ArticlePublisher preview available

Non-linear Wiener process–based cutting tool remaining useful life prediction considering measurement variability

Authors:
Huibin Sun at Northwestern Polytechnical University
Huibin Sun
Junlin Pan
Junlin Pan
Junlin Pan
  • This person is not on ResearchGate, or hasn't claimed this research yet.
Jiduo Zhang at The University of Manchester
Jiduo Zhang
Dali Cao
Dali Cao
Dali Cao
  • This person is not on ResearchGate, or hasn't claimed this research yet.
Read publisher preview
Download citation
To read the full-text of this research, you can request a copy directly from the authors.

Abstract and Figures

Remaining useful life prediction is essential for cutting tool utilization evaluation and replacement decision-making. However, it is very difficult to build a mechanism model for the time-varying and non-linear cutting tool wear and life decreasing process. Based on big samples, artificial intelligence–based models have weak interpretability and un-quantized uncertainty. In fact, the cutting tool degradation is a stochastic process, and cutting tool remaining useful life is a random variable. Then, a non-linear Wiener-based cutting tool wear and remaining useful life prediction model is proposed for a specific cutting tool. The probability density function of remaining useful life is derived to quantize uncertainty. On the basis of Bayesian model, unknown parameters are estimated and updated by using history data and real-time data, respectively. Measurement variability is also considered to improve the accuracy and reliability. Experimental study verifies the approach’s effectiveness and accuracy. Detailed comparisons validate the approach’s advantages over existing models. By quantizing the uncertainty of cutting tool remaining useful life prediction with confidence intervals, the model is meaningful for cutting tool selection and replacement decision-making.
Figures - available from: The International Journal of Advanced Manufacturing Technology
This content is subject to copyright. Terms and conditions apply.
A preview of this full-text is provided by Springer Nature.
Learn more
Content available from The International Journal of Advanced Manufacturing Technology
This content is subject to copyright. Terms and conditions apply.
ORIGINAL ARTICLE
Non-linear Wiener process–based cutting tool remaining useful life
prediction considering measurement variability
Huibin Sun
1
&Junlin Pan
1
&Jiduo Zhang
1
&Dali Cao
1
Received: 8 May 2019 /Accepted: 30 March 2020
#Springer-Verlag London Ltd., part of Springer Nature 2020
Abstract
Remaining useful life prediction is essential for cutting tool utilization evaluation and replacement decision-making. However, it
is very difficult to build a mechanism model for the time-varying and non-linear cutting tool wear and life decreasing process.
Based on big samples, artificial intelligence–based models have weak interpretability and un-quantized uncertainty. In fact, the
cutting tool degradation is a stochastic process, and cutting tool remaining useful life is a random variable. Then, a non-linear
Wiener-based cutting tool wear and remaining useful life prediction model is proposed for a specific cutting tool. The probability
density function of remaining useful life is derived to quantize uncertainty. On the basis of Bayesian model, unknown parameters
are estimated and updated by using history data and real-time data, respectively. Measurement variability is also considered to
improve the accuracy and reliability. Experimental study verifies the approach’s effectiveness and accuracy. Detailed compari-
sons validate the approach’s advantages over existing models. By quantizing the uncertainty of cutting tool remaining useful life
prediction with confidence intervals, the model is meaningful for cutting tool selection and replacement decision-making.
Keywords Cutting tool .Remaining useful life prediction .Non-linear Wiener process .Measurement variability
1 Introduction
As important components in the machining system, cutting
tools always attract great attentions from both academia and
industry. Cutting tool wear affects machining quality, cost,
efficiency, and environmental cost. During the machining pro-
cess, cutting tool wear develops gradually. Worn cutting tools
may deteriorate surface integrity and produce unqualified
parts. However, too frequent cutting tool replacement in-
creases the machining cost with redundant production suspen-
sion. In order to make most use of every cutting tool accurate-
ly, cutting tool remaining useful life (RUL) prediction is es-
sential. A cutting tool’s RUL is defined as the time from the
current moment to its failure. Based on accurate cutting tool
RUL prediction, cutting tools could be used precisely by
avoiding overuse or underuse. Both poor machining quality
and high machining cost could be avoided to the maximum
extent with high machining efficiency.
