Search

The quality of machined products is largely dependent on the status of machines in various aspects. Vibration status monitoring is a fundamental step for reliability evaluation, machine fault diagnosis, and prognosis. For achieving such a task, miniature low-cost but effective sensors such as MEMS accelerometers should be used. However, due to the nature of machine tool characteristics, both high sensitivity and bandwidth are required and this imposes a challenging constraint on current MEMS devices. In this work, a tri-axial module by integrating three types and totally five MEMS accelerometers is designed and realized as the first step to realize the measurement module for machine tool applications. The performance of the module has been validated by a piezoelectric accelerometer, and the result indicated that the outputs of the proposed design essentially agree with those obtained by the standard piezoelectric accelerometer very well. Finally, the module has been applied on both a five-axis milling machine and a six-axis robot arms for detecting the process-induced vibration for demonstrating its possible applications in both abnormal condition monitoring and machine prognosis. In the near future, the performance will be improved by reducing noise and the size would be further reduced for fitting into critical location of machine tools. [ABSTRACT FROM AUTHOR]

To fundamentally improve the energy efficiency of computer numerical control (CNC) machine tools, it is essential to study the energy saving in the design stage. The spindle system is one of the most important energy consumers of CNC lathe, and the optimization design for spindle is considered an effective energy saving strategy. However, little research has focused on this issue, and most previous optimization design methods aimed at improving the machining performance such as deformation and vibration. Since previous studies paid little attention to energy saving design, we attempt to consider the energy consumption in the spindle design stage. First, the energy consumption function of the spindle is built, and the optimization objectives are selected. The structure parameters that significantly affect the optimization objectives are determined as the design variables. Then, on the basis of the effective fitting method and dimensionality reduction, a comprehensive objective is defined. Finally, the biogeography-based optimization (BBO) algorithm is adopted to optimize the proposed model. The results indicate that the proposed method can reduce the energy consumption and improve the static and dynamic performance. The impact of the design variables on the objectives is discussed, which provides the guidance for the spindle optimization design. [ABSTRACT FROM AUTHOR]

Ensemble learning is a system that combines a set of base learners to improve the performance in machine learning, where accuracy and diversity of base learners are two important factors. However, these two factors are usually contradictory. To address this problem, in this paper, we propose a novel ensemble learning algorithm based on fitness Euclidean-distance ratio differential evolution, to train the neural network ensemble. FEFERR_ELA employs a multimodal evolutionary algorithm that is capable of producing diverse solutions to search for optimal solutions corresponding to parameters of base learners, where each optimal solution leads to one trained model. A dynamic ensemble selection scheme is applied to select appropriate individuals for the ensemble. The proposed algorithm is evaluated on several benchmark problems and compared with some related ensemble learning models. The experimental results demonstrate that the proposed algorithm outperforms the related works and can produce the neural network ensembles with better generalization. [ABSTRACT FROM AUTHOR]

The geometric complexity of 3D-printed parts increases the difficulty of surface finishing with the conventional grinding technique. In this study, a rotary ultrasonic-assisted abrasive flow finishing (RUA-AFF) method is proposed towards improving the performance of the AFF process by providing rotary motion and ultrasonic vibration to the equipment. Finishing experiments are carried out involving Al6061 to investigate the capabilities of the proposed RUA-AFF. The obtained results show that (1) the higher ultrasonic vibration amplitude results in the greatest improvement in the work surface quality and a higher increase in material removal rate (MRR); (2) with the higher ultrasonic frequency and rotational speed, the surface roughness and MRR are decreased marginally; (3) there are two material remove modes in the RUA-AFF process: abrasive impingement and micro-cutting. These indicate that the high efficiency and high-quality finishing of Al6061 can be performed with the proposed RUA-AFF technique. [ABSTRACT FROM AUTHOR]

Recent transcriptomics and bioinformatics studies have shown that ncRNAs can affect chromosome structure and gene transcription, participate in the epigenetic regulation, and take part in diseases such as tumorigenesis. Biologists have found that most ncRNAs usually work by interacting with the corresponding RNA-binding proteins. Therefore, ncRNA-protein interaction is a very popular study in both the biological and medical fields. However, due to the limitations of manual experiments in the laboratory, machine-learning methods for predicting ncRNA-protein interactions are increasingly favored by the researchers. In this review, we summarize several machine learning predictive models of ncRNA-protein interactions over the past few years, and briefly describe the characteristics of these machine learning models. In order to optimize the performance of machine learning models to better predict ncRNA-protein interactions, we give some promising future computational directions at the end. [ABSTRACT FROM AUTHOR]

Objective: To investigate the effect of coronary calcification morphology and severity on the diagnostic performance of machine learning (ML)–based coronary CT angiography (CCTA)–derived fractional flow reserve (CT-FFR) with FFR as a reference standard. Methods: A total of 442 patients (61.2 ± 9.1 years, 70% men) with 544 vessels who underwent CCTA, ML-based CT-FFR, and invasive FFR from China multicenter CT-FFR study were enrolled. The effect of calcification arc, calcification remodeling index (CRI), and Agatston score (AS) on the diagnostic performance of CT-FFR was investigated. CT-FFR ≤ 0.80 and lumen reduction ≥ 50% determined by CCTA were identified as vessel-specific ischemia with invasive FFR as a reference standard. Results: Compared with invasive FFR, ML-based CT-FFR yielded an overall sensitivity of 0.84, specificity of 0.94, and accuracy of 0.90 in a total of 344 calcification lesions. There was no statistical difference in diagnostic accuracy, sensitivity, or specificity of CT-FFR across different calcification arc, CRI, or AS levels. CT-FFR exhibited improved discrimination of ischemia compared with CCTA alone in lesions with mild-to-moderate calcification (AUC, 0.89 vs. 0.69, p < 0.001) and lesions with CRI ≥ 1 (AUC, 0.89 vs. 0.71, p < 0.001). The diagnostic accuracy and specificity of CT-FFR were higher than CCTA alone in patients and vessels with mid (100 to 299) or high (≥ 300) AS. Conclusion: Coronary calcification morphology and severity did not influence diagnostic performance of CT-FFR in ischemia detection, and CT-FFR showed marked improved discrimination of ischemia compared with CCTA alone in the setting of calcification. Key Points: • CT-FFR provides superior diagnostic performance than CCTA alone regardless of coronary calcification. • No significant differences in the diagnostic performance of CT-FFR were observed in coronary arteries with different coronary calcification arcs and calcified remodeling indexes. • No significant differences in the diagnostic accuracy of CT-FFR were observed in coronary arteries with different coronary calcification score levels. [ABSTRACT FROM AUTHOR]

