Search

• A model of general range measurement model is proposed. • The model obtains the initial parameters through layer-by-layer unsupervised learning. • The model uses the back-propagation algorithm to fine-tune the entire model. • The model uses SoftMax for classification and recognition. The focus of this article is that it is difficult to effectively measure the universal distance through the use of conventional parameters. This paper uniquely proposes the use of depth-limiting machine learning methods to model the universal distance measurement model, which can effectively perform multi-modal online measurement of the data measurement machine learning model. The feature extraction is unique to this article. The model uses the Softmax software to fine-tune the logarithm to verify the correctness of the propagation algorithm. It can effectively verify the measured distance and obtain multimodal data online. Recognition accuracy and strong robustness are also one of the characteristics of this article. This article has high academic value and innovative features, and is of great significance for the study of machine learning models. [ABSTRACT FROM AUTHOR]

• The characteristics of different posture signals have been studied. • A multi-level hierarchical recognition algorithm for human pose is designed. • The recognition rate of the designed algorithm has been verified. Wireless wearable technology is widely used in health monitoring, rehabilitation training and other fields. In this paper, motion capture sensors are used to collect human motion data, and the collected data format is processed accordingly. We study the characteristics of different posture signals, and select the signal feature sequence that can identify the posture of the signal. According to the signal feature sequence of the pose in the pose database, a multi-level hierarchical recognition algorithm for human pose is designed. Only three-layer recognition algorithm is used to accurately recognize various poses. Through simulation experiments, it is verified that the recognition rate of the designed algorithm meets the monitoring requirements. Experiments were carried out on the physiological parameters of the experimenters. The experimental results show that the entire system has applicability and reliability, and can meet the needs of physicians for real-time monitoring of patient physiological parameters during medical health monitoring. [ABSTRACT FROM AUTHOR]

• A new method for the in-process measuring the size and roundness of workpiece with discontinuous surface. • The proposed method is based on the machine learning technology. • An efficient approach for replacing the invalid data of the groove is also presented. • The roundness evaluation model established meets geometrical product specifications (GPS). • Experimental results validate the feasibility of the proposed on-line measuring method. On-line metrology is widely applied in manufacturing which not only contributes to improving the working efficiency but also helps to ensure the surface quality of machined parts. However, for cylindrical workpieces with discontinuous surface such as spline shaft, the accurate measurement method is still the key challenge. This paper proposes a novel in-process measuring method of geometrical characteristics of discontinuous surface based on extreme value memory strategy and predictive regression algorithm. The invalid measured size data in the keyways can be effectively replaced by the computed data. Based on the in-process measuring method and the established evaluation mathematical models, the roundness of the manufacturing workpieces can be obtained by using the continuous and valid measurement data. Experimental investigations for machining multi-keyway shaft are carried out based on the proposed method. The experimental results show that the proposed measurement method has a high accuracy for discontinuous surfaces. [ABSTRACT FROM AUTHOR]

• A new developed 4-DOF parallel robot implemented as Coordinate Measuring Machine. • An adaptive computer-aided planning method for touch scanning developed. • A linear cost-effective contact probe for scanning free-form surfaces employed. • The measuring algorithm commensurate normal-to-surface contacts. • The developed system eliminates from occurring contact probe errors. Nowadays, digitizing precise parts consisting of free-form surfaces is essential in geometrical inspections and reverse engineering. Despite the ease of use and high speed of non-contact techniques, contact methods have higher precision. However, using touch probes confront a few difficulties. Various compensation approaches that are often post-processing techniques have been introduced for contact probe errors consisting of ball-tip-radius error, stylus tilt, pre-travel, and probe lobbing. Moreover, in precise applications, achieving satisfying results is difficult and time-consuming. The current paper proposes a computer-aided planning method by implementing a new 4-DOF parallel robot and a cost-effective linear contact probe. This parallel structure exploits three translational DOF and one rotational DOF. Furthermore, the robot is equipped with a precise turn-table. A measuring algorithm for free-form surfaces by normal-to-surface contact has been developed. Evaluating the simulation and the practical measuring results shows excellent feasibly of the method on elimination or great reduction of errors. [ABSTRACT FROM AUTHOR]

