A real-time predictive maintenance system for machine systems

This paper describes a novel real-time predictive maintenance system for machine systems based upon a neural network approach. The ability of a neural network to learn non-linear mapping functions has been used for the prediction of machine system parameters using the motion current signature. This approach avoids the need for costly measurement of system parameters. Unlike many neural network based condition monitoring systems, this approach is validated in an off-line proof of concept procedure, using data from an experimental test rig providing conditions typical of those used on production machines. The experiment aims to classify five distinct motor loads using the motion current signature, irrespective of changing tuning parameters. Comparison of the predicted and actual loads shows good agreement. Generation of data covering all anticipated machine states for neural network training, using a production machine, is impractical, and the use of simulated data, generated by an experimentally validated simulation model, is effective. This paper demonstrates the underlying structure of the developed simulation model.     

BJEST: A reverse algorithm for the real-time predictive maintenance system

An algorithm, to estimate the machine system parameters from the motion current signature, based upon non-linear time series techniques for use in the real-time predictive maintenance system is presented in this paper. Earlier work has introduced the use of a neural-network approach to learn non-linear mapping functions for condition monitoring systems. However, the performance of the neural-network largely depends upon the quality of the training data, and that of the quality and type of the pre-processing of the input data. A reverse algorithm called BJEST (Bansal–Jones Estimation), for estimating the machine input parameters using the motion current signature, has been designed and proven to be successful in estimating the macro-dynamics of the motion current signature. This motivated the enhancement of the predictive analysis to incorporate non-linear characteristic of the motion current signature. The results show considerable improvement in the estimation of the parameters using the enhanced BJEST algorithm due to estimation consistency, hence, improving the real-time predictive maintenance system.     

基于神经网络的虚拟轴机床轨迹跟踪控制方法的研究

针对永磁直线电机驱动的六自由度虚拟轴机床提出基于神经网络的轨迹跟踪控制方法.采用神经网络来实现永磁直线电机伺服系统的位置和速度控制,将负载扰动、杆间耦合扰动等效地看成电动机模型参数的变化,通过调整权值来补偿不确定性扰动对轨迹跟踪控制的影响,保证了系统的鲁棒性.仿真实验结果表明,用该方法设计的虚拟轴机床控制系统具有良好的跟踪给定和抑制扰动的效果,从而保证机床运动协调、姿态合理.     

基于数字化车间的改进图像识别系统

数字化车间的主要特点是对于普通车间的重新定义,是通过虚拟的方式将传统车间构建到虚拟环境中,对整个生产过程进行虚拟仿真推进生产。针对数字化车间虚拟信息化的识别精度问题,传统的图像识别系统对于数字化车间中精密仪器的识别准确率偏低,文章中的图像识别系统是通过针对数字化车间与卷积神经网络^[1]的特点进行改进,通过将卷积神经网络与支持向量机相结合,将分类模型加入到图像识别中,最终提高识别准确率,稳定高效的帮助用户进行仿真从而推进生产。     

人体下肢多关节连续运动的肌电预测方法

为了准确捕捉人体下肢关节在不同运动模式下的运动状态，提出一种利用肌电信号进行下肢多关节连续运动预测的方法。采集人体在蹲起运动、膝屈伸运动和上下阶梯运动时的肌电和运动数据进行处理分析，利用肌肉骨骼几何建模软件Opensim建立人体骨骼肌肉仿真模型，并进行逆运动学分析，提取人体下肢关节运动曲线。建立人体下肢在矢状面内的运动与肌电的映射关系，利用麻雀搜索算法优化Elman神经网络，实现对踝、膝和髋关节角度连续变化的预测，与传统的反向传播神经网络、支持向量机回归和Elman神经网络预测结果进行对比。结果表明：利用麻雀搜索算法优化的Elman神经网络在预测下肢关节角度变化中具有更高的精度，且该预测模型在不同运动模式下关节运动预测值与测量值均表现出一定的相关性，相关系数均大于0.89,证明利用肌电信号进行下肢多关节连续运动预测是可行的。     

神经网络在覆盖件模具表面激光硬化虚拟过程中的应用

基于人工神经网络技术对覆盖件模具表面激光硬化虚拟过程的仿真建模,结合几何因素分析了模型的主要影响参数,对BP网络的结构和训练进行了说明。预测了激光表面硬化的加工效果(表面硬度、硬化层深、相对耐磨性和表面粗糙度),实现激光加工工艺参数的优化,为实际生产和加工提供了依据。并以C语言为开发语言,利于实现各平台间的集成。     

基于细胞神经网络刀具磨损图像的预处理

提出了一种基于细胞神经网络的刀具磨损图像处理方法，通过设计细胞神经网络参数，运用细胞神经网络对刀具的二值图像平滑滤波，边缘提取，通过仿真证明该方法是有效的，由于细胞神经网络易于用VLSI实现并且并行处理速度快，因此该方法对刀具的磨损状态机器视觉检测中的图像处理具有实用意义。     