The present work intends to investigate the application potential of coconut oil, as cutting fluid, during metal machining of a nickel-based superalloy. In doing so, machining performance of difficult-to-cut aerospace superalloy Inconel 718 is studied under dry, minimum quantity lubrication (MQL), and nanofluid MQL (NFMQL) using uncoated WC–Co tool, operated at varied cutting speeds. MQL environment is created by supplying air-oil mist in which commercially available coconut oil is used as cutting fluid. On the other hand, nanofluid is prepared by dispersing a specific concentration of multi-walled carbon nanotubes within coconut oil. Machining performance as observed in MQL and NFMQL is compared to that of dry machining. The following machining performance indicators are considered herein: cutting force magnitude, tool-tip temperature, and width of flank wear progression. In addition, studies on dominant modes and mechanisms of cutting tool wear, chip morphology (macro/micro), and surface roughness of the machined work part are also carried out. In the purview of machined surface roughness, it is concluded that NFMQL performs better than dry machining and conventional MQL. Flank wear is witnessed more severe during dry machining than MQL/NFMQL. However, beyond 83 m/min cutting speed, conventional MQL outperforms NFMQL (with 0.2 wt% nano-additives concentration) machining, from the viewpoint of flank wear width. [ABSTRACT FROM AUTHOR]

Electrochemical discharge machining (ECDM) is a triumphant process for producing micro-holes in glass materials and yet holds the potential for improvement. In ECDM, electrolyte selection is crucial as it controls the spark pattern and flushes debris from the machining zone. Despite numerous experimental studies, few analytical studies were reported for ECDM modeling and finite element modeling-based electrolyte's comparative study is not reported yet. This study focuses on developing a thermal model to compare the ECDM performance in terms of material removal rate (MRR) for different electrolytes, viz. NaOH, KOH, and NaCl. The plots of temperature distributions are obtained underneath the spark and further processed to evaluate the MRR. Predicted results are observed to be in accordance with the experimental results. The effects of electrolyte concentration, applied voltage, spark radius, and duty ratio on MRR are analyzed through simulations. Results show that MRR decreases with the increase in spark radius while increases with the increase in other parameters. NaOH electrolyte produces higher MRR; an increase of 8.72 mg and 10.18 mg is observed when compared to KOH and NaCl at 60 wt% concentration. Moreover, experimental studies are performed to evaluate the electrolyte's effect (includes NaNO3) on radial overcut (ROC) and hole circularity. KOH electrolyte provides the least overcut and better hole circularity due to stable and smooth spark pattern. An improvement of 32.1% in overcut and 40.8% in hole circularity was observed using the KOH when compared to NaNO3. NaOH and KOH are revealed as the successful electrolytes for machining with ECDM. [ABSTRACT FROM AUTHOR]

Malaria is a disease affecting hundreds of millions of people across the world, mainly in developing countries and especially in sub-Saharan Africa. It is the cause of hundreds of thousands of deaths each year and there is an ever-present need to identify and develop effective new therapies to tackle the disease and overcome increasing drug resistance. Here, we extend a previous study in which a number of partners collaborated to develop a consensus in silico model that can be used to identify novel molecules that may have antimalarial properties. The performance of machine learning methods generally improves with the number of data points available for training. One practical challenge in building large training sets is that the data are often proprietary and cannot be straightforwardly integrated. Here, this was addressed by sharing QSAR models, each built on a private data set. We describe the development of an open-source software platform for creating such models, a comprehensive evaluation of methods to create a single consensus model and a web platform called MAIP available at https://www.ebi.ac.uk/chembl/maip/. MAIP is freely available for the wider community to make large-scale predictions of potential malaria inhibiting compounds. This project also highlights some of the practical challenges in reproducing published computational methods and the opportunities that open-source software can offer to the community. [ABSTRACT FROM AUTHOR]

Disc cutters are the most commonly used cutting tools in the hard rock tunneling by tunnel boring machines (TBMs). During operation of a TBM in any project, disc cutters wear out and should be replaced to maintain high machine performance. This is especially true in very strong and abrasive rocks. In the process of planning and estimation of completion time and cost for a given tunneling project, the lifetime of the disc cutter is often predicted using the empirical models due to their simplicity and availability. But understanding the background of the models and the relative difference between both predicted and actual lifetimes is critical for improving the accuracy of estimated cutter life. In this study, the basis and input parameters of the most frequently used empirical models for prediction of disc wear life are examined. The result of estimating cutter life by using these models is compared in similar working conditions, including TBM specifications, operating conditions, and rock abrasivity. The results show that in slightly abrasive rocks, the empirical models should be carefully chosen, because the estimated disc life may spread over a wide range. Furthermore, some models are very conservative, while others offer more compatible estimates with the rock abrasivity. Also, in the Kerman water conveyance tunnel, the predicted and actual cutter lives show good correlation with the ratio of the rock mass and intact rock strengths, in addition to laboratory measures of rock abrasivity. [ABSTRACT FROM AUTHOR]

In this paper, after a brief description of engineering geological properties of Ghomrood water conveyance tunnel project (lots 3 and 4 with a total length of 18 km), the actual performance of the machine during the excavation of tunnel through sedimentary and low to medium grade metamorphic rocks has been analyzed. Subsequently, using data collected during pre-construction and construction phases of the project, the relationships between rock mass properties and measured machine performance parameters have been investigated and proposed as a prediction model which can be used in future projects in which similar geological conditions are encountered. Results of analyses indicated that there are strong relationships between TBM performance and engineering geological conditions of the tunnel alignment. [ABSTRACT FROM AUTHOR]

Amongst all the working elements of single grip harvesters, the working elements of felling and processing play an important role within the cut-to-length (CTL) harvester's working phases. This includes the felling cutting and cross-cutting of stems within the felling and processing operation. The detailed investigation of such individual machine activities may help to analyse and improve the performance of forest machines. The objective of this study was to investigate the cutting duration and other performance parameters, including the fuel consumption, of a CTL harvester's sawing unit under real working conditions. Detailed information on the felling cutting and cross-cutting performance was collected at short intervals using CAN bus data of two single grip harvesters in final felling from two different sites in Eastern Finland. As a result, models for effective time consumption in the work phase of cutting as a function of stem size were developed, both for felling cutting and cross-cutting. Felling cutting and cross-cutting durations were somewhat identical until the cutting diameters of 400–450 mm, depending on the site. Thereafter, the cutting time difference increased and was higher in felling cutting. At the site with large diameters of 550–650 mm, the difference varied between 15% and 28%, between the comparisons of formulae. In addition, other performance parameters, including the respective fuel consumption of this working phase, were part of the study. The study revealed a higher hourly based fuel consumption for the entire guide bar movement time compared to the pure cutting time in cross-cutting with stem diameters below 400 mm. Detailed knowledge of the performance of the sawing unit's activities might help the planning process of future studies, as well as support the future development of efficient and intelligent machinery. [ABSTRACT FROM AUTHOR]

