Machining Techniques Research Articles

Continuous Speech Commands Recognition with Thai Language used Support Vector Machine Technique: A case study of speech commands control for mobile robots

Continuous Speech Commands Recognition with Thai Language used Support Vector Machine Technique: A case study of speech commands control for mobile robots

Enhancing human sight perceptions to optimize machine vision: Untangling object recognition using deep learning techniques

The goal of machine vision is to develop human-like visual abilities; however, it is unclear whether understanding human vision will advance machines. Here, it exemplifies two key conceptual advancements: It first shows that the majority of computer vision models consistently differ from the way that individuals perceive objects. To do this, a significant dataset of human perceptions of the separations of isolated things was acquired, and it was then examined to see if a well-known machine vision algorithm can predict these perceptions. The best algorithms can account for the majority of the volatility in the intuitive data, but every algorithm we verified repeatedly misjudged several different object types. Second, it shows that removing these systemic biases can considerably increase classification accuracy. For instance, machine techniques overestimated detachments between symmetric objects in comparison to human vision. These results illustration that methodical differences between human and machine vision can be identified and improved.In order to improve the machine vision, employing a deep learning algorithm Visual Geometry Group (VGG 16) with planar reflection symmetry (PRS-Net) technique. VGG 16 is a convolutional neural network with 16 deep layers. VGG pre-trained architecture can point out visual features present in the image. The planar reflection symmetry concept is appended with VGG to create a hybrid environment that can improve machine vision significantly by 90%.

Hybrid Additive and Subtractive Manufacturing Method Using Pulsed Arc Plasma.

In this study, a novel hybrid additive and subtractive manufacturing method using pulsed arc plasma (PAP-HASM) was developed to better integrate additive and subtractive processes. The PAP-HASM process is based on the flexible application of pulsed arc plasma. In this PAP-HASM method, wire arc additive manufacturing using pulsed arc plasma (PAP-WAAM) and dry electrical discharge machining (EDM) milling were used as additive and subtractive techniques, respectively; both are thermal machining processes based on pulsed arc plasma, and both are dry machining techniques requiring no working fluids. The PAP-HASM can be easily realized by only changing the pulsed power supply and tool electrodes. A key technological challenge is that the recast layer on the part surface after dry EDM milling may have a detrimental effect on the component fabricated by PAP-HASM. Here, the hybrid manufacturing method developed in this study was validated with commonly used 316L stainless steel. Preliminary experimental results showed that the PAP-HASM specimens exhibited excellent tensile properties, with an ultimate tensile strength of 539 ± 8 MPa and elongation of 46 ± 4%, which were comparable to the PAP-WAAM specimens. The recast layer on the surface after dry EDM milling has no significant detrimental effect on the mechanical properties of the parts fabricated by PAP-HASM. In addition, compared with components fabricated by PAP-WAAM, those fabricated by PAP-HASM showed significantly better surface roughness.

Design and Optimization of Conformal Cooling Channels for Increasing Cooling Efficiency in Injection Molding

Injection molding is widely used in the manufacturing process of plastic products. The injection molding process aims to increase productivity, which is impacted by factors such as cooling time, the temperature distribution of plastic parts, thermal stress, warpage, etc. The cooling stage of the forming process is a critical factor that significantly influences the quality and cost-effectiveness of the final product in injection molding. With the development of additive manufacturing, the fabrication of conformal cooling channels becomes easier and more affordable compared to traditional machining techniques. Well-designed conformal cooling channels can substantially reduce cooling time, resulting in increased productivity and improved efficiency of the injection molding process. While conformal cooling channels offer improved cooling efficiency and part quality, their design process is more intricate than that of conventional channels due to the demand to account for the complex geometry of the part and to accommodate manufacturing constraints. In this paper, the conformal cooling channels are designed, and the variables for the channels are optimized to increase the cooling efficiency, and 3D printing technology is applied to produce the conformal cooling channel. A simplified finite element cooling analysis was used to compare traditional and hybrid solutions for designing cooling channels to reduce injection cycle time for plastic parts, and the design and simulation of the cooling system were based on UG and Moldflow. After comparing the simulation results of the cooling performance, we found that conformal cooling channels can reduce cooling time by 30% compared to conventional cooling channels. Additionally, they result in a more uniform temperature distribution throughout the plastic products.