However, RUL of a cutting tool is not directly accessible to
sensor measurements. Due to the influence of various factors
in the machining process, including cutting fluid, machine
tools, tool clamping length, and cutting speed, it decreases
non-linearly, individually, and dynamically. Considering the
complex coupling relationship among these factors, it is very
difficult to build a precise mechanism model for the cutting
tool wear process. Then, data-driven models are widely used
to predict cutting tool RUL. For example, machine learning–
based approaches fit the cutting tool wear curve well. But, big
samples are needed to obtain results with weak interpretability
and un-quantized uncertainty. Many researchers argued that
cutting tool wear was a stochastic process, and a cutting tool’s
RUL was a random variable. Then, the stochastic process
theory was believed to be suitable for cutting tool wear model-
ling and RUL prediction. However, the linear cutting tool
wear curve is not accurate enough. In order to improve the
reliability of cutting tool RUL prediction, variability in mea-
sured data should be considered.
Therefore, this work aims to improve the accuracy and
reliability of cutting tool RUL prediction. A non-linear
Wiener-based cutting tool wear and RUL prediction model
*Huibin Sun
sun_huibin@nwpu.edu.cn; 380885779@qq.com
1
Key Laboratory of High Performance Manufacturing for Aero
Engine, Ministry of Industry and Information Technology,
Northwestern Polytechnical University, Xi’an 710072, China
https://doi.org/10.1007/s00170-020-05264-3
The International Journal of Advanced Manufacturing Technology (2020) 107:4493–4502
/ Publishedonline: 28 2020
April
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
... At any moment t k during the life cycle, the stochastic parameter a of the degradation model can be estimated from all the observed data y 1:k = {y 1 , y 2 , · · · , y n } before t k . According to the concept of Bayesian updating [25], the updated value of the stochastic parameter a of the degradation model of the performance of the rail circuit equipment at the moment t k is: ...
... Due to the nonlinearity of the drift coefficients, it is often difficult to derive an exact solution for the lifetime T probability density. In this paper, we adopt the approach in the literature [25], where the PDF of the amount of performance degradation {Y(t), t ⩾ 0} reaching or exceeding the failure threshold w is, under the condition that the stochastic parameters are determined: ...
... Step 5: Online remaining life prediction. The PDF of the residual life of the in-service device at any moment t k can be obtained by substituting the parameter estimation results obtained in steps 3 and 4 into equation (25). ...
Remaining useful life prediction method for jointless track circuits based on multivariate feature fusion and nonlinear Wiener process
Article
Full-text available
Predicting the remaining useful life (RUL) of track circuits is essential to ensure the safe and reliable operation of high-speed railways. In response to the challenges faced by current machine-learning-based RUL prediction methods, which struggle to represent the uncertainty in the probability distribution of RUL predictions, this paper suggests a hybrid-driven method for estimating remaining life. Firstly, the track circuit Health Index is constructed by feature dimensionality reduction and fusion of the original multivariate monitoring data through kernel principal component analysis and Autoencoder; Secondly, the degraded state of the rail circuit is modelled using a nonlinear Wiener degradation model. Finally, the principle of first hitting time is used to derive the probability density function of the anticipated RUL. The efficacy and superiority of the approach presented in this paper are validated by experimental research on the track circuit monitoring dataset. The method enhances forecast accuracy and reduces prediction uncertainty, offering robust technical support for track circuit maintenance decision-making.
View
... Nevertheless, nonlinearity is prevalent in practical applications such as bearings, turbine engines, and fatigue cracks. This issue is addressed in [28,32,47], which develop a nonlinear diffusion model from which a time-space transformation is used to approximate the RUL PDF. ...
... Upon reviewing the extant research literature, it becomes evident that while some studies have attempted to resolve the strong Markovianity with uncertainty problem, each of [13,25,33,36,41] only focuses on the case with linear degeneracy. [6,12,28,32,39,42,46,47] only focuses with the nonlinear degradation case. A comprehensive stochastic degradation model that addresses three-source uncertainty and partially mitigates Markovianity has yet to be proposed by any researcher. ...