Silicon carbide particle-reinforced aluminum matrix composites (SiCp/Al) are increasingly applied in high-end industries because of their excellent comprehensive material properties. However, as a typical difficult-to-cut material, machining of SiCp/Al using traditional processes faces serious challenges. Jet electrochemical machining (Jet-ECM) is a new-generation electrochemical machining (ECM) technology, which exhibits improved process flexibility and machining accuracy, in addition to the basic advantages of the ECM process. Jet-ECM appears to be a potential alternative machining technology for SiCp/Al to achieve high machining rate and damage-free surface. In the present work, the polarization curves obtained by electrochemistry study showed that a passivation-tanspassivation transition behavior with increasing the voltage occurs in NaNO3 electrolyte, while NaCl electrolyte causes active behavior. Thus, compared to NaCl electrolytes, the NaNO3 electrolyte is relatively suitable for Jet-ECM of SiCp/Al because of its potential to improve the machining localization. Besides, the effects of process parameters (including voltage, electrolyte type, and concentration) on the machining performance were systematically investigated. It was found that the voltage has the most significant influence on the machined dimple profile. A higher voltage helps in achieving a large depth. The surface roughness is dependent on the selection of the electrolyte and voltage. The NaNO3 electrolyte is preferred to realize smooth surface. Additionally, it was found that with increasing voltage, the surface roughness increases. Finally, this work was extended to be Jet-ECM with linear movement of the nozzle, and well-formed SiCp/Al grooves can be obtained using the selected process parameters. [ABSTRACT FROM AUTHOR]

Following the machine tool with demands of high-speed, high-precision machining and multi-axis, the heat generation of machine tool will lead to thermo growth of spindle, thermal error, etc. Therefore, the machine tool oil cooler has been frequently adopted for thermal management of heat generation for machine tool. However, it will be quite challenging to develop high-precision machining without suitable temperature control scheme. And, the machine tool oil cooler with on-off control is difficult to control the oil temperature accurately. Therefore, in this paper, the machine tool oil cooler with hot-gas bypass temperature control scheme and inverter temperature control scheme based on the PID control logic have been proposed to solve the oil temperature control problem for machine tool oil cooler. Furthermore, the temperature control performance (i.e. thermo growth of spindle, oil temperature control accuracy, and EER) of hot-gas bypass temperature control scheme and inverter temperature control scheme has been compared by means of experiment. The experimental results shows that the inlet oil temperature control accuracy can be maintained ±0.2 °C during spindle operation by using the hot-gas bypass temperature control scheme (HGB cooler) and inverter temperature control scheme (INV cooler). Furthermore, the thermo growth of spindle of HGB cooler is about 14 % smaller than the INV cooler because of hot-gas bypass temperature control scheme with thermal compensation function. However, the total power consumption of HGB cooler is about 0.18 kW higher than the INV cooler. Hence, the energy efficiency ratio (EER) of INV cooler reveals about 10.9 % higher than the HGB cooler at steady state. Furthermore, when oil flow rate declined to about half, the temperature difference between initial and final steady state temperature of the spindle is increased around 94.3 %, and the thermo growth of spindle is increased 58.7 %. [ABSTRACT FROM AUTHOR]

The reliability of boom crane structure determines the safety and performance of the whole machine. Due to the sample information of uncertain parameter is lacking, non-probabilistic reliability of boom crane structure under the condition of limited samples is studied. Considering the time dependent deterioration of structure resistance and the load time effect, the time-dependent uncertain parameters of boom structures are comprehensively analyzed. A time-dependent reliability model of boom structures based on interval non-probabilistic is proposed. Based on the theory of interval non-probabilistic time-varying reliability, an engineering example is analyzed by using VC++ programming. The influence of different weight and uncertain parameters on the structure reliability of the boom is analyzed. The results show that the resistance, attenuation coefficient, average resistance and non-probabilistic time-varying reliability indicators of the boom structure attenuate seriously after decades. The conclusion verifies the scientific of the proposed method on the time accumulation effect. [ABSTRACT FROM AUTHOR]

The application of cutting fluids to the machining of titanium alloys has received increasing interest due to the severe tribological conditions occurring during this process. Cutting fluids are applied during the machining process to reduce tool wear, friction, and heat generation. This study was carried out to investigate the effect of polymer-based additives and phosphorus-based additives on the cutting fluid performance during the drilling of a titanium alloy, Ti-6Al-4V, under a constant metal removal rate (MRR) of 4.2 mm3/s. The tool wear and specific cutting energy (SCE) were evaluated to assess the performance of the additives in a cutting fluid. The performance of the additives was dependent on the drilling conditions. At the lower limit of the drilling conditions tested (500 RPM under 4.2 mm3/s MRR), the cutting fluid with high phosphorus concentration was able to achieve better lubricity. While at the higher limit drilling conditions (700–2500 RPM under 4.2 mm3/s MRR), the cutting fluid with primarily polymer-based additives was more effective at providing improved drilling performance. The improvement in the drilling performance of the cutting fluids was related to the formation of boundary layers associated with additives employed. An optimum range (providing the lowest SCE value) within the tested drilling conditions was noted for the cutting fluids. [ABSTRACT FROM AUTHOR]

The micro-hole array machined by microelectrical discharge machining (micro-EDM) has been widely used in industrial fields, such as inkjet nozzle, oil nozzle, spinneret, and so on. However, it is difficult to guarantee the dimensional consistency accuracy of micro-hole array. In the present study, the characteristic and compensation strategy of microelectrode wear were explored to meet the requirements of dimensional consistency accuracy of micro-hole array machining. The effect of pulse generator parameters on microelectrode wear was studied, and the microelectrode wear was reduced by optimizing pulse generator parameters. The geometry evolution rule of microelectrode was investigated, and the microelectrode wear model was established based on the wear characteristic of microelectrode. By analyzing the wear characteristic of microelectrode, single-hole quantitative compensation strategy was adopted during the machining process. Meanwhile, grouping detection variable compensation strategy was proposed to adjust axial position of microelectrode periodically in order to eliminate the cumulative compensation errors. The array of 256 (16 × 16) micro-holes with an average diameter of 45.28 μm was machined on a 50-μm-thick stainless steel. The dimensional consistency accuracy of micro-hole array was assured by showing a standard deviation of 0.15 μm. In addition, the array of 64 (8 × 8) micro-holes and the array of 25 (5 × 5) micro-holes with an average diameter of less than 20 μm were machined to explore the performance of micro-hole array machining with smaller size, and the dimensional consistency accuracy was shown by the standard deviation of 0.25 and 0.4 μm, respectively. [ABSTRACT FROM AUTHOR]

Tunnel boring machine (TBM) performance prediction is often a critical issue in the early stage of a tunnelling project, mainly due to the unpredictable nature of some important factors affecting the machine performance. In this regard, deterministic approaches are normally employed, providing results in terms of average values expected for the TBM performance. Stochastic approaches would offer improvement over deterministic methods, taking into account the parameter variability; however, their use is limited, since the level of information required is often not available. In this study, the data provided by the excavation of the Maddalena exploratory tunnel were used to predict the net and overall TBM performance for a 2.96 km section of the Mont Cenis base tunnel by using a stochastic approach. The preliminary design of the TBM cutterhead was carried out. A prediction model based on field penetration index, machine operating level and utilization factor was adopted. The variability of the parameters involved was analysed. A procedure to take into account the correlation between the input variables was described. The probability of occurrence of the outcomes was evaluated, and the total excavation time expected for the tunnel section analysed was calculated. [ABSTRACT FROM AUTHOR]

