Data-driven approach to predict the fatigue properties of ferrous metal materials using the cGAN and machine-learning algorithms

The stress-life curve(S–N)and low-cycle strain-life curve(E–N)are the two primary representations used to characterize the fatigue behavior of a material.These material fatigue curves are essential for structural fatigue analysis.However,conducting material fatigue tests is expensive and time-intensive.To address the challenge of data limitations on ferrous metal materials,we propose a novel method that utilizes the Random Forest Algorithm and transfer learning to predict the S–N and E–N curves of ferrous materials.In addition,a data-augmentation framework is introduced using a conditional generative adversarial network(cGAN)to overcome data deficiencies.By incorporating the cGAN-generated data,the accuracy(R2)of the Random Forest Algorithm-trained model is improved by 0.3–0.6.It is proven that the cGAN can significantly enhance the prediction accuracy of the machine-learning model and balance the cost of obtaining fatigue data from the experiment.

Optimal tool design in micro-milling of difficult-to-machine materials

The limitations of significant tool wear and tool breakage of commercially available fluted micro-end mill tools often lead to ineffective and inefficient manufacturing,while surface quality and geometric dimensions remain unacceptably poor.This is especially true for machining of difficult-to-machine(DTM)materials,such as super alloys and ceramics.Such conventional fluted micro-tool designs are generally down scaled from the macro-milling tool designs.However,simply scaling such designs from the macro to micro domain leads to inherent design flaws,such as poor tool rigidity,poor tool strength and weak cutting edges,ultimately ending in tool failure.Therefore,in this article a design process is first established to determine optimal micro-end mill tool designs for machining some typical DTM materials commonly used in manufacturing orthopaedic implants and micro-feature moulds.The design process focuses on achieving robust stiffness and mechanical strength to reduce tool wear,avoid tool chipping and tool breakage in order to efficiently machine very hard materials.Then,static stress and deflection finite element analysis(FEA)is carried out to identify stiffness and rigidity of the tool design in relation to the maximum deformations,as well as the Von Mises stress distribution at the cutting edge of the designed tools.Following analysis and further optimisation of the FEA results,a verified optimum tool design is established for micro-milling DTM materials.An experimental study is then carried out to compare the optimum tool design to commercial tools,in regards to cutting forces,tool wear and surface quality.

Predictive defect detection for prototype additive manufacturing based on multi-layer susceptibility discrimination

This paper presents a predictive defect detection method for prototype additive manufacturing(AM)based on multilayer susceptibility discrimination(MSD).Most current methods are significantly limited by merely captured images,disregarding the differences between layer-by-layer manufacturing approaches,without combining transcendental knowledge.The visible parts,originating from the prototype of conceptual design,are determined based on spherical flipping and convex hull theory,on the basis of which theoretical template image(TTI)is rendered according to photorealistic technology.In addition,to jointly consider the differences in AM processes,the finite element method(FEM)of transient thermal-structure coupled analysis was conducted to probe susceptible regions where defects appeared with a higher possibility.Driven by prior knowledge acquired from the FEM analysis,the MSD with an adaptive threshold,which discriminated the sensitivity and susceptibility of each layer,was implemented to determine defects.The anomalous regions were detected and refined by superimposing multiple-layer anomalous regions and comparing the structural features extracted using the Chan-Vese(CV)model.A physical experiment was performed via digital light processing(DLP)with photosensitive resin of a non-faceted scaled V-shaped engine block prototype with cylindrical holes using a non-contact profilometer.This MSD method is practical for detecting defects and is valuable for a deeper exploration of barely visible impact damage(BVID),thereby reducing the defect of prototypical mechanical parts in engineering machinery or process equipment via intellectualized machinevision.

Intelligent modular design with holonic fuzzy agents

Presently,modular designs use various technologies accompanied by multiple models.Although no integral solution is known,a plethora of approaches is used to resolve this trans disciplinary challenge,often by using local intelligence.However,the effective utilization of multiple models requires proper integration for them to work together as a cohesive system.This requirement calls for the development of intelligent models and tools that can be used for the development of intelligent modular products.Modular design based on these intelligent models and tools is called intelligent modular design.Intelligent modular design requires to be considered both dynamically and holistically by combining customer requirements,product functions,solutions,service specifications,and their fuzziness in order to structure a product into intelligent modules.This paper proposes the use of holonic fuzzy agents to fulfill both the properties of intelligent models and the requirements of intelligent modular design.The set of fuzzy function agents and their corresponding fuzzy solution agents are found from customization of the product-service system in the fuzzy function agent-fuzzy solution agent sub-network.On the basis of attractor agent recognition,the fuzzy function and fuzzy solution agents interact to form the holonic fuzzy module agents.Selfembedding of holonic fuzzy module agents,which is the fundamental property of the holonic structure,is also characterized by vertical and horizontal communication.The flexibility and agility of the software agent make the holonic structure of intelligent modules adaptable.An application illustrates the proposed intelligent modular design.

