On-line prevention of wire breakage in wire electro-discharge machining

Wire electro-discharge machining (WEDM) is a fully extended and competitive machining process widely used to produce dies and moulds. However, the risk of wire breakage affects adversely the full potential of WEDM since the overall process efficiency is considerably reduced. The present paper discusses the results of the analyses of an exhaustive experimental database that reproduces unexpected disturbances that may appear during normal operation. The results of the analyses reveal new symptoms that allow one to predict wire breakage. These symptoms are especially related to the occurrence of an increase in discharge energy, peak current, as well as increases and/or decreases in ignition delay time. The differences observed in the symptoms related to workpiece thickness are also studied. Another contribution of this paper is the analyses of the distribution of the anticipation time for different validation tests.Based on the results of the analyses, this paper contributes to improve the process performance through a novel wire breakage monitoring and diagnosing system. It consists of two well differentiated parts: the virtual instrumentation system (VIS) that measures relevant magnitudes, and the diagnostic system (DS) that detects low quality cutting regimes and predicts wire breakage. It has been successfully validated through a considerable number of experimental tests performed on an industrial WEDM machine for different workpiece thickness. The efficiency of the supervision system has been quantified through an efficiency rate defined in this paper.     

Improved PI neural network-based tension control for stranded wire helical springs manufacturing

During the winding process of stranded wire helical springs (SWHSs), uneven wire tension always results in high rejection rate and non-compliance service life of SWHSs. Combining the proportion integral neural network (PINN) with a simplified actuator model, this paper presents a new control scheme for the SWHS CNC machine to keep the wire tension uniform. The PINN is improved by introducing an error variance ratio, accounting for the interaction between wires, as a modifying factor in the second hidden layer. The actuator model is simplified based on the analysis of the dynamic characteristics of the actuator. The output value of the improved PINN is transferred into control voltage value by the simplified model. The tension of each wire is controlled by an improved PINN. In order to enhance the control performance, the network parameters are updated using the gradient-based back-propagation method. The validity and consistency of the improved PINN are verified by experiments. The results indicate that (1) the computation load is slight; (2) the rising time of the step response is within 1 s; (3) 89%-96% of tension deviation values of the wire 1 and wire 3 under different process parameters are within 10% of the reference tension value; (4) the standard deviation of the wire 2 with large disturbance is 8.24 N. Compared with other algorithms (incremental PI, multiple PIDNN, PI based particle swarm optimization), the control scheme based on the improved PINN has less computation load, faster response speed and better performance in the time-varying and nonlinear system with larger disturbance.     

Solid modeling of 4-axis wire EDM cut geometry

NC verification has been widely used in milling and turning machining, but the implementation of NC verification for wire electrical discharge machine (wire EDM) has received little attention, made difficult by the characteristics of geometry cut by the thin wire. For an NC verification system, the speed and the accuracy of are two of the most critical issues. In this paper, a novel solid modeling method, which utilizes a dual quadtree structure and a boundary representation, is applied to modeling the parts cut by a wire EDM. In this implementation, two extended quadtrees on the top and bottom surfaces of the stock are first created. Special quadtree nodes are introduced into the dual quadtree structure to model the features associated with the almost coincident surfaces left by a thin wire. The region swept by a wire in each move is generated based on the diameter of the wire, spark gap size and the paths. The overall geometry of the part is represented by a boundary representation which is updated by means of efficient Boolean operations of extended quadtrees. The free-falling objects which may damage the machine tool can be detected. Dimensional inspection and feature detection can be made with the solid model representing the machined part. The system has demonstrated very promising speed and accuracy.     

Robotized recognition of a wire harness utilizing tracing operation

In assembly lines, deformation of the objects makes many efforts on automating the corresponding processes difficult. Mating of a wire harness onto a car body is among the representative tasks. Academically, this kind of problems are known as automatic manipulation of deformable linear objects. Previous technical approaches for solving the problem is characterized by employment of multiple contactless sensors and complex sensor based algorithms. These attempts are verified to be feasible only for simplified environments. The dependence of the approaches on real-time measurement to the state of the deformable targets limits their application in practical factory environment. In this paper, it is proposed to utilize wire tracing operation in recognizing the wire harness for the purpose of automatic mating. This method is inspired from the behaviors of humans demonstrated when they deal with similar problems. The proposed method is implemented and verified in a robot system. The feasibility and effectiveness demonstrated in the preliminary experiments show the potential of the proposed method for substantially simplifying the wire recognition problems which is thought to be crucial for practices in factory environments.     