This research aimed to describe a fuzzy optimal energy control strategy, so the electrical discharge machining (EDM) process can be optimized as a whole. The purpose is to develop a fuzzy logic-based multi-input–multi-output optimal control system to characterize the discharge energy efficiency of the EDM process. Methodology used in this paper is specific experimental design covering final and rough operations. Two controllable parameters were selected, and they were discharge current and pulse duration. The discharge energy and material removal rate were chosen as quality characteristics. The proposed approach uses the concept of fuzzy optimal control which is formulated on the basis of fuzzy logic technology with the use of knowledge-based system control. The finding that resulted in this research is that the effect of the discharge energy developed was observed on the performance of material removal rate. The material removal of the electrical discharge is directly proportional to the amount of the discharge energy applied during the on time of the pulse. The experimental results revealed that the effect of the EDM is increased as electrical discharge input parameters increase, but their interaction is limited and still somewhat unclear. The fuzzy optimal control model provides a highly advantageous option of the EDM input parameters, to achieve the best machining performance: the minimum discharge energy and the maximum material removal rate. [ABSTRACT FROM AUTHOR]

The present study examines dry machining performance of AA7075 T6 alloy using uncoated and MT-CVD TiCN–Al2O3-coated carbide inserts. Machining performance is assessed with regard to tangential cutting force, tool tip temperature and depth of flank wear. The performance of coated tool is compared to that of uncoated insert. In addition, different modes of tool wear and chip morphology are studied in detail. It is experienced that coated tool causes lower tool tip temperature and lesser flank wear than untreated counterpart. Apart from abrasion, adhesion and built-up layer formation; attrition wear is distinctly visible in case of uncoated tool. On the contrary, in addition to common wear modes, coated tool also experiences diffusion wear. [ABSTRACT FROM AUTHOR]

In metal machining, minimum quantity lubrication (MQL) refers to supply of an optimal amount of cutting fluid into tool-work interfacial region. As compared to traditional dry machining, machining process executed under MQL has several advantages. Properties of cutting fluid play important role toward machining performance. In the present work, application potential of rice bran oil is compared with coconut oil in the context of MQL machining of AISI 304 stainless steel. Cutting force magnitude, approximate tool-tip temperature, flank wear depth, etc., are studied as machining performance indicators. In addition, different tool wear mechanisms, chip morphology, and surface roughness of the machined work part are studied in detail. It is experienced that MQL (rice bran oil) outperforms MQL (coconut oil). MQL (rice bran oil) exhibits lower cutting force, lesser tool-tip temperature, reduced extent of tool wear, and better machined surface finish than dry as well as MQL (coconut oil) machining. [ABSTRACT FROM AUTHOR]

The droplet dynamic behavior of spraying during the manufacturing process affects tool life, cooling, and lubrication. A detailed understanding of droplet dynamic behavior is required to improve the surface quality while reducing the cutting fluid consumption. In this study, quantitative measurements for droplet diameter, number concentration, and axial velocity were carried out through a phase Doppler particle analyzer technique. The droplet characteristic spatial distributions of a standard twin-fluid nozzle (STN) and a novel twin-fluid nozzle (NTN) under different gas-to-liquid mass ratio (GLR) values were analyzed in detail. The results showed that the high-speed internal gas flow of an atomizing core has a significant effect on improving the synergistically enhanced atomization behavior, and more smaller droplets and an increased droplet momentum are easily obtained. Furthermore, a better spatial distribution of atomization characteristics can be achieved under a higher GLR of 4.3%. Furthermore, the effects of the novel atomizing core structural parameters on the droplet mean velocity spatial distribution, droplet turbulence, and root mean square velocity fluctuation were investigated. The results also showed that the improved NTN can effectively improve the spray atomization performance and droplet dynamic characteristics. In addition, comparative studies of the effect of the atomizing core structure on the machining performance were performed. Compared with the STN, the improved NTN achieves a significant reduction in cutting temperature, tool wear, and surface roughness of 17.25%, 33.96%, and 37.83%, respectively, owing to the better droplet dynamic characteristics and spatial distribution of spray atomization characteristics. [ABSTRACT FROM AUTHOR]

In recent years, machine learning has developed rapidly, enabling the development of applications with high levels of recognition accuracy relating to the use of speech and images. However, other types of data to which these models can be applied have not yet been explored as thoroughly. Labelling is an indispensable stage of data pre-processing that can be particularly challenging, especially when applied to single or multi-model real-time sensor data collection approaches. Currently, real-time sensor data labelling is an unwieldy process, with a limited range of tools available and poor performance characteristics, which can lead to the performance of the machine learning models being compromised. In this paper, we introduce new techniques for labelling at the point of collection coupled with a pilot study and a systematic performance comparison of two popular types of deep neural networks running on five custom built devices and a comparative mobile app (68.5–89% accuracy within-device GRU model, 92.8% highest LSTM model accuracy). These devices are designed to enable real-time labelling with various buttons, slide potentiometer and force sensors. This exploratory work illustrates several key features that inform the design of data collection tools that can help researchers select and apply appropriate labelling techniques to their work. We also identify common bottlenecks in each architecture and provide field tested guidelines to assist in building adaptive, high-performance edge solutions. [ABSTRACT FROM AUTHOR]

Machine translation has become an irreplaceable application in the use of mobile phones. However, the current mainstream neural machine translation models depend on continuously increasing the amount of parameters to achieve better performance, which is not applicable to the mobile phone. In this paper, we improve the performance of neural machine translation (NMT) with shallow syntax (e.g., POS tag) of target language, which has better accuracy and latency than deep syntax such as dependency parsing. In particular, our models take less parameters and runtime than other complex machine translation models, making mobile applications possible. In detail, we present three RNN-based NMT decoding models (independent decoder, gates shared decoder and fully shared decoder) to jointly predict target word and POS tag sequences. Experiments on Chinese-English and German-English translation tasks show that the fully shared decoder can acquire the best performance, which increases the BLEU score by 1.4 and 2.25 points respectively compared with the attention-based NMT model. In addition, we extend the idea to transformer-based models, and the experimental results also show that the BLEU score is further improved. [ABSTRACT FROM AUTHOR]

Electrical discharge machining (EDM) is a non-conventional machining process which has been widely used to machine difficult-to-cut materials. However, the efficiency of EDM hole making processes is low, and the quality of drilled holes is a concern especially when machining deep blind holes due to the poor removal of debris in the discharging gap. In this study, a new strategy based on shape modification of the electrode was developed to address this issue by indirectly changing the flow of dielectric. The electrodes of five different shapes with identical machining area were applied to improve the debris removal during the machining process. The effects of various electrode profiles on the machining performance were investigated. Machine time, material removal rate (MRR) and electrode wear rate (EWR) were used to evaluate the machining performance. The computational fluid dynamic (CFD) software Fluent was used to simulate the flow of dielectric caused by different electrode profiles. The results show that, compared with the straight electrode, the maximum effective cutting depth of curved electrodes and long-curved electrodes increased 13.793% and 132.184%, respectively, and the MRR of stepped electrodes was significantly higher after the machining depth reached 2 mm and 4 mm for electrodes with the working length of 2 mm and 4 mm, respectively. The surface roughness was identical for holes machined by all types of electrode. However, obvious carbon adhesion can be observed at lower cutting depth for straight electrode, followed by curved and long-curved electrodes. [ABSTRACT FROM AUTHOR]