Residual stress model of pre-stressed dry grinding considering coupling of thermal, stress, and phase transformation

Pre-stressed dry grinding can result in a hardened layer on the part surface while the surface residual stress is controlled.Considering the factors of the thermal field,pre-stress,and microstructural transformation,a proximate model of surface residual stress for pre-stressed dry grinding is established using the ANSYS finite element simulation method and verified through experiment.The variation laws and mechanisms of the residual stress along with the grinding parameters are revealed.Under the comprehensive effect of pre-stress and phase transformation,the residual stress of pre-stressed dry grinding is revealed mainly as compressive stress.This increases as the pre・stress and grinding depth increase.Under the coupling effect,pre-stress has larger influence on the residual stress than the grinding depth.The model can analyze and predict the residual stress of pre-stressed dry grinding in general.

Dynamic interplay between dislocations and precipitates in creep aging of an Al-Zn-Mg-Cu alloy

Abstract Creep age forming (CAF) is an advanced forming technology used for manufacturing large complex integrated panel components. However, in creep aging (CA), unlike in sole creep or aging procedure, the dislocation movement and the precipitation process occur simultaneously, leading to difficulty in understanding of the dynamic interplay between these two phenomena. In this work, taking 7050 Al alloy, a typical Al-Zn-Mg-Cu alloy, as the test material, an experimental scheme combining pre-deformation, artificial aging (AA), and tensile/ compressive CA is designed to decouple and reveal the dynamic interaction mechanism of both phenomena. From AA experiments, the static interaction between dislocations and precipitates is studied, and then their dynamic interactions in CA and each evolution are comparatively investigated. The research shows that both total strain and strain rate increase with the increase in pre-deformation in tensile and compressive CA. However, the total creep strain in compressive CA is larger than that in tensile CA. In additional, the more the dislocations are induced, the sparser and more heterogeneous the overall distribution of precipitates becomes. For dynamic interplay, in the first stage of CA (I), under thermal-mechanical loading, the GP zones and η phases gradually nucleate and grow, while the effect of dislocation multiplication is dominant compared with dislocation annihilation, leading to an increase in total dislocation density. Soon, the dislocation movement is gradually hindered by tangling, pile-up, and the precipitates that have grown on the dislocation lines, this decreases the mobile dislocation density and results in a significant decrease in creep rate. In the second stage (II), the precipitates grow further, especially those lying on the dislocation lines;the effects of pinning and hindrance are enhanced until the dislocation multiplication and annihilation reach a dynamic equilibrium, and the total and mobile dislocation densities tend to be roughly unchanged, thus, the creep rate remains relatively constant in this stage.

Application of grey relational analysis based on Taguchi method for optimizing machining parameters in hard turning of high chrome cast iron

High chrome white cast iron is particularly preferred in the production of machine parts requiring high wear resistance. Although the amount of chrome in these materials provides high wear and corrosion resistances, it makes their machinability difficult. This study presents an application of the grey relational analysis based on the Taguchi method in order to optimize chrome ratio, cutting speed, feed rate, and cutting depth for the resultant cutting force (Fr) and surface roughness (Ra) when hard turning high chrome cast iron with a cubic boron nitride (CBN) insert. The effect levels of machining parameters on Fr and Ra were examined by an analysis of variance (ANOVA). A grey relational grade (GRG) was calculated to simultaneously minimize Fr and Ra. The ANOVA results based on GRG indicated that the feed rate, followed by the cutting depth, was the main parameter and contributed to respo ses. Optimal levels of parameters were found when the chrome ratio, cutting speed, feed rate, and cutting depth were 12%, 100m/min, 0.05mm/r, and 0.1mm, respectively, based on the multiresponse optimization results obtained by considering the maximum signal to noise (SIN) ratio of GRG. Confirmation results were verified by calculating the confidence level within the interval width.