A Response Surface Modeling Study on Effects of Powder Rate and Machining Parameters on Surface Quality of CoCrMo Processed by Powder Mixed Electrical Discharge Machining

Due to the high mechanical strength of the metals used in implant manufacture, which makes them difficult to work with using other machining techniques, electrical discharge machining (EDM) is frequently employed in the production of implants. In this study, the effect of powder ratio and other EDM parameters used in the machining of CoCrMo alloy, which used in implant production widely, with powder-mixed EDM on the surface roughness of the machined part was investigated through the response surface methodology. AISI 316L stainless steel was chosen as the electrode material, and Ti6V4Al was chosen as the additive powder, taking into account their biocompatibility properties. Using a Taguchi L16 array, an experimental design was created by selecting 4 levels for each parameter of additive ratio, discharge current, pulse on time (Ton), and pulse off time (Toff). The response surface method was used, along with the experimental data, to estimate how the parameters affected the arithmetic average roughness (Ra) and mean roughness depth (Rz).

Predicting cutting forces in 5-axis milling of sculptured surfaces directly from a CAM tool path

The present work is motivated by the fact that for predicting the cutting forces which arise under various cutting conditions, workpiece-tool pairs, machining depths and tool orientation, numerical methods are slower and less efficient than analytical methods. In addition, recent developments in Computer Assisted Machining (CAM) techniques have enabled analytical methods to be applied even with complex workpiece geometry. The method developed in this paper is both practical and inexpensive. It is based on an analytical method using only few approximations and uses the toolpath file as the main information source for the machined surface. This method takes into account both tool position and orientation in a 5-axis finishing milling process with ball-end cutter, independent of the machining parameters and the couple of tool-workpiece materials. However, this calculation must be supplemented by a mechanistic approach in order to obtain results equivalent to those of the experiment. The result is a model which is implemented in a program allowing the visualization of the cutting forces along a given path. The program is easy to apply in practical cases with the possibility of implementing it in a machining software.

The semiproximal SVM approach for multiple instance learning: a kernel-based computational study

The semiproximal Support Vector Machine technique is a recent approach for Multiple Instance Learning (MIL) problems. It exploits the benefits exhibited in the supervised learning by the Support Vector Machine technique, in terms of generalization capability, and by the Proximal Support Vector Machine approach in terms of efficiency. We investigate the possibility of embedding the kernel transformations into the semiproximal framework to further improve the testing accuracy. Numerical results on benchmark MIL data sets show the effectiveness of our proposal.

Hydrologic Design of Soil and Water Conservation Structures Using Probability Analysis and Machine Learning Techniques in Saurashtra Region of Gujarat, India

The soil and water conservation structures are constructed to overcome water scarcity as a result of interannual rainfall variability and paucity of the perennial source of water. The present study was aimed to estimate the design runoff for the efficient hydrologic design of various soils and water conservation structures using machine techniques for enabling efficient harvesting of available rainfall with economical design which can support in developing climate resilience for the Saurashtra region of Gujarat, India. The design rainfall at various return periods was predicted by Annual One Day Maximum Rainfall (ADMR) using three technics i.e. Probability Distribution Fitting, Artificial Neural Network (ANN) and Gaussian Process Regression (GPR) for 11 stations. Various goodness of fit tests revealed that ADMR was efficiently predicted by log-logistic (3P) distribution for six stations, generalized extreme value distribution for two stations and lognormal (3P), gamma (3P) and lognormal distribution for one station each. Among ANN and GPR, the performance indicators revealed that GPR has shown a higher capability to predict ADMR as compared to ANN with correlation coefficient ranging from 0.97 to 0.99, mean absolute error from 15 mm to 411 mm and root mean squared error from 40 mm to 494 mm for various stations. The design runoff estimation was demonstrated based on predicted ADMR for return periods suitable for various soil and water conservation structures like field bunding, terrace outlets and vegetative outlets, field diversion, permanent masonry gully control structures, earthen dam, etc. using SCS-Curve Number method for curve number 70 and 85. The study is useful for researchers, planners and engineers to implement the economical, efficient and safe design of various soil and water conservation structures.