Compared with minimum quantity lubrication, electrostatic spraying can greatly improve machining and environmental performances. However, the cooling and lubricating performance of electrostatic spraying needs to be further improved. In this paper, the composite electrostatic spray cutting approach is proposed, and the environment-friendly machining of titanium alloy by using this approach is investigated. An atomization test device for composite electrostatic spray cutting and a composite electrostatic spray milling system are developed. The atomization tests for composite electrostatic spray cutting are performed to study its atomization morphology and to determine the atomization parameters that guarantee a stable composite electrostatic spraying. The atomization current is measured by using a picoammeter under stable composite electrostatic spraying conditions, and the charging performance and atomizing stability of composite electrostatic spraying are evaluated. Milling experiments of titanium alloy and ambient oil mist concentration tests are also carried out under electrostatic and composite electrostatic spraying conditions. The effects of composite electrostatic spraying on milling force, tool wear, chip morphology, and oil mist concentration are analyzed. Results show that compared with electrostatic spraying, composite electrostatic spraying can effectively reduce tool wear and oil mist concentration. Under composite electrostatic spraying condition, an increase in the flow rates of internal and external fluids can further reduce tool wear. An increase in the flow rate of internal fluid does not affect ambient oil mist concentration. But an increase in the flow rate of external fluid significantly increases ambient oil mist concentration. [ABSTRACT FROM AUTHOR]

The ongoing research is hard to predict the machining accuracy of machine tools in the preliminary design phase. Hence, a rapid and quantitative prediction method seems very necessary and imperative. In this study, a new machining accuracy prediction strategy for a five-axis machine tool (FAMT) based on tolerance is proposed. Firstly, a characterization model of surface error profiles for key parts of the FAMT is established, and GESP prediction models based on tolerance for a translational axis and a rotary axis are established by utilizing the mapping relationship between tolerance, surface error profiles, and GESPs of key parts for the FAMT. Then, the kinematic equation of the FAMT is established based on GESPs by using the multi-body system (MBS) theory. As a basis, the actual machining position for the cone frustum based on GESPs by considering the tool setting position and the relative NC machining instruction are solved. And, a machining accuracy prediction model for the cone frustum based on tolerance is established. To enhance the prediction efficiency and intuitively evaluate the machining performance of the FAMT, some accuracy indexes of the cone frustum, such as roundness, angularity, and concentricity, are described. Thus, the machining accuracy prediction model based on key parts' tolerance parameters of the FAMT is established substantially. Finally, a simulation prediction scheme is conducted. Simulation results show that predicted machining accuracy can meet the requirements of accuracy. To further demonstrate the practicability of the method, experimental verification is implemented. Experimental results show measuring values are very close to predictive ones. Therefore, a reasonable conclusion can be drawn that the proposed method can quickly and effectively predict the machining accuracy of machine tools based on tolerance. [ABSTRACT FROM AUTHOR]

We study the equipment developed at the AO AKhK VNIIMETMASh and intended for the built-in electromagnetic stirring in the molds of bloom CCM performed with an aim to improve the quality of manufactured tubular billets. [ABSTRACT FROM AUTHOR]

In this work, a cascaded doubly fed induction machine with a cage rotor is studied. To investigate the machine performance, a model is provided. In this model, the spatial harmonics of the rotor magnetomotive force are exactly considered. For this purpose, each rotor loop is selected as an individual circuit. The winding function theory is applied to compute the self- and mutual inductances of the rotor and stator circuits. To reduce the required simulation time, complex vector representation is adopted. Using the obtained voltage equations, the vector form of the machine state-space equations is yielded. Experimental tests are carried out to show the accuracy of the provided model. In addition, the adequacy of the proposed machine is verified by the appropriate tests. [ABSTRACT FROM AUTHOR]

The overhead of data transfer to the GPU poses a bottleneck for the performance of CUDA programs. The accurate prediction of data transfer time is quite effective in improving the performance of GPU analytical modeling, the prediction accuracy of kernel performance, and the composition of the CPU with the GPU for solving computational problems. For estimating the data transfer time between the CPU and the GPU, the current study employs three machine learning-based models and a new analytical model called λ -Model. These models run on four GPUs from different NVIDIA architectures and their performance is compared. The practical results show that the λ -Model is able to anticipate the transmission of large-sized data with a maximum error of 1.643%, which offers better performance than that of machine learning methods. As for the transmission of small-sized data, machine learning-based methods provide better performance and a predicted data transfer time with a maximum error of 4.52%. Consequently, the current study recommends a hybrid model, that is, the λ -Model for large-sized data and machine learning tools for small-sized data. [ABSTRACT FROM AUTHOR]

In a large-scale data center, it is vital to precisely recognize the termination statuses of applications at an early stage. In recent years, many machine learning techniques have been applied to this issue, which is beneficial for optimizing the scheduling policy and improving the efficiency of resource utilization. However, if the application's dynamic information is insufficient at the early stage, the generalization performance of the machine learning model will be lessened, and the prediction accuracy could be low. To overcome this problem, a novel failure prediction method that is based on the association relationships between similar jobs is proposed in this paper to jointly predict task's termination statuses at an earlier stage. The similar jobs whose tasks have similar changing modes of consumed resources, an inherent structural correlation may exist, and the correlation information is significant for improving the prediction model's generalization performance. First, a job clustering algorithm is proposed for identifying the jobs with higher similarity from jobs that have various numbers of tasks. Second, based on the job clustering results, the robust multi-task learning algorithm is introduced to effectively utilize the domain information among jobs (i.e. interactional relationship among jobs on the termination statuses of task). Experiments are conducted on a Google cluster workload traces dataset. The results show that the proposed method can realize higher prediction accuracy, lower misjudgment rate, and higher predictive stability than several state-of-the-art methods at 1/3 the running time of the tasks. [ABSTRACT FROM AUTHOR]

Thermal behaviors significantly affect the contact stiffness and position accuracy of the rolling linear guide, which worsen the performance of the precision machinery. In this paper, based on the modified lumped capacitance method (MLCM) and finite element method (FEM), a thermal model of rolling linear guides has been formulated in terms of the geometric parameters, friction, gyroscopic moment, and loading conditions. Experimental results show that the thermal model can well predict the real-time temperature variation under different working conditions. Further analysis shows that, the applied vertical load, carriage velocity, and toque load significantly influence the temperature variation of linear guides, while the influence of the torque load on temperature variation decreases with increasing vertical load levels. Besides, the reasons of the steady-state temperature variation considering the load distribution model are also discussed. [ABSTRACT FROM AUTHOR]