Advanced adaptive feed control for CNC machining

In computer numerical control (CNC) machining, the tool feed rate is crucial for determining the machining time. It also affects the degree of tool wear and the final product quality. In a mass production line, the feed rate guides the production cycle. On the other hand, in single-time machining, such as for molds and dies, the tool wear and product quality are influenced by the length of machining time. Accordingly, optimizing the CNC program in terms of the feed rate is critical and should account for various factors, such as the cutting depth, width, spindle speed, and cutting oil. Determining the optimal tool feed rate, however, can be challenging given the various machine tools, machining paths, and cutting conditions involved. It is important to balance the machining load by equalizing the tool's load, reducing the machining time during no-load segments, and controlling the feed rate during high load segments. In this study, an advanced adaptive control method was designed that adjusts the tool feed rate in real time during rough machining. By predicting both the current and future machining load based on the tool position and time stamp, the proposed method combines reference load control curves and cutting characteristics, unlike existing passive adaptive control methods. Four different feed control methods were tested including conventional and proposed adaptive feed control. The results of the comparative analysis was presented with respect to the average machining load and tool wear, the machining time, and the average tool feed speed. When the proposed adaptive control method was used, the production time was reduced up to 12.8% in the test machining while the tool life was increased.     

Online tool wear prediction system in the turning process using an adaptive neuro-fuzzy inference system

Tool wear is a detrimental factor that affects the quality and tolerance of machined parts. Having an accurate prediction of tool wear is important for machining industries to maintain the machined surface quality and can consequently reduce inspection costs and increase productivity. Online and real-time tool wear prediction is possible due to developments in sensor technology. Recently, various sensors and methods have been proposed for the development of tool wear monitoring systems. In this study, an online tool wear monitoring system was proposed using a strain gauge-type sensor due to its simplicity and low cost. A model, based on the adaptive network-based fuzzy inference system (ANFIS), and a new statistical signal analysis method, the I-kaz method, were used to predict tool wear during a turning process. In order to develop the ANFIS model, the cutting speed, depth of cut, feed rate and I-kaz coefficient from the signals of each turning process were taken as inputs, and the flank wear value for the cutting edge was an output of the model. It was found that the prediction usually accurate if the correlation of coefficients and the average errors were in the range of 0.989–0.995 and 2.30–5.08% respectively for the developed model. The proposed model is efficient and low-cost which can be used in the machining industry for online prediction of the cutting tool wear progression, but the accuracy of the model depends upon the training and testing data.     

人工神经网络预测刀具磨损和切削力

刀具磨损和切削力预测与控制是切削加工过程中需要考虑的重要问题.本文介绍了利用人工神经网络模型预测刀具磨损和切削力的步骤并且针对产生误差的因素进行分析.首先将切削速度、切削深度、切削时间、主轴转速和不同频带的能量值通过归一化法处理,作为输入特征值,对改进的神经网络模型进行训练.然后利用训练完成的神经网络模型预测刀具磨损和切削力.结果表明:神经网络模型能够综合考虑加工过程中更多的影响因素,与经验公式结果对比,具有更高的预测精度.研究结果表明神经网络模型预测刀具磨损和切削力具有可行性和准确性,为刀具结构的优化及加工参数的选择提供了依据.     

