Tool Life Prediction Research Articles

Service life prediction models for exterior stone cladding

The use of service life prediction tools can allow a more rational management of the maintenance of a building and its components by supporting reduced life cycle costs of constructed assets. A study was conducted to evaluate the degradation of facades of buildings located in Lisbon, Portugal, and in which visual in-situ inspections of 140 stone cladding facades were completed. Based on this information, different models were proposed to describe the degradation of masonry stone-clad buildings and a novel approach to predicting service life was developed. A description is provided for the use of artificial neural networks to predict the service life of stone cladding of the type that is directly adhered to a building substrate. With the use of artificial neural networks an analytical expression was established to estimate the degradation of this type of cladding system. The viability of this method was then compared with other methods used for service life prediction including multiple linear regression and a graphical method. Some of the statistical parameters used to assess the performance of the various methods are described. Of these three methods, those variables that best explain the degradation of stone cladding, are identified: cladding age, distance from the sea, type of finish, and size of the stone plates. L'utilisation d'outils de prévision de la durée de vie peut permettre une gestion plus rationnelle de l'entretien d'un bâtiment et de ses composants en contribuant à des coûts de cycle de vie réduits des biens immobiliers construits. Une étude a été menée afin d'évaluer les dégradations des façades de bâtiments situés à Lisbonne, au Portugal, et dans lesquels des contrôles in situ de 140 façades à parement en pierre ont été réalisés. Sur la base de ces informations, il a été proposé différents modèles pour décrire la dégradation des bâtiments en maçonnerie à revêtement de pierre et il a été développé une approche nouvelle de prévision du cycle de vie. Il est fourni un descriptif pour utiliser des réseaux neuraux artificiels afin de prévoir la durée de vie des revêtements de pierre qui se collent directement sur un substrat de construction. Grâce à l'utilisation de réseaux neuraux artificiels, il a été établi une expression analytique afin d'évaluer les dégradations de ce type de système de revêtement. La viabilité de cette méthode a ensuite été comparée aux autres méthodes utilisées pour prévoir la durée de vie, dont la régression linéaire multiple et une méthode graphique. Certains des paramètres statistiques utilisés pour évaluer les performances des différentes méthodes sont décrits. Sur ces trois méthodes, les variables qui expliquent le mieux la dégradation des revêtements de pierre sont identifiées: âge du revêtement, distance de la mer, type de finition, et taille des plaques de revêtement de pierre. Mots clés: réseaux neuraux artificiels bâtiments dégradation durabilité façades entretien prévision de la durée de vie revêtement de pierre

Bearing fault prognosis based on health state probability estimation

In condition-based maintenance (CBM), effective diagnostic and prognostic tools are essential for maintenance engineers to identify imminent fault and predict the remaining useful life before the components finally fail. This enables remedial actions to be taken in advance and reschedule of production if necessary. All machine components are subjected to degradation processes in real environments and they have certain failure characteristics which can be related to the operating conditions. This paper describes a technique for accurate assessment of the remnant life of bearings based on health state probability estimation and historical knowledge embedded in the closed loop diagnostics and prognostics system. The technique uses the Support Vector Machine (SVM) classifier as a tool for estimating health state probability of machine degradation process to provide long term prediction. To validate the feasibility of the proposed model, real life fault historical data from bearings of High Pressure-Liquefied Natural Gas (HP-LNG) pumps were analysed and used to obtain the optimal prediction of remaining useful life (RUL). The results obtained were very encouraging and showed that the proposed prognosis system based on health state probability estimation has the potential to be used as an estimation tool for remnant life prediction in industrial machinery.

An adaptive procedure for tool life prediction in face milling

Accurate prediction of tool life is essential to guarantee surface quality and economics of cutting operations in face milling. This article presents a procedure for tool life prediction through in-process adaptation of tool wear rate based on indirect measures. The procedure effectively accounts for the uncertainty of tool wear progress owing to the complexity of the machining process. First, sensor fusion of spindle motor current AC and DC portions is taken to estimate the actual tool wear through relevance vector machine. Then, a tool life prediction model relating flank wear with cutting time is proposed for tracking the progress of tool wear under certain cutting settings. Further, a recursive least square algorithm is developed to update the parameters of the tool life prediction model by considering the error between the predicted tool wear and the estimated tool wear. Finally, the updated model capturing the uncertainty of tool wear progress is used to predict tool life in face milling. Tool life experiments validate that the adaptive procedure can quickly track the progress of tool wear, and make more accurate prediction of tool life compared with the procedure with constant model parameters.