• Physiological parameter detection in the data processing platform through the built-in algorithm. • The physiological parameter array is obtained and transmitted to ZigBee network by platform. • It can realize the remote and real-time information interaction between doctors and patients. In this paper, the data processing platform for the detection of wearable Sports Physiological Parameters Based on the medical Internet of things can be connected with the network to build a medical Internet of things platform for remote monitoring of patients' lives, which can process physiological data in real time. Finally, the important physiological parameter array of human body is obtained and transmitted to ZigBee network in real time. If the patient needs further diagnosis, the collected data can also be sent to the home gateway through ZigBee and then sent to the telemedicine center through internwt for diagnosis by the doctor and the diagnosis results will be sent back to the patient. The data processing platform can realize the remote and real-time information interaction between doctors and patients. [ABSTRACT FROM AUTHOR]

• The crucial issue in wireless sensor network is to maximize the network lifetime. • To achieve this, partitioning the set of all sensors into disjoint subsets. • Sensors in the each subset can monitor the entire targets. • Activating these subsets one after another will maximize the network lifetime. • Cuckoo search algorithm is used to identify the maximum number of subsets. The primary challenge of a remotely located wireless sensor network in an unmanned application is to maximize the network lifetime, and it is achieved by effective usage of available resources. The efficient way to use the available resources in wireless sensor network is to partition the deployed sets of all sensors into disjoint subsets or sensor covers such that each sensor cover can monitor the entire target at any time. Furthermore, this way of maximizing sensor lifetime by activating sensor cover one after another is found to be effective without exploiting the resource availability. In this paper, the cuckoo search algorithm is proposed to identify the maximum number of sensor covers. It is observed that the proposed algorithm outperforms than existing algorithms by eliminating redundant sensors and supplementing with a sensor when it is required. The simulation results of various problem instances show the superiority of proposed algorithm in terms of network lifetime, and it is validated by comparison with other few existing algorithms. [ABSTRACT FROM AUTHOR]

• A distributed algorithm for density estimation and data clustering. • Minimizing the effects of uncertainties in sensor measurements. • The proposed algorithm analytically converges to the maximum likelihood estimates. In this paper, a novel distributed averaging-based evidential Expectation-Maximization algorithm is presented for data clustering and probability density function estimation in unreliable sensor networks in the presence of uncertain observations. All observations of the sensor network are statistically modelled by a Gaussian mixture model. A generalized likelihood function is maximized in a distributed manner for density estimation. For this purpose, the parameters of the mixture model such as weighting coefficients, mean vectors, and covariance matrices are updated by using a distributed averaging method. The proposed algorithm can be applied to any network with a mesh topology. The convergence of algorithm is also investigated and guaranteed by using a mathematical analysis. In the simulation result section, the proposed algorithm has been applied to a simulated uncertain sensor network with a mesh topology. The simulations confirm the favourable performance of the distributed averaging-based evidential EM algorithm. [ABSTRACT FROM AUTHOR]

• Applying iterative closest point algorithm to stitch point clouds. • Employing semi-global block matching algorithm for stereo correspondence problems. • Incorporating structural information of objects to improve point cloud registration accuracy. • Interpolating surfaces onto point clouds of construction material for volume estimation. Bucket fill factor is of paramount importance in measuring the productivity of earth-moving machines which is the amount of material loaded in the bucket within one scooping. In this paper, a stereo vision-based method is proposed and a novel perception system is developed. The system tested on a wheel loader is able to provide an accurate estimation of the fill factor. Disparity is computed for a pair of stereo images to reconstruct high fidelity 3-D point clouds. Preliminary processing is then used to extract regions of interest. In addition to applying existing state-of-the-art techniques including point cloud registration, stitching, and surface interpolation, our system also incorporates the structural information of wheel loaders, achieving excellent performance of 94.72 % accuracy even under the unavailability of visual information due to occlusion, and under different environment conditions. This research paves the way for the arrival of fully autonomous earth-moving machines in the future. [ABSTRACT FROM AUTHOR]

• Novel pre-processing procedure of harmonic removal for operational modal analysis. • Synchronous use of the Discrete Random Separation algorithm for multiple time histories. • Application of the harmonic removal procedure to data recorded in industry. • Detection of structural causes of excessively high vibration levels of rotary machines. One of the important issues concerning the dynamic analysis of rotating machines is the separation of deterministic and random components. In industrial conditions, the determination of dynamic parameters is performed using Operational Modal Analysis (OMA). In order to correctly identify the natural frequencies, it is necessary to separate the components associated with them from rotational speed harmonics. The presented article shows the application of in-operational measurements with a new data pre-processing procedure for OMA carried out on a rotating machine operating in a large industrial plant. An innovative procedure for removing harmonics from Cross-Power Spectral Density (CSD) was used in the data preparation process. The novel aspect in the paper is the application of the Discrete Random Separation (DRS) algorithm to remove harmonics from data for OMA. A procedure for the synchronous use of DRS for multiple time histories has been developed so that modal analysis is possible. [ABSTRACT FROM AUTHOR]

