The spindle system is a critical part of a machine tool structure and its dynamic properties are important for the performance of the whole machining system. Currently the only way to extract the dynamic properties of a given structure is via experimental modal analysis. This approach, however, can only be employed on idle systems and is performed with the assumption that the dynamics of a system are independent of rotational speed. The latter assumption cannot be applied to spindle systems. This paper introduces a novel testing system for analysing machine tool spindles dynamic properties, consisting of real-time recursive estimation of modal and operational dynamic parameters, employed alongside a contactless excitation and response system. The presented approach allows analysing the spindle system condition and dynamic properties not only at discrete rotational speed intervals but also during continuous sweep of rotational speed.

This paper presents a novel test concept for the evaluation of the accuracy of NC machine tools. The evaluation of machine tools deformations is performed by help of a device similar to the double ball bar (DBB) with the difference that an adjustable load generated by the device can be applied between spindle nose and machine tool table. This load eliminates the play existing in machine tool joints, thus reproducing the testing conditions that exist during machining. Collected data are used to plot diagrams displaying characteristic aspects of achine tool performance and a number of key figures such as static stiffness may be etermined. The data can also be used for trend analysis; to predict any accuracy deviations, and further to conduct preventive maintenance instead of emergency calls. The determined static behaviour could also be used to improve digital models for process simulations and compensation of errors that are caused by deflection.

With advances in material technology come challenges to productivity. New materials are, in fact, more difficult to machine with regards to tool wear and especially machine tool stability. This paper proposes to extend the stability limits of the machining system by enhancing the structure’s damping capability. The aim of the research work presented here is to introduce a unified concept based on the distribution of damping within the machining system components exploiting the dynamic properties of the existing joints. To maintain a high level of static stiffness, it was chosen to adapt hydrostatic clamping systems to the tools. Damping is designed in the structure via high damping interfaces (HDI), intentionally introduced interfaces where the damping ratio is enhanced by introduction of viscoelastic polymer metal composites between the two metallic surfaces composing the interface. In this paper HDI are introduced at two joints, between tool and turret and between turret and lathe. The tests show that the designed-in damping is effective and allows extending the stability limits of the machining system. The implementation of designed-in damping allows the end user to select the most suitable parameters in terms of productivity avoiding the hassle of tuning the devices, having to acquire a deep knowledge in structural dynamics or having to use additional control systems. In addition to this, the enhanced machine tool system becomes less sensitive to stability issues provoked by difficult-to-machine materials or even fluctuations of the work material properties that might occur in everyday production processes.

Machine tool joints have significant influence on the dynamic characteristics of the machine tool and therefore on the response of the machining system to excitations from the cutting process. In cases of unstable response, generally described as chatter, surface quality of a machined workpiece and tool life deteriorate significantly. This paper presents a novel way of exploiting joints in order to control the dynamic response of the system, by integrating a mechatronic tool holder (Joint Interface Module - JIM) in the machine tool. This system has a purposely designed joint interface with controllable natural characteristics (stiffness and damping). These characteristics are controlled by altering the applied pre-load on the internal joint interface of the tool holder. The preload on the joint interface is controlled by pneumatic means. In doing so, a milling process during which the stability limit was exceeded became stable during the machining process, without alteration of the process parameters.

Multi-layered nanostructured Cu and Cu-CNx composites synthesized by plasma-enhanced chemical vapour deposition were applied in the clamping area of a milling tool to suppress regenerative tool chatter. Scanning electron microscopy analysis showed a multi-layered nanostructure with excellent conformality, i.e. coating is not only uniform on planar surfaces but also around corners of the substrate. Cu:CuCNx nanostructured multilayers with thicknesses of approximately 0.5:1.6 mu m were obtained. With a diameter of 20 mm, the milling tool performed slotting processes at an overhang length of 120 mm. Modal analysis showed that a coating, with a thickness of approximately 300 mu m, can add sufficient damping without losing stiffness of the tool, to increase the critical stability limit by 50% or 100% depending on cutting direction.