In recent years, there is an increasing demand for fabricating microstructures such as micro grooves, micro holes, micro dimples, and micro channels on various materials. Micro machining is an alternative method to manufacture microstructures due to its high efficiency and flexibility, and the micro end mill preparation is an essential step. In this study, the low-speed wire electrical discharge machining (LS-WEDM) is used to fabricate ball micro end mills with four cutting edges and radius of less than 200 μm, and the optimized parameters and multiple cutting strategies are adopted to improve machining surface quality and edge sharpness of ball micro end mills. Then, the micro dimples on soft and hard materials are formed by the designed and prepared ball micro end mills made of cemented carbide, respectively. The micro machining performance of ball micro end mills was evaluated by investigating the surface quality and microstructure of micro dimples, cutting force, and tool wear, and the influence of micro machining parameters on micro dimples is discussed in detail. Experimental results indicated that tool wear patterns mainly are adhesion friction and cutting edge damage, and the LS-WEDM-fabricated ball micro end mills with four cutting edges are potential to manufacture microstructures on hard and soft materials with acceptable surface quality and tool wear. [ABSTRACT FROM AUTHOR]

The nature of electrical discharge machining makes it difficult to predict or even measure machining performance, which is why great attention has been paid to methodologies for measuring these performances. In this work, an experimental approach for measuring machining electrical discharge performance and geometric errors was presented. This can improve the authenticity of the parameters measured. A technique for identifying machining parameters using a multi-output system based on fuzzy logic has been proposed. The objective was to determine the influence of machining parameters on the machining performance and associated geometric errors. It is shown that the fuzzy model is capable of giving results providing a good correlation between the real and predicted values. The average error of the model was approximately 1.51% for material removal rate, 3.386% for tool wear rate, 2.924% for wear rate, 5.285% for surface roughness, 4.004% for radial overcut, 4.381% for circularity, and 2.937% for cylindricity. [ABSTRACT FROM AUTHOR]

Recently, sparse representation and dictionary learning have shown significant performance in machine vision. In particular, several supervised dictionary learning methods have been proposed for classification aim and increasing its accuracy. Among them, structured dictionary learning is an interesting approach which captures the discriminative properties of each class and common features among all classes in class-specific sub-dictionaries and a distinct shared sub-dictionary, respectively. It extracts the structural information that exists in samples of each class to increase the classification accuracy. Therefore, in this paper, a group-based structured dictionary learning method is proposed that captures structural information in each class and learns class-specific and shared sub-dictionaries based on mixed l 2 , 1 norm. Also, mixed l 2 , 1 norm is used for acquiring the sparse coefficients of data samples based on the learned sub-dictionaries. Then, classification is done by finding the class with (1) minimum reconstruction error or (2) maximum number of nonzero groups based on l 1 , 0 norm. The proposed method is evaluated by conducting experiments on Extended YaleB, AR and CMU-PIE face databases and the USPS handwritten digits database. The experimental results demonstrate the effectiveness of the proposed method in data representation and classification. [ABSTRACT FROM AUTHOR]

Low commodity prices have forced many mining companies to explore new strategies to minimize operating costs. One cost-saving strategy is to increase drilling efficiency and performance, specifically in open pit mines; drilling operations are expensive, and they directly and indirectly affect most aspects of the mining process. A substantial portion of drilling costs is associated with drill bit consumption due to bit wear. Bit wear decreases the rate of penetration, which reduces drilling efficiency, but changing the bit unnecessarily early increases drilling costs. In addition to this decision-making dilemma, the inventory management strategy is also crucial to cost minimization. The innovative aspect of this paper is to propose an approach for determining the number of bits required in a given period and compute the number of holes to be drilled in drilling operations through reliability analysis and discrete event simulation under uncertainty. Reliability analysis is an effective method to monitor the efficiency and performance of mining equipment and ensure that performance goals and quality criteria are met. In this paper, reliability analysis of drilling machines and drill bits was performed and the relationship between reliability and machine performance was established. The range of drillable holes was also determined. Results of 100 simulations showed that the proposed approach can be an effective tool to facilitate production scheduling and asset management. [ABSTRACT FROM AUTHOR]

Many techniques have been developed to improve the machinability of aeronautical materials titanium and nickel-based alloys such as ultrasonic-assisted turning, laser-assisted turning, and cryogenic-assisted turning. This collaborative scientific investigation presents the steps taken to gain insight into the phenomena of machining Nimonic 90 (a nickel-based alloy) alloy using ultrasonically assisted turning. The cutting speed, feed rate, depth of cut, and frequency are taken as input parameters and average surface roughness (Ra), power consumption (P), and chip formation are considered as output parameters. The experiments are carried out with the full factorial design. The UAT (ultrasonically assisted turning) process gives a significant improvement in average surface roughness and power consumption because of the intermittent cutting action of the cutting tool. UAT process shows a 70–80% reduction in average surface roughness (Ra) and a 6–15% reduction in power consumption as compared with CT (conventional turning) process. Ultrasonically assisted turning also resulted in the thin and smoother chips as compared with CT process which helps to achieve a more superior machining effect. Finite element modeling shows that the quasi-static nature of the stress induced in the UAT process leads to lower force and ultimately lower power generation. Moreover, a sustainability assessment model is implemented to investigate the effect of UAT in terms of machining performance as well as sustainability effectiveness in a single integrated approach. The novelty of this work lies in providing an integrated concept that combines experimental analysis and sustainability assessment when using ultrasonic vibrational energy during turning of Nimonic 90. [ABSTRACT FROM AUTHOR]

In EDM drilling, the accumulation of debris is the main factor leading to abnormal discharges and process instability, which prevent the processing of microholes with a high aspect ratio. To fabricate microholes with an ultrahigh aspect ratio, bubbles produced by an electrochemical reaction were applied to assist electrical discharge machining (B-EDM) in tungsten materials. The mechanism and machining performances of the B-EDM method were investigated. The results showed that bubbles produced by electrochemical reaction on both the surface of the electrode and the workpiece prevented the accumulation of debris at the bottom of the hole, which improved machining stability. This process is effective in enhancing drilling performance in deep microholes by improving such parameters as the material removal rate, electrode tool wear, longitudinal section topography, and achieved aspect ratio. Experimental results showed that the processing time and electrode wear of B-EDM were approximately one third those of EDM in the processing of 20 mm through-holes. A hole with a diameter of 0.22 mm and depth of 120 mm was successfully fabricated by B-EDM on tungsten within 130 min. [ABSTRACT FROM AUTHOR]

