Full Machine Research Articles

Modeling of laser-assisted micro-milling Inconel718

Nickel-based superalloy Inconel718 is a typical difficult-to-machine material with the characteristic of high strength and hardness. Demands for micro parts of nickel-based superalloy Inconel718 are growing in aerospace, biomedical and other fields. Laser-assisted micro-milling (LAMM) can realize precision machining of the workpiece because of its lower cutting force. However, the machining mechanism and the change law of cutting force during LAMM of nickel-based superalloy Inconel718 are still unclear. There is a crucial need for modeling of LAMM Inconel718. This paper explores LAMM of nickel-based superalloy Inconel718 and realizes the simulation of LAMM process. A three-dimensional transient heat transfer finite element (FE) model is established to predict the temperature distribution of a Gaussian heat source in LAMM. Multiple sets of simulations of laser pre-heating with various laser power are conducted to analyze the temperature fields of workpiece and determine the appropriate laser parameters. A novel 3D model of the full machining process of LAMM Inconel 718 is constructed by ABAQUS with multi-subroutine, a fully coupled thermal-mechanical process is achieved. A modified Johnson-Cook constitutive model is used to reflect the size effect and flow stress of LAMM Inconel 718. Finally, the validity of the models is verified by experiments. Based on the verified model, the appropriate laser pre-heating parameters are obtained, and the prediction of cutting forces of LAMM Inconel718 is achieved.

Geodesics on an arbitrary ellipsoid of revolution

The algorithms given in Karney (J. Geodesy 87:43–55, 2013), to compute geodesics on terrestrial ellipsoids, are extended to apply to ellipsoids of revolution with arbitrary eccentricity. For the direct and inverse geodesic problems, this entails implementing the formulation in terms of elliptic integrals given by Legendre and Cayley. The integral for the area of geodesic polygons is computed in terms of the discrete sine transform of the integrand. In all cases, accuracy close to full machine precision is achieved. An open-source implementation of these algorithms is available.

Numerical modelling of HTS tapes under arbitrary external field and transport current via integral method: review and application to electrical machines *

Superconductors in practical use can be subjected to spatially non-uniform and time-varying external magnetic field as well as carrying a transport current, for example, in an electrical machine. This paper verifies that the integral method can be used in situations in which the external field is spatially non-uniform, by providing both theoretical reasoning and simulation results. Variations in the integral method are reviewed, such as how to impose transport current. Further, the integral method is applied to calculate ac loss in superconducting tapes in an air-cored electrical machine in a two-stage process: the external field is calculated in a COMSOL model without superconducting tapes, and exported into the integral method model that consists of the tapes only. The time taken by the integral method is a small fraction of the time taken by the full machine model in COMSOL, which uses the T-A formulation. There are good agreements between the full COMSOL model and two-stage method incorporating the integral method.

The Right Software To Power Production

Aktiv Solutions has reduced programming time, increase machining productivity and improved toolpath simulation for full machine collision avoidance with Tebis CAD/CAM software

Analysis of the Effect of Machining of the Surfaces of WAAM 18Ni 250 Maraging Steel Specimens on Their Durability.

It is now well-known that the interaction between surface roughness and surface-breaking defects can significantly degrade the fatigue life of additively manufactured (AM) parts. This is also aptly illustrated in the author's recent study on the durability of wire and arc additively manufactured (WAAM) 18Ni 250 Maraging steel specimens, where it was reported that failure occurred due to fatigue crack growth that arose due to the interaction between the surface roughness and surface-breaking material defects. To improve the durability of an AM part, several papers have suggested the machining of rough surfaces. However, for complex geometries the fully machining of the entire rough surface is not always possible and the effect of the partial machining on durability is unknown. Therefore, this paper investigates if partial machining of WAAM 18Ni 250 Maraging steel surfaces will help to improve the durability of these specimens. Unfortunately, the result of this investigation has shown that partial machining may not significantly improve durability of WAAM 18Ni 250 Maraging steel specimens. Due to the order of surface roughness seen in WAAM 250 Maraging steel, the improvement to durability is only realized by full machining to completely remove the remnants of any print artefacts.

