基于机器学习的回流焊焊点形貌预测

提出一种基于机器学习预测回流焊焊点形貌的方法,通过该方法建立一个针对钽电容回流焊焊点形貌的预测模型,该模型为现有实验方式提供了新的思路。通过峰值温度、降温速率和焊膏厚度3种影响因素以及焊点厚度、焊点宽度和焊料爬高3种评价焊点形貌的评价标准,分别基于BPNN和LightGBM算法建立钽电容回流焊焊点形貌预测模型。对比实验证明,通过LightGBM算法建立的预测模型优于通过BPNN建立的预测模型,并通过实际测试帮助实验人员减少实验次数,节约大量时间成本。     

Prediction of TC11 single-track geometry in laser metal deposition based on back propagation neural network and random forest

Laser metal deposition process usually involves the nonlinear interaction of multiple factors, such as process parameters and ambient temperature. In this study, random forest (RF) and multilayer back propagation neural network (BPNN) algorithms were employed to investigate the coupling relationship between process parameters and single-track geometry in laser metal deposition for TC11 alloy. With laser power, scanning speed, and powder feeding rate as inputs and track width and height as outputs, 30 different groups of experimental results were adopted as training groups. Their geometries were also predicted. The maximum relative errors of track width and height predictions based on BPNN model were 0.007 % and 0.029 %, respectively, which were lower than those based on RF model. Then, the two models were used to predict the geometry under four new sets of process parameters. Experimental results showed that the maximum error of BPNN model is lower than that of RF model. BPNN model also showed potential to improve cladding quality and efficiency.

     

基于量子微粒群的BPNN在转炉炼钢静态模型中的应用

针对转炉炼钢静态模型终点命中率较低的问题,首先分析了影响转炉炼钢终点命中率的各种因素,确定了BP神经网络(BPNN)的拓扑结构,并依此建立了转炉炼钢静态模型.然后把量子微粒群算法(QPSO)应用于BP网络的学习中,并比较了QPSO、基本微粒群优化算法(PSO)、梯度下降法的学习性能.最后,基于某炼钢厂的历史数据进行了仿...     

液压挖掘机臂杆结构疲劳寿命预测方法研究

液压挖掘机臂杆结构承受复杂的冲击载荷,其疲劳寿命存在许多不确定性因素。首先采用Miner准则,依据实验载荷谱和有限元方法对液压挖掘机工作装置寿命进行了预测。其次研究了结构疲劳寿命变化过程和灰色理论预测模型内在规律的一致性,建立了液压挖掘机工作装置疲劳寿命的灰色预测GM模型,并分别运用GM模型的两种形式--GM(1,1)线性模型及GM(1,1)幂模型对液压挖掘机工作装置进行疲劳寿命预测。分析比较Miner准则、GM(1,1)线性模型及GM(1,1)幂模型三种预测方法。结果表明,三种预测方法结果基本一致,灰色系统模型同Miner准则模型相比误差明显减小且非线性幂模型具有更高的预测精度。基于灰色理论的GM(1,1)幂模型考虑了非线性因素,更适合于液压挖掘机工作装置结构疲劳寿命预测。     

Springback prediction of high-strength sheet metal under air bending forming and tool design based on GA�CBPNN

Based on orthogonal test for air bending of high-strength steel sheets, 125 values of sheet thickness (t), tool gap (c), punch radius (r), ratio of yield strength to Young??s modulus (?? _y /E), and punch displacement (e) are used to model the springback for air bending of high-strength sheet metal using the genetic algorithm (GA) and back propagation neural network (BPNN) approach, where the positive model and reverse model of springback prediction are established, respectively, with GA and BPNN. Adopting the ??object-positive model?Creverse model?? learning method, air bending springback law is studied with positive model and punch radius is predicted by reverse model. Manifested by the experiment for air bending forming of a workpiece used as crane boom, the prediction method proposed yields satisfactory effect in sheet metal air bending forming and punch design.     