Development of a high speed and precision wire clamp with both position and force regulations

This paper presents the mechanism and robust control of a monolithic wire clamp to achieve fast and precision operations for strong and robust micro device packaging. The wire clamp is piezoelectrically actuated and a two-stage flexure-based amplification was designed to obtain large and parallel jaw displacements. The grasping forces of the wire clamp were evaluated based on finite element analysis (FEA), and the force measurement was presented. The wire clamp was manufactured using wire EDM technique and the position and force transfer functions were obtained based on the frequency response approach. The position/force switching control strategy was employed to regulate the motion position and grasping force, and the position/force switching controller composed of a PID controller for position control and a sliding model controller (SMC) for force control was designed. Experimental tests were carried out to investigate the performance of wire clamp with the position/force switching controller during the grasping and releasing operations. The results show that the wire clamp exhibits good performance and demonstrate that high speed and precision grasping operations can be realized through the developed wire clamp and the control strategy.     

Numerical simulations of the laminar flow in pipes with wire coil inserts

Helical wire coils fitted inside a round pipe is a simple and well-known heat transfer enhancement technique in order to improve the overall performance of heat exchangers. Three-dimensional numerical simulations of the incompressible laminar flow that develops into smooth round pipes of diameter, d, with wire coil inserts of helical pitch, p, and diameter, e, have been accomplished with the finite volume method. In particular, we describe the behaviour of the Fanning friction factor, f, as a function of the Reynolds number, Re = ρUd/μ, where, U = 4Q/πd², is the mean velocity based in the flow rate, Q, and ρ and μ the density and dynamic viscosity of the fluid, respectively. For a wire coil of 40 pitches in length with dimensionless pitch p/d = 2.5 and dimensionless wire diameter e/d = 0.074, both pitch-periodic and full domain numerical results have been validated with experiments. We have found an excellent agreement with both numerical models and experimental results for Re < 500, showing the friction factor a quasi-linear dependence on Re when is plotted in log–log axes. For 500 < Re < 600 both experimental and full domain numerical results of the values for the friction factor leave the quasi-linear trend observed for Re < 500. Our full domain numerical calculations reveal the onset of a linear instability into the range 500 < Re < 550 that becomes the flow unsteady and breaks the periodic axial pattern of the flow. The friction factor becomes constant in the range, 600 < Re < 850, and only the full numerical model shows a good agreement with the experimental results, but periodic numerical simulations fail. For 850 < Re, even the full domain laminar model fails due to the onset of turbulent outbreaks. Finally, the effect of the pitch on the friction factor has been addressed by performing a parametrical study with a pitch-periodic computational domain for wire coils within the dimensionless pitch range, 1.50 ? p/d ? 4.50, and dimensionless wire diameter, e/d = 0.074, showing that the increase of the nondimensional pitch, p/d, decreases the friction factor.     

A novel approach to CNC machining center processing parameters optimization considering energy-saving and low-cost

NC machining is currently a machining method widely used in mechanical manufacturing systems. Reasonable selection of process parameters can significantly reduce the processing cost and energy consumption. In order to realize the energy-saving and low-cost of CNC machining, the cutting parameters are optimized from the aspects of energy-saving and low-cost, and a process parameter optimization method of CNC machining center that takes into account both energy-saving and low -cost is proposed. The energy flow characteristics of the machining center processing system are analyzed, considering the actual constraints of machine tool performance and tool life in the machining process, a multi-objective optimization model with milling speed, feed per tooth and spindle speed as optimization variables is established, and a weight coefficient is introduced to facilitate the solution to convert it into a single objective optimization model. In order to ensure the accuracy of the model solution, a combinatorial optimization algorithm based on particle swarm optimization and NSGA-II is proposed to solve the model. Finally, take plane milling as an example to verify the feasibility of this method. The experimental results show that the multi-objective optimization model is feasible and effective, and it can effectively help operators to balance the energy consumption and processing cost at the same time, so as to achieve the goal of energy conservation and low-cost. In addition, the combinatorial optimization algorithm is compared with the NSGA-II, the results show that the combinatorial optimization algorithm has better performance in solving speed and optimization accuracy.     

