Higher Training Efficiency Research Articles

A Novel Bearing Fault Diagnosis Method Using Spark-Based Parallel ACO-K-Means Clustering Algorithm

K-Means clustering algorithm is a typical unsupervised learning method, and it has been widely used in the field of fault diagnosis. However, the traditional serial K-Means clustering algorithm is difficult to efficiently and accurately perform clustering analysis on the massive running-state monitoring data of rolling bearing. Therefore, a novel fault diagnosis method of rolling bearing using Spark-based parallel ant colony optimization (ACO)-K-Means clustering algorithm is proposed. Firstly, a Spark-based three-layer wavelet packet decomposition approach is developed to efficiently preprocess the running-state monitoring data to obtain eigenvectors, which are stored in Hadoop Distributed File System (HDFS) and served as the input of ACO-K-Means clustering algorithm. Secondly, ACO-K-Means clustering algorithm suitable for rolling bearing fault diagnosis is proposed to improve the diagnosis accuracy. ACO algorithm is adopted to obtain the global optimal initial clustering centers of K-Means from all eigenvectors, and the K-Means clustering algorithm based on weighted Euclidean distance is used to perform clustering analysis on all eigenvectors to obtain a rolling bearing fault diagnosis model. Thirdly, the efficient parallelization of ACO-K-Means clustering algorithm is implemented on a Spark platform, which can make full use of the computing resources of a cluster to efficiently process large-scale rolling bearing datasets in parallel. Extensive experiments are conducted to verify the effectiveness of the proposed fault diagnosis method. Experimental results show that the proposed method can not only achieve good fault diagnosis accuracy but also provide high model training efficiency and fault diagnosis efficiency in a big data environment.

Comparing Reinforcement Learning Methods for Real-Time Optimization of a Chemical Process

One popular method for optimizing systems, referred to as ANN-PSO, uses an artificial neural network (ANN) to approximate the system and an optimization method like particle swarm optimization (PSO) to select inputs. However, with reinforcement learning developments, it is important to compare ANN-PSO to newer algorithms, like Proximal Policy Optimization (PPO). To investigate ANN-PSO’s and PPO’s performance and applicability, we compare their methodologies, apply them on steady-state economic optimization of a chemical process, and compare their results to a conventional first principles modeling with nonlinear programming (FP-NLP). Our results show that ANN-PSO and PPO achieve profits nearly as high as FP-NLP, but PPO achieves slightly higher profits compared to ANN-PSO. We also find PPO has the fastest computational times, 10 and 10,000 times faster than FP-NLP and ANN-PSO, respectively. However, PPO requires more training data than ANN-PSO to converge to an optimal policy. This case study suggests PPO has better performance as it achieves higher profits and faster online computational times. ANN-PSO shows better applicability with its capability to train on historical operational data and higher training efficiency.

Comparison of different machine learning algorithms for prediction of wind evolution

Wind evolution, i.e. the evolution of turbulence structures over time, has become an increasingly interesting topic in recent years. Part of this increasing interest is resulted from the development of lidar-assisted wind turbine control, which requires accurate prediction of wind evolution to avoid unnecessary or even harmful control actions. Moreover, study of wind evolution can deepen our understanding of turbulence characteristics and, for example, could make 4D stochastic wind field simulations possible by integrating wind evolution into 3D simulations. This study compares different machine learning algorithms for the prediction of wind evolution, including three variants of Gaussian process regression, regression tree, support vector machine regression, and shallow neural network. The comparison is done using lidar data measured by a nacelle-mounted lidar, and it focuses on the prediction accuracy and the computational time of models, to provide some insights of the trade-off between these factors. In terms of computational time, both the training time of the final model and the time for hyperparameter optimization are taken into account. Moreover, up-scaling of the computational time with data size is also investigated. It is found that Gaussian process regression provides high prediction accuracy but needs long computational time, increasing significantly with data size. For cases where long computational time might be a critical problem, shallow neural network could be a good alternative thanks to its high training efficiency, although its prediction accuracy is lower than Gaussian process regression. The other methods have no advantages in wind evolution prediction compared to these two methods.

Distributed Learning for Automatic Modulation Classification in Edge Devices

Automatic modulation classification (AMC) is a typical technology for identifying different modulation types, which has been widely applied into various scenarios. Recently, deep learning (DL), one of the most advanced classification algorithms, has been applied into AMC. However, these previously proposed AMC methods are centralized in nature, i.e., all training data must be collected together to train the same neural network. In addition, they are generally based on powerful computing devices and may not be suitable for edge devices. Thus, a distributed learning-based AMC (DistAMC) method is proposed, which relies on the cooperation of multiple edge devices and model averaging (MA) algorithm. When compared with the centralized AMC (CentAMC), there are two advantages of the DistAMC: the higher training efficiency and the lower computing overhead, which are very consistent with the characteristics of edge devices. Simulation results show that there are slight performance gap between the DistAMC and the CentAMC, and they also have similar convergence speed, but the consumed training time per epoch in the former method will be shorter than that on the latter method, if the low latency and the high bandwidth are considered in model transmission process of the DistAMC. Moreover, the DistAMC can combine the computing power of multiple edge devices to reduce the computing overhead of a single edge device in the CentAMC.