Multi-variable measurements and optimization of GMAW parameters for API-X42 steel alloy using a hybrid BPNN–PSO approach

This research addresses multi criteria modeling and optimization procedure for Gas Metal Arc Welding (GMAW) process of API-X42 alloy. Experimental data needed for modeling are gathered as per L₃₆ Taguchi matrix. Model inputs include work piece groove angle as well as the five main GMAW process parameters. The proposed back propagation neural network (BPNN) simultaneously predicts weld bead geometry (WBG) and heat affected zone (HAZ). Image processing technique along with Bridge Cam and AWS gauges are used to take accurate measurements of WBGs and HAZs. The adequacy of the developed BPNN is established through comparisons against measured process outputs. Measurements indicate that the BPNN model simulates GMAW process with average errors of 0.33 to 0.82%. Next, the BPNN model is implanted into a particle swarm optimization (PSO) algorithm to simultaneously optimize HAZ and WBG characteristics. The hybrid BPNN–PSO determines process parameters values and groove angle so as a desired WBG is achieved while HAZ is minimized. Verification tests demonstrate that the proposed BPNN–PSO is quite efficient for in multi-criteria modeling and optimization of GMAW.     

Performance prediction of a nozzled and nozzleless mixed-flow turbine in steady conditions

This paper presents a meanline model to predict the performance parameters of a turbocharger turbine under steady state conditions. The turbine was developed at Imperial College and the design was based on a commercial nozzleless unit that was modified into a variable geometry single-entry turbine.The wide range of tests data from the Imperial College Turbocharger Group dynamometer enabled the evaluation of the model in the areas of the turbine map where currently no previous comparison had been made in the literature. This facility is designed to allow testing over a wide range of velocity ratios (0.3-1.1) previously unavailable with conventional test stands.The nozzleless turbine model was validated against experimental results spanning an equivalent speed range of 27.9 and 53.8 rev/s √K while for the nozzled case the model was validated against one single speed (43.0 rev/s √K) and three different vane angle settings (40°, 60° and 70°).The results of the model simulation showed that the performance can be predicted with excellent accuracy for different turbine speeds and vane angles. Based on the model prediction, a breakdown aerodynamic loss was performed.     

A microstructure evolution model for numerical prediction of austenite grain size distribution

In this study, a mathematical model has been developed to predict austenite grain size (AGS) of hot rolled steel. Using the compression test, the static (SRX) and metadynamic (MDRX) recrystallization characteristics of medium carbon steel were studied. Compression tests were carried out at various temperatures in the range 900-1100 °C with strain rates ranging from 0.1 to 10 s⁻¹. The time required for 50% recrystallization for the SRX and MDRX was determined by carrying out double compression tests, respectively. Grain growth equation after full recrystallization was also derived by compression tests with various interpass times. The currently determined microstructure model has been integrated with a three-dimensional non-isothermal finite element program. The predicted results based on the model proposed in the present investigation for hot bar rolling processes were compared with the experimental data available in the literature. It was found that the proposed model was beneficial to understand the effect of recrystallization behavior and control the microstructure evolution during the hot bar rolling.     

Thermostatting System Using a Single-Chip Microcomputer and Thermoelectric Modules Based on the Peltier Effect

A flexible temperature-controlling system matched with the liquid thermostatting circuit is based on a single-chip microcomputer. Thermoelectric modules based on the Peltier effect are used for sample cooling and heating. The accuracy of the temperature maintenance is better than 0.1°C in a range of –20 to +70°C. The system is designed for NMR relaxometers but can be also used to control the temperatures of any volume-comparable objects (50 cm³).     

