Remaining Useful-Life Prediction of the Milling Cutting Tool Using Time-Frequency-Based Features and Deep Learning Models.

The milling machine serves an important role in manufacturing because of its versatility in machining. The cutting tool is a critical component of machining because it is responsible for machining accuracy and surface finishing, impacting industrial productivity. Monitoring the cutting tool's life is essential to avoid machining downtime caused due to tool wear. To prevent the unplanned downtime of the machine and to utilize the maximum life of the cutting tool, the accurate prediction of the remaining useful life (RUL) cutting tool is essential. Different artificial intelligence (AI) techniques estimate the RUL of cutting tools in milling operations with improved prediction accuracy. The IEEE NUAA Ideahouse dataset has been used in this paper for the RUL estimation of the milling cutter. The accuracy of the prediction is based on the quality of feature engineering performed on the unprocessed data. Feature extraction is a crucial phase in RUL prediction. In this work, the authors considers the time-frequency domain (TFD) features such as short-time Fourier-transform (STFT) and different wavelet transforms (WT) along with deep learning (DL) models such as long short-term memory (LSTM), different variants of LSTN, convolutional neural network (CNN), and hybrid models that are a combination of CCN with LSTM variants for RUL estimation. The TFD feature extraction with LSTM variants and hybrid models performs well for the milling cutting tool RUL estimation.

Fault Detection in Induction Motor Using Time Domain and Spectral Imaging-Based Transfer Learning Approach on Vibration Data.

The induction motor plays a vital role in industrial drive systems due to its robustness and easy maintenance but at the same time, it suffers electrical faults, mainly rotor faults such as broken rotor bars. Early shortcoming identification is needed to lessen support expenses and hinder high costs by using failure detection frameworks that give features extraction and pattern grouping of the issue to distinguish the failure in an induction motor using classification models. In this paper, the open-source dataset of the rotor with the broken bars in a three-phase induction motor available on the IEEE data port is used for fault classification. The study aims at fault identification under various loading conditions on the rotor of an induction motor by performing time, frequency, and time-frequency domain feature extraction. The extracted features are provided to the models to classify between the healthy and faulty rotors. The extracted features from the time and frequency domain give an accuracy of up to 87.52% and 88.58%, respectively, using the Random-Forest (RF) model. Whereas, in time-frequency, the Short Time Fourier Transform (STFT) based spectrograms provide reasonably high accuracy, around 97.67%, using a Convolutional Neural Network (CNN) based fine-tuned transfer learning framework for diagnosing induction motor rotor bar severity under various loading conditions.

A Low-Cost Multi-Sensor Data Acquisition System for Fault Detection in Fused Deposition Modelling.

Fused deposition modelling (FDM)-based 3D printing is a trending technology in the era of Industry 4.0 that manufactures products in layer-by-layer form. It shows remarkable benefits such as rapid prototyping, cost-effectiveness, flexibility, and a sustainable manufacturing approach. Along with such advantages, a few defects occur in FDM products during the printing stage. Diagnosing defects occurring during 3D printing is a challenging task. Proper data acquisition and monitoring systems need to be developed for effective fault diagnosis. In this paper, the authors proposed a low-cost multi-sensor data acquisition system (DAQ) for detecting various faults in 3D printed products. The data acquisition system was developed using an Arduino micro-controller that collects real-time multi-sensor signals using vibration, current, and sound sensors. The different types of fault conditions are referred to introduce various defects in 3D products to analyze the effect of the fault conditions on the captured sensor data. Time and frequency domain analyses were performed on captured data to create feature vectors by selecting the chi-square method, and the most significant features were selected to train the CNN model. The K-means cluster algorithm was used for data clustering purposes, and the bell curve or normal distribution curve was used to define individual sensor threshold values under normal conditions. The CNN model was used to classify the normal and fault condition data, which gave an accuracy of around 94%, by evaluating the model performance based on recall, precision, and F1 score.

Application of the Nadaraya-Watson estimator based attention mechanism to the field of predictive maintenance.

Attention mechanism has recently gained immense importance in the natural language processing (NLP) world. This technique highlights parts of the input text that the NLP task (such as translation) must pay "attention" to. Inspired by this, some researchers have recently applied the NLP domain, deep-learning based, attention mechanism techniques to predictive maintenance. In contrast to the deep-learning based solutions, Industry 4.0 predictive maintenance solutions that often rely on edge-computing, demand lighter predictive models. With this objective, we have investigated the adaptation of a simpler, incredibly fast and compute-resource friendly, "Nadaraya-Watson estimator based" attention method. We develop a method to predict tool-wear of a milling machine using this attention mechanism and demonstrate, with the help of heat-maps, how the attention mechanism highlights regions that assist in predicting onset of tool-wear. We validate the effectiveness of this adaptation on the benchmark IEEEDataPort PHM Society dataset, by comparing against other comparatively "lighter" machine learning techniques - Bayesian Ridge, Gradient Boosting Regressor, SGD Regressor and Support Vector Regressor. Our experiments indicate that the proposed Nadaraya-Watson attention mechanism performed best with an MAE of 0.069, RMSE of 0.099 and R2 of 83.40 %, when compared to the next best technique Gradient Boosting Regressor with figures of 0.100, 0.138, 66.51 % respectively. Additionally, it produced a lighter and faster model as well.•We propose a Nadaraya-Watson estimator based "attention mechanism", applied to a predictive maintenance problem.•Unlike the deep-learning based attention mechanisms from the NLP domain, our method creates fast, light and high-performance models, suitable for edge computing devices and therefore supports the Industry 4.0 initiative.•Method validated on real tool-wear data of a milling machine.