Development and solving of optimization model for WEDM of AISI D3 using TLBO algorithm

Manufacturing companies employ advanced machining techniques to create complex and accurate products, which require selecting optimum process parameters to achieve the desired quality and cost-effectiveness. Traditional techniques such as Design of Experiments and Response Surface Methodology optimize locally and may not be sufficient for complex systems. This study aims to optimize the Wire Electrical Discharge Machining process of AISI D3 material using Response Surface Methodology and Teaching and Learning-Based Optimization (TLBO), which can optimize globally. The proposed model can improve the quality and cost-effectiveness of the machining process, making it more suitable for various manufacturing applications.

Face mask detection in foggy weather from digital images using transfer learning

ABSTRACT Community mask use is an efficacious non-pharmacologic way to minimize viral infection spread. It is a recommendation that individuals wear face masks as protective gear. Under ideal weather conditions, machine and artificial intelligence techniques can typically determine if a person is wearing a mask properly. Identification becomes more difficult under inclement weather such as fog, clouds, haze or rain. In this work, we propose a technique that can detect a human face wearing a mask even in adverse weather. For this, homogeneous foggy images have been considered. The main challenge with this problem is that video quality degrades because of fog. Here, diverse Deep learning models train regular datasets containing digital pictures of persons with facial and non-facial masks. The training and validation parameters ensure 97% accuracy in classifying faces wearing a mask.

Multivariate analysis of following and filtering manoeuvres of Motorized Two Wheelers in mixed traffic conditions

Mixed traffic conditions present a complex problem for analysis by transportation engineers and policymakers, due to the inherent heterogeneity and lane indiscipline. The chaotic arrangements make it difficult to identify and analyze basic riding behaviour such as following and filtering. Moreover, the smaller lateral dimensions of Motorized Two Wheelers (MTWs) and their higher manoeuvreability add to the difficulty. Presence of different types of vehicles in traffic mix, following the leader with or without obliqueness, filtering through tight pores between leader vehicles, etc. are some of the aspects that require special attention. In this respect, this study is aimed at investigating the attributes related to following and filtering manoeuvres of MTWs in such disordered traffic conditions. Real world data from two Indian cities was used for the analysis, which showed that the behaviour of MTWs is heavily influenced by the type of leader vehicle(s), Clear Lateral Gap (CLG), speed, etc. Safety analysis carried out using Time-To-Collision (TTC) showed that about 8.2% of the interactions were risky. Support Vector Machine (SVM) technique was used to investigate the choice of filtering based on Clear Lateral Gap (CLG) and relative speed. Moreover, analysis of the observed parameters was conducted to obtain their specific distributions based on leader vehicle type, following regime and choice of filtering. The analysis will give directions for further research on developing driving behaviour models of MTWs in mixed traffic. The results will also find potential application in traffic flow theories, safety studies, microsimulation, implementation of MTW infrastructure, etc.