Surface roughness plays an indispensable and fundamental role as a leading indicator of the surface quality of machined parts in the manufacturing process. The precise and effective monitoring and prediction of surface roughness is crucial for surface quality control. In this regard, the development of an in-process surface quality monitoring system is necessary, which has the promising potential to achieve this goal. Such a system typically comprises data-driven models for decision-making and sensing techniques for detecting associated process information. However, some challenges still exist in building such systems. Firstly, the architecture design and deployment of data-driven models, specifically deep learning (DL)-based models, demand adequate domain knowledge. Secondly, most models trained on specific tasks with limited datasets are prone to suppressing their versatility and generalization across different machining conditions. Additionally, in most cases, reliance on handcrafted features to represent dynamic information on various signals during model training necessitates extensive expertise in selecting appropriate feature types. Furthermore, due to the nature of their low dimensionality, handcrafted features have difficulty in capturing of overall process-related underlying patterns from dynamics signatures, which is time-varying and often occurs in transient events. To address these challenges, this paper proposes the regression-based pre-trained convolutional neural network (pre-trained CNN) combined with Mel-spectrogram images based on the transfer learning method for surface roughness prediction. Within the context, the architecture of the transfer model is slightly adapted from already well-trained CNNs. Initial weights in each layer of the CNN model are directly inherited and then fine-tuned through the Bayesian optimization tuning method. Besides, the audible sound signals are captured and subsequently converted into 2D Mel-spectrogram images with variant time lengths, which are separately engaged to retrain and validate four existing pre-trained CNN models (VGG16, VGG19, ResNet50V2 and InceptionResNetV2). Eventually, the effectiveness of proposed models and comparison of their predictive capabilities are further validated through a case study in the turning process. The results demonstrate that each applied pre-trained CNN model is capable of effectively predicting surface quality with satisfactory prediction results. Therefore, the proposed method can facilitate the establishment of a machining monitoring system concerning its accuracy, reliability, and robustness.

In the turning process, the surface roughness of the machined part is considered a critical indicator of quality control. Provided the conventional offline quality measurement and control is time-consuming, with slow feedback and an intensive workforce, this paper presents an online monitoring and prediction system for the effective and precise prediction of surface roughness of the machined parts during the machining process. In this system, the audible sound signal captured through the microphone is employed to extract the features related to surface roughness prediction. However, owing to the nonlinear phenomena and complex mechanism causing surface quality in the whole process, the selection of statistical features of the sound signal in both the time and frequency domains varies from one case to another. This variation may lead to false prediction results as sufficient domain knowledge is required. Therefore, the versatile and knowledge-independent features extraction method is proposed, which exploits deep transfer learning to automatically extract sound signal features in the time-frequency domain through pre-trained convolution neural networks (pre-trained CNN). The performance of prediction models based on two feature extraction methods – statistical feature extraction and automatic feature extraction was further tested and validated in the case study. The results demonstrate that the performances of the prediction model built on the automatically extracted features outperformed that developed with the statistical feature method concerning the accuracy and generalization of the prediction model. In addition, this study also provides solid theoretical and experimental support for developing a more precise and robust online surface quality monitoring system.

Machining components out of tolerances is of no use in competitiveproduction. The machining system sets the limitations of dimensionalaccuracy and surface quality of a machined component. The capabilityof the machining system describes these limits in terms of specifiedvalues. This thesis deals with machining system capability analysismainly focused on machine tool static and dynamic stiffness.The influence of stiffness and flexibility on machining systemcapability is analytically and experimentally investigated. Theexperimental work presented in the thesis complies with the theoriesand shows the relation between machine tool capability and theoutcome on the machine component.The concept of capability analysis by elastic linked system andthe currently available tools for such an evaluation is presented anddiscussed. The basis of elastic linked system analysis is the use ofmeasurements under loaded condition. The machine tool is loadedwith a known force creating a test condition closer to real machining,compared to current methods of unloaded machine tools. Twomeasurement tools for elastic linked system capability analysis areexplained in the thesis: Loaded Double Ball Bar and ContactlessExcitation and Response System.The thesis consists of an analytical base and an experimental casestudy on spiral bevel gear face milling. The experiments are discussedand compiled with the given theories.