Development of microcontroller-based microwave system to measure solid content of hevea rubber latex

This paper describes the development of digital, portable, simple to operator, cheap and accurate microwave instrument to determine the total solid content (TSC) or moisture content (MC) of hevea rubber latex by using microcontroller, dual directional coupler, and open-ended coaxial probe. A low-cost stripline dual directional coupler and open-ended coaxial probe are simulated, designed and connected to microwave source and detectors. TSC parameter is obtained from the relationship between moisture content in the sample and the value of reflection coefficient from coaxial probe. Experimental results are presented and the empirical equation was implemented by the microcontroller for the calibration and followed the calculation setup. The whole system was tested by using diluted rubber latex with different TSCs at room temperature (25 °C) and the results of standard oven drying method and this study were found with the accuracy and reproducibility at the level of less than 1%.     

基于深度信念网络的风机主轴承状态监测方法

提出了一种基于深度信念网络（DBN）的风电机组主轴承状态监测方法。为了降低建模难度并减少训练时间,首先利用相关系数法选取建模变量,进而建立主轴承正常行为的DBN温度模型并用于主轴承温度预测。该模型克服了传统神经网络随机初始化网络权重、易陷入局部最小值等缺点,能有效提高主轴承温度的预测精度。然后采用指数加权移动平均法（EWMA）对主轴承温度残差序列进行分析,并利用核密度估计方法确定故障阈值。最后基于实测的数据采集与监视控制（SCADA）系统数据对主轴承故障进行模拟。结果表明,与传统预测方法相比,该方法能有效地实现主轴承的异常状态监测。     

基于类神经网络注塑成型翘曲和收缩值之预测

以熔融温度、模具温度、射出时间、保压压力、保压时间等5个制程参数作为控制因子。利用Moldflow来模拟塑料薄壳挡板不同的成型制程参数下的翘曲与收缩值。基于仿真所得翘曲及收缩值数据,使用田口方法结合倒传递神经网络5-14-14-2建立预测模型。再利用测试样本来验证的倒传递神经网络模型的准确性。运用所建立的倒传递神经网络模型预测其他成型制程参数的翘曲及收缩值。结果证明,田口法结合倒传递神经网络,不仅可以有效的优化倒传递神经网络,而能成功的预测翘曲及收缩值,与Moldflow仿真值相比平均误差都在±1%内。     

Oxidation-abrasion of TiC-based cermets in SiC medium

The aim of the current study was to investigate the effect of oxidation on abrasive wear behaviour of TiC based cermets at temperatures ranging from 20 to 900 °C. Three types of material performance maps were constructed: oxidation rate maps, wear rate maps and maps showing the effect of oxidation on abrasion. Discussion on the performance of different cermet grades is supported by the SEM images combined with EDS and XRD analysis. The results should facilitate the selection of TiC-based cermets providing optimum composition of cermets for high temperature applications.     

有源相控阵雷达阵面热变形预测建模理论

目前大部分研究均通过软件仿真的方法来预测有源相控阵雷达阵面热变形,但仿真结果与真实情况存在一定偏差,为此,提出了有源相控阵雷达热变形预测建模理论。通过实验测量得到阵面热变形和温度变化,依据模糊聚类结合灰色关联算法对温度测点进行优化,建立各方向热变形量的预测模型。实验结果表明：该模型具有较高的预测精度和稳健性,从而可实现阵面热变形的准确预测。     

基于主动学习GASVM分类器的连铸漏钢预报

针对在小样本数据情况下训练的连铸漏钢预报模型难以获得较高预报准确率的问题,提出了一种基于主动学习遗传算法-支持向量机（GASVM）分类器的漏钢预报算法。该算法首先将采集到的连铸结晶器坯壳温度数据进行预处理,并将有效数据进行标注;然后利用标注后的小样本数据和遗传算法来优化SVM的经验参数,训练并得到支持向量机模型;最后利用某钢厂采集到的连铸结晶器坯壳温度数据进行测试。测试结果表明,在利用小样本数据进行训练的情况下,所提出的基于主动学习GASVM分类器的连铸漏钢预报算法具有较高的漏钢预报率（预报精度）和100%的漏钢报出率,验证了所提漏钢预报算法的有效性。     