A method for computing inverse parametric PDE problems with random-weight neural networks

We present a method for computing the inverse parameters and the solution field to inverse parametric partial differential equations (PDE) based on randomized neural networks. This extends the local extreme learning machine technique originally developed for forward PDEs to inverse problems. We develop three algorithms for training the neural network to solve the inverse PDE problem. The first algorithm (termed NLLSQ) determines the inverse parameters and the trainable network parameters all together by the nonlinear least squares method with perturbations (NLLSQ-perturb). The second algorithm (termed VarPro-F1) eliminates the inverse parameters from the overall problem by variable projection to attain a reduced problem about the trainable network parameters only. It solves the reduced problem first by the NLLSQ-perturb algorithm for the trainable network parameters, and then computes the inverse parameters by the linear least squares method. The third algorithm (termed VarPro-F2) eliminates the trainable network parameters from the overall problem by variable projection to attain a reduced problem about the inverse parameters only. It solves the reduced problem for the inverse parameters first, and then computes the trainable network parameters afterwards. VarPro-F1 and VarPro-F2 are reciprocal to each other in some sense. The presented method produces accurate results for inverse PDE problems, as shown by the numerical examples herein. For noise-free data, the errors of the inverse parameters and the solution field decrease exponentially as the number of collocation points or the number of trainable network parameters increases, and can reach a level close to the machine accuracy. For noisy data, the accuracy degrades compared with the case of noise-free data, but the method remains quite accurate. The presented method has been compared with the physics-informed neural network method.

ANALISIS PELAKSANAAN KURIKULUM MA PLUS KETRAMPILAN DI MA NEGERI 1 KARANGANYAR

The curriculum is the spirit of educational institutions. 4 curriculum components, namely objectives, material/content, teaching and learning process and evaluation. The implementation of the madrasa curriculum is adjusted to the characteristics of the madrasa. This study aims to determine; Implementation of the 2013 MA Plus Skills Curriculum and Curriculum Components at MAN 1 Karanganyar. This study used a qualitative approach with case study design. The research location is MAN 1 Karanganyar. Research data were collected through observation, in-depth interviews and documentation. Data analysis was carried out with the following procedures: collecting data, reducing, presenting, and verifying data. The results of the study show: (1) It is a research-based madrasa and plus skills, (2) the curriculum implemented is the 2013 curriculum (3) Organizes regular science/IPS/religious classes and skills classes (welding techniques, fashion, machining techniques, motorcycle banking and business techniques, modeling techniques and building information and multimedia);(4) Curriculum implementation focuses on preparing superior, intelligent, skilled, independent, noble and global-minded people (5) Evaluations are carried out regularly to measure the achievement of goals curriculum.

Edge chipping damage in lithium silicate glass-ceramics induced by conventional and ultrasonic vibration-assisted diamond machining.

Diamond machining of lithium silicate glass-ceramics (LS) induces extensive edge chipping damage, detrimentally affecting LS restoration functionality and long-term performance. This study approached novel ultrasonic vibration-assisted machining of pre-crystallized and crystallized LS materials to investigate induced edge chipping damage in comparison with conventional machining. The vibration-assisted diamond machining was conducted using a five-axis ultrasonic high-speed grinding/machining machine at different vibration amplitudes while conventional machining was performed using the same machine without vibration assistance. LS microstructural characterization and phase development were performed using scanning electron microscopy (SEM) and x-ray diffraction (XRD) techniques. Machining-induced edge chipping depths, areas and morphology were also characterized using the SEM and Java-based imaging software. All machining-induced edge chipping damages resulted from brittle fractures. The damage scales, however, depended on the material microstructures; mechanical properties associated with the fracture toughness, critical strain energy release rates, brittleness indices, and machinability indices; and ultrasonic vibration amplitudes. Pre-crystallized LS with more glass matrix and lithium metasilicate crystals yielded respective 1.8 and 1.6 times greater damage depths and specific damage areas than crystallized LS with less glass matrix and tri-crystal phases in conventional machining. Ultrasonic machining at optimized amplitudes diminished such damages by over 50 % in pre-crystallized LS and up to 13 % in crystallized LS. This research highlights that ultrasonic vibration assistance at optimized conditions may advance current dental CAD/CAM machining techniques by significant suppression of edge chipping damage in pre-crystallized LS.