Warehouse scheduling performance analysis considering LHRL

Warehouse scheduling efficiency has to do with the length-height ratio of location （LHRL） to some extent, which hasn＇t been well investigated until now. In this paper a mathematic model is built by analyzing the relation between the travel time of the stacker and LHRL. Mean- while, warehouse scheduling strategy is studied combining with the project on the automatic production line of an enterprise, and a warehouse scheduling strategy is pro- posed based on index of quality （IoQ） parameters. Besides, the process of getting the value of IoQ is also simplified with the idea of sparse matrix. Finally, the IoQ scheduling strategy is compared with random strategy and First Come First Out strategy in different LHRLs. The simulation results show that the IoQ scheduling strategy not only improves the quality of the product effectively, but also improves the efficiency of the scheduling substantially.

Progress of remanufacturing engineering and future technology expectation

After development for decades, abroad remanufacturing has formed a complete industrial system. At present, the research emphases are on marking logistics management and market cultivation theory of remanufacturing products, and so on. The Chinese remanufacturing starts fairly late. After 10 years of development, it formed a remanufacturing mode with Chinese characteristics that is sustained by high-tech industries, using the surface engineering technology to restore the size and improve properties, and combining manufacturing, study and research together. The remanufacturing mode is not only circular but also economic. With the development of science and technology, future remanufacturing technology will break the previous limits, explore and understand the limits of micro machining. It will carry out the waste product remanufacturing in the micro-nano scale, and extend the remanufacturing industry to a more broad space.

Programming time-dependent behavior in 4D printing by geometric and printing parameters

Smart structures realize sequential motion and self-assembly through external stimuli.With the advancement of four-dimensional(4D)printing,the programming of sequential motions of smart structures is endowed with more design and manufacturing possibilities.In this research,we present a method for physically programming the timescale of shape change in 4D-printed bilayer actuators to enable the sequential motion and self-assembly of smart structures.The effects of the geometric and printing parameters on the time-dependent behavior of 4D-printed bilayer actuators are investigated.The results show that the thickness of the active layer directly affects the timescale of motion,and increasing the thickness leads to faster motion until the thickness ratio is close to 4:6.Similarly,a higher printing speed results in faster motion.Conversely,a higher printing temperature and a greater layer height result in a slower shape change.The effects of the length-width ratio,line width,and filling ratio on the timescale of motion are not as straightforward.Finally,we demonstrate several smart structures that exhibit sequential motion,including a labyrinth-like self-folding structure that is choreographed to achieve multi-step self-shaping and a flower-shaped structure where each part completes its movement sequentially to avoid collisions.The presented method extends the programmability and functional capabilities of 4D printing.

Hybrid genetic algorithm for a type-Ⅱ robust mixed-model assembly line balancing problem with interval task times

The typemixed-model assembly line balancing problem with uncertain task times is a critical problem. This paper addresses this issue of practical significance to production efficiency. Herein, a robust optimization model for this problem is formulated to hedge against uncertainty. Moreover, the counterpart of the robust optimization model is developed by duality. A hybrid genetic algorithm (HGA) is proposed to solve this problem. In this algorithm, a heuristic method is utilized to seed the initial population. In addition, an adaptive local search procedure and a discrete Levy flight are hybridized with the genetic algorithm (GA) to enhance the performance of the algorithm. The effectiveness of the HGA is tested on a set of benchmark instances. Furthermore, the effect of uncertainty parameters on production efficiency is also investigated.

System integration for on-machine measurement using a capacitive LVDT-Iike contact sensor

In this study, an air-bearing capacitive linear variable differential transformer （LVDT）-like contact sensor with a rounded diamond tip was mounted to a desktop machine tool to construct an on-machine （OM） measuring system. The measuring system was capable of decoding the digital signals of linear encoders mounted on the machine tool and acquiring the analog signal of the contact sensor. To verify the measuring system, experimental examinations were performed on an oxygen-free copper （OFC） convex aspheric mold with a diameter of 5 mm and a curve height of 0.46 mm. The acquired signals were processed by the implemented Gaussian regression filter （GRF）, removing the tilt of measured profile, and compensating for the radius of probe tip. The profile obtained was compared to that measured using a commercially available device, and a maximum deviation of 14.6μm was found for the rough cutting. The compensation cutting was then performed according to the form error of OM measurement. As a result, the PV form error compared with the designed profile was reduced from 19.2μm over a measured diameter of 4 mm to 9.7 μm over a measured diameter of 3.1 mm, or a percentage improvement of 35.4% in form accuracy. Through the examination for aspheric machining, the effectiveness of the implemented OM measuring system was demonstrated, and the technical details of system implementation were presented. Further improvement was suggested to reduce the diameter of probe tip and measuring force.