CBN Tool Life Prediction by Artificial Neural Network

For economic application of CBN tools it is important to know how the tool life changes as a function of the most important technological parameters (cutting speed, feed rate, depth of cut and work piece diameter). These technological parameters have significant effects on the productivity of machining. The extended Taylor equation takes the effects of feed rate, depth of cut into account besides the cutting speed, but this equation is not valid in the whole range of cutting speed. Beyond the publicated tool life equations in the professional articles, the tool life can be determined by Artificial Neural Network (ANN). ANNs have huge flexibility and can handle different number of inputs thus they are suitable for determining tool life as a function of several cutting parameters.

Residual life prediction of repairable systems subject to imperfect preventive maintenance using extended proportional hazards model

Proportional hazards model (PHM) is a representative tool for equipment residual life (RL) prediction and preventive maintenance (PM) scheduling in a condition-based maintenance (CBM) program. However, the use of PHM in a CBM program is currently limited to irreparable systems or repairable systems receiving perfect PM acts. Motivated by the classical lifetime distribution-based imperfect PM model and the PHM, our recent study proposes an extended PHM (EPHM) for PM scheduling of systems subject to imperfect PM acts and variable operational conditions. This study further proposes the use of EPHM (with time-varying covariates) for RL prediction of repairable systems subject to imperfect PM acts. A numerical experiment based on a typical degradation model illustrates the use of the EPHM. The experimental results provide evidence that the EPHM provides reasonably accurate and reliable RL prediction after imperfect PM acts, as opposed to the overestimation by the classical PHM which does not take the effect of imperfect PM acts into account.

Tool Life Prediction for Ceramic Tools in Intermittent Turning of Hardened Steel Based on Damage Evolution Model

Al2O3-based ceramic is one of the most widely used materials for tools employed in hardened steel turning applications due to its high hardness, wear resistance, heat resistance and chemical stability. The objective of this work is to predict the lives of Al2O3-(W, Ti)C ceramic tools in intermittent turning of hardened AISI 1045 steel by means of damage evolution model taking into account the mechanical loading and thermal effect in the cutting process. A damage evolution model analyzing the RVE with uniformly distributed interacting cracks is constructed based on micromechanics. The calculated results of the proposed damage evolution model are compared with the lives of two kinds of Al2O3-(W, Ti)C ceramic tools obtained through experiments. It is found that the proposed model can be used to predict the lives of the ceramic cutting tools in intermittent turning operation.

Life prediction under tension of titanium alloys that develop an oxygenated brittle case during use

A simple model was used to predict the tendency of an (α + β) titanium alloy to crack under tension after exposure to air at temperature. The model was experimentally validated for minimum stress to crack and saturation crack spacing. A diffusion model for oxygen ingress as a function of time, temperature and oxygen partial pressure is outlined; together with the surface cracking model and empirical crack growth models, this enables the framework for a life prediction tool for arbitrary time–temperature–environment history.

Experimental Study on Tool Wear in Cutting Titanium Alloy Ti6Al4V

Tool wear experiments in turning titanium alloy Ti6Al4V in the solution-treated and aged condition were carried out; the influence rules of cutting parameters such as cutting speed, feed rate and cutting depth on the tool life were obtained. Experimental formula of tool wear was regressed based on multi-variable linear regression analysis, it could predict tool life under certain conditions.

Tool Life Prediction in Face Milling Machining of 7075 Al by Using Artificial Neural Networks (ANN) and Taguchi Design of Experiment (DOE)

Tool Life Prediction in Face Milling Machining of 7075 Al by Using Artificial Neural Networks (ANN) and Taguchi Design of Experiment (DOE)

New machine prognostics approach based on health state probability estimation

This paper presents an innovative prognostic model based on health state probability estimation embedded in the closed loop diagnostic and prognostic system to provide comprehensive timely analysis for effective decision making in industrial asset management. To apply an appropriate classifier in health state probability estimation for prediction of machine remnant life, a comparative study on intelligent fault classification for four fault conditions with five different fault progressed data from high pressure cryogenic pumps were conducted. Five different classifiers, such as support vector machines (SVMs), radial basis function neural networks, random forest and linear regression, were used to evaluate their effectiveness for the health state estimation process. Two sets of impeller-rub data were employed for the prediction of pump remnant life based on estimation of discrete health state probability with feature selection technique using the outstanding capability of SVM. The results obtained were very encouraging and showed that the proposed prognosis system has the potential to be used as an estimation tool for machine remnant life prediction in real life industrial applications.