Dynamic characteristics of machining systems are analysed for improved understanding of both structural and process properties. The thesis stresses the use of testing methods under operational like conditions as these are more representative of closed loop systems, such as machining systems, as compared to conventional testing methods.

The test instrument proposed is a contactless excitation and response system, developed for testing of machine tool spindles under load and with rotating spindle. The instrument uses electromagnetic excitation and displacement sensors for analysis of rotating milling tools subject to load. A graphical tool for displaying and analysing rotor displacement was developed in conjunction with this.

A modelling procedure for both off-line and on-line estimation of dynamic properties of mechanical structure and process information is presented. The proposed auto-regressive moving average models enable calculation of operational dynamic parameters and they can be estimated in a recursive manner, thus enabling real-time monitoring. The discrimination between stable and unstable processes, both in turning and milling, was performed by analysing the damping obtained from the operational dynamic parameters.

This paper presents a study where acoustic imaging technology was employed for problem identification and high damping interfaces (HDI) were implemented in the machine tool structure to improve performance of process machine interaction, in bevel gear cutting, in terms of resistance to instability. Chatter marks at the gear root and tool fracturing represent a serious and widespread problem in the studied process. The acoustic image approach showed good agreement with modal analysis and also allowed to identify the chatter frequency. Once identified the problem, the paper also presents a novel approach for improving resistance to cutting instability by implementing HDI in the structure. The HDI consists of a viscoelastic composite material and is used to damp out the stick blade motion relative to the clamp block by transformation of vibratory energy, into heat. The primary contributions of this paper are the introduction of acoustic imaging for identification of chatter in gear cutting and employment of HDI for performance improvement for rotating tools.

Dynamic characteristics of machine tools and cutting tools have gained much attention from researchers and industry as it is one of the major factors limiting productivity due to excessive vibrations such as chatter during the cutting process. Numerous factors have to be taken into consideration when selecting material of the cutter body. This paper presents a comparison between two cutter bodies with the same geometry but made out of different alloys. Pre-hardened steel and conventional tool steel were investigated in order to highlight the advantages of selecting correct material to achieve high performance tools with respect of chatter resistance. The experimental part of this paper consists of impact testing, machining tests and surface integrity measurements. Operational dynamic parameters obtained through auto-regressive moving average model estimates from machining tests under stable and unstable conditions was used to characterise the performance. The findings are in correlation with material damping research and chatter analysis, thus giving a strong coupling to material selection in tool holders for enhanced process stability. The research also shows that operational dynamic properties obtained through indirect measurements is a valuable tool for process stability characterisation.

Vibration in metal cutting processes has been studied to a great extent resulting in for instance stability lobe diagrams under which stable machining parameters can be selected. One limitation of accurately estimated stability diagrams is the change in process and dynamic characteristics of the machine tool under operation. The machine tool dynamic response is often analysed with experimental modal analysis under off operational conditions. One drawback with this approach is the large number of measurements required to fully describe a machine tool and workpiece in different positions and time of machining. Another drawback is that the change of dynamic characteristics under operation is excluded. Operational modal analysis has been applied in machining under different conditions resulting in successfully improved stability lobe prediction. This research includes operational modal analysis of the workpiece, fixed on a stiffness controllable joint and stability prediction to stress the importance of various machining conditions.

Rotor vibration and stationary displacement patterns observed in rotating machineries subject to local harmonic excitation are analysed for improved understanding and dynamic characterization. The analysis stresses the importance of coordinate transformation between rotating and stationary frame of reference for accurate results and estimation of dynamic properties. A generic method which can be used for various rotor applications such as machine tool spindle and turbo machinery vibration is presented. The phenomenon shares similarities with stationary waves in rotating disks though focuses on vibration in shafts. The paper further proposes a graphical tool, the displacement map, which can be used for selection of stable rotational speed for rotating machinery. The results are validated through simulation of dynamic response of a milling cutter, which is a typical example of a variable speed rotor operating under different load conditions.