Research on Identification Technology of Explosive Vibration Based on EEMD Energy Entropy and Multiclassification SVM

In this study, the authors introduced energy entropy as a reference feature into the field of blast vibration recognition classification and achieved good results. On the basis of the previous experimental database, 4 kinds of typical vibration signals were selected to form the sample group (building collapse vibration, surface rock blast vibration, underground tunnel blast vibration, and natural gas pipeline explosion vibration). EEMD (ensemble empirical mode decomposition) algorithm was used to calculate the energy entropy of each signal. Taking eigenvector composed of CEE (components of energy entropy) as input, multiclassification SVM algorithm was used for training and prediction. Prediction accuracy was more than 80%. Compared with BP (backpropagation) neural network algorithm, SVM (support vector machine) algorithm has higher training efficiency. The research results can be used in urban vibration monitoring, identify the nature of vibration source in time, and provide technical support for rapid response of emergency rescue.

Research on Plant Leaf Disease Identification Based on Transfer Learning Algorithm

Plant disease is one of the important threat factors that hinder the normal growth and development of plants. The intelligent identification of plant disease species has become increasingly important in the agricultural field. In This paper, the open-source data set including Black rot, bacterial spot, rust, and healthy leaves are used to train the ResNet model. And the transfer learning algorithm is applied on ResNet to establish a plant disease recognition model with good versatility and high training efficiency. The experiment results show that the disease identification accuracy of the transfer learning model is 83.75%, which is much higher than that of the traditional ResNet-101 model. Therefore, the plant disease recognition model based on transfer learning algorithm is highly feasible.

Flat random forest: a new ensemble learning method towards better training efficiency and adaptive model size to deep forest

The known deficiencies of deep neural networks include inferior training efficiency, weak parallelization capability, too many hyper-parameters etc. To address these issues, some researchers presented deep forest, a special deep learning model, which achieves some significant improvements but remain poor training efficiency, inflexible model size and weak interpretability. This paper endeavors to solve the issues in a new way. Firstly, deep forest is extended to the densely connected deep forest to enhance the prediction accuracy. Secondly, to perform parallel training with adaptive model size, the flat random forest is proposed by achieving the balance between the width and depth of densely connected deep forest. Finally, two core algorithms are respectively presented for the forward output weights computation and output weights updating. The experimental results show, compared with deep forest, the proposed flat random forest acquires competitive prediction accuracy, higher training efficiency, less hyper-parameters and adaptive model size.

Improved Traffic Sign Detection and Recognition Algorithm for Intelligent Vehicles.

Traffic sign detection and recognition are crucial in the development of intelligent vehicles. An improved traffic sign detection and recognition algorithm for intelligent vehicles is proposed to address problems such as how easily affected traditional traffic sign detection is by the environment, and poor real-time performance of deep learning-based methodologies for traffic sign recognition. Firstly, the HSV color space is used for spatial threshold segmentation, and traffic signs are effectively detected based on the shape features. Secondly, the model is considerably improved on the basis of the classical LeNet-5 convolutional neural network model by using Gabor kernel as the initial convolutional kernel, adding the batch normalization processing after the pooling layer and selecting Adam method as the optimizer algorithm. Finally, the traffic sign classification and recognition experiments are conducted based on the German Traffic Sign Recognition Benchmark. The favorable prediction and accurate recognition of traffic signs are achieved through the continuous training and testing of the network model. Experimental results show that the accurate recognition rate of traffic signs reaches 99.75%, and the average processing time per frame is 5.4 ms. Compared with other algorithms, the proposed algorithm has remarkable accuracy and real-time performance, strong generalization ability and high training efficiency. The accurate recognition rate and average processing time are markedly improved. This improvement is of considerable importance to reduce the accident rate and enhance the road traffic safety situation, providing a strong technical guarantee for the steady development of intelligent vehicle driving assistance.

Learning semantic dependencies with channel correlation for multi-label classification

Multi-label image classification is a fundamental and challenging task in computer vision. Although remarkable success has been achieved by applying CNN–RNN pattern, such method has a slow convergence rate due to the existence of RNN module. Instead of utilizing the RNN modules, this paper proposes a novel channel correlation network which is fully based on convolutional neural network (CNN) to model the label correlations with high training efficiency. By creating a new attention module, the image features obtained by CNN are further convoluted to obtain the correspondence between the label and the channel-wise feature map. Then we use the SE and the convolution operation alternately to eliminate the irrelevant information to better explore the label correlation. Experiments on PASCAL VOC 2007 and MIRFlickr25k show that our model can effectively exploit the dependencies between multiple tags to achieve better performance.