A new multivariate control chart based on the isolation forest algorithm

Statistical process control (SPC) is an extensively utilized technique in quality management for service and manufacturing industries. Traditional SPC control charts rely on normality assumptions and analyze data in one dimension. Machine learning methods have been increasingly used to address limitations of traditional SPC techniques, such as one-class classification (OCC) procedures, using support vector machine (SVM), K-nearest neighbor (KNN), and other techniques. Nonetheless, there are drawbacks associated with these methods. For instance, SVM-based approaches may require significant computational resources, whereas KNN-based approaches may encounter difficulties in identifying minor variations. Here, these limitations are addressed using a new SPC framework that incorporates the isolation forest (iForest) algorithm, an unsupervised anomaly detection technique developed for machine learning applications. Our presented method examines how the number of trees impacts the charting scheme and offers several advantages: (1) it has a low constant time complexity that results in fast computing speeds; (2) it outperforms competitors in detecting small shifts in various distributions; and (3) while dealing with challenging distributions with heavy tails, it proves to be effective. We provide two examples that demonstrate the performance of our proposed control chart using Monte Carlo simulations.

Using supervised machine learning algorithms in pavement degradation monitoring

• Dynamic pavement monitoring system using vibration-based method. • Noise cancelling and vibration data smoothing. • Degradation detection and classification using machine learning. Pavement monitoring plays a vital role in maintaining sustainable road network conditions and provides road users with satisfactory comfort riding. Regular and routine monitoring provide clear information on road conditions and the level of damage to pavement surfaces. In this study, a vibration-based method was used as a monitoring technique to evaluate the pavement surface conditions on local roads. Pre-processing techniques were used to cancel noise and prepare the data for feature extraction and prediction. Two multi-classification Machine Learning (ML) models were used, including Random Forest (RF) and Decision Tree (DT), for the automated classification and detection of different types of pavement distresses. In addition, a Support Vector Machine (SVM) technique was used to develop a binary ML model for the same classification and detection purposes. The results showed that the developed ML models provide high accuracy in predicting the road degradation classification with about 93% accuracy using the RF and 90% accuracy using the DT. Using the SVM model, the overall average accuracy of detection and classification of pavement defects was about 96%.

Application and development of Artificial Neural Networks to predict structural damage subjected to earthquake-A review

The complexity of earthquake incidents supports efforts to use machine learning methods as an alternative to conventional methods because they can simply capture difficult relations between the input and output parameters without requiring a particular functional structure. Machine learning is a branch of artificial intelligence techniques that applies the capacity of machines to recreate intelligent human behavior and has been receiving increasing consideration as an effective solution for damage detection due to earthquakes. Artificial neural networks (ANNs) are one of the most important machine techniques, inspired by the human brain. There hasn’t been a lot of published research on the use of ANNs for structural damage identification to predict earthquakes on various structures up until now. This paper presents a comprehensive review of recent studies on the applications and development of ANNs to predict earthquakes and damage caused by earthquakes on different structures. ANNs can learn from their experience and have drawn significant consideration in damage detection due to their capability of pattern identification and error tolerance in forming a nonlinear modeling between the inputs and outputs. According to this research, the use of ANNs can prevent unpredictable failures and reduce maintenance costs, increasing the safety and functionality of structures. A brief introduction to the ANN algorithm is presented first. Following that, the benefits, and limitations of ANNs, as well as some new research trends, have been discussed.