Online Tool Life Prediction in the Machining of Titanium Alloys

In order to reduce the risk of expensive tool failure in the machining of Titanium alloys, the paper presents a tool life prediction approach based on the analysis of cutting forces. Regression analysis was applied to develop the prediction model. Detailed steps of implementation are presented. Prediction logics and criteria are introduced. Cutting tests were carried out to validate the reliability of the proposed method. When compared with empirical methods the proposed approach which is based on the analysis of cutting force measured in the machining process appears far more effective in predicting tool life.

Study on Tool Life by Breakage in Micro Endmilling

To obtain some knowledge to predict the tool life caused by breakage, a series of slotting experiments was performed at various feed rates for high carbon steel by using coated carbide square end mills of 0.5mm in diameter. The main results obtained are as follows, (1) There found a possibility to obtain an optimum feed rate for the maximum slotting distance. (2) Lower feed rate caused chatter but higher feed rate rarely caused chatter. (3) A newly defined cumulative damage seems to be useful in some extent to predict tool life by monitoring cutting force.

Relationship between the Punch–Die Clearance and Shearing Quality of Progressive Shearing Die

This study performed experiments for the progressive shearing die of thin phosphorous bronze sheet to investigate effects of punch materials, punch–die clearance, sheared angle, and number of shearing strokes on burnish band of a sheared surface and burr width of shearing corner. It also analyzed the relationship between height of the burnish band and gravimetric wear rate. The concept of manufacturing tolerance was applied to establish an analytical model of the dimensional tolerance between punch and punch base. It was concluded that after assembling the punch and punch base under the worst conditions, the possible offset of the punch center and the punch–die clearance can be analyzed, and the burr width of the shearing corner can be accessed. Based on the experimental data, four linear predictive models elucidating the characteristics of shearing qualities such as punch–die clearance, punch gravimetric wear rate, and height of the burnish band of a sheared surface can be deduced. Besides, a predictive model of burr width was built by using the least-square method, exponential method, and polynomial regressive method. The result is highly referential to the assessment of sheared products as well as the prediction of tool life.

Preliminary Review of Physically Based Methodologies for Predicting the Strength of Visibly Damaged Composite Laminates

Approaches to detect, assess, monitor and repair damage in critical aircraft components fabricated from composite materials are essential for safe and cost effective operation. In metallic aircraft structures, it is common to leave some fractures in situ for a prescribed period until it is convenient to repair, provided strict inspection and verification processes are in place. Under current military aircraft structural management guidelines, visible damage to critical composite components requires either immediate repair or replacement. Much has been learnt about the behaviour of damaged composite structures, but further investigation is required to develop validated residual strength and life prediction tools. A preliminary review of an early physically based, residual strength prediction method was conducted. The accuracy of this method for use in predicting the strength of composite following a complex damage was tested by comparing the results with compression-after-impact test data for a composite laminate representative of F/A 18 fracture critical structure.

A computational tool for the life prediction of GFRP laminates under irregular complex stress states: Influence of the fatigue failure criterion

A new computational tool for the fatigue life prediction of composite materials, designated CCfatigue, is presented in this paper. This new software has been developed based on a well-established phenomenological fatigue life prediction methodology. It is capable of performing the calculations necessary to estimate the fatigue life of composite materials under complex cyclic stress states of variable amplitude. Although the software allows the benchmarking of several modules encountered in a complete fatigue life prediction methodology, this paper focuses on the presentation of the software framework and the demonstration of its applicability through the evaluation of the effect of the failure criterion on the fatigue life prediction of the examined materials. Different methods were chosen, programmed and added to the CCfatigue library in order to cover most of the existing concepts regarding the problem of multiaxial fatigue modeling and prediction. The theoretical results produced by the developed software, when compared to available experimental data, proved that there is a prominent effect of the selected fatigue failure module on the estimated lifetime of composite materials loaded under irregular loading patterns.