A novel vision-based method for 3D profile extraction of wire harness in robotized assembly process

Automating stages for deformable objects in the production line, in which assembling a wire harness into a predefined position is a complex task owing to the specialized characteristics of the objects. Besides a few automatized systems proposed in the other studies to implement this task under simplified setup conditions, a significant portion of this process remains to be completed manually in industrial environments. To construct an automatic wire harness assembly system, the development of a method that can automatically detect the wire harness profile in a 3D environment and, consequently, guide robot arms to implement assembly tasks is indispensable. Therefore, this study presents an approach that satisfies this requirement, which not only proposes a deep learning-based system to detect the wire profile, but also improves the accuracy of the detected results through a correction method according to the depth values of contiguous areas. The verification of the approach in a robot system that highlights its usefulness and practicality demonstrates the potential of the proposed method to replace people and consequently, reduce labour costs in factory environments.     

High aspect ratio meso-scale parts enabled by wire micro-EDM

Micro-electro discharge machining (EDM) is a subtractive meso-scale machining process. The Agie Excellence 2F wire micro EDM is capable of machining with a 25 micron diameter wire electrode and positioning the work piece to within ±1.5 microns. The over-burn gap can be controlled to within 3 microns to obtain a minimum feature radius of about 16 microns while achieving submicron surface finish and an imperceptible recast layer. For example, meso-scale gears that require vertical sidewalls and contour tolerances to within 3 microns can be wire EDMed into a variety of conductive materials. Material instabilities can affect the dimensional precision of machined meso-scale parts by material relaxation during the machining process. A study is done to investigate the machining performance of the wire micro EDM process by machining a high aspect ratio meso-scale part into a variety of metals (e.g. 304L stainless steel, Nitronic 60 Austentic Stainless, Beryllium Copper, and Titanium). Machining performance parameters such as, profile tolerance, perpendicularity, and repeatability are compared for the different materials. Pertinent inspection methods desirable for meso-scale quality assurance tasks are also evaluated. Sandia National Laboratories is developing meso-scale electro-mechanical components and has an interest in the assembly implications of piece parts fabricated by various meso-scale manufacturing processes. Although the wire EDM process is typically used to fabricate 2½ dimensional features, these features can be machined into a 3 dimensional part having other features such as hubs and chamfers to facilitate assembly.     

Computer aided contouring operation for traveling wire electric discharge machining (EDM)

This paper develops a computational method for numerical control (NC) of traveling wire electric discharge machining (EDM) operation from geometric representation of a desired cut profile in terms of its contours. Normalized arc length parameterization of the contour curves is used to represent the cut profile and a subdivision algorithm is developed together with kinematic analysis to generate the required motions of the machine tool axes. In generating the tool motions for cutting sections with high curvatures such as corners with small radii, a geometric path lifting method is presented that increases the machining gap and prevents gauging or wire breakage.     

Spindle thermal error prediction approach based on thermal infrared images: A deep learning method

It is essential to precisely model the spindle thermal error due to its dramatic influence on the machining accuracy. In this paper, the deep learning convolutional neural network (CNN) is used to model the axial and radial thermal errors of horizontal and vertical spindles. Unlike the traditional CNN model that relies entirely on thermal images, this model combines the thermal image with the thermocouple data to fully reflect the temperature field of the spindle. After pre-processing and data enhancement of the thermal images, a multi-classification model based on CNN is built and verified for accuracy and robustness. The experimental results show that the model prediction accuracy is approximately 90 %–93 %, which is higher than the BP model. When the spindle rotation speed changes, the model also shows good robustness. Real cutting tests show that the deep learning model has good applicability to the spindle thermal error prediction and compensation.     

SMA驱动丝的建模与仿真

研究SMA动力学优化模型,针对形状记忆合金(SMA)驱动丝具有强非线性、迟滞效应等特性,为设计SMA驱动丝的自适应结构,提出建立SMA驱动丝模型并提供高效的仿真方法。采用有限元软件实现了受轴向载荷的SMA驱动丝的仿真建模。对本构模型是根据自由能的一维热-力学耦合模型,可以同时复现形状记忆效应和超弹性。数值仿真能够引起材料相变的非均匀温度和应变分布。仿真结果表明,建立热-力学耦合模型,可为设计SMA驱动丝的自适应结构计算提供可靠依据。     

三维集成车削力传感器数据采集与分析系统设计

切削力与切屑形成、切削热、刀具磨损和切削振动等现象有着密切联系,是影响加工精度、刀具寿命和切削效率的重要因素.通过实时测量切削力,及时调整切削参数、优化切削工艺,对于保证加工质量、延长刀具寿命、提高切削效率等有着重要意义.切削力的准确测量和处理离不开优良的数据采集与分析系统,针对基于MEMS压阻式芯片的三维集成车削力传感器,以微处理器STM32为控制核心研制了一种三维集成车削力传感器数据采集与分析系统,实现了三维车削力的标定、实时采集和数据分析功能.     

