Tool life prediction using Bayesian updating. Part 2: Turning tool life using a Markov Chain Monte Carlo approach

According to the Taylor tool life equation, tool life reduces with increasing cutting speed following a power law. Additional factors can also be added, such as the feed rate, in Taylor-type models. Although these models are posed as deterministic equations, there is inherent uncertainty in the empirical constants and tool life is generally considered a stochastic process. In this work, Bayesian inference is applied to estimate model constants for both milling and turning operations while considering uncertainty.     

CHMM for tool condition monitoring and remaining useful life prediction

To solve the problems of tool condition monitoring and prediction of remaining useful life, a method based on the Continuous Hidden Markov Model (CHMM) is presented. With milling as the research object, cutting force is taken as the monitoring signal, analyzed by wavelet packet theory to reduce noise and extract the energy feature of the signal as a basis for diagnosis. Then, CHMM is used to diagnose tool wear state. Finally, a Gaussian regression model is proposed to predict the milling tool’s remaining useful life after the test sample data are verified to be consistent with the Gaussian distribution based on a reliable identification of the milling tool wear state. The probability models of tool remaining useful life prediction could be established for tools with different initial states. For example, when an unknown state of milling force signal is delivered to the milling tool online diagnostic system, the state and the existing time of this state could be predicted by the established prediction model, and then, the average remaining useful life from the present state to the tool failure state could be obtained by analyzing the transfer time between each state in the CHMM. Compared to the traditional probabilistic model, which requires a large amount of test samples, the experimental cost is effectively reduced by applying the proposed method. The results from the experiment indicate that CHMM for tool condition monitoring has high sensitivity, requires less training samples and time, and produces results quickly. The method using the Gaussian process to accurately predict remaining life has ample potential for application to real situations.     

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.     

Effect of tool stiffness upon tool wear in high spindle speed milling using small ball end mill

Longer tool life can be tentatively achieved at a higher feed rate using a small ball end mill in high spindle speed milling (over several tens of thousands of revolutions per minute), although the mechanism by which tool life is improved has not yet been clarified. In the present paper, the mechanism of tool wear is investigated with respect to the deviation in cutting force and the deflection of a ball end mill with two cutting edges. The vector loci of the cutting forces are shown to correlate strongly with wear on both cutting edges of ball end mills having various tool stiffnesses related to the tool length. The results clarified that tool life can be prolonged by reducing tool stiffness, because the cutting forces are balanced, resulting in even tool wear on both cutting edges as tool stiffness is lowered to almost the breakage limit of the end mill. A ball end mill with an optimal tool length showed significant improvement in tool life in the milling of forging die models.     

Fatigue life estimation of FBGA memory device under vibration

This paper predicts the fatigue life of fine-pitch ball grid array (FBGA) solder joints in memory devices due to harmonic excitation through experiments and finite element analysis. Finite element models of the memory device with simplified solder joints and with detailed solder joints were developed as a global model and a local model, respectively. A global-local modeling technique was used in the finite element simulation to calculate the stress magnitude of solder joints in the memory device under vibration. Stress versus life (S-N) curve was generated for the memory devices under various vibration levels to derive the fatigue constants of solder material. The fatigue life of the memory device was then determined by using the Basquin equation and Miner’s rule. It was experimentally verified that the predicted fatigue life of the memory device under cumulative damage conditions matches the experimental results within reasonable accuracy.     

概率循环应变—寿命曲线及其估计方法

引入了概率循环应变—寿命曲线描述低周疲劳循环应变—寿命试验数据的分散性。应用线性回归方法和极大似然法原理 ,发展了曲线及其置信限的估计方法。方法采用经验证的良好假设分布即对数正态分布模拟疲劳寿命的分散性。基于Coffin Manson律 ,将曲线近似表征均值和均方差循环应变—寿命曲线的形式。任意可靠度下的疲劳分析可方便地根据正态分布进行。与现有独立随机变量观点不同 ,材料常数视为关联随机变量。1Cr18Ni9Ti不锈钢管道焊接头试验数据分析验证了方法的有效性。     

Feasibility of mild hard turning of stainless steel using coated carbide tool

Hard turning has been explored as an alternative to the traditional processing technique used to manufacture parts made of hardened steels. However, advanced cutting tool materials for hard turning applications are relatively expensive. The continuous developments in carbide tool material and its coating technology have offered inexpensive cutting tool alternatives for a mild range of hard turning operations. Commercially available TiAlN-coated carbide tool is utilized in this study to perform hard turning of stainless steel within the mild range (47–48 HRC) at various cutting parameters, i.e., cutting speed and feed. Empirical models to measure its performance by quantifying the effect of the cutting parameters to the tool’s service lifetime and the machined workpiece’s surface roughness are developed. The coated carbide tool performed hard turning with fair tool life and fine surface finish, especially at low cutting parameters as shown by the models’ solutions for the optimized input selection.     