• A novel SAR image despeckling is proposed using Local and Non-Local features. • Our method reconstructs the missing information with edge linking algorithm. • Eight directional Kirsch filter is used for better edges extraction. • Patch shape selection to remove the noise causes a greater efficiency. • Our proposed algorithm outperforms many existing approaches on despeckling. Removal of speckle noise from Real SAR images is a significant task for the better segmentation, target detection, target recognition or other processing. In this paper, a novel complex strategy is presented based on carefully analyzing the effect of the speckle noise on the histogram. The histogram of a patch is expanded in the image corrupted with speckle noise. To overcome this, we introduced some rules to remove both noise and preserve the crucial features (edges). Owing to the use of patch shape selection automatically, the patch's histogram modification is more effective. Our method, also, is being able to reconstruct some missing edges by using the kirsch filter and edge linking approach. Our satisfactory results indicate that the proposed approach has a competitive despeckling performance in terms of both visual quality assessment and preserving vital details. [ABSTRACT FROM AUTHOR]

• GOA-MSVM method can identify both fault type and fault severity of rolling bearings. • GOA-MSVM is proved to be able to identify weak faults of planetary gearbox. • VMD SE is used as a part of feature vector and compared with VMD AE and EMD SE. • GOA-MSVM is more suitable to identify rotating machinery faults than other 5 SVM-based methods. Identifying fault of rotating machinery under different load conditions with high accuracy is a remaining challenge for vibration signal based fault diagnosis. Aiming at this challenge, this paper proposes a comprehensive strategy of combining mixed kernel-support vector machine (MSVM) with grasshopper optimisation algorithm (GOA) to identify typical faults of rotating machinery subject to different load levels. The basic idea of the proposed strategy can be summarized as the following three steps. Firstly, a feature vector that uses multi-domain indexes containing the sample entropy (SE) of variational mode decomposition (VMD) is constructed to characterize the fault information. Secondly, a MSVM model containing six design variables is established and then optimized by GOA. Finally, the optimized MSVM model is adopted to train the fault feature vectors to fulfill fault pattern recognition. In order to verify the identification accuracy of the proposed strategy, two sets of fault signal generated from a rolling bearing test rig and a laboratory planetary gearbox operating under different load conditions are analyzed. The diagnostic results manifest that the proposed strategy can fully identify different level faults of rolling bearing and weak faults of planetary gearbox as well. More than 99% identification accuracy of the proposed strategy is further highlighted by comparisons with other five SVM-based methods. The present method provides a promising solution for high fault identification accuracy in rotating machinery working under different loads. [ABSTRACT FROM AUTHOR]

• 3D similarity datum transformation. • Multiple outlier detection in the generalized Errors-In-Variables (EIV) model. • Reduction of the influence of the multiple outliers. • Accurate and reliable transformation parameters. Estimated transformation parameters based on the three-dimensional (3D) similarity datum transformation are affected or even severely distorted when the observed coordinates are contaminated by gross errors or outliers. In our paper, the problem of 3D similarity datum transformation is described as a generalized Errors-In-Variables (EIV) model based on a least squares solution. Then, an advanced multiple outlier detection algorithm that uses the L 1 penalty function on the mean-shift model is proposed in the generalized EIV model. A general thresholding rule and a least trimmed squares estimator are invoked in the proposed algorithm. The results of the real and simulated experiments indicate that the proposed algorithm can effectively reduce the influence of multiple outliers and yield reliable transformation parameters compared with the iteratively reweighted total least squares, data snooping and robust outlier removal approaches. [ABSTRACT FROM AUTHOR]

• An improved fast calculation method for TSST is proposed. • The proposed method still has a high concentration of the TF representation. • The proposed method is verified with both simulations and experiments. Time-reassigned synchrosqueezing transform (TSST) provides a high concentration of the time–frequency (TF) representation of impulsive-like signals. However, when TSST is used to process non-stationary real-world signals, it often faces a large number of data points which will cause huge computational overhead. To address this problem, an improved algorithm for TSST is proposed in this paper. The improved algorithm mainly includes the following ideas: increasing the window function shifting point number H and decreasing the point number N f of Fourier transform combining fast matrix operation and time–frequency band selection and reassignment. Moreover, in order to ensure phase consistency in the time-reassigned operation, phase correction is carried out for the TF matrix. The proposed method is validated with numerical simulation and then applied in mechanical fault diagnosis. The results show that the method can greatly improve computational efficiency while ensuring a high concentration of the TF representation. [ABSTRACT FROM AUTHOR]