Machining error model for full machining process of thread milling

Thread milling is increasingly used in engineering manufacturing because of its advantages, but there must be some error in the thread due to the machining principle. In response to this problem, this paper discussed the causes of thread milling error and established the machining error model for the full machining process of thread milling. Firstly, the geometric model of thread milling is developed based on the parametric definitions of the cutter tooth surface, thread surface, and tool-path. Then, based on the cutter motion characteristics of the thread milling penetration and retraction segments and machining segment, the machining error model is established. The calculation of machining error is transformed into solving for the relationship between the position of the intersection of the cutter profile line and the thread profile line generated in the cross-section, which means that the three-dimensional problem is transformed into the two-dimensional problem. Finally, the proposed model is proven via thread milling experiments. The simulation results of machining error agree well with the experimental results, which prove the correctness and validity of the model. The model provides theoretical support for further research into thread milling machining error.

Full wCDM analysis of KiDS-1000 weak lensing maps using deep learning

We present a full forward-modeled $w\mathrm{CDM}$ analysis of the KiDS-1000 weak lensing maps using graph-convolutional neural networks (GCNN). Utilizing the cosmogrid, a novel massive simulation suite spanning six different cosmological parameters, we generate almost one million tomographic mock surveys on the sphere. Due to the large dataset size and survey area, we perform a spherical analysis while limiting our map resolution to HEALPix ${n}_{\mathrm{side}}=512$. We marginalize over systematics such as photometric redshift errors, multiplicative calibration and additive shear bias. Furthermore, we use a map-level implementation of the nonlinear intrinsic alignment model along with a novel treatment of baryonic feedback to incorporate additional astrophysical nuisance parameters. We also perform a spherical power spectrum analysis for comparison. The constraints of the cosmological parameters are generated using a likelihood-free inference method called Gaussian process approximate Bayesian computation (GPABC). Finally, we check that our pipeline is robust against choices of the simulation parameters. We find constraints on the degeneracy parameter of ${S}_{8}\ensuremath{\equiv}{\ensuremath{\sigma}}_{8}\sqrt{{\mathrm{\ensuremath{\Omega}}}_{M}/0.3}=0.7{8}_{\ensuremath{-}0.06}^{+0.06}$ for our power spectrum analysis and ${S}_{8}=0.7{9}_{\ensuremath{-}0.05}^{+0.05}$ for our GCNN analysis, improving the former by 16%. This is consistent with earlier analyses of the 2-point function, albeit slightly higher. Baryonic corrections generally broaden the constraints on the degeneracy parameter by about 10%. These results offer great prospects for full machine learning based analyses of ongoing and future weak lensing surveys.

Solving Fredholm second-kind integral equations with singular right-hand sides on non-smooth boundaries

A numerical scheme is presented for the solution of Fredholm second-kind boundary integral equations with right-hand sides that are singular at a finite set of boundary points. The boundaries themselves may be non-smooth. The scheme, which builds on recursively compressed inverse preconditioning (RCIP), is universal as it is independent of the nature of the singularities. Strong right-hand-side singularities, such as 1/|r|α with α close to 1, can be treated in full machine precision. Adaptive refinement is used only in the recursive construction of the preconditioner, leading to an optimal number of discretization points and superior stability in the solve phase. The performance of the scheme is illustrated via several numerical examples, including an application to an integral equation derived from the linearized BGKW kinetic equation for the steady Couette flow.