Inconel-800 is a superalloy having moderate strength, sufficient resistance to oxidation at high temperature and carburization. Work hardening nature of the material makes it very hard to machine using traditional methods. Wire-cut electrical discharge machining (WEDM) is most widely applied for difficult-to-machine materials because of its capability to produce jobs with minute accuracy and precision. Time taken to process job is the major drawback of the WEDM technique. Therefore, the need arose for formulating a stable predictive model which reduces human effort and maximizes production. The comprehensive intention of this study is to develop robust and stable multi-objective decision-making and soft computing models for selecting the best machining parameters and forecasting the machining performance of Inconel-800 superalloy. Taguchi's method is applied for designing the experimental phase of the study. The computational part is a three-stage process. The first stage is selecting the best machining parameters of WEDM by multi-criteria decision-making (MCDM) technique. A novel MCDM technique is proposed that is based on the concept of risk minimization. The second stage involves the formulation of predictive model using soft computing technique like response surface methodology (RSM). The performance of the technique was scrutinized on a statistical platform to check the gullibility of the model. The final stage is the trade-off analysis by multi-objective RSM (MORSM) which strategically draws balance among the multiple outputs of the problem. Finally, the proposed MCDM and MORSM models are applied for selecting the best and optimal combination of machining Inconel-800 by WEDM technique. [ABSTRACT FROM AUTHOR]

The static stiffness of machine tools plays an important role in ensuring the performance of machine tools and improving the machining accuracy. In the design stage, stiffness matching provides a theoretical basis for determining the stiffness of each component. This study proposed a stiffness matching design method for machine tools. First, the deformation model of machine tools is established on the basis of the state space model. Through force analysis, the relationships between the deformations and stiffness of each part are determined. Second, the stiffness model of the machine tool is established by combining these relationships with the deformation model. Third, the objective function is determined by performing sensitivity analysis after obtaining the stiffness range. The stiffness matching design results are obtained through multi-objective linear programming. Lastly, the stiffness simulations, theoretical model, and experimental results are analyzed. The maximum errors among the experimental, simulation, and theoretical results (18.8 % and 19.9 %) are within the acceptable range. In conclusion, the proposed method is suitable for designing ball screw feed drive systems that requires rigidity. [ABSTRACT FROM AUTHOR]

The electrochemical micromachining (ECMM) system is a modern equipment well thought out as the phenomenon of conventional electrochemical machining (ECM) system in microscale. The ECM system and ECMM system are similar, but the practice method and mechanism involved in the ECM system are analogous to those of the ECMM system, i.e., the differences in machining conditions and machining performance give rise to a factor called the scale effect. In this paper, ECMM has been applied with the advantageous concept called scale effect. Single factor experiments are conducted in which voltage, duty ratio, and tool feed rate are considered as the input factors for the exploration of the different grades of scale effect through the material removal rate, circularity, and overcut. The scale effects that occurred during machining are found using the similarity precision of the micro- and macrosystems based on the similarity theory. Quantitative evaluation of the scale effects is done for the ECMM process. The evaluation showed greater similarity precision indicating a larger significant scale effect. The output values of MRR, circularity, and overcut were normalized which also showed a positive scale effect. The research on scale effects delivers benefits in engineering such as optimizing the processing parameters involved and improving the machining performances of ECMM. [ABSTRACT FROM AUTHOR]

This study is focused on developing and fabricating ball micro end mills with spiral blades by using low speed wire electrical discharge machining (LS-WEDM). Firstly, the new type ball micro end mill is designed and the mathematical model of wire movement trajectory is established for optimizing machining parameters and realizing visual prediction. Then, the diameter, spiral width, and the length consistent accuracy of ball micro end mills fabricated by LS-WEDM is investigated in detail, and more importantly, the three-spiral blades ball micro end mill with the diameter of about 278 μm and the length of 1860 μm is firstly and efficiently manufactured by LS-WEDM. Besides, the micro machining performance and tool wear of LS-WEDM fabricated ball micro end mill are evaluated by conducting micro milling experiments on nickel-based single-crystal superalloy DD5 material, and experimental results revealed that the main tool wear of the new type ball micro end mill is friction wear and adhesion wear rather than diffusion wear. Furthermore, the surface roughness of machined surface can be reduced to 183 nm, which indicates that the ball micro end mills with spiral blades fabricated by LS-WEDM are significantly potential to achieve high-quality machining for complex micro structures on hard material. [ABSTRACT FROM AUTHOR]

Host-side flash caching emerges as an effective solution for improving the performance of virtual machines (VMs) in cloud computing environment. However, for VMs with the commonly used Copy-on-Write (COW) virtual disks, flash caching in fact has negative impacts since it brings lots of unnecessary cache writes, hurting both the VM performance and the flash endurance. This paper presents COWCache, a COW-aware caching solution that addresses this problem by co-designing flash caching with COW-based virtual disks. First, COWCache designs a new architecture that bridges the semantic gap between flash caching and virtual disk management for realizing the VMs' real data locality. Second, it separately manages COW metadata with fine-grained caching and journaling to improve the metadata caching efficiency. Third, it provides a novel decoupled Copy-on-Write mechanism, which decouples the amplified data requests from the critical I/O path and only admits the data with real VM locality into flash cache. COWCache also introduces a new data structure, the virtual cache map, to mitigate the memory footprint overhead for indexing the cached data in flash. Evaluations show that COWCache improves the application performance by up to 122.7% and reduces the flash cache writes by up to 78.5% compared to traditional flash caching solutions. [ABSTRACT FROM AUTHOR]

Objective: To investigate the effect of image quality of coronary CT angiography (CCTA) on the diagnostic performance of a machine learning-based CT-derived fractional flow reserve (FFRCT).Methods: This nationwide retrospective study enrolled participants from 10 individual centers across China. FFRCT analysis was performed in 570 vessels in 437 patients. Invasive FFR and FFRCT values ≤ 0.80 were considered ischemia-specific. Four-score subjective assessment based on image quality and objective measurement of vessel enhancement was performed on a per-vessel basis. The effects of body mass index (BMI), sex, heart rate, and coronary calcium score on the diagnostic performance of FFRCT were studied.Results: Among 570 vessels, 216 were considered ischemia-specific by invasive FFR and 198 by FFRCT. Sensitivity and specificity of FFRCT for detecting lesion-specific ischemia were 0.82 and 0.93, respectively. Area under the curve (AUC) of high-quality images (0.93, n = 159) was found to be superior to low-quality images (0.80, n = 92, p = 0.02). Objective image quality and heart rate were also associated with diagnostic performance of FFRCT, whereas there was no statistical difference in diagnostic performance among different BMI, sex, and calcium score groups (all p > 0.05, Bonferroni correction).Conclusions: This retrospective multicenter study supported the FFRCT as a noninvasive test in evaluating lesion-specific ischemia. Subjective image quality, vessel enhancement, and heart rate affect the diagnostic performance of FFRCT.Key Points: • FFRCTcan be used to evaluate lesion-specific ischemia. • Poor image quality negatively affects the diagnostic performance of FFRCT. • CCTA with ≥ score 3, intracoronary enhancement degree of 300-400 HU, and heart rate below 70 bpm at scanning could be of great benefit to more accurate FFRCTanalysis. [ABSTRACT FROM AUTHOR]

