Autonomic point cloud-based surface reconstruction using SVR

A surface reconstruction framework based on support vector regression (SVR) to generate a three-dimensional (3D) model is proposed in this paper. It can reduce the noise in sampled data as well as repair the holes by handling the missing data during the acquisition phase. SVR is quite efficient for surface reconstruction using parameter tuning and selective data sampling. Automatic parameter tuning of SVR is proposed using two techniques: particle swarm optimization (PSO) and genetic algorithm (GA). Independent component analysis (ICA) is a feature-preserved non-uniform simplification method which is applied to simplify point set by optimal attribute selection. First, under-sample the data, remove the redundancy, reduce the features using ICA and construct the surface using SVR. Both theoretical analysis and experimental results show that the performance of the proposed method yields an average SVR error ≈ 3% on the publicly available datasets. For majority of standard datasets, PSO–SVR is found superior to GA–SVR in convergence speed. Details of the surface are also preserved well which makes it suitable for 3D surface reconstruction.     

基于粒子群支持向量回归的原位自生工艺参数优化

综合应用激光熔覆和原位反应增强金属基复合材料,是当前金属基复合材料研究领域的一个热点,本文采用该工艺制备铁基表面复合材料,重点考虑该工艺参数的确定问题.根据在不同工艺参数下合成的铁基表面的WC体积分数实测数据集,提出建立不同工艺参数下WC体积分数的支持向量回归预测模型,并与基于人工神经网络模型(ANN)的预测结果进行比较.结果显示:对于相同的训练样本和检验样本,SVR预测模型比ANN预测模型具有更强的泛化能力.最后根据建立的预测模型,应用粒子群算法寻优得到最优工艺参数,该工艺参数在实际实验过程中的应用,验证了该方法的有效性.     

基于相空间重构和遗传优化SVR的机械设备状态趋势预测

针对机械设备振动信号序列的非线性、非平稳性特点,提出了一种基于相空间重构与遗传优化支持向量回归机的设备状态趋势预测方法。首先,采用相空间重构技术将一维振动信号时间序列转化成矩阵形式,自适应地选取特征,以相点作为输入特征训练SVR预测器;然后应用自适应遗传算法对惩罚因子、不敏感系数以及高斯核宽度进行同步优化,自动获取最佳的建模参数;最后构建SVR预测模型,并将其应用于某机组振动信号预测。实验结果表明,无论是单步还是24步预测,本文所提遗传优化SVR模型的预测精度都要比标准SVR模型的预测精度高,说明该方法对机械设备的运行状态趋势具有较好的预测能力。     

Results from trials of a planar, incoherent, synthetic aperture sonar for underwater imaging

Synthetic aperture sonar (SAS) techniques can yield high-resolution images with a small physical array. Their application in the underwater environment is usually confined to the deployment and synthesis of linear apertures. Incoherent SAS processing is a suboptimal approach compared with coherent SAS processing as the absence of phase information results in an inferior along-track resolution and a higher sidelobe level. The absence of phase information implies that incoherent processing is not constrained by phase-aliasing. Incoherent SAS can accommodate non-uniform inter-ping spacing and is tolerant to trajectory estimation errors of the order of the range-compressed pulse length. This presents new opportunities for a robust surveying system. Apertures of nonlinear shapes can be synthesised depending on the nature and requirement of the specific application and environment. Surveys can be conducted with minimal hardware deployment, such as by a diver. On the basis of this concept, tank trial results of a 3-dimensional incoherent SAS technique utilising the synthesis of 2D apertures are presented. Broadband pulses are employed to achieve optimal survey resolution. The feasibility of this technique is demonstrated for both monostatic and single-transmit, multiple-receive configurations. Practical results are shown for arbitrary-surface apertures sampled at non-uniformly separated positions.     

Non-uniform Sampling in Optical Processing

The use of non-uniform sampling in optical processing is considered. The filtering operations needed to recover the spectrum of a non-uniformly sampled band-limited function are examined. Serious limitations are shown to exist for markedly non-uniform sampling.     

Forecasting systems reliability based on support vector regression with genetic algorithms

This study applies a novel neural-network technique, support vector regression (SVR), to forecast reliability in engine systems. The aim of this study is to examine the feasibility of SVR in systems reliability prediction by comparing it with the existing neural-network approaches and the autoregressive integrated moving average (ARIMA) model. To build an effective SVR model, SVR's parameters must be set carefully. This study proposes a novel approach, known as GA-SVR, which searches for SVR's optimal parameters using real-value genetic algorithms, and then adopts the optimal parameters to construct the SVR models. A real reliability data for 40 suits of turbochargers were employed as the data set. The experimental results demonstrate that SVR outperforms the existing neural-network approaches and the traditional ARIMA models based on the normalized root mean square error and mean absolute percentage error.     

