Sensing tool breakage in face milling with a neural network

A new approach using a neural network to process the features of the cutting force signal for the recognition of tool breakage in face milling is proposed. The cutting force signal is first compressed by averaging the cutting force signal per tooth. Then, the average cutting force signal is passed through a median filter to extract the features of the cutting force signal due to tool breakage. With the back propagation training process, the neural network memorizes the feature difference of the cutting force signal between with and without tool breakage. As a result, the neural network can be used to classify the cutting force signal with or without tool breakage. Experiments show this new approach can sense tool breakage in a wide range of face milling operations.     

Sweeping filters and tooth rotation energy estimation (TREE) techniques for machine tool condition monitoring

This paper presents a method for the condition monitoring of the milling cutting process based upon a combination of two techniques; sweeping filters and tooth rotation energy estimation (TREE). Existing spindle speed and spindle load signals from the machine are used thus avoiding the need for any additional sensors. The sweeping filter technique determines the frequency components of the spindle signal using low cost hardware. The filter's cut off frequency is swept across a range of frequencies and its output is acquired and analysed in real time. The variations of individual tooth energies estimated by the TREE technique in the time domain are used to verify the results. The hybrid approach created is based on the verification of any indicated faults before making a final conclusion about the health of the cutting tool. This provides a robust and reliable tool monitoring system that is able to identify tool breakage in real time during machining operations.     

Study of milling cutting force pulsation applied to the detection of tool breakage

A study of the cutting force pulsation due to tool breakage is presented. Monitoring algorithms extracting the cutting force signal changes caused by tool breakage and further processing the extracted cutting force signal to recognize tool breakage are proposed. Theoretical studies and experimental results performed in milling operations have proven the feasibility of the algorithms proposed.     

In-process drill wear and breakage monitoring for a machining centre based on cutting force parameters

There are many physical parameters in the cutting process that could be used for the in-process monitoring of cutting tools' working conditions. Of these parameters, cutting forces are strongly recommended by the investigators to be used because of their higher sensitivity and more rapid response to the changes in cutting states. In this paper, the variation in the dynamic behaviour of drilling thrust force and torque with drill wear, breakage and other kinds of fault forms have been investigated experimentally. The results and associated theoretical investigations indicate that, to make the monitoring system more reliable and suitable for a wider range of cutting conditions, both thrust force F_z and torque T arising from drilling operations should be taken as associated monitoring quantities, rather than choosing just one of them. The methodology of signal signature extraction and data processing techniques are discussed and a practical monitoring strategy proposed.     

An automated monitoring solution for avoiding an increased number of surface anomalies during milling of aerospace alloys

The demand for automated machining systems to enable the increase of process productivity and quality in milling of aerospace safety critical components requires advanced on-line monitoring and supervision systems to identify and reduce the number of workpiece surface anomalies caused by faulty tooling. It is well known that chipping or breakage of only one tooth of a milling cutter can lead to extensive damage to the machined surface. Therefore, implementing a process monitoring solution that can detect workpiece surface anomalies associated with the cutting tooth that generated them could be beneficial for proposing more efficient process supervision systems. The system presented in this paper is directed towards real-time control of the contact length of each tooth of a milling cutter with the scope of reducing the amount of workpiece surface anomalies. The proposed control solution consists of: (i) an automated process monitoring part, that, through signal analysis, automatically improves the reliability of fault detection; (ii) a signal based decision and supervision system for the avoidance of surface anomalies or tool malfunctions. The novelty of this supervision system comes from the fact that it is automatically acquiring Acoustic Emission (AE) and cutting forces, adjusts the monitoring parameters accordingly and transmits decision commands to the machine. The supervision system makes use of the sensor fusion and an original methodology for the detection of process malfunctions, to control the movement of the tool. This is directed towards the adjustment of the individual feed per tooth for each cutting edge, and if necessary reduce it to zero for a cutting tooth that is creating the surface damage. The efficiency of the proposed process supervision system is proven to avoid the generation of surface anomalies in milling of a Ni-based alloy used for the manufacture of parts for gas turbine engines.     

Radial immersion angle estimation using cutting force and predetermined cutting force ratio in face milling

Radial immersion ratio is an important factor to determine the threshold for tool conditioning monitoring and automatic force regulation in face milling. In this paper, a method of on-line estimation of the radial immersion angle using cutting force is presented. When a tooth finishes sweeping, a sudden drop of cutting force occurs. This force drop is equal to the cutting force that acts on a single tooth at the swept angle of cut and can be obtained from the cutting force signal in feed and cross-feed directions. The ratio of cutting forces in feed and cross-feed directions acting on the single tooth at the immersion angle is a function of the immersion angle and the ratio of radial-to-tangential cutting force. In this study, it is found that the ratio of radial-to-tangential cutting force is not affected by cutting conditions and axial rake angle. Therefore, the ratio of radial-to-tangential cutting force determined by just one preliminary experiment can be used regardless of the cutting conditions for a given tool and workpiece material. Using the measured cutting force during machining and a predetermined ratio, the radial immersion ratio is estimated in the process. Various experiments show that the radial immersion ratio and instantaneous ratio of the radial to tangential direction cutting force can be estimated very well by the proposed method.     