Digitalization and intelligentization of manufacturing industry

1 Relying on innovation to realize the historical leapChina has entered a new historical period in her development. In order to achieve scientific development and to accelerate transformation of economic development pattern, the most fundamental issue is to rely on the power of science and technology, and the most crucial element is to improve the capability of independent innovation. The key for promoting China＇s economic and social development is to embark on the innovation-driven track as soon as possible.

A Fast and Accurate Method for the Analysis of Multi-conductor Coplanar Wave-Guide Transmission Lines

This paper introduces the complex image concept, and uses the method to analyze multi conductor coplanar waveguides. The method of spectral domain Green's function for modeling point charge and line charge structures is studied, in which Chebyshev polynomials are used as basis functions to solve the integral equation by the Galerkin's method. It is believed that the complex method has the features of accuracy and rapid convergence, and it is possible to make the technique useful as CAD tool for coplanar waveguide design.

A Generalized Variational Principle for 2-D Piezoelectricity with Surface Electrodes

It is difficult to establish a classical variational model for piezoelectricity. Following the semi inverse method of establishing generalized variational principles, an energy like trial functional with a certain unknown function is constructed. The unknown function is easily identified step by step. A family of variational principles for the static behavior of the elastic and electric variables in the vicinity of a surface electrode attached to a piezoelectric ceramic is established directly from its field equations and boundary conditions.

Unsteady Inverse Problem of Type B for 2-D Transonic Flow: A Variational Formulation

A family of variational principles (VP) has been developed for the unsteady inverse problem of the second type I B. It opens new ways for the inverse shape design of unsteady airfoils and can serve as key basis of multipoint inverse shape design of steady airfoils and cascades.

Fuzzy Set Identification of Bone Marrow Cells

Using rubricytes and lymphocytes as examples,this paper presents a fuzzy set theory and method to identify human bone marrow hematopoiesis system cells (BMCs).On the basis of the Cauchy’s distribution function,this paper sets up a series of membership function formulae of the BMC feature fuzzy subsets,general identification formulae of fuzzy sets for the BMCs,as well as identification formulae of fuzzy sets for rubricytes and lymphocytes.These formulae will assist with the quantitation of unknown cells compared to standard cells.

Optimization of Membership Function for Fuzzy Control Based on Genetic Algorithm and Its Applications

In this paper, a simple and practicable algorithm for optimization of membership function (MF) is proposed. As it is known that MF is very important in the fuzzy control. Unfortunately, to find, especially to optimize MF is always rather complex even difficult. So, to study and develop an effectual aglorithm for MF optimization is a good topic. Allow for the inner advantages of genetic algorithm (GA), it is adopted in the algorithm .The principle and executive procdeure are first presented. Then it is applied in the fuzzy control system of a typical plant. Results of real time run show that the control strategy is encouraging, and the developed algorithm is practicable.

An Experimental Setup for Study of Photoresponse of High T_c Superconductors

We have developed an experimental setup that has been proven suitable for the study of photoresponse of high Tc superconductors. The distinguish feature of this experimental setup lies mainly in the data acquisition system which is equipped with computer as well as the IEEE-488 interface bus. which ensures the accuracy to experimental results. Using the experimental setup, the optical response to laser radiation in high-Tc superconductors has been examined, both of bolometric effect and nonequilibrium optical response are revealed.

Effect of cooling media on bead geometry,microstructure,and mechanical properties of wire arc additive manufactured IN718 alloy

This work aims to present and explore thermal management techniques for the wire arc additive manufacturing(WAAM)of IN718 components.Excessive heat can be mitigated via air or water cooling.In this study,the material was deposited under four different heat-input conditions with air or water cooling.In air cooling,the layer is deposited in a normal atmospheric air environment,whereas with water cooling,the material is deposited inside a water tank by varying the water level.To validate the air and water cooling thermal management techniques,IN718 single-pass and multilayer linear walls were deposited using the bidirectional gas metal arc welding based WAAM setup under four different heat input conditions.During the deposition of single layers,the temperature profiles were recorded,and the geometric and microstructural features were explored.For multilayer wall structures,the mechanical properties(hardness,tensile strength,and elongation)were determined and assessed using the corresponding microstructural features explored through scanning electron microscopy(SEM),energy dispersive spectroscopy(EDS),and electron backscatter diffraction(EBSD)analyses.The microstructure observed through SEM analysis in the building direction was found to be nonhomogenous compared with that in the deposition direction.Moreover,water cooling was found to govern bead characteristics,such as wall width and height.The grain size and anisotropy of the mechanical properties also decreased in the water-cooled case.Hence,water cooling is an economical and efficient method to mitigate excessive heat accumulation in WAAM-deposited IN718.