Efficient Self-Tuned Fuzzy Logic Based Power System Stabilizer

A mechanism for efficient self-tuning of a fuzzy logic-based PSS in real-time is presented. The proposed controller with a self-tuning mechanism can reduce low-frequency oscillations regardless of magnitude or type of disturbance, whether sudden or gradual. It is not necessary to set the initialization or learning factor for each disturbance and its implementation is simple. Speed deviation and its derivative are used as fuzzy controller inputs. The main idea, based on the properties of the conventional PID controller, is to use a fuzzy controller similar to a PID controller but for the nonlinear power system. Since the objective is implementation on a physical system, nonlinearity without any simplification of the system is very important. Therefore, in this mechanism, the full order machine model is considered. The optimal performance is obtained by determining in real-time the appropriate gains for each input. Tests for a wide range of disturbances show the robustness of the controller. The proposed approach has fewer fuzzy rules than similar methods. Therefore, its hardware implementation can be simpler and will require less computation time. Very good damping of all types of disturbances is a prominent feature of this method.

Machine learning in medicine: what clinicians should know.

With the advent of artificial intelligence (AI), machines are increasingly being used to complete complicated tasks, yielding remarkable results. Machine learning (ML) is the most relevant subset of AI in medicine, which will soon become an integral part of our everyday practice. Therefore, physicians should acquaint themselves with ML and AI, and their role as an enabler rather than a competitor. Herein, we introduce basic concepts and terms used in AI and ML, and aim to demystify commonly used AI/ML algorithms such as learning methods including neural networks/deep learning, decision tree and application domain in computer vision and natural language processing through specific examples. We discuss how machines are already being used to augment the physician's decision-making process, and postulate the potential impact of ML on medical practice and medical research based on its current capabilities and known limitations. Moreover, we discuss the feasibility of full machine autonomy in medicine.

Evolving Fully Automated Machine Learning via Life-Long Knowledge Anchors

Automated machine learning (AutoML) has achieved remarkable progress on various tasks, which is attributed to its minimal involvement of manual feature and model designs. However, most of existing AutoML pipelines only touch parts of the full machine learning pipeline, e.g., neural architecture search or optimizer selection. This leaves potentially important components such as data cleaning and model ensemble out of the optimization, and still results in considerable human involvement and suboptimal performance. The main challenges lie in the huge search space assembling all possibilities over all components, as well as the generalization ability over different tasks like image, text, and tabular etc. In this paper, we present a first-of-its-kind fully AutoML pipeline, to comprehensively automate data preprocessing, feature engineering, model generation/selection/training and ensemble for an arbitrary dataset and evaluation metric. Our innovation lies in the comprehensive scope of a learning pipeline, with a novel "life-long" knowledge anchor design to fundamentally accelerate the search over the full search space. Such knowledge anchors record detailed information of pipelines and integrates them with an evolutionary algorithm for joint optimization across components. Experiments demonstrate that the result pipeline achieves state-of-the-art performance on multiple datasets and modalities. Specifically, the proposed framework was extensively evaluated in the NeurIPS 2019 AutoDL challenge, and won the only champion with a significant gap against other approaches, on all the image, video, speech, text and tabular tracks.

Influence of the Preformed Coil Design on the Thermal Behavior of Electric Traction Machines

Preformed coils are used in electrical machines to improve the copper slot fill factor. A higher utilization of the machine can be realized. The improvement is a result of both, low copper losses due to the increased slot fill factor and an improved heat transition out of the slot. In this study, the influence of these two aspects on the operational improvement of the machine is studied. Detailed simulation models allow a separation of the two effects. A preform wound winding in comparison to a round wire winding is studied. Full machine prototypes as well as motorettes of the two designs are built up. Thermal finite element models of the stator slot are developed and parameterized with the help of motorette microsections. The resulting thermal lumped parameter model is enlarged to represent the entire electric machine. Electromagnetic finite element models for loss calculation and the thermal lumped parameter models are parameterized using test bench measurements. The developed models show very good agreement in comparison to the test bench evaluation. The study indicates that both, the improvements in the heat transition path and the advantages of the reduced losses in the slot contribute to the improved operational range in dependency of the studied operational point.