枕式包装机伺服控制系统的研究与优化

为了确保枕式包装机各项功能能够得到精确高效的运行,对神经网络控制策略进行了具体的研究。通过建模仿真,对比分析经典PID控制与神经网络PID控制的优劣,论证了神经网络PID的控制方法更有优势。在优化完成的基础上,开发出色标切功能。对包装机进行实际生产测试结果表明,整套系统在完成相应功能的过程中满足了稳定性和包装精度的要求,具有良好的应用价值。     

Optimization of electric discharge machining using simulated annealing

This paper proposes an optimization methodology for the selection of best process parameters in electro-discharge machining. Regular cutting experiments are carried out on die-sinking machine under different conditions of process parameters. The system model is created using counter-propagation neural network using experimental data. This system model is employed to simultaneously maximize the material removal rate as well as minimize the surface roughness using simulated annealing scheme. Finally consistency of the method is tested with several initial trail values. Results are shown in the form of tables and figures.     

电子束焊机数字化控制系统

以PCI数据采集与运动控制卡为控制核心,设计了一套电子束焊机数字化控制系统。系统通过PCI数据采集卡和运动控制卡实现了采样的模拟量输入、给定的模拟量输出、开关数字量的输入、开关数字量的输出、串口数据的采集以及伺服电机的闭环控制。设计出相应的软件控制系统,使电子束焊机具备了真空操作、焊接操作、运动控制操作、工艺参数管理、焊接实时数据的采集与处理以及日志管理等功能,实现了电子束焊机的数字化控制。在此基础上实现焊接工艺数据库的专家系统及多台电子束焊机的网络化管理。     

CO2焊电流波形在线自适应控制系统

介绍了一种针对短路过渡的CO2焊接过程的电流波形在线自适应控制系统。它通过以短路过渡频率和电弧声能量，作为表征和传感焊接过程稳定性和飞溅的参数，建立了焊接规范和波形控制参数的神经网络优化模型，采用神经网络自适应调整PID参数的双闭环波形控制算法，实现对焊接过程的在线自优化实时控制。试验结果表明，采用该控制系统的焊机，实现了焊接规范及波形控制参数的自动匹配和优化，同时降低了焊接过程飞溅，提高了熔滴过渡的稳定性。     

Thermal error measurement and modelling in machine tools. Part II. Hybrid Bayesian Network—support vector machine model

Prediction accuracy of machine tool thermal error significantly depends on the structure of the error model. Machine tool thermal error varies considerably depending upon the specific operating parameters adopted. Most error models developed thus far generally employ neural networks to map temperature data against thermal error. However, it is very important to account for the specific conditions as well within the model. This paper presents a hybrid Support Vector Machines (SVM)–Bayesian Network (BN) model that seeks to address this issue. The experimental data is first classified using a BN model with a rule-based system. Once the classification has been effected, the error is predicted using a SVM model. The hybrid thermal error model thus predicts the thermal error according to the specific operating conditions. This concept leads to a more generalised prediction model than the conventional method of directly mapping error and temperature irrespective of conditions. Such a model is especially useful in a production environment wherein the machine tools are subject to a variety of operating conditions.     

DSX5—70型三杆五自由度并联虚轴机床的运动仿真

基于开发研制的三杆五自由度并联虚轴机床，概述了实现虚轴机床三维实体模型数控加工的运动仿真方法，利用开发的虚轴模块，解决了在虚轴机床上实现CAD／CAM集成的难题，实现了三杆虚拟轴机床CAD／CAM的一体化、软件的模块化设计，使仿真系统具有良好的扩展性和通用性，操作方便。     

基于RBF-ARX模型的绞吸挖泥船横移过程模型设计

由于绞吸挖泥船横移过程的影响因素众多,具有明显的非线性和时变性特征,难以精确建模来描述其动态特性。因此在对挖泥船产量形成过程详细分析的基础上,针对传统神经网络收敛速度慢、隐含层较多、训练阶段无法自适应调整网络结构的缺点,采用RBF-ARX模型结构进行绞吸挖泥船产量和横移速度的建模。通过结构化非线性参数优化方法SNPOM,离线识别出RBF-ARX模型的线性和非线性参数,并利用仿真对所建模型和真实数据的误差进行对比。仿真结果表明:所建立的绞吸式挖泥船横移过程模型能够比较准确描述系统在全局范围内的动态特性,与挖泥船的实际产量输出拟合较好。     

Intelligent adaptive control and monitoring of band sawing using a neural-fuzzy system