A multi-sensor strategy for tool failure detection in milling

A multi-sensor monitoring strategy for detecting tool failure during the milling process is presented. In this strategy, both cutting forces and acoustic emission signals are used to monitor the tool condition. A feature extracting algorithm is developed based on a first order auto-regressive (AR) model for the cutting force signals. This AR(1) model is obtained by using average tooth period and revolution difference methods. Acoustic emission (AE) monitoring indices are developed and used in determining the setting threshold level on-line. This approach was beneficial in minimizing false alarms due to tool runout, cutting transients and variations of cutting conditions. The proposed monitoring system has been verified experimentally by end milling Inconel 718 with whisker reinforced ceramic tools at spindle speeds up to 3000 rpm.     

Tool wear monitoring and breakage detection based on “intelligent filtering”

Several physical quantities can be used for indirect tool wear monitoring and breakage detection. Cutting forces are appropriate, since they determine the suitability of a tool for cutting. However, disturbances make the accurate and fast tool condition monitoring based upon force analysis difficult. To eliminate disturbances averaging or filtering within a predetermined bandwidth has often been applied. A new method of decomposing the force signal into components having close relationships with physical phenomena taking place during cutting is presented. The term “intelligent filtering” denotes decomposition based on on-line identification of model(s) of the cutting process. Application of “intelligent filtering” for the milling process with two cutting insert tools is discussed.     

铣削过程中基于伺服电流信号的切削力监测

刀具破损和刀具磨损的自动检测是数控机床自动加工中的一个重要环节。基于伺服电机电流的切削力监控是一个有效的无传感器方法,该方法的关键是如何从伺服电流信号中准确提取切削刃的力波动表征信号。在对伺服电流信号进行时域、频域分析的基础上,基于可变加工参数给出实时提取切削刃力波动表征信号的方法,并通过相关实验验证了该方法的合理性。     

Measurement of quasi-mean resultant force using the vibrational signal of spindle in milling

The quasi-mean resultant force has been proven to be useful on the real-time process control and tool monitoring in milling operations. This paper presents a new way to measure the quasi-mean resultant force using the vibrational displacement signal of spindle. The quasi-mean resultant force can be obtained by subtracting the spindle run-out pattern from the average displacement signal per tooth period, then multiplying a constant, k^*. This new approach is illustrated by computational simulations and experimental cutting tests.     

The shape characteristic detection of tool breakage in milling operations

Detection of tool failure is very important in automated manufacturing. All previously developed tool breakage detection approaches in milling operations have adopted the strategy of parameter detection in which the detection of tool breakage was carried out according to values of specific parameters selected to reflect tool state (with or without tool breakage). In this paper the new concept of shape characteristic detection of tool breakage in milling operations is proposed. The detection of tool breakage is conducted according to the shape characteristics of discrete dyadic wavelet decomposition of cutting force. By means of the proposed method, the influence caused by the variation of cutting parameters and transients is eliminated. The proposed method is conducted in two steps. In the first step, cutting force signals are decomposed by discrete dyadic wavelet, with the shape characteristic vectors then being generated by the proposed shape characteristic vector-generating algorithm. In the second step, the shape characteristic vectors are fast classified by the ART2 neural networks. The accuracy and effectiveness of the proposed method are verified by numerous experiments.     

An approach based on current and sound signals for in-process tool wear monitoring

This paper presents a tool condition monitoring system (TCMS) for on-line tool wear monitoring in turning. The proposed TCMS was developed taking into account the necessary trade-off between cost and performance to be applicable in practice, in addition to a high success rate. The monitoring signals were the feed motor current and the sound signal. The former was used to estimate the feed cutting force using the least squares version of support vector machines (LS-SVM). Singular spectrum analysis (SSA) was used to extract information correlated with tool wear from the sound signal. The estimated feed cutting force and the SSA decomposition of the sound signal alone with the cutting conditions constitute the input data to the TCMS. Again LS-SVM was used to estimate tool condition and its reliability for on-line implementation was validated by experiments using AISI 1040 steel. The results showed that the proposed TCMS is fast and reliable for tool condition monitoring.     

On-line detection of the breakage of small diameter drills using current signature wavelet transform

This paper presents on-line tool breakage detection of small diameter drills by monitoring the AC servo motor current. The continuous wavelet transform was used to decompose the spindle AC servo motor current signal and the discrete wavelet transform was used to decompose the feed AC servo motor current signal in time–frequency domain. The tool breakage features were extracted from the decomposed signals. Experimental results show that the proposed monitoring system possessed an excellent on-line capability; in addition, it had a low sensitivity to change of the cutting conditions and high success rate for the detection of the breakage of small diameter drills.     