Automatic control method of construction quality of geological survey in geotechnical engineering

In view of the low efficiency and low automation of the traditional geotechnical engineering geological survey construction quality control method, this paper puts forward the construction quality automatic control method of geotechnical engineering geological survey. The finite element model of geotechnical engineering geological investigation is established by using CRM geological model transformation method. According to the numerical analysis method, the boundary area is cut, and according to the characteristics of the geological model, the volume mesh model is generated by multi-spatial domain method, and the generated model is imported into the finite element analysis system. The finite element model of automatic control platform of full hydraulic core drilling machine is designed. The drilling pressure control method is used to control the drilling pressure in the process of measurement, and the rammed expanded pile with composite carrier is used for construction. The experimental results show that the method can establish the finite element model with high speed and precision on the basis of realizing the automatic control of the construction quality of geological survey in geotechnical engineering. The utility model has the advantages of high precision, small size, strong anti-interference ability, and low cost.

Failure analysis and modernization of high-pressure hydraulic press for drilling tubes testing

This research paper represents the results of abrupt failure investigation in the structure of industrial plant for hydrostatic pressure testing of tubes. The significant damages of testing machine elements are observed after the moving tube impact when the opposite tube end cap has been suddenly released because of the screw-thread defect. The water pressure inside the tube at the time of the incident was only 20 MPa while the maximal pressure of 125 MPa is required by tubes tests specification. Therefore, based on admitted assumptions, the detailed analysis is conducted of static and dynamic forces acting on elements of structure to avoid full machine destruction in case of the next incidents under higher pressures. Conditions are determined of the sharp cap end penetration into thrust plate and possible damage of upper protection casing from the water jet and tube pieces impacts under conditions of the crack opening in the tested tube. The two options are proposed of plant modernization to reinforce structure for safety work under pressure from 70 MPa to 125 MPa depending on tube size and testing conditions. The developed improvements of machine structure allow simultaneously increasing the productivity of testing plant up to 14–20 tubes per hour depending on their sizes.

Laminar Burning Velocity Model Based on Deep Neural Network for Hydrogen and Propane with Air

The aim of the study was to develop deep neural network models for laminar burning velocity (LBV) calculations. The present study resulted in models for hydrogen–air and propane–air mixtures. An original data-preparation/data-generation algorithm was also developed in order to obtain the datasets sufficient in quality and quantity for models training. The discussion about the current analytical models highlighted issues with both experimental data and methodology of creating those analytical models. It was concluded that there is a need for models that can capture data from multiple experimental techniques with ease and automate the model design and training process. We presented a full machine learning based approach that fulfills these requirements. Not only model development, but also data preparation was described in detail as it is crucial in obtaining good results. Resulting models calculations were compared with popular analytical models and experimental data gathered from literature. The calculations comparison showed that the models developed were characterized by the smallest error with regards to the experiments and behaved equally well for variable pressure, temperature, and equivalence ratio. The source code of ready-to-use models has been provided and can be easily integrated in, for example, CFD software.

Securing Vulnerabilities in Docker Images

Docker is an alternative application development and publishing infrastructure tool to various virtualization environments such as Virtual box and the like. The most popular containerization platform is Docker which is the area where Docker images are run. Container is a lightweight contrasting option to full machine virtualization that includes exemplifying an application in a container with its own working condition. These two concepts, virtualization and containerization are competing in the cloud-based environments. When virtualization became the mainstream, VM security concerns was common. IT Security experts are discussing the potential weaknesses of a virtualized environment for a long time. In this paper, focusing on Docker container, its vulnerabilities and possible measurements against security concerns, we have provided information about assessment of risks and vulnerabilities of containerization and the main differences between these two concepts via vulnerability analysis.

Deep Learning-Based Partial Domain Adaptation Method on Intelligent Machinery Fault Diagnostics

In the past years, deep learning-based machinery fault diagnosis methods have been successfully developed, and the basic diagnostic problems have been well addressed where the training and testing data are collected under the same operating conditions. When the training and testing data are from different distributions, domain adaptation approaches have been introduced. However, the existing methods generally assume the availability of the target-domain data in all the health conditions during training, which is not in accordance with the real industrial scenarios. This article proposes a deep learning-based fault diagnosis method to address the partial domain adaptation problems, where the unsupervised target-domain training data do not cover the full machine health state label space. The conditional data alignment and unsupervised prediction consistency schemes are proposed to achieve partial domain adaptation. The experimental results on two rotating machinery datasets suggest the proposed method offers a promising tool for this practical industrial problem.