中国微米纳米-三维集成车削力传感器数据采集与分析系统设计

切削力与切屑形成、切削热、刀具磨损和切削振动等现象有着密切联系,是影响加工精度、刀具寿命和切削效率的重要因素。通过实时测量切削力,及时调整切削参数、优化切削工艺,对于保证加工质量、延长刀具寿命、提高切削效率等有着重要意义。切削力的准确测量和处理离不开优良的数据采集与分析系统,本文针对基于MEMS压阻式芯片的三维集成车削力传感器,以微处理器STM32为控制核心研制了一种三维集成车削力传感器数据采集与分析系统,实现了三维车削力的标定、实时采集和数据分析功能。     

Selection of cutting conditions for power constrained parallel machine scheduling

Most previous studies on machining optimization focused on aspects related to machining efficiency and economics, without accounting for environmental considerations. Higher cutting speed is usually desirable to maximize machining productivity, but this requires a high power load peak. In Taiwan, electricity prices rise sharply if instantaneous power demand exceeds contract capacity. Many studies over the previous decades have examined production scheduling problems. However, most such studies focused on well-defined jobs with known processing times. In addition, traditional sequencing and scheduling models focus primarily on economic objectives and largely disregard environmental issues raised by production scheduling problems. This study investigates a parallel machine scheduling problem for a manufacturing system with a bounded power demand peak. The challenge is to simultaneously determine proper cutting conditions for various jobs and assign them to machines for processing under the condition that power consumption never exceed the electricity load limit. A two-stage heuristic approach is proposed to solve the parallel machine scheduling problem with the goal of minimizing makespan. The heuristic performance is tested by distributing 20 jobs over 3 machines with four possible cutting parameter settings.     

A standard deviation based firefly algorithm for multi-objective optimization of WEDM process during machining of Indian RAFM steel

Non-conventional machining processes always suffer due to their low productivity and high cost. However, a suitable machining process should improve its productivity without compromising product quality. This implies the necessity to use efficient multi-objective optimization algorithm in non-conventional machining processes. In this present paper, an effective standard deviation based multi-objective fire-fly algorithm is proposed to predict various process parameters for maximum productivity (without affecting product quality) during WEDM of Indian RAFM steel. The process parameters of WEDM considered for this study are: pulse current (I), pulse-on time (T _on), pulse-off time (T _off) and wire tension (WT).While, cutting speed (CS) and surface roughness (SR) were considered as machining performance parameters. Mathematical models relating the process and response parameters had been developed by linear regression analysis and standard deviation method was used to convert this multi objective into single objective by unifying the responses. The model was then implemented in firefly algorithm in order to optimize the process parameters. The computational results depict that the proposed method is well capable of giving optimal results in WEDM process and is fairly superior to the two most popular evolutionary algorithms (particle swarm optimization algorithm and differential evolution algorithm) available in the literature.

     

Hybrid modeling of wire cable vibration isolation system through neural network

A hybrid modeling method based on neural network (NN) is developed and used to model the hysteretic restoring force of a wire cable vibration isolation system for electronic equipment. Firstly, a knowledge-based model for the nonlinear hysteretic restoring force is identified using the measured data obtained from period loading tests. Secondly, the remaining characteristic of hysteretic restoring force, which cannot be modeled in an easy way, is identified using the NN method through network training. By building up a parallel hybrid NN model for the nonlinear hysteretic restoring force, the dynamic responses of the vibration isolation system under harmonic and broad band random excitations are predicted. The predicted results are compared with the measured ones to validate the effectiveness and prediction accuracy of the model. The comparative studies show the developed hybrid NN model possesses good prediction accuracy and generalization capability in contrast with the pure black box NN model.