In bandsaw machines, it is desired to feed the bandsaw blade into the workpiece with an appropriate feeding force in order to perform an efficient cutting operation. This can be accomplished by controlling the feed rate and thrust force by accurately detecting the cutting resistance against the bandsaw blade during cutting operation. In this study, a neural-fuzzy-based force model for controlling band sawing process was established. Cutting parameters were continuously updated by a secondary neural network, to compensate the effect of environmental disturbances. Required feed rate and cutting speed were adjusted by developed fuzzy logic controller. Results of cutting experiments using several steel specimens show that the developed neural-fuzzy system performs well in real time in controlling cutting speed and feed rate during band sawing. A material identification system was developed by using the measured cutting forces. Materials were identified at the beginning of the cutting operation and cutting force model was updated by using the detected material type. Consequently, cutting speed and feed rate were adjusted by using the updated model. The new methodology is found to be easily integrable to existing production systems.     

基于PSO-LSSVM算法的表面粗糙度预测模型与应用

为便于选取合适的切削参数,以满足期望的加工表面质量要求,提出一种最小二乘支持向量机(LSSVM)和粒子群优化(PSO)相结合的表面粗糙度预测模型。以预测精度和收敛速度为指标,对比PSO-LSSVM模型与支持向量机、人工神经网络和遗传算法优化BP神经网络模型的优劣。结果表明:PSO-LSSVM模型具有较高的预测精度和较快的收敛速度。基于MATLAB GUI搭建了表面粗糙度预测与参数优化应用系统。该系统具有较好的实用性,可实现简单、快速预测表面粗糙度,帮助决策人员灵活选取切削参数。     

基于BP神经网络的板料折弯件下料尺寸的计算方法

在板料成形工艺中 ,弯曲成形是一种主要的工艺方法。而板料下料尺寸将直接影响到零件的成形精度以及后续工序的成败。针对工厂积累的大量生产和实验数据 ,提出一种基于人工神经网络计算弯曲件板料展开长度的方法。在概述解析法确定板料展开长度的基础上 ,建立了计算展开长度的人工神经网络模型。以板料厚度、成形半径及成形角度作为输入参数 ,中性层位移系数作为输出参数 ,用已积累的大量数据作为训练样本 ,对神经网络进行训练 ,得到这些参数和中性层位移系数的隐性规律 ,实现了弯曲成形板料展开长度的快速计算。实验结果表明 ,利用人工神经网络能够快速准确地计算出板料展开长度。     

Prediction of bead geometry in pulsed GMA welding using back propagation neural network

This paper presents the development of a back propagation neural network model for the prediction of weld bead geometry in pulsed gas metal arc welding process. The model is based on experimental data. The thickness of the plate, pulse frequency, wire feed rate, wire feed rate/travel speed ratio, and peak current have been considered as the input parameters and the bead penetration depth and the convexity index of the bead as output parameters to develop the model. The developed model is then compared with experimental results and it is found that the results obtained from neural network model are accurate in predicting the weld bead geometry.     

应用人工神经网络建立Ti-22Al-25Nb合金高温本构关系模型

本构方程是描述材料变形的基本信息和有限元模拟中不可缺少的数学模型,反映了流动应力与应变、应变速率以及温度之间的相互关系。文章运用Gleeble-1500热模拟机对Ti-22Al-25Nb钛合金试样进行等温压缩变形试验,以试验所得数据(变形温度940℃～1030℃,应变速率0.001s-1～1s-1)为基础,采用BP神经网络的方法建立了该合金的高温本构关系,并与传统回归拟合的方法计算出的结果进行了对比。结果表明,BP神经网络本构关系模型的预测精度明显优于传统公式的计算结果,而且模型还可以很好地描述该合金在高温变形时,各热力学参数之间的复杂非线性关系,为该合金本构关系方程模型的建立,提供了一种便捷有效的方法。     

Apriori算法及神经网络在数控机床中应用研究

数控机床加工精度受到机床零部件、外部环境等因素的影响，从而需要添加适当的补偿参数确保加工精度的稳定性，另外，不同车床不同时刻的补偿参数实时变化。为此，提出一种基于关联规则及神经网络方法的智能误差补偿模型。以实际生产中产生的数据集为基础，通过Apriori算法对数据集进行筛选；对各个特征值与补偿参数进行归一化处理，以提高数据的收敛速度；利用神经网络模型为不同情形下的车床搜寻最佳补偿参数模型，从而构建起最佳的智能误差补偿模型；经过智能误差补偿后,对生产的物件进行图像识别，分析其是否符合精度要求。仿真测试结果表明:针对训练集数据和测试集数据，车床稳定性分别提高了0.695和0.713。实测结果显示:利用上述方法，对30个产品进行雕刻，精度均符合要求。因此，智能误差补偿模型能够提高车床加工稳定性，提升产品合格率。