A generalized diffusion model for remaining useful life prediction with uncertainty
Article
Full-text available
  • Jan 2025
Forecasting the remaining useful life (RUL) is a crucial aspect of prognostics and health management (PHM), which has garnered significant attention in academic and industrial domains in recent decades. The accurate prediction of RUL relies on the creation of an appropriate degradation model for the system. In this paper, a general representation of diffusion process models with three sources of uncertainty for RUL estimation is constructed. According to time-space transformation, the analytic equations that approximate the RUL probability distribution function (PDF) are inferred. The results demonstrate that the proposed model is more general, covering several existing simplified cases. The parameters of the model are then calculated utilizing an adaptive technique based on the Kalman filter and expectation maximization with Rauch-Tung-Striebel (KF-EM-RTS). KF-EM-RTS can adaptively estimate and update unknown parameters, overcoming the limits of strong Markovian nature of diffusion model. Linear and nonlinear degradation datasets from real working environments are used to validate the proposed model. The experiments indicate that the proposed model can achieve accurate RUL estimation results.
View
... In contrast, artificial intelligence-based models have some advantages in bearing RUL prediction. These methods analyze large amounts of bearing operational data and utilize machine learning (ML), deep learning, and other technologies to learn the complex features and underlying patterns of bearings without relying on detailed physical rules and model assumptions [9]. Therefore, they offer higher flexibility, adaptability to different equipment and environments, higher prediction accuracy, and robustness while reducing the dependence on expert knowledge [10,11]. ...
Bearing remaining life prediction method based on ARAD -ELN and multi-stage wiener process
Article
Full-text available
Stochastic process-based models are extensively utilized in health assessments and Remaining Useful Life (RUL) predictions of bearings. Nevertheless, bearings in actual operation undergo multiple degradation stages, each characterized by a unique trend of degradation. The application of a singular stochastic process for RUL prediction falls short of achieving optimal performance. Consequently, this paper introduces a multi-stage Wiener process-based approach for the prediction of bearings’ RUL. Initially, to address the challenge of imbalanced sample sizes across different degradation stages of bearings, an ensemble learning-based neural network, enhanced by ARIMA Residual Anomaly Detection for identifying bearing degradation stages, is proposed. Subsequently, considering the temporal, unit-to-unit, and nonlinear variabilities of the degradation process at each stage, a Wiener process-based multi-stage degradation model for bearings is developed. A method for parameter estimation and updating, utilizing Kalman filtering and Maximum Likelihood Estimation (K-M), is introduced. Finally, the proposed model is validated using both simulated data and the XJTU-SY bearing dataset. Experimental results from three RUL predictions show that the proposed method outperforms the benchmark model with root mean square error values of 3.61, 2.92 and 7.24, respectively, affirming that the proposed model can precisely classify equipment degradation stages and predict RUL with high accuracy and stability.
View
... He et al. [40] developed a cross-domain adaptation network with an attention mechanism to predict tool wear during the milling of AISI 1045 steel. Sun et al. [41] proposed a non-linear Wiener-based model for predicting cutting tool wear and RUL. Liu et al. [42] developed an Attention-LSTM based method for milling tool wear prediction. ...
A Domain-Aware Federated Learning Study for CNC Tool Wear Estimation
Chapter
Full-text available
  • Aug 2024
This study proposes a cutting tool condition monitoring platform for CNC machines used in metal part manufacturing to estimate tool wear values. The PHM 2010 Dataset, along with operational and situational data from CNC machines and sensors, were analyzed using artificial intelligence algorithms to support total equipment performance with current tool wear values. The innovation lies in developing an artificial intelligence application that incorporates the Federated Learning method with artificial neural networks. This application is among the first to monitor machine cutting tools using Federated Learning. An efficient and accurate predictive tool wear estimation method is presented through the application of Federated Learning with Long-Short Term Memory models. This novel approach holds great potential for industrial applications, optimizing CNC cutting processes and reducing operational costs through enhanced tool wear prediction.
View
View
View
View
View
View
View
Force-based reliability estimation of remaining cutting tool life in titanium milling
Article
Full-text available
  • Feb 2020
  • INT J ADV MANUF TECH
Reliability estimation and prediction of the remaining life of a cutting tool is paramount for ensuring economic and damage proof machining. Monitoring of the status of the cutting tool and especially flank wear disrupts the process and thus avoided. In the present study a novel reliability estimation approach to the cutting tools remaining life based on advanced approximation methods is proposed. An approximation model that is based on response surface and a surrogate model is used for the representation of flank wear as a function of process parameters and cutting forces. For the establishment of this model, the relationship of cutting forces with flank wear is investigated experimentally. The results show that the proposed method is an efficient method for assessing the reliability of the cutting tool based on a minimum number of experimental runs. Experimental verification for the case of titanium milling confirms the findings of the present study.
