基于聚焦式模糊综合评判的汽轮发电机组振动故障诊断系统

针对目前汽轮发电机组振动故障的模糊诊断系统精度较低这一问题，本文提出了基于聚焦式模糊综合评判的汽轮发电机组振动故障诊断方法。该方法在文献采用扩展隶属函数和变权重模糊综合评判模型的基础上，提出采用聚焦式计算方法对初始扩展隶属函数进行修正，使得初始扩展隶属曲线中正常与不正常的转折区域被更加细致地刻划，从而大大提高了转折处的分辨率，使故障诊断系统的精度得到明显提高。采用新的诊断策略设计和实现了600MW汽轮发电机组振动故障的模糊诊断系统。用于广东省某电厂的诊断结果表明，这一新的故障诊断方法对于大型汽轮发电机组的振动故障诊断是有效的。     

基于图论诊断法与小波包变换的数控机床进给系统故障诊断研究

在复杂机械系统未知故障诊断中,单一故障诊断方法诊断精度低、推理速度慢,不利于工程实际应用。针对这一问题,提出了基于小波包变换与图论诊断法的数控机床进给系统故障联合诊断方法。首先利用图论诊断法建立数控机床进给系统的故障模型,并根据故障原因定位算法快速确定引起故障发生的可能原因,缩小诊断范围;然后对振动信号进行小波包变换,进一步精确定位故障原因,最终实现数控机床进给系统中未知故障的准确识别与诊断。试验证明,基于图论诊断法与小波包变换的故障联合诊断方法,弥补了图论诊断法故障冗余解多和诊断精度低的缺陷,实现了未知故障的快速筛选与精确定位,可以高效、准确的诊断数控机床进给系统中的未知故障。     

故障诊断中的计算机辅助测试

本文阐述了复杂技术系统故障诊断特点,认为在复杂技术系统中对于故障的测试是一个有穷递归过程,根据这一特点我们采用专家系统方法进行故障诊断。文中提出测试是故障诊断过程中的基本元素,在复杂技术系统故障诊断模型的每一个结点所进行的“收集症状→检测→决策”是整个诊断过程的基础与驱动力。另外还论述了建造故障诊断专家系统过程中的一些技术问题。     

基于多值关联矩阵扩展的诊断策略设计

对于复杂系统,在实际测试中测试不通过的情况可能有多种,如果依然采用基于确定性关联矩阵的测试序列生成算法,将损失大量信息,导致测试精度和效率降低．提出了基于关联矩阵扩展的诊断策略设计方法,将多值关联矩阵扩展为二值关联矩阵,然后利用AO*算法得出诊断策略．实例验证表明,算法提高了故障隔离精度,为不确定系统的诊断策略设计提供一种高效算法．     

基于多故障假设的诊断策略研究与应用

 针对工程中最优单故障序贯测试难以隔离的多故障问题,利用Grunberg等人提出的多故障模糊组的概念,以最优单故障测试策略为基础,构造了一种多故障假设下的序贯测试算法．该算法以多故障状态集的最小碰集作为生成最优单故障策略的故障状态,生成最优单故障策略;以诊断决策树中各叶节点的多故障状态集为评判标准,确定系统的状态,完成多故障假设下的系统故障诊断;与连续使用单故障策略的方法相比,该方法提高了测试效率．以某型机载电子设备为例给出算例分析,验证了该算法隔离多故障的有效性．     

MHSACAE-CNN在噪声下的电机轴承故障诊断

电机的运行情况复杂,实际运行工况下会有大量的噪声,导致其轴承故障诊断精度下降。为了改善这一问题,提出了一种基于多头自注意力机制的一维全卷积自编码网络（One-dimensional Fully Convolutional Autoencoding Network Basedon Multi-head Self-attention,MHSACAE）与卷积神经网络（Convolutional Neural Network,CNN）结合的轴承故障诊断方法。该方法先采用MHSACAE网络进行降噪,再通过CNN进行故障诊断。其中MHSACAE去噪网络采用无监督训练的方式,充分考虑了实际工况和序列数据内在联系,在实现对噪声的滤除效果的同时,最大限度地保留下了原始的故障信息,使得CNN可以实现在噪声情况下对电机轴承故障的高精度诊断。通过与其他轴承故障诊断方法在噪声情况下进行对比,证明提出的方法具有更好的效果。     

基于多分支深度可分离卷积神经网络的滚动轴承故障诊断研究

针对传统滚动轴承故障诊断方法存在抗噪性差、需要人工特征提取、计算量较大、对运行设备要求高的问题,提出一种基于多分支深度可分离卷积神经网络(MBDS-CNN)的滚动轴承故障诊断方法,利用深度可分离卷积和权重剪枝技术对模型尺寸进行压缩,通过多分支结构保证模型的精度,避免梯度消失现象的发生。使用模型尺寸、诊断精度、预测速度作为评价指标对模型进行评估。试验结果证明,基于多分支深度可分离卷积神经网络的滚动轴承故障诊断,可以在噪声环境下有效识别轴承不同部位故障程度,提高了诊断效率,降低了对运行设备性能的要求。     

基于双DSP的导引头数字控制系统设计

分析了导引头伺服系统的结构及其预置、稳定、搜索和跟踪等回路的功能.设计了基于双DSP结构的导引头伺服系统控制器,以实现导引头的通信、扰动隔离以及目标跟踪等功能,介绍了其硬件与软件实现方法.分析了导引头伺服系统的扰动隔离与目标跟踪原理,并以方位轴稳定回路为例,提出了基于dSPACE半实物仿真系统的数字控制器设计方法.扰动隔离仿真与目标跟踪试验结果表明:所设计的导引头数字控制系统具有较高精度,基本满足导引头的扰动隔离与目标跟踪功能.     