A comparative study of just-in-time-learning based methods for online soft sensor modeling

Most traditional soft sensors are built offline and only to be used online. In modern industrial processes, the operation condition is changed frequently. For these time-varying processes, online soft sensor modeling is required, since the prediction result is highly related to other components of the process control system. In the present paper, a comparative study of three different just-in-time-learning (JITL) methods for online soft sensor modeling is carried out, which are based on partial least squares (PLS), support vector regression (SVR) and least squares support vector regression (LSSVR). Different from traditional soft sensors which model the process through a global and offline manner, the JITL-based method exhibits an online local model structure, depending on which the change of the process can be well tracked. Besides, the process nonlinearity can also be addressed under this modeling framework. As a further contribution of this paper, a real-time performance improvement strategy is proposed to enhance the online modeling efficiency of the JITL-based soft sensor. For performance evaluation, two industrial case studies are provided.     

A two-stage procedure for forecasting freight inspections at Border Inspection Posts using SOMs and support vector regression

The number of goods which passes through a border inspection post (BIP) may cause important congestion problems and delays in the port system, having an effect in the level of service of the port. Therefore, a prediction of the daily number of goods subject to inspection in BIPs seems to be a potential solution. This study proposes a two-stage procedure to better predict freight inspections. In the first stage, a Kohonen self-organising map (SOM) is employed to decompose the whole data into smaller regions which display similar statistical characteristics. In the second stage, support vector regression (SVR) is used to forecast the different homogeneous regions individually. The results obtained are compared with the single SVR technique. The experiment shows that SOM–SVR models outperform the single SVR models in the inspection forecasting. The application of the proposed technique may become a supporting tool for the prediction of the number of goods subject to inspection in BIPs of other international seaports or airports, and provides relevant information for decision-making and resource planning.     

Saturation Magnetic Induction Prediction for Amorphous Magnetic Alloys by Using Support Vector Regression

This paper illustrates an application of support vector regression (SVR) approach in forecasting the saturation magnetic induction (B _s ) of amorphous magnetic alloys. SVR was trained and tested with an experimental data set comprised of five input variables, comprising the average number of valence electrons of amorphous magnetic alloys, mixed entropy, ratio of radii, difference of electron density, and difference of work function. The prediction performance of SVR was compared with that of artificial neural networks’ (ANN) model. The results demonstrate that the prediction ability of SVR is superior to that of ANN. This investigation indicates that SVR-based modeling is a practically useful tool in prediction of the saturation magnetic induction of amorphous alloys. This study provides a novel methodology to foresee the saturation magnetic induction in sintering/development of novel amorphous magnetic alloys possessing high saturation magnetic induction.     

汽轮发电机定子绕组参数化建模方法研究及系统开发

针对汽轮发电机定子端部绕组结构的复杂性提出了用圆锥面展开图实现空间三维形状的线棒建模方法.通过分析绕组的结构及设计成型和组装原理,建立了线棒参数化建模及装配的数学模型,并给出了三维建模的关键步骤,在此基础上结合二次开发技术、计算机编程技术以及数据库技术设计开发了基于UG/NX平台的汽轮发电机定子端部绕组参数化自动建模系统.以某百万千瓦级汽轮发电机定子端部绕组为例,验证了该建模方法及参数化自动建模系统的有效性和可行性.     

The application of SVR model in the improvement of QbD: a case study of the extraction of podophyllotoxin

In order to make a further optimization of process design via increasing the stability of design space, we brought in the model of Support Vector Regression (SVR). In this work, the extraction of podophyllotoxin was researched as a case study based on Quality by Design (QbD). We compared the fitting effect of SVR and the most used quadratic polynomial model (QPM) in QbD, and an analysis was made between the two design spaces obtained by SVR and QPM. As a result, the SVR stayed ahead of QPM in prediction accuracy, the stability of model and the generalization ability. The introduction of SVR into QbD made the extraction process of podophyllotoxin well designed and easier to control. The better fitting effect of SVR improved the application effect of QbD and the universal applicability of SVR, especially for non-linear, complicated and weak-regularity problems, widened the application field of QbD.     

Solving the Inverse Heat Conduction Problem in Using Long Square Pulse Thermography to Estimate Coating Thickness by Using SVR Models Based on Restored Pseudo Heat Flux (RPHF) In-Plane Profile

This research develops a method to estimate opaque coating thickness based on time-and-space-resolved thermography. Thermography is a viable technique for measuring coating thickness. However, time-resolved thermography becomes unreliable when uncontrolled constant thermal stimulation amplitude appears. Therefore, a time- and space-resolved thermography technique for coating thickness measurement called restored pseudo heat flux (RPHF) has been developed by using Fourier–Hankel transform. A non-dimensional analysis was conducted and the results show RPHF curves with different coating thicknesses converging as the thermal diffusivity ratio or thermal conductivity ratio goes to one. For large thermal conductivity ratio values, the RPHF curves have two inflection points along the non-dimensional radius. Fifty-nine samples were tested using the proposed method. Support vector regression (SVR) models were constructed with the in-plane distribution of RPHF and the temporal distribution of the measured surface temperature as inputs. To avoid overfitting, cross validation was applied to all the models. Later, another twenty-eight samples were tested to validate the SVR models. The results suggest that a support vector regression model with in-plane profiles of RPHF handles uncontrolled heat flux variation better and yields a better performance in coating thickness measurement than the temporal profile does. With in-plane RPHF profile as input, the SVR model can evaluate coating thickness with a relative root mean square error at 25.3% even when heat amplitude varies 50%.     