View
Prediction of PCBN tool life in hard turning process based on the three-dimensional tool wear parameter
Article
Full-text available
  • Jan 2020
  • INT J ADV MANUF TECH
The ideal scenario for the implementation of a machining process is to be able to predict tool performance without the need to conduct practical experiments. However, in an industrial environment, each set of machining conditions is unique, since the machine-tool conditions, machined material, cutting tool, and fixture system can vary. This can lead to differences between the predicted values and practical results. In this context, the aim of this research was to show and discuss a tool performance test methodology and a tool-life prediction model using the three-dimensional (volumetric) wear parameter WRM (volume of material removed from the tool) applied to hard turning with PCBN tools. The wear parameter WRM is measured at the beginning of the tool life (up to 25%) by focus variation microscopy (FVM). The tool wear rate (WRRM) is then calculated based on the ordinary least squares (OLS) method, and the tool life is estimated (TWRM) adopting the volume of material removed from the tool (WRMmax) as the criteria for the end of tool life. The tool-life model developed was capable of predicting the tool life with errors below 4% at the higher values of cutting speed adopted (vc = 150–187.5 m/min), that is, the cutting speeds applied industrially. The methodology adopted and the model developed represent a significant time reduction in the experimental machining tests, streamlining the research and development of the cutting tool grades, as well as the machining process optimization.
View
Tool remaining useful life prediction method based on LSTM under variable working conditions
Article
Full-text available
  • Oct 2019
  • INT J ADV MANUF TECH
Tool remaining useful life prediction is important to guarantee processing quality and efficient continuous production. Tool wear is directly related to the working conditions, showing a complex correlation and timing correlation, which makes it difficult to predict the tool remaining useful life under variable conditions. In this paper, we seek to overcome this challenge. First, we establish the unified representation of the working condition, then extract the wear characteristics from the processing signal. The extracted wear features and corresponding working conditions are combined into an input matrix for predicting tool wear. Based on this, the complex spatio-temporal relationship under variable working conditions is captured. Finally, using the unique advantages of the long short-term memory (LSTM) model to solve complex correlation and memory accumulation effects, the tool remaining useful life prediction model under variable working conditions is established. An experiment illustrates the effectiveness of the proposed method.
View
Tool life prediction based on Gauss importance resampling particle filter
Article
Full-text available
  • Aug 2019
  • INT J ADV MANUF TECH
Effective tool remaining useful life (RUL) prediction can greatly improve the quality of workpiece and reduce the cost of production. An online tool RUL prediction framework is constructed based on Bayesian inference and sensory signals. In this method, Pairs model is adopted to depict the tool degradation process during cutting process and Gaussian importance resampling (GIR) method is proposed to update model parameters iteratively. Therefore, the future tool wear status can be predicted and RUL can be estimated correspondingly. To testify the effectiveness of the proposed method, milling experiment was carried out and relative waveform features are extracted to depict the relationship between sensory signal and tool wear value. The analysis and comparison show that Gaussian importance resampling method can avoid the problem of particle degradation and impoverishment effectively so as to realize accurate RUL prediction.
View
Hybrid data-driven physics-based model fusion framework for tool wear prediction
Article
Full-text available
  • Apr 2019
  • INT J ADV MANUF TECH
An integral part of modern manufacturing process management is to acquire useful information from machining processes to monitor machine and tool condition. Various models have been introduced to detect, classify, and predict tool wear, as a key parameter of the machining process. In more recent developments, sensor-based approaches have been attempted to infer the tool wear condition from real-time processing of the measurement data. Experiments show that the physics-based prediction models can include large uncertainties. Likewise, the measurement-based (or sensor-based) inference techniques are affected by sensor noise and measurement model uncertainties. To manage uncertainties and noise of both methods, a hybrid framework is proposed to fuse together the results of the prediction model and the measurement-based inference data in a stepwise manner. The fusion framework is an extension to the regularized particle filtering technique, used to facilitate updating the state prediction with a numerical inference model, when measurement models alone are not satisfactory. The results show significant improvement in tool wear state estimation, reducing the prediction errors by almost half, compared to the prediction model and sensor-based monitoring method used independently.
