacement strategy based on applying limits to the predicted hole diameter was also developed. Adjusting these limits allows the strategy to be tuned for either hole accuracy or tool life depending on the requirements of a specific application. Experiments of drilling of 39.0-mm-diameter holes in 2205 Duplex stainless steel in an industrial field were designed and performed with the results to illustrate the effectiveness of developed TCM and strategies. Specifically, the TCM system ensured that none of over tolerance holes would have been drilled, which is critically important since any over-tolerance hole can result in the failure of an entire finished product of tubesheet; the replacement strategy for tool life resulted in a 44 % increase in tool life and a non-trivial reduction in machine down time due to fewer tool changes while holding a hole diameter tolerance of ±0.1 mm.

Tool Condition Monitoring Using the Electromotive Force from the Chip-Tool Thermocouple

Tool condition monitoring systems are extensively study. However, the machining processes are non-stationary and comprise many details that interfere in its monitoring. Aiming to develop a simple, low cost and efficient tool condition monitoring system, this study analyzed the electromotive force (EMF) from a chip-tool thermocouple in turning tests with AISI 1045. Since EMF comprises time and frequency variations related to machining conditions a Wavelet Packet Transform extracted the signals features from EMF. These signals features fed inputs of a neural network that aimed to evaluate the cutting tool maximum flank wear. The maximum error of the neural network was 1.88% for tested signals. Moreover, EMF showed changes that allow the detection of cutting tool breakage. Therefore, the chip-tool thermocouple may be a promising method for tool condition monitoring. This is the first report of electromotive force analysis in time-frequency domain aiming to quantify the wear of the cutting tool and evaluate its condition.

On-machine tool prediction of flank wear from machined surface images using texture analyses and support vector regression

In this paper, a method for on-machine tool condition monitoring by processing the turned surface images has been proposed. Progressive monitoring of cutting tool condition is inevitable to maintain product quality. Thus, image texture analyses using gray level co-occurrence matrix, Voronoi tessellation and discrete wavelet transform based methods have been applied on turned surface images for extracting eight useful features to describe progressive tool flank wear. Prediction of cutting tool flank wear has also been performed using these eight features as predictors by utilizing linear support vector machine based regression technique with a maximum 4.9% prediction error.