• Our goal is to estimate the parameters of solar photovoltaic models using the Salp Swarm Algorithm. • The uncertainties of the parameters are considered by using the meta-heuristic method. • The results obtained are compared with well-established algorithms in literature. • The estimation errors is of the order of 10 e−8 On the other hand the other works publish are 10e−4. This paper deals with the problem of estimating the parameters for single and double diode solar photovoltaic models. To solve this problem the Salp Swarm Algorithm (SSA) will be used. The proposed approach takes into account the uncertainties of the measures. This approach comprises three steps such as retrieving the parameters in a conventional manner and then determining the uncertainties of each parameter. finally, the determination of the instantaneous parameters taking into account the results of the first two steps. For the validation of the theoretical developments proposed, the algorithm is applied to 3 different PV parameter estimation problems. The results obtained are compared with well-established algorithms to confirm its effectiveness. These comparisons have shown that the proposed method presents largely effective performances compared to existing methods in the literature. [ABSTRACT FROM AUTHOR]

• Precise terrain occlusion measurement and prediction. • Precise intrinsic and external parameters of the stereo cameras. • Reduction of the rotation angle errors of the mast mechanism. • Reduction of the accumulated errors of the rover pose. • Visual terrain occlusion in Chang'e-4 mission. This paper proposes a precise terrain occlusion prediction algorithm based on stereo image sequences from the China's Chang'e-4 Yutu-2 rover' visual system, which includes an on orbit stereo camera calibration method and a camera pose estimation method. The main advantage of the proposed algorithm is that the above two methods consider the problem that the coefficient matrix and observation vector should also be considered with regard to the error during optimization. In this case, the total least squares solution is applied instead of the ordinary least squares solution. Experimental results show that the proposed algorithm has higher precision (with average differences in the altitude angles and azimuthal angles of −0.37 degrees and −0.29 degrees) and application stability than the ordinary least squares method. The proposed algorithm has been successfully used in Chang'e-4 mission operations. [ABSTRACT FROM AUTHOR]

• Unbalance parameters of multi discs rotor can be detected by artificial intelligent algorithms. • Used K-Nearest Neighbor Algorithm shows average accuracy as 86% for detecting unbalance parameters. • Decision Tree Algorithm represents lower average accuracy as 70% for detecting unbalance parameters. • So, in this case, K-Nearest Neighbor Algorithm reveals better accuracy in comparison to Decision Tree Algorithm. Multi discs rotors are widely used in the industry. Shaft unbalance in multi-discs' rotors is the main failure origin that leads to global failures in rotary systems. Unbalance parameters that must be detected in the shaft are focused on this study. Unbalance parameters are eccentric mass value, eccentric radius, and disc number which are presenting an unbalance location. The main aim of the current paper is to identify unbalance parameters of a rotating shaft having multi-discs by artificial intelligent methods namely KNN and Decision Tree Algorithm. For both algorithms, data derived from a fabricated test rig consists of a shaft in which four discs are mounted on that. The results show that the KNN presents more accuracy in estimating of unbalance parameters compared to the Decision Tree in terms of unbalance locating. [ABSTRACT FROM AUTHOR]

• A subtle feature extraction algorithm for large-scale irregular surface is proposed. • The L 1-median is calculated to optimize the mean median value. • New feature description of the point cloud is obtained. • The subtle feature is extracted by multi-threshold. • A feature extraction test on large irregular workpiece is performed. In reverse engineering, feature extraction of point cloud data is a key process for the precision machining of the large-scale complex workpieces. Because of the large numbers of the point cloud data and the difficult recognition of the complex feature, incomplete feature recognition will have a serious impact on the accuracy of the machining. Thus, this paper proposes a subtle feature extraction algorithm, which can be used for the laser scanning measurement of the large-scale irregular surface. First, the L 1-median point is calculated as the center point of the neighborhood. Second, the k + 1 neighbors are introduced to compute the feature description of the point cloud. Then, the feature is extracted by multi-threshold based on Poisson region growth algorithm. Last, the proposed algorithm is applied to feature extraction experiment of point cloud data for the large spherical crown workpiece. Compared with the traditional algorithms, the proposed algorithm not only can identify the subtle feature information quickly, but also can locate the weld position more accurately. [ABSTRACT FROM AUTHOR]