Prediction of tool wear using regression and ANN models in end-milling operation

Tool wear prediction plays an important role in industry for higher productivity and product quality. Flank wear of cutting tools is often selected as the tool life criterion as it determines the diametric accuracy of machining, its stability and reliability. This paper focuses on two different models, namely, regression mathematical and artificial neural network (ANN) models for predicting tool wear. In the present work, flank wear is taken as the response (output) variable measured during milling, while cutting speed, feed and depth of cut are taken as input parameters. The Design of Experiments (DOE) technique is developed for three factors at five levels to conduct experiments. Experiments have been conducted for measuring tool wear based on the DOE technique in a universal milling machine on AISI 1020 steel using a carbide cutter. The experimental values are used in Six Sigma software for finding the coefficients to develop the regression model. The experimentally measured values are also used to train the feed forward back propagation artificial neural network (ANN) for prediction of tool wear. Predicted values of response by both models, i.e. regression and ANN are compared with the experimental values. The predictive neural network model was found to be capable of better predictions of tool flank wear within the trained range.     

Model of tool life based on fuzzy sets

The influence of the statistical distribution of technological factors on the life of a cutting tool is investigated. A mathematical model is developed for determining the mathematical expectation and distribution of the tool life as a function of the distribution of the technological factors.     

An economic and reliable tool life estimation procedure for turning

The conventional methods of tool life estimation take a long time and consume a lot of work piece material. In this paper, a quicker method for the estimation of tool life is proposed, which requires less consumption of work piece material and tools. In this method the tool life is estimated by fitting a best-fit line on the data falling in the steady wear zone and finding the time till tool failure by extrapolation. Neural networks are used to predict lower, upper and most likely estimates of the tool life. Comparison between neural networks and multiple regression shows the superiority of the former. The paper also proposes a methodology for continuous monitoring of tool use in the shop floor and updating/obtaining the tool life estimates based on the shop floor feed back.     

Fretting fatigue life prediction using the extended finite element method

In this work, an efficient procedure to predict fatigue lives in fretting fatigue problems is presented. This is accomplished by means of a combined initiation-propagation approach in which the extended finite element method (X-FEM) is used. The procedure is verified by modelling several fretting fatigue tests available in the literature. The application of the X-FEM enables to numerically evaluate the stress intensity factors (SIFs) for cracks of different lengths emanating at the end of the contact zone and to estimate the propagation life corresponding to each of the tests. This propagation life is combined with the initiation life calculated using a multiaxial fatigue criterion (Fatemi-Socie). The predicted lives are then compared with the reported experimental lives, showing that the consideration of the crack-contact interaction through the numerical models tends to improve the life estimation when compared with a fully analytical approach. The procedure can be applied to more general fretting problems for which analytical solutions are not available.     

Survival life analysis applied to tool life estimation with variable cutting conditions when machining titanium metal matrix composites (Ti-MMCs)

A survival analysis methodology is employed through a novel approach to model the progressive states of tool wear under different cutting conditions during machining of titanium metal matrix composites (Ti-MMCs). A proportional hazards model (PHM) with a Weilbull baseline is developed to estimate the reliability and hazard functions of the cutting inserts. A proper criterion is assigned to each state of tool wear and used to calculate the tool life at the end of each state. Accounting for the machining time and different stages of tool wear, in addition to the effect of cutting parameters, an accurate model is proposed. Investigating the results obtained for different states, it was shown that the evolution of the time-dependent phenomena, such as different tool wear mechanisms, throughout the whole machining process were also reflected in the model. The accuracy and reliability of the predicted tool lives were experimentally validated. The results showed that the model gives very good estimates of tool life and the critical points at which changes of states take place.     

航空难加工材料切削刀具磨损与剩余寿命预测研究进展

航空制造领域因轻量化、强度等特殊的应用需求,大量使用钛合金、镍基合金等难切削材料,刀具磨损速率快,刀具过度磨损会影响产品质量,在保证产品质量的前提下,为了充分发挥刀具使用价值,亟需监测刀具磨损状态和预测刀具剩余寿命。针对刀具剩余寿命预测的定义、分类和主要预测方法进行了阐述,同时,刀具磨损监测作为刀具寿命预测的基础和先决条件,简述了其重要的流程和常见模型。刀具剩余寿命预测模型主要包括基于物理模型的预测、基于数据驱动的预测和混合预测三大类,对不同预测方法的优缺点和适用场景进行总结,并讨论了刀具剩余寿命预测的未来研究方向。     

Impact evaluation of rolling contact fatigue life models

Since the accurate prediction of fatigue life has a significant value, many researchers have attempted to develop a reliable fatigue life model. Recently, rolling contact fatigue life models incorporating machining impact were developed. These models have contributed to a significant improvement in prediction accuracy as compared with earlier models, thus representing a major step forward in the modeling effort. This paper compares the prediction accuracy of these models with that of the prediction method in International Standards. When α is set to 0.25, the observed improvement of prediction accuracy as measured by variance of prediction errors due to these models over that due to prediction method in International Standards is statistically significant. Impact analyses of such improvement are conducted to illustrate its value. It is further noted that while difference was observed between the variance of prediction errors due to the crack initiation life model based on a dislocation model and that due to the crack initiation life model based on a local stress-life curve, the observed difference is not statistically significant.