View
A Hybrid Approach to Cutting Tool Remaining Useful Life Prediction Based on the Wiener Process
Article
  • Jun 2018
Accurate remaining useful life prediction is meaningful for cutting tool usability evaluation. Over the years, experience-based models, data-driven models, and physics-based models have been used individually to predict cutting tool remaining useful lives. In order to improve prediction performances, different prognostics models can be combined to leverage their advantages. In this paper, a hybrid cutting tool remaining useful life prediction approach is proposed by combining a data-driven model and a physics-based model. By using force, vibration and acoustic emission signals, the data-driven model monitors cutting tool wear conditions based on empirical mode decomposition and back propagation neural network. On the basis of the Wiener process, the physics-based model builds a cutting tool condition degradation model to predict cutting tool remaining useful lives. Experimental study verifies the approach's effectiveness, accuracy, and robustness. Then, cutting tool remaining useful lives can be predicted more accurately during the machining process.
View
Prediction of remaining useful life of cutting tools: a comparative study using soft computing methods
Article
  • Jan 2018
Predicting the remaining useful life of the partially degraded components and putting them to use will help to save natural resources to a great extent. This will reduce the overall cost, energy and protects environment. High productivity cutting tools used in manufacturing industry are generally expensive. As such, the accurate assessment of remaining useful life (for reuse) of any given tool is of great significance in any manufacturing industry. The main objective of this research is to develop a comprehensive methodology to assess the reuse potential of carbide-tipped tools. This paper presents the use of three soft computing methods, namely, artificial neural network, neuro fuzzy logic and support vector regression methods for the assessment of remaining useful life (RUL) of cutting tools. In this work, experiments are conducted based on Taguchi approach and tool life values are obtained. Tool life values are predicted using the aforesaid three soft computing methods and RUL obtained from these values are compared. It is found that the predictive neuro fuzzy method is capable of giving a better prediction of remaining useful tool life than the other methods.
View
Degradation Data Analysis and Remaining Useful Life Estimation: A Review on Wiener-Process-Based Methods
Article
  • Feb 2018
  • EUR J OPER RES
Degradation-based modelling methods have been recognized as an essential and effective approach for lifetime and remaining useful life (RUL) estimations for various health management activities that can be scheduled to ensure reliable, safe, and economical operation of deteriorating systems. As one of the most popular stochastic modelling methods, the previous several decades have witnessed remarkable developments and extensive applications of Wiener-process-based methods. However, there is no systematic review particularly focused on this topic. Therefore, this paper reviews recent modelling developments of the Wiener-process-based methods for degradation data analysis and RUL estimation, as well as their applications in the field of prognostics and health management (PHM). After a brief introduction of conventional Wiener-process-based degradation models, we pay particular attention to variants of the Wiener process by considering nonlinearity, multi-source variability, covariates, and multivariate involved in the degradation processes. In addition, we discuss the applications of the Wiener-process-based models for degradation test design and optimal decision-making activities such as inspection, condition-based maintenance (CBM), and replacement. Finally, we highlight several future challenges deserving further studies.
View
View
Machinery health prognostics: A systematic review from data acquisition to RUL prediction
Article
  • May 2018
Machinery prognostics is one of the major tasks in condition based maintenance (CBM), which aims to predict the remaining useful life (RUL) of machinery based on condition information. A machinery prognostic program generally consists of four technical processes , i.e., data acquisition, health indicator (HI) construction, health stage (HS) division, and RUL prediction. Over recent years, a significant amount of research work has been undertaken in each of the four processes. And much literature has made an excellent overview on the last process, i.e., RUL prediction. However, there has not been a systematic review that covers the four technical processes comprehensively. To fill this gap, this paper provides a review on machinery prognostics following its whole program, i.e., from data acquisition to RUL prediction. First, in data acquisition, several prognostic datasets widely used in academic literature are introduced systematically. Then, commonly used HI construction approaches and metrics are discussed. After that, the HS division process is summarized by introducing its major tasks and existing approaches. Afterwards, the advancements of RUL prediction are reviewed including the popular approaches and met-rics. Finally, the paper provides discussions on current situation, upcoming challenges as well as possible future trends for researchers in this field.
View