Tribological and Hardness Analyses of Friction-Stir-Processed Composites Using the Taguchi Approach.

The friction stir process (FSP) is becoming a highly utilized method to manufacture composites since it refines the microstructure and improves the physical characteristics like hardness, strength, and wear resistance of their surfaces. In this study, the hardness and wear behaviours of Al6061-based surface composites prepared by the FSP were investigated and compared for the influences of various parameters-FSP tool geometry, reinforcement composition, number of FSP passes, pin load, etc. The Taguchi design with an L27 orthogonal array was developed to analyze the influence of five input parameters on the output parameter, i.e., wear rate during wear tests. The hardness of the composite samples for different reinforcement compositions was investigated, and the results were statistically compared with the obtained wear rates. It was concluded from the results that various parameters influenced the surface wear and hardness of the composites. Tool geometries cylindrical pin and square pin had the maximum and minimum wear rates, respectively. Additionally, the optimal composition of the reinforcements copper and graphene as 1:3 possessed the maximum wear rate and minimum hardness. However, the reinforcement composition 3:3 (Cu:Gr) by weight had the minimum wear rate and maximum hardness. The higher the FSP pass numbers, the lesser the wear rate and the higher the hardness, and vice-versa. This work helps identify the influence of numerous factors on the wear and hardness aspects of surface composites prepared by the FSP. In the future, this study can be modified by combining it with thermal analysis, sensor data analysis of the composites, and optimization of the parameters for desirable microstructure and physical properties.

Wear Characterization of Phenol-Formaldehyde-Based Saguvani Wood-Polymer Composite-An ANOVA Approach.

In this part of the research work, the Taguchi approach is used to analyze the weight wear loss of PF-based 10% chemically treated saguvani wood-polymer composite under dry sliding conditions. The fabrication of PF-based wood-polymer composite consisting of 10% chemically treated saguvani wood particles as reinforcement material filled with coconut shell powder is used. The rotary-drum-type blender is used for uniform mixing of reinforcement materials with resin as per the calculated volume ratio. The inclusion of coconut shell powder as secondary particles in the PF-based wood plastic composite minimizes the wearability of the composite. The Taguchi method is used successfully to analyze the wear behavior of the PF-based wood-polymer composite with sliding speed, load, and sliding distance as control parameters. The experimental work reveals that the composite C1 shows minimum wear loss compared to the other composite specimens, C2 and C3. And the most influential parameter that causes more wear is the sliding distance among the three control parameters.

Development of an Oxide Layer on Al 6061 Using Plasma Arc Electrolytic Oxidation in Silicate-Based Electrolyte.

The plasma electrolytic method is one of the techniques which can be used to form an oxide layer on the substrate material surface. This technique employs ion exchange by developing an electrolytic arc between the cathode and the anode. The strong bond at high temperatures promotes the formation of an oxide layer on the metal surface. The electrolyte composition has a strong influence on the metal surface characteristics. Hence, the addition of certain nanoparticles in an adequate amount can improve the surface properties like wear and corrosion resistance. In this study, a plasma electrolytic technique based on using a direct current and voltage approach is investigated. The plasma electrolytic technique is utilized to develop an oxide layer on the Al 6061 alloy substrate surface using a DC voltage input on a silicate-based electrolyte. The substrate surface is then investigated for the thickness of the oxide layer formed and the amount of carbon element absorbed, using the SEM and XRD analysis. The experimentation and the study of the results confirmed the presence of a substantial oxide layer on the surface. The influence of the process on the output parameters-direct voltage and electrode distance is studied with the significant changes obtained in the weight percentage of elements like C, Al, Si, and O as supported by SEM and EDAX analysis. Most changes occurred when using a 197 V and in the current range of 0.3 A to 1 A. This can be useful further to improve the mechanical properties of the metal alloy using the plasma arc oxidation method.

Optimization and Modeling of Material Removal Rate in Wire-EDM of Silicon Particle Reinforced Al6061 Composite.

The mechanical, physical and interfacial properties of aluminum alloys are improved by reinforcing the silicon carbide particles (SiCp). Machinability of such alloys by traditional methods is challenging due to higher tool wear and surface roughness. The objective of research is to investigate the machinability of SiCp reinforced Al6061 composite by Wire-Electrical Discharge Machining (wire-EDM). The effect of wire-EDM parameters namely current (I), pulse-on time (Ton), wire-speed (Ws), voltage (Iv) and pulse-off time (Toff) on material removal rate (MRR) is investigated and their settings are optimized for achieving the high MRR. The experiments are designed by using Taguchi L16 orthogonal arrays. The MRR obtained at different experiments are analyzed using statistical tools. It is observed that all the chosen process parameters showed significant influence of on the MRR with contribution of 27.39%, 22.08%, 21.32%, 15.76% and 12.94% by I, Iv, Toff, Ton and Ws, respectively. At optimum settings, the Wire-EDM resulted in MRR of 65.21 mg/min and 62.41 mg/min for samples with 4% and 8% SiCp. The results also indicated reinforcing SiCp upto 8% showed marginally low influence on MRR. Microstructural investigation of the cut surface revealed the presence of craters with wave pattern on its surface. The top surface of the crater is featured by the recast layers connecting adjacent craters. Further, the statistical model is developed using linear regression to predict the MRR (?2-73.65%) and its predicting accuracy is verified by the confirmation trials. The statistical model is useful for predicting the MRR for different settings of the process parameters. The optimized settings can be used to improve the machining productivity by increasing the MRR while machining of Al6061-SiCp (upto 8 wt. %) alloy by wire-EDM industries.