基于KDE-FPBT的模拟电路测试性分析与故障诊断方法研究

针对传统测试性模型故障分辨率低的问题,提出一种基于故障对布尔表的模拟电路测试性分析与故障诊断方法。根据故障对布尔表的特点构建新的测试性模型,研究相应的故障检测率和隔离率预计方法。基于核密度估计实现实数域样本数据到故障对布尔表的转换,并采用"一对一"的投票策略完成对实测数据的故障诊断推理。以Sallen-Key带通滤波电路为例开展仿真实验,通过与经典的相关性模型以及整数编码字典进行比较,表明所提方法具有更加突出的故障检测与隔离能力。     

离心泵故障诊断的对称极坐标变换和图像形状特征方法

针对基于时频变换图像特征的机械故障诊断方法耗时长、诊断精度不高等问题，提出基于对称极坐标变换和图像形状特征的离心泵故障诊断新方法。该方法以对称极坐标变换（Symmetrized Dot Pattern，SDP）为基础，通过简单的公式将一维振动信号转换为二维极坐标下的对称花瓣图像，而后进行对称花瓣图像的形状特征提取，以此作为离心泵故障特征，实现离心泵故障诊断。SDP方法计算简单，且故障辨识度高，因此可以在耗时更短的情况下得到更高的故障诊断精度。离心泵故障诊断实例表明，所提方法诊断精度更高，耗时更短。为离心泵的故障诊断提供一种新方法。     

New distortion measurement method for booster amplifiers

The usual method of distortion measurement of an amplifier in a feedback loop involves breaking the loop, which may change the dc and (nonlinear) interstage loading conditions. A new method is presented and proven with theory and experiment, permitting distortion measurement of a final or booster stage within an amplifier. The new method does not require breaking of the feedback loop. Measurements are carried out with linear and nonlinear loudspeaker load     

A BEST-FIRST SEARCH STRATEGY FOR ENERGY RELAXATION IN MER HEAT EXCHANGER NETWORKS

The pinch design method efficiently generates a maximum energy recovery (MER) network which meets the utility targets for a given value of the minimum approach temperature difference. However, this MER design usually contains a significantly greater number of heat exchanger units than the theoretical minimum. Loop breaking and energy relaxation may be used to eliminate these additional units. At each stage of loop breaking, it has been recommended that the unit with the smallest heat load should be removed. However, this study shows that this heuristic can lead to suboptimal designs with respect to energy consumption. An alternative systematic method is presented which reduces the number of units such that the energy penalty is a minimum. A mixed integer non-linear program (MINLP) is formulated with the objective of minimizing the energy consumption for a given number of units. Subsequent loop-network interaction analysis helps in identifying the exchanger units which are good candidates for removal. A lower bound on the consequent energy penalty is also evaluated. These bounds are employed in a ‘best-first’ search strategy to solve the proposed model. On removal of the candidate unit/units, the resulting topology transforms the MINLP to a non-linear/linear program (NLP/LP) which is solved by conventional algorithms. A loop and path identification algorithm (LAPIT), based on graph theory, has been developed as an aid to these computations.     

A robust backward search method based on walk‐through for point location on a 3D surface mesh

Point location is one of the most basic searching problems in computational geometry. It has been largely studied on the aspect of query time in the worst case, and many methods have been proposed, such as counter clockwise wise search and barycentric coordinates search. However, most of them aim at the convex field. For a non‐convex problem, such as a concave field or a convex field with holes, these searching schemes may fail. The numerical experience shows that, even for a convex problem, the searching path may lead to an infinite loop for some special case and may not find the element containing the query point. In this paper, a robust backward search method based on Walk‐through algorithm is proposed to deal with the searching problems in non‐convex fields and to avoid the problems of infinite loop. Another important improvement is to locate the query point on a 3D surface mesh. Several examples demonstrate that the present method is efficient and robust for the workpieces of complex geometry. Copyright © 2007 John Wiley & Sons, Ltd.     