Machine ensemble approach for simultaneous detection of transient and gradual abnormalities in end milling using multisensor fusion

In a fully automated manufacturing environment, instant detection of the cutting tool condition is essential for the improved productivity and cost effectiveness. This paper studies a tool condition monitoring system (TCM) via machine learning (ML) and machine ensemble (ME) approach to investigate the effectiveness of multisensor fusion technique when machining 4340 steel with multilayer coated and multiflute carbide end mill cutter. In this study, 135 different features are extracted from multiple sensor signals of force, vibration, acoustic emission and spindle power in the time and frequency domain by using data acquisition and signal processing module. Then, a correlation-based feature selection technique (CFS) evaluates the significance of these features along with machining parameters collected from machining experiments. Next, an optimal feature subset is computed for various assorted combinations of sensors. Finally, machine ensemble methods based on majority voting and stacked generalization are studied for the selected features to classify not only flank wear but also breakage and chipping. It has been found in this paper that the stacked generalization ensemble can ensure the highest accuracy in tool condition monitoring. In addition, it has been shown that the support vector machine (SVM) outperforms other ML algorithms in most cases tested.     

Driver current analysis for sensorless tool breakage monitoring of CNC milling machines

Tool wear and breakage detection is one of the most important problems found during manufacture in automated CNC machines. From several techniques devoted to sense tool condition, driver current monitoring has been used for a sensorless approach. In order to efficiently use the driver current monitoring technique an exhaustive analysis on the nature of the real components of the signal is required. The novelty of this paper is to present a driver current signal analysis to estimate the influence of the most important spurious signal components in order to determine the optimal parameters for signal conditioning. Beside the cutting force signal, the spurious signals considered in the analysis are high-frequency noise, current control commutation and ball screw effects. The analysis is compared with experimental data in order to validate the model and a case study is presented to show the general procedure.     

Sensing Cutter Breakage in Milling

The paper describes a system for sensing milling cutter breakage. It is based on recognizing changes in the pattern of the cutting force. The regular periodic variation per tooth is filtered out by synchronized sampling, averaging per tooth and first differencing and the change due to the broken tooth stands out. The threshold for alarm is set with respect to the mean force per revolution (moving average) to make it independent of changes in radial and axial depths of cut and in chip load. Due to practical difficulties with force measurements the use of vibrational signal is explored and found feasible.     

An investigation of cutting forces in machining with worn ball-end mill

Reliable tool wear monitoring technique is one of the important aspects for achieving an integrated and self-adjusting manufacturing system. In this study, an analytical model is proposed to estimate the cutting forces, the tool geometry, and the chip geometry in relation to the flank wear, when milling with a ball-end mill. Modeling is based on thermomechanical modelling of oblique cutting. The worn tool geometry is decomposed into a series of axial elementary cutting edges. At any active tooth element, the flank wear geometry is calculated and the chip formation is obtained from an oblique cutting process characterised by local undeformed chip section and local cutting angles. Coated carbide ball-end tool, and a titanium workpiece material have been considered in this paper. The results found by using developed models have shown good agreement with experimental results.     

Development of a tool failure detection system using multi-sensors

Tool monitoring and machine diagnosis in real machining have been crucial to the realization of fully automated machining. Also, the on-line detection technique of the tool breakage in machining should be supported. The effect of tool breakage is usually revealed from an abrupt change in the processed measurements, which is in excess of a threshold value. Although these techniques are generally effective for a specific cutting condition, they are often not sufficiently reliable for use in production due to the inability of single measurement to reflect tool breakage under various cutting conditions. In order to enhance the reliability of tool breakage signatures obtained from a single sensor, an integrated approach based on measurements from several sensors has been put forward. In this study, the tool breakage detection method using multi-sensors is proposed and the sensor fusion algorithm is developed to integrate and make decisions from data measured through the multisensors. Also, the performances of this scheme are compared and evaluated with real cutting process.     

Tool condition monitoring in turning using fuzzy set theory

This paper presents a study on tool condition monitoring in turning using the fuzzy set theory. The tool conditions considered include tool breakage, several states of tool wear, and chatter. Force, vibration, and power sensors are used in this study to monitor the three components of the cutting force, i.e. acceleration of the tool holder in two perpendicular directions, and the spindle motor current respectively. A total of 11 monitoring indices (signature features) are selected to describe the signature characteristics of various tool conditions. A linear fuzzy equation is proposed to describe the relationship between the tool conditions and the monitoring indices. The proposed methodology is verified experimentally using a total of 396 cutting tests performed at 52 different cutting conditions. The proposed methodology is also compared with that of several classification schemes, including the K-mean and the Fisher's pattern recognition methods, the nearest neighbor method and the fuzzy C-mean method. The results indicate an overall 90% reliability of the proposed methodology for detecting tool conditions regardless of the variation in cutting conditions.     

Real time monitoring of tool breakage in a milling operation using a digital signal processor

A monitoring system that can detect tool breakage and chipping in real time was developed using a digital signal processor (DSP) board in a face milling operation. An autoregressive (AR) model and a band energy method were used to extract the features of tool states from cutting force signals. Then, two artificial neural networks, which have a parallel processing capability, were embedded on the DSP board to discriminate different malfunction states from features obtained by each of the two methods of signal processing. In experiments, we found that feature parameters extracted by AR modeling were more accurate indicators of malfunctions in the process than those from the band energy method, although the computing speed is slower. By using the selected features, we were able to monitor malfunctions in real time.