Research of Web-Page Multi-Class Classification System Based on Support Vector Machine

A Chinese web-page classification algorithm based on SVM including the important aspects of text preprocessing, feature selection and multiple-Classification algorithm. In this paper, based on the analyses of features of Web documents, this paper does research the approach of classification in Support Vector Machine (SVM) and select of Kernel function. Furthermore, a web-page classification model and algorithm that is based on Binary Tree SVM is presented. The experiments show that it not only reduces the size of train set, but also has very high training efficiency. Its precision and recall are better.

Laplacian forests: semantic image segmentation by guided bagging.

This paper presents a new, efficient and accurate technique for the semantic segmentation of medical images. The paper builds upon the successful random decision forests model and improves on it by modifying the way in which randomness is injected into the tree training process. The contribution of this paper is two-fold. First, we replace the conventional bagging procedure (the uniform sampling of training images) with a guided bagging approach, which exploits the inherent structure and organization of the training image set. This allows the creation of decision trees that are specialized to a specific sub-type of images in the training set. Second, the segmentation of a previously unseen image happens via selection and application of only the trees that are relevant to the given test image. Tree selection is done automatically, via the learned image embedding, with more precisely a Laplacian eigenmap. We, therefore, call the proposed approach Laplacian Forests. We validate Laplacian Forests on a dataset of 256, manually segmented 3D CT scans of patients showing high variability in scanning protocols, resolution, body shape and anomalies. Compared with conventional decision forests, Laplacian Forests yield both higher training efficiency, due to the local analysis of the training image space, as well as higher segmentation accuracy, due to the specialization of the forest to image sub-types.

Effects of Plyometric Training Volume and Training Surface on Explosive Strength

The purpose of this study is to examine the effects of different volume and training surfaces during a short-term plyometric training program on neuromuscular performance. Twenty-nine subjects were randomly assigned to 4 groups: control group (CG, n = 5), moderate volume group (MVG, n = 9, 780 jumps), moderate volume hard surface group (MVGHS, n = 8, 780 jumps), and high volume group (HVG, n = 7, 1,560 jumps). A series of tests were performed by the subjects before and after 7 weeks of plyometric training. These tests were measurement of maximum strength (5 maximum repetitions [5RMs]), drop jumps (DJs) of varying heights (20, 40, and 60 cm), squat and countermovement jumps (SJ and CMJ, respectively), timed 20-m sprint, agility, body weight, and height. The results of the present study suggest that high training volume leads to a significant increase in explosive performance that requires fast stretch-shortening cycle (SSC) actions (such as DJ and sprint) in comparison to what is observed after a moderate training volume regimen. Second, when plyometric training is performed on a hard training surface (high-impact reaction force), a moderate training volume induces optimal stimulus to increase explosive performance requiring fast SSC actions (e.g., DJ), maximal dynamic strength enhancement, and higher training efficiency. Thus, a finding of interest in the study was that after 7 weeks of plyometric training, performance enhancement in maximal strength and in actions requiring fast SSC (such as DJ and sprint) were dependent on the volume of training and the surface on which it was performed. This must be taken into account when using plyometric training on different surfaces.

Optimal Design of Adaptive Wavelet Neural Network Based on Hierarchy Genetic Algorithm

Based on the study of adaptive wavelet neural networks, a hierarchy genetic algorithm is proposed to training network. Compared with standard genetic algorithm, the method can not only optimize network parameters such as scale factor, transform factor parameter and connection weights, but also solve structure problem of adaptive wavelet neural network. The result of simulation indicates that the algorithm can efficiently determinate the parameter and structure of adaptive wavelet neural network, and has better higher training efficiency and forecasting precision.

Research on Key Problems of Simulation Training for Practical Software Entry Based on IP Filter Hook

Higher training efficiency and realistic of equipment operation simulation are solved by adopting practical software entry based on IP filter hook. Key problems of realization details such as data interception and data submission are also presented within this approach, which has general feasibility for operation simulation training of some software equipment based on wide area network. And it possesses favorable practical value due to its advantages such as verisimilar man-machine interaction, network condition unconstrained, training flexibly and conveniently.

Speaker adaptation of acoustic models using correlations of training transfer vectors

The authors proposed for acoustic models based on the hidden Markov model a method that involves applying constraints to the model structure and tying the model's parameters in order to improve the training efficiency. Conventionally, the tied structure of an acoustic model is, mostly, defined by tying several adjacent parameters and expressing them with a single representative parameter. This method can be regarded as a tying method based on the parameters' values, under the assumption that adjacent parameters, usually, exhibit similar behavior. As opposed to this concept, the current study proposes