Modelling of tool life, torque and thrust force in drilling: a neuro-fuzzy approach

This paper presents the application of neuro-fuzzy approach for modelling tool life, torque and thrust force in drilling operation for set of given process parameters, namely cutting speed, feed rate and drill diameter. The proposed approach uses a hybrid-learning algorithm i.e., combination of the back-propagation gradient descent method and least squares method, to identify premise and consequent parameters of the first-order Sugeno-fuzzy inference system. The least square method is used to optimize the consequent parameters with the premise parameters fixed. Once the optimal consequent parameters are found, the back-propagation algorithm gradient descent method is used to adjust optimally the premise parameters corresponding to the fuzzy set in the input domain. The predicted tool life, torque and thrust force values obtained from neuro-fuzzy system were compared with the experimental data. This comparison indicates that the proposed approach can produce optimal knowledge base of fuzzy system for predicting tool life, torque and thrust force in drilling operation. (Received in August 2009, accepted in November 2009. This paper was with the author 1 month for 1 revision.)

Tool life modeling for evaluating the effects of cutting speed and reinforcements on the machining of particle reinforced metal matrix composites

The wear of cutting tools in the machining of 2024Al alloy composites reinforced with Al2O3 particles using varying sizes and volume fractions of particles up to 23.3vol% was investigated by a turning process using coated carbide tools K10 and TP30 at different cutting speeds. Machining tests were performed with a plan of experiments based on the Taguchi method. The tool life model was developed in terms of cutting speed, size, and volume fraction of particles by multiple linear regressions. The analysis of variance (ANOVA) was also employed to carry out the effects of these parameters on the cutting tool life. The test results show that the tool life decreases with the increase of cutting speed for both cutting tools K10 and TP30, and the tool life of the K10 tool is significantly longer than that of the TP30 tool. For the tool life, cutting speed is found to be the most effective factor followed by particle content and particle size, respectively. The predicted tool life of cutting tools is found to be in very good agreement with the experimentally observed ones.

Optimised tool life by partial swarm optimisation

The aim of the this paper is to develop the tool life prediction model for P20 tool steel with aid of statistical method and find the optimisation values with partial swarm optimisation (PSO), using coated carbide cutting tool. By using Response Surface Method (RSM), first and second order models were developed with 95% confidence level. The tool life model was developed in terms of cutting speed, feed rate, axial depth and radial depth, using RSM. It was found that the feedrate, cutting speed, axial depth and radial depth played a major role. Tool life increases with a reduction in cutting speed and feedrate. For end-milling of P20 tool steel, the optimum conditions obtained from PSO are: cutting speed of 100 m/s, federate of 0.1 mm/tooth, axial depth of 1.9596 mm and radial depth of 2 mm. Using these parameters, a tool life of 40.52 min was obtained.

Integrated optimisation of surface roughness and tool performance when face milling 416 SS

Tool life is an important factor to be considered during the optimisation of a machining process since cutting parameters can be adjusted to optimise tool changing, reducing cost and time of production. Also the performance of a tool is directly linked to the generated surface roughness and this is important in cases where there are strict surface quality requirements. The prediction of tool life and the resulting surface roughness in milling operations has attracted considerable research efforts. The research reported herein is focused on defining the influence of milling cutting parameters such as cutting speed, feed rate and axial depth of cut, on three major tool performance parameters namely, tool life, material removal and surface roughness. The research is seeking to define methods that will allow the selection of optimal parameters for best tool performance when face milling 416 stainless steel bars. For this study the Taguchi method was applied in a special design of an orthogonal array that allows studying the entire parameter space with only a number of experiments representing savings in cost and time of experiments. The findings were that the cutting speed has the most influence on tool life and surface roughness and very limited influence on material removal. By last tool life can be judged either from tool life or volume of material removal.

Construction of Typical Shell Ring Cutting Database System

According to the process characteristics of shell ring, the cutting parameters have been optimized by adopting appropriate boundary conditions at different machining stages and the optimal cutting parameters have been achieved. The objective function and constraints boundary have been stored into the database, which form basic data structures of database. Cutting parameters have been obtained for new type tools using rule-based reasoning approach. According to the characteristics of large number of changing tool for shell ring, tool life prediction is designed, which is very favorable for large shell ring processing. In addition, on-site processing recorded video will be added to the database by multi-media technology, which achieves the diversity of data structure.