High Precision Robust Automatic Alignment Method for Rotating Shaft

Rotating Shaft is widely used in various high precision instruments on scientific and industrial metrology. This paper proposes a high precision and robust method to automatically align a rotating shaft perpendicular to the horizontal plane. Firstly, an alignment model consisting of a dual-axis inclinometer and four motors is designed. Then the rotating shaft tilt angle between the practical and the ideal shaft is projected to two orthogonal planes, and generates a rotation angle in each plane. The two rotation angles are calculated by differential measurement method based on the inclinometer outputs, and then the relationship of shaft tilt angle against inclinometer outputs is obtained. According to these two rotation angles, the relative heights of the three supporting points of the platform attached to the rotating shaft are calculated. Employing an angle closed loop strategy; the rotating shaft tilt angle is aligned by three linear stepping motors. Experiment results show an alignment precision of 0.003º is achieved using an inclinometer with the resolution of 0.0005º and a linear stepping motor with the regulation precision of 0.89 μm. The proposed method is suitable for a wide variety of precision machines that require the use of rotating shaft.     

摄象机参数高精度求解方法

提出了一种新的高精度的摄象机参数的求解方法，文章利用透视变换理论建立了摄象机数学模型，模型中考虑了摄象机镜头的径向畸变和切向畸变，首先利用牛顿－高斯非线性搜索方法求争摄象机畸变系数和一个中间矩阵，然后再从中间矩阵中分离出其它模型参数；     

Blood glucose concentration control for type 1 diabetic patients: a multiple‐model strategy

In this study, a multiple‐model strategy is evaluated as an alternative closed‐loop method for subcutaneous insulin delivery in type 1 diabetes. Non‐linearities of the glucose–insulin regulatory system are considered by modelling the system around five different operating points. After conducting some identification experiments in the UVA/Padova metabolic simulator (accepted simulator by the US Food and Drug Administration (FDA)), five transfer functions are obtained for these operating points. Paying attention to some physiological facts, the control objectives such as the required settling time and permissible bounds of overshoots and undershoots are determined for any transfer functions. Then, five PID controllers are tuned to achieve these objectives and a bank of controllers is constructed. To cope with difficulties of the presence of delays in subcutaneous blood glucose (BG) measuring and in administration of insulin, a glucose‐dependent setpoint is considered as the desired trajectory for the BG concentration. The performance of the obtained closed‐loop glucose–insulin regulatory system is investigated on the in silico adult cohort of the UVA/Padova metabolic simulator. The obtained results show that the proposed multiple‐model strategy leads to a closed‐loop mechanism with limited hyperglycemia and no severe hypoglycemia.Inspec keywords: blood, patient diagnosis, medical control systems, biochemistry, three‐term control, closed loop systems, diseases, patient treatment, drugs, sugarOther keywords: blood glucose concentration control, type 1 diabetic patients, multiple‐model strategy, alternative closed‐loop method, subcutaneous insulin delivery, type 1 diabetes, transfer functions, control objectives, PID controllers, subcutaneous blood glucose measuring, glucose‐dependent setpoint, closed‐loop glucose–insulin regulatory system, closed‐loop mechanism     

基于混合粒子群的多目标数字印刷智能排活系统的优化研究

通过分析多目标数字印刷智能排活系统中各要素的相互关系,提出了PSO与局部搜索策略的混合算法,并引入了新的学习策略进行分层局部优化,用多目标分散搜索,逐步缩小复杂的搜索空间,改善了PSO算法的早熟收敛缺限,并取得了较高的求解质量。采用了一种随机键的编码方式,利用析取图编码将有序表作为优先决策,来决定发生冲突时各印刷活件的排列顺序。仿真实验验证了混合算法的有效性。     

Robust medical image elastic registration using global optimisation strategy in frequency domain

A new global optimisation strategy in frequency domain (GOFD) is presented and applied in medical image elastic registration. The method is consists of a global optimisation phase for rough searching and a local optimisation phase for fine searching. Rough searching is based on the random sampling technique in the frequency domain. According to the sampling theory, when the sampling frequency is higher than twice the maximum frequency of a function, the function can be completely reconstructed from these finite sampling points. The maximum (or minimum) value of the function at these finite sampling points is approximately in the global extreme. To obtain the exact global extreme, fine searching is performed in the small neighbourhood of the point corresponding to the approximate global maximum value. The new method presented can theoretically ensure that the global optimisation solution is found. The experiments show that our new method is more robust and accurate than other elastic registration algorithms.     

基于实验的钛合金优化动态本构模型与有限元模拟

为确定钛合金材料的Johnson-Cook模型中5个待定本构参数,克服传统单因素分析法的弊端并提高参数识别效率和精度,采用拉丁超立方抽样、Spearman秩相关分析的参数敏感度整体分析方法,并在参数敏感度分析结果和基本遗传算法的基础上,建立了一种基于改进小生境算法、可疑峰值点判断策略和局域精确搜索技术的改进遗传算法,基于实验数据建立了精细的钛合金Johnson-Cook本构模型。采用隐式应力积分法将该模型嵌入到ABAQUS用户材料子程序UMAT中,并通过隐函数求导方法推导出一致切线刚度矩阵。通过ABAQUS有限元软件对钛合金材料的动态响应进行数值模拟分析,计算结果与已有的实验数据吻合良好,UMAT子程序的准确性得到验证,可用于钛合金材料的动态响应预测分析中。