In pursuit of sustainability in machining thin walled α‑titanium tubes: An industry supported study

Cryogenic machining has emerged as a major sustainable machining techniques especially for difficult-to-cut alloys, including titanium, Inconel alloys in terms of its impact on cost savings for high-value manufacturing projects reduced energy consumption, enhanced worker safety, and eliminating the infrastructure and disposal needed for flood coolants. However, further study is needed to quantify the sustainable elements of machining processes to have a clearer vision for industries to transition from traditional techniques to such advanced techniques. In the currently known literature, the topic of high-speed machining of titanium tubes with thin walls is yet to be explored under advanced cooling technique. The studies on the machinability indicators including tool wear, power consumption etc. are the backbone to the assessment of sustainability aspect of machining. The quantification of sustainability indicators in terms of machining costs and carbon emissions will provide better insights to the industries to maximise their profits and have least negative impacts on the environment. Thus, this study focuses on analyzing sustainability of the machining processes in terms of economic and environmental factors while turning thin walled α‑titanium tubes under different cutting environments which include dry, wet and cryogenic machining under two different high cutting speeds of 150 m/min and 200 m/min. The economical aspect comprises the assessment of the different machining costs occurring during the machining. The environmental aspect includes the assessment of carbon emissions occurring during the machining operation. The backbone of the sustainability analysis is the machinability analysis which involves the investigation of power consumption and cutting tool-wear. A comprehensive study of tool wear and total power consumption has been presented to get a better perspective in the machinability of thin walled α‑titanium tubes while working in the cryogenic environment in comparison with dry and wet environments. The total machining cost is relatively less (approximately 27%) under cryogenic environment in comparison with a wet environment making it a more economical process. The total emissions of carbon are also found to be lowered (up to 9%) while machining under cryogenic environment in comparison with a wet environment making it a more environmentally friendly process. The cutting tool wear is found to be relatively low (up to 38.68% and 72.24%) under cryogenic environment in comparison with wet and dry environments respectively. Power consumption was compared for the three different environments to have detailed analysis in terms of productivity and sustainability. A significant reduction in total power consumption (up to 19.25%) was observed under cryogenic environment in comparison with wet environment.

EFFECT OF ZINC COATED BRASS AND LN2 COOLED ZINC COATED BRASS ELECTRODES IN WEDM PERFORMANCE ON INCOLOY

Due to their superior mechanical qualities, such as corrosion resistance and carburization resistance, particularly at high temperatures, nickel-based superalloys are being used more often in the aviation, marine, and nuclear industries. Because of their great strength, low heat conductivity, and strain hardening properties, nickel-based superalloys are particularly difficult to process conventionally. Wire electric discharge machining, one of the non-traditional machining techniques, is utilised in this study to cut Inconel 800. (Incoloy). Wire cut EDM uses electrodes made of zinc-coated brass wire that have been cryogenically chilled. The RSM approach is used to arrange trials using Servo voltage and Wire feed as constant variables and Wire tension, Pulse on time, and Wire feed as changeable factors. In order to maximizes responses, lower is better and higher is better criteria are used to assess material removal and surface roughness for zinc-treated brass and cryogenically chilled (LN2) zinc-treated brass.

Advanced Marine Craft Model Identification via Multi-Kernel Weighted Least Square Support Vector Machine and Characteristic Model Techniques

This paper combines the piecewise Cubic Hermite (CH) interpolation algorithm and the weighted least square support vector machine (WLS-SVM) to improve identification accuracy for marine crafts built based on the characteristic model. The characteristic model is first used to describe the heading dynamics of marine crafts and is a superior model to the traditional response model in both accuracy and complexity. Especially in order to improve identification accuracy, a CH-based data preprocessing strategy is utilized to densify and smooth data for further accurate identification. Subsequently, the combination of the linear kernel function and the Gaussian kernel function is introduced in the conventional WLS-SVM method, which renders global and local performance improvements compared with the conventional WLS-SVM method. Finally, informative maneuvers composed of Zigzag and Sine are carried out to test the performance of the improved identification method. Compared to the conventional LS-SVM method based on the response model, the root mean square error of the proposed CH-MK-WLS-SVM method based on the characteristic model is reduced by an order of magnitude in the presence of sensor noise.