• A framework of Remaining Useful Life estimation for railway turnout is proposed. • An effective feature selection and fusion strategy is conducted. • Genetic Programming is applied for RUL prediction. • The proposed methodology achieves more precise results than state-of-the-art algorithms. The remaining useful life (RUL) prediction of railway turnout systems (RTS) is very important to avoid unplanned shutdowns and reduce labor costs for the normal operation of railways. One key challenge on RUL prediction is how to construct an appropriate health indicator (HI) that can be utilized to infer conditions of RTS. Existing methods usually adopt some inherit merits (e.g., monotonicity, trendability, and robustness), and their prediction results lack real-world physical meaning due to their "black-box-like" property. In this paper, we present a novel feature fusion method for RUL prediction, which is able to capture the relationship between RUL and HI. A variant correlation-based feature selection method is utilized to extract features, which has the potential to depict the degradation process optimally, and then the selected features are fused by Auto-Associative Kernel Regression (AAKR) for prediction. To reduce the noise interference, the extracted features and the combined HI are all smoothed by using the locally weighted regression. Finally, a genetic programming (GP) algorithm is employed to predict the RUL of RTS. The proposed method is extensively tested on two turnout machine degradation datasets, and the results show that the proposed approach is effective for RUL prediction of RTS. [ABSTRACT FROM AUTHOR]

• An accurate method for detecting complex component surface defects is proposed. • The defect detection accuracy is high to 99.3% for the complex components. • Normal feature misdetection can be guaranteed by machine learning method. • Spreading algorithm is proposed to classify and recognize the number of defects. • Sizes of defects can be detected by the covariance matrix 3D measurement method. Detecting the surface defects of complex components exhibiting different visual properties in positions, shapes, number and sizes is a challenging problem. In this paper a novel computational framework is developed to accurately detect the component surface defects through three steps. In the framework, the positions and shapes of the components surface defects are extracted based on the support vector machine and the point cloud model. Then a novel unsupervised classification method termed as spreading algorithm is proposed to classify the defects for the recognition of the number of defects, and finally the sizes of defects are calculated using the covariance matrix 3D measurement method. Experimental cases on two typical complex components, the blade with depression and protrusion defects and the transmission case with defect and normal features, are investigated. The results are particularly compared with the ones with the state-of-the-art methods, showing the practicality and effectiveness of the proposed technique. [ABSTRACT FROM AUTHOR]

• The network is divided into several clusters. • Both the inter-cluster and intra-cluster correlation is utilized. • The measurement matrix is formulated based on the gathered data. • The proposed scheme maintains reconstruction accuracy under unreliable wireless links. • Network life is prolonged effectively. Packet loss is one of the most critical factors affecting the accuracy of compressed sensing (CS)-based data gathering algorithms. In this paper, a data gathering algorithm is proposed to decrease energy consumption and resist packet loss. Each cluster head formulates a sparsest random measurement matrix (SRMM) via the received data to avoid the measurement of the lost node and decrease the number of measurements. To employ spatial correlation between clusters, the sink constructs a block diagonal matrix (BDM) as a measurement matrix via SRMMs and reconstructs the entire network data. Additionally, the optimal number of clusters is discussed under this framework to reach the minimum power consumption. The SR-BDM is evaluated on the emulated data and the real sensor data from GreenOrbs, respectively. The simulation results indicate the proposed algorithm reaches high precision, both with reliable links and with a 60% packet loss rate link, without causing increased energy consumption. [ABSTRACT FROM AUTHOR]

• A strategy for intelligent prognostics of NASA batteries was proposed. • Locations of capacity regeneration phenomena were automatically and probabilistically detected. • NASA battery capacity degradation data with capacity regeneration phenomena were modeled. • Detection of capacity regeneration phenomena improved battery state of health prediction accuracies. State of health prediction of rechargeable batteries is an important topic in battery health management systems to infer remaining charge-discharge cycles (RCDC) and ensure high reliability of rechargeable batteries. To investigate various prognostic algorithms, battery life-cycle fade experiments were conducted in the NASA Ames Prognostic Center of Excellence. Even though some prognostic algorithms were proposed in the past years to predict RCDC, most of them omitted the influence of battery capacity regeneration phenomena (CRP) on RCDC prediction. Battery CRP cause some sudden capacity increments at unexpected charge-discharge cycles, which may further result in inaccurate RCDC prediction if such CRP are not considered. In this paper, piecewise model based intelligent prognostics for RCDC prediction of rechargeable batteries with CRP is proposed. Results show that the proposed prognostic model can precisely estimate NASA battery capacity fade data, automatically detect CRP and accurately predict RCDC. [ABSTRACT FROM AUTHOR]