Particle modeling of random crack patterns in epoxy plates

This paper employs particle modeling for simulation of dynamic fragmentation in (elastic–brittle) epoxy plates (8.25 cm×33.02 cm), containing non-uniformly distributed circular holes [Al-Ostaz A, Jasiuk I. Crack initiation and propagation in materials with randomly distributed holes. Eng Fract Mech 1997;58:395–420]. Since the experiments on nominally identical specimens resulted in a range of different crack patterns, the model focuses on matching the most dominant experimentally observed cracks. Indeed, this is achieved with lattices having several different mesh resolutions. Next, by introducing very weak, microscale perturbations in the material properties, it is found that the stiffness has a stronger effect on the deviation from the dominant crack pattern than does the tensile strength.     

Fatigue residual life estimation of jib structure based on improved v‐SVR algorithm obtaining equivalent load spectrum

Fatigue tests of truck crane jib structure have great difficulty, long testing cycle, and high cost. Therefore, with the portfolio strategy for “collection, prediction, measurement and simulation” being introduced, an improved support vector regression (v‐SVR) algorithm is proposed to predict the equivalent load spectrum based on the small measured load spectrum and the fatigue residual life evaluation model is built. First, the v‐SVR algorithm corrected in kernel function, decision function, and parameter optimization is utilized to acquire the equivalent load spectrum. Second, the distribution of critical points is determined by the case‐based reasoning technique. The simulation model of first main stress‐time histories for critical points is established, and the two‐dimensional stress spectrum is obtained by the rain‐flow counting method. Finally, the fatigue residual life of jib structure is estimated by the Forman formula. Taking ZLJ5551JQZ110V truck crane jib structure as an example, the effectiveness of the above method is validated.     

反射型超声衍射CT迭代重建算法

本文以傅立叶衍射定理为基础,将非均匀傅立叶变换和迭代法相结合,用正则化方法处理迭代的收敛问题,建立了反射型超声衍射成像算法。数据直接在频域中的非均匀频率点上比较,避免了频域内插引入的误差。本算法也减少了采样数据量,降低了运算的复杂度。实验结果表明,在迭代次数不多情况下,重建图像可以达到较好的视觉效果。     

Post-Processing of Phased-Array Ultrasonic Inspection Data with Parallel Computing for Nondestructive Evaluation

Phased-array ultrasonic nondestructive evaluation is an effective tool of safety assurance for key structural components. The paper presents a general post-processing methodology for phased-array ultrasonic inspection data. The methodology is developed to integrate three components: mapping of sampling points to structure model, re-sampling from non-uniformly distributed sampling points in phased-array to a uniform volume, and data fusion strategies for multiple channels. An adaptive method called spatially adaptive Gaussian splatting is proposed for data re-sampling and fusion considering the reconstruction resolution and local characteristics of ultrasonic sound paths. This adaptivity provides a viable approach to minimize the effects of under-sampling, over-sampling, and holes which are introduced by the non-uniformly distributed sampling points. The processing of large scale data through segmentation and parallelization techniques is discussed in detail. The effectiveness and performance of the proposed methodology are investigated using actual phased-array ultrasonic testing data.     

Multivariate performance reliability prediction in real-time

This paper presents a technique for predicting system performance reliability in real-time considering multiple failure modes. The technique includes on-line multivariate monitoring and forecasting of selected performance measures and conditional performance reliability estimates. The performance measures across time are treated as a multivariate time series. A state–space approach is used to model the multivariate time series. Recursive forecasting is performed by adopting Kalman filtering. The predicted mean vectors and covariance matrix of performance measures are used for the assessment of system survival/reliability with respect to the conditional performance reliability. The technique and modeling protocol discussed in this paper provide a means to forecast and evaluate the performance of an individual system in a dynamic environment in real-time. The paper also presents an example to demonstrate the technique.     

水电工程施工总布置三维图形建模

水电工程施工总布置是一个复杂的系统,文中对水电工程施工总布置三维可视化建模技术进行了深入研究。首先将施工总布置系统分解为各个子系统,然后基于GIS平台,对各个子系统采用不同的建模技术分别建立模型,最后通过可视化技术将其模型在屏幕上显示输出。在大型水电工程中的实际应用,证明了此方法的实用性和高效性。