精密数控车床主轴热误差建模

开展了精密数控车床主轴系统热误差补偿的实验与建模方法的研究。建立了精密数控车床主轴系统轴向与径向偏转热误差补偿模型以增强其误差补偿能力,并提高机床加工精度。构建了主轴系统热误差测试平台,应用五点法测试主轴系统热误差,使用热电偶与红外热像仪测量主轴系统温升关键点温度变化数据,应用灰色综合关联分析法实现温度敏感测点辨识。构建了基于粒子滤波重采样粒子群算法的热误差预测模型,对模型预测效果进行评价。结果表明:基于粒子滤波重采样粒子群热误差补偿模型得到的轴向热误差预测残差为-1.29μm~1.55μm,建模精度为95.04%;y向热偏转误差预测残差为-4.68×10~(-6°)~9.66×10~(-6°),建模精度为91.26%;z向热偏转误差预测残差为-5.83×10~(-6°)~8.59×10~(-6°),建模精度为93.24%。实验结果证明该热误差补偿模型具有较高的预测精度,具有较强的工程应用价值。     

Aero-engine blade fatigue analysis based on nonlinear continuum damage model using neural networks

Fatigue life and reliability of aero-engine blade are always of important significance to flight safety.The establishment of damage model is one of the key factors in blade fatigue research.Conventiona...     

基于FARIMA模型的Internet时延预测

针对Internet时延具有自相似性这一特点,采用自回归分数滑动平均模型(fractal autoregressive integrated moving aver-age,FARIMA)对Internet时延建模,提出了基于概率上限的Internet时延预报方法,即保证实际时延按一定概率在预测时延范围之内。通过对实测时延数据进行预测对比,结果表明基于FARIMA模型的预测效果要优于基于ARMA(auto regnessive and mov-ing average)模型的预测效果。     

Parameter prediction in laser bending of aluminum alloy sheet

Based on the basic platform of BP neural networks, a BP network model is established to predict the bending angle in the laser bending process of an aluminum alloy sheet (1–2 mm in thickness) and to optimize laser bending parameters for bending control. The sample experimental data is used to train the BP network. The nonlinear regularities of sample data are fitted through the trained BP network; the predicted results include laser bending angles and parameters. Experimental results indicate that the prediction allowance is controlled less than 5%–8% and can provide a theoretical and experimental basis for industry purpose. __________ Translated from Optics and Precision Engineering, 2007, 15(6): 915–921 [译自: 光学精密工程     

A wrapper approach-based key temperature point selection and thermal error modeling method

A wrapper approach-based key temperature point selection and thermal error modeling method is proposed to concurrently screen the optimal key temperature points and construct the thermal error model. This wrapper approach can strengthen the intrinsic relation between the key temperature points and the thermal error model to ensure the strong prediction performance. On the whole, the least squares support vector machine (SVM) is used as the basic thermal error modeling method and the binary bat algorithm (BBA) is used as the optimization algorithm. The selection status of temperature points and the values of hyperparameters γ and σ2 of SVM are coded in separate binary parts of the artificial bat’s position vector of BBA. The cost function is designed by balancing the prediction error and the number of key temperature points. For verification, the thermal error experiment was conducted on a horizontal machining center. Feeding the collected experimental temperature data and thermal error data to the proposed method, three optimal key temperature points were screened out and the corresponding optimal hyperparameters were simultaneously searched. To verify the superiority of the proposed method, the prediction performance comparison analysis was conducted with the conventional filter-based method. Specifically, in the conventional method, the key temperature points were screened by combining fuzzy c means (FCM) clustering and correlation analysis, and the multiple linear regression (MLR), the backpropagation neural network (BPNN), and the SVM were used to build the thermal error model, respectively. Comparison results showed that the prediction accuracy of the proposed method increased by up to 44.0% compared to the conventional method, which suggests the superior prediction performance of the proposed method.