     

A weighted hidden Markov model approach for continuous-state tool wear monitoring and tool life prediction

Tool wear is one of the important indicators to reflect the health status of a machining system. In order to obtain tool’s wear status, tool condition monitoring (TCM) utilizes advanced sensor techniques, hoping to find out the wear status through those sensor signals. In this paper, a novel weighted hidden Markov model (HMM)-based approach is proposed for tool wear monitoring and tool life prediction, using the signals provided by TCM techniques. To describe the dynamic nature of wear evolution, a weighted HMM is first developed, which takes wear rate as the hidden state and formulates multiple HMMs in a weighted manner to include sufficient historical information. Explicit formulas to estimate the model parameters are also provided. Then, a particular probabilistic approach using the weighted HMM is proposed to estimate tool wear and predict tool’s remaining useful life during tool operation. The proposed weighted HMM-based approach is tested on a real dataset of a high-speed CNC milling machine cutters. The experimental results show that this approach is effective in estimating tool wear and predicting tool life, and it outperforms the conventional HMM approach.     

Application of particle swarm optimisation in artificial neural network for the prediction of tool life

In an advanced manufacturing system, accurate assessment of tool life estimation is very essential for optimising the cutting performance in turning operations. Estimation of tool life generally requires considerable time and material and hence it is a relatively expensive procedure. In this present work, back-propagation feed forward artificial neural network (ANN) has been used for tool life prediction. Speed, feed, depth of cut and flank wear were taken as input parameters and tool life as an output parameter. Twenty-five patterns were used for training the network. Recently there have been significant research efforts to apply evolutionary computational techniques for determining the network weights. Hence an evolutionary technique named particle swarm optimisation has been used instead of a back-propagation algorithm and it is proven that the experimental results matched well with the values predicted by both artificial neural network with back-propagation and the proposed method. It is found that the computational time is greatly reduced by this method .     

Metal forming oils: How their application increases tool life

Metal forming oils can contribute considerably to increased tool life, thus reducing production costs. Low tool wear, good surface finish of the workpiece, no clinging of waste material to the tool in punching, no bulging of the side wall in deep drawing, easy removal of the oil, no workpiece and tool assembly corrosion, and minimum health risks are the most important performance characteristics to be met by a metal forming oil. Low tool wear and thus a long tool life are best realized if elastohydrodynamic (ehd) lubrication conditions are present in almost all stages of the process. Several examples of metal forming operations are given in which a long tool life has been achieved     

Application of particle swarm optimisation in artificial neural network for the prediction of tool life

In an advanced manufacturing system, accurate assessment of tool life estimation is very essential for optimising the cutting performance in turning operation. Estimation of tool life generally requires considerable time and material and hence it is a relatively expensive procedure. In this present work, back-propagation feed forward artificial neural network (ANN) has been used for tool life prediction. Speed, feed, depth of cut and flank wear were taken as input parameters and tool life as an output parameter. Twenty-five patterns were used for training the network. Recently there have been significant research efforts to apply evolutionary computational techniques for determining the network weights. Hence an evolutionary technique named particle swarm optimisation has been used instead of the back-propagation algorithm and it is proved that the experimental results matched well with the values predicted by both artificial neural network with back-propagation and the proposed method. It is found that the computational time is greatly reduced by this method.     

A predicted modelling of tool life of high-speed milling for SKD61 tool steel

This paper presents an abductive network for predicting tool life in high- speed milling (HSM) operations. The abductive network is composed of a number of functional nodes. These functional nodes are well organised to form an optimal network architecture by using a predicted squared error criterion. Once the cutting speed, feed per tooth, and axial depth of cut are given, tool life can be predicted based on the developed network. Experimental results have shown that the abductive network can be used to predict HSM end mill life under varying cutting conditions and the prediction error of HSM tool life is less than 10%.     

Effect of cutting parameters on the degree of work hardening and tool life during high-speed machining of Inconel 718

The use of superalloy Inconel 718 is increasing in most of the sophisticated applications like aircraft engines, industrial gas turbines, rocket engines, space vehicles, submarines, etc. Hence, in-depth understanding of this material helps to determine the ability of this material to withstand severe conditions of stress, temperature, corrosion, and controls its longevity and reliability. In the present work, an attempt has been made to study the relationship of degree of work hardening and tool life as a function of cutting parameters like cutting speed, feed, depth of cut, untreated tungsten carbide and postcryogenic-treated tool. Work hardening and tool life are the major factors which need to be controlled/improved to enhance the machinability characteristics of superalloy Inconel 718. A significant performance in tool life was observed due to cryogenic treatment given to tungsten carbide tool. Moreover, it was observed that optimized cutting parameters not only minimized/controlled work hardening characteristics but also improved tool life while high-speed machining of Inconel 718.