Sub‐assembly experiments to determine PWM supply‐induced losses in electric machines

This paper presents a low-cost approach to experimentally characterise the effect of the inverter supply on the losses in an electrical machine. The set-up comprises of representative subassemblies of the stator and winding, referred to as ‘motorettes’, and a single phase, fast switching silicon-carbide-based converter. The sub-assembly-based thermal analysis has been shown to be a time and cost-effective method to examine winding loss, over a range of operating conditions. In addition to the slot to stator thermal resistance and capacitance. This provides critical information for thermal model calibration and produces more accurate thermal and loss predictions of the full machine behaviour. A test arrangement has been devised to replicate the magnetic excitation of the electrical machine stator during normal operation. This builds upon previous techniques, to allow iron loss to be investigated. Through multiple tests using DC, AC, and PWM-AC supplies, a 20% increase in iron loss is observed due to PWM effects.

Numerical Investigation of the Effect of Tower Dam and Rotor Misalignment on Performance and Loads of a Large Wind Turbine in the Atmospheric Boundary Layer

A modern horizontal axis wind turbine was simulated by means of computational fluid dynamics (CFD) simulations. The analyzed machine has a diameter of 100 m and is immersed in the atmospheric boundary layer (ABL). The velocity and turbulence stratification of the ABL is correctly preserved along the domain by the adoption of modified wall functions. An overset technique is employed to handle the rotation of the turbine rotor throughout the operation of the machine. The ABL induces periodically oscillating loads and generated torque on the rotor blades. Several configurations are analyzed. First, the results of a rotor-only simulation are compared to the ones obtained from the simulation of the full machine in order to evaluate the effect of the supporting structures on the produced torque and on the loads acting on the blades. Then, a tilt angle is introduced on the analyzed rotor and its effect on the oscillating loads of each blade is highlighted by comparing the results to the untilted configuration. Lastly, a yaw misalignment is also introduced and the results are compared to the unyawed configuration.

MPI vs Fortran coarrays beyond 100k cores: 3D cellular automata

Abstract Fortran coarrays are an attractive alternative to MPI due to a familiar Fortran syntax, single sided communications and implementation in the compiler. Scaling of coarrays is compared in this work to MPI, using cellular automata (CA) 3D Ising magnetisation miniapps, built with the CASUP CA library, https://cgpack.sourceforge.io , developed by the authors. Ising energy and magnetisation were calculated with MPI_ALLREDUCE and Fortran 2018 co_sum collectives. The work was done on ARCHER (Cray XC30) up to the full machine capacity: 109,056 cores. Ping-pong latency and bandwidth results are very similar with MPI and with coarrays for message sizes from 1B to several MB. MPI halo exchange (HX) scaled better than coarray HX, which is surprising because both algorithms use pair-wise communications: MPI IRECV/ISEND/WAITALL vs Fortran sync images . Adding OpenMP to MPI or to coarrays resulted in worse L2 cache hit ratio, and lower performance in all cases, even though the NUMA effects were ruled out. This is likely because the CA algorithm is network bound at scale. This is further evidenced by the fact that very aggressive cache and inter-procedural optimisations lead to no performance gain. The sampling and tracing analysis shows good load balancing in compute in all miniapps, but imbalance in communication, indicating that the difference in performance between MPI and coarrays is likely due to parallel libraries (MPICH2 vs libpgas) and the Cray hardware specific libraries (uGNI vs DMAPP). Overall, the results look promising for coarray use beyond 100k cores. However, further coarray optimisation is needed to narrow the performance gap between coarrays and MPI.