To comprehensively evaluate the accuracy and performance of machine tools, in this paper, a novel measuring method for the body diagonal error of a machine tool using a multibeam laser interferometer as the measuring instrument is proposed. On this basis, a three-dimensional evaluation method for the body diagonal error of a machine tool is proposed. First, a method of measuring the body diagonal error of a machine tool by using a multibeam laser interferometer is proposed. Six errors for the body diagonal of a machine tool can be measured simultaneously, including three position deviations and three angle errors. The mathematical model of the relationship between the measured data and the volumetric error is established. Second, a new evaluation method for the body diagonal error is presented that takes the position deviation and attitude error into consideration. The advantage of this evaluation method is that it fills in the blanks in the evaluation method for the body diagonal error. The method can also be applied to the evaluation of linear axis error. Finally, the proposed measurement method is applied to a vertical three-axis machining centre, and the experimental results show that this measuring method is quite convenient and effective for measuring the body diagonal error of a machine tool. The proposed method is used to evaluate the body diagonal error from measurement data. The results show that the positioning error of the machine tool is 49.37 μm; the repeatability of the positioning error is 36.85 μm; and the angle errors are − 22.47 μm/m, − 70.9 μm/m and 29.7 μm/m. The traditional method (ISO230-6) is used to evaluate the body diagonal error from measurement data. The results show that the positioning error of the machine tool is 49.4 μm. Compared with the traditional method, the information from the proposed evaluation method is more abundant, and the evaluation method is systematic and universal. [ABSTRACT FROM AUTHOR]

Improving the machining dynamics performance and efficiency of machine tools is extremely important to finish the processing of complex channel parts with high quality in a short time. In this paper, a three-step frame is adopted to achieve this goal by planning the tool orientations and feed rate, which includes C-spaces determination, collaborative optimization of toolpath and feed rates, and the adjustment for over-limitation dynamic parameters. Firstly, a conical ball-end cutter is used and the feasible C-spaces computation method is provided based on the envelop theory. Secondly, a multi-objective mathematical model with multi-constraint conditions for optimizing the dynamics performance and machining efficiency is built. Thirdly, the solving algorithms for the model but simplified is designed to generate the optimal toolpath and initial variable feed rate sequence. To design the efficient algorithms, the Closed-region Parametric Encoding (CPE) method for data structure and the Parametric Data Algorithm Design (PDAD) for algorithm design are proposed, which can transform the geometric constraints into the parametric surfaces in C-space and compute with the continuous parameters to implement the adaptive local subdivision. The algorithms can automatically generate the different shape of toolpath curve to obtain better optimization results. Finally, considering the dynamic parameters limitation, an adjustment algorithm is designed to adjust the feed rates for reducing the over-limitation dynamic parameters. At the end of the paper, two sets of comparative experiments are given for verification. The first set experiment shows that our approach can efficiently improve the machining dynamics performance and efficiency for complex channel parts. The second set experiment shows that the solving algorithm designed by the CPE method and PDAD method has the better computation efficiency and optimization effect. [ABSTRACT FROM AUTHOR]

Forest resource data is important in targeting the forestry operations, and it is in the hearth of the precision forestry concept. The forest resource data can be produced with many techniques, and the number of existing forest data sources has increased during the years. In addition to the forest resource data, other data describing the circumstances of the forest site, such as trafficability and weather conditions, are available. In Finland, a forest data platform gathers the data sources under a single service for easier implementation of the precision forestry applications. This data is useful in operations planning, but it also describes the conditions that prevail when the forest machine arrives to the forest site. This study proposes data fusion between fieldbus time series of the forest machine and the forest data. The fused dataset enables explorative statistical analysis for examining the relationship between the machine performance and the forest attributes and provides data for building predictive models between the two. The presented methods are applied into a dataset generated from a field test data. The results show that some fieldbus time series features are predictable from forest attributes with R 2 value over 0.80, and clustering methods help in interpreting the machine behavior in different environments. In addition, an idea for generating a new forest data source to the forest data platform based on the fusion is discussed. [ABSTRACT FROM AUTHOR]

In order to realize the historical sequence establishment of abstract dynamics in the cooperative translation of Korean language under machine learning and generate an abstract representation dynamically of the translation decoding tree in the recursive model during translation decoding. Combining the advantages of the two kinds of neural networks, this paper constructs a recursive recurrent neural network model, which can not only model the translation process by using the traditional machine translation features but also gradually construct the abstract representation of translation candidates in the process of translation, mining the important language model and other global features in machine translation effectively. This paper has trained the number of Korean-Chinese translation vocabulary, sentence length, and language pairs. Based on the test results, the model can effectively improve the performance of the machine translation model. In addition, based on the adjustment of pre-order word order to optimized the recursive recurring neural network model, and improved the performance of machine translation significantly. [ABSTRACT FROM AUTHOR]

Temperature changes have widespread impacts on the environment, economy, and municipal planning. Generating accurate climate prediction at finer spatial resolution through downscaling could help better assess the future effects of climate change on a local scale. Ensembles of multiple climate models have been proven to improve the accuracy of temperature prediction. Meanwhile, machine learning techniques have shown high performance in solving various predictive modeling problems, which make them a promising tool for temperature downscaling. This study investigated the performance of machine learning (long short-term memory (LSTM) networks and support vector machine (SVM)) and statistical (arithmetic ensemble mean (EM) and multiple linear regression (MLR)) methods in developing multi-model ensembles for downscaling long-term daily temperature. A case study of twelve meteorological stations across Ontario, Canada, was conducted to evaluate the performance of the proposed ensembles. The results showed that both machine learning and statistical techniques performed well at downscaling daily temperature with multi-model ensembles and had similar performance with relatively high accuracy. The R2 of 12 stations ranged between 0.756 and 0.820 and RMSE ranged between 4.318 and 7.063 °C. Both machine learning and statistical ensembles for downscaling had difficulty in predicting extreme values for temperature below − 10 °C and above 20 °C. The results provided technical support for using statistical and machine learning methods to generate high-resolution daily temperature prediction. [ABSTRACT FROM AUTHOR]

In abrasive waterjet (AWJ) process, abrasive properties are of critical factors that influence machining performances. Abrasive process parameters including abrasive material, size and flow rate, exert a significant impact on cutting ability and quality of AWJ. In this paper, monolithic abrasive material (garnet, alumina and silicon carbide) and corresponding mixing abrasive materials are adopted to evaluate the effect of abrasive process parameters on the cutting ability and quality, respectively. Two kinds of cutting experiments are conducted with a monolithic abrasive to explore the influence of variation in abrasive material, size and flow rate on the machining performance. The cutting depth experiments for evaluating the cutting ability in terms of cutting depth are conducted with a right trapezoid aluminium alloy block. The kerf cutting experiments are taken to evaluate the cutting quality by the kerf taper angle, kerf width and surface roughness. On this basis, AWJ experiments with mixing abrasive materials are taken to probe into the composition of mixing abrasives' effect on cutting performance. It is found that the abrasives consisting of 75% alumina and 25% garnet in mass fraction lead to a maximum cutting depth of 41.7 mm and the smallest surface roughness of Ra 2.1 μm under the flow rate of 130 g/min. [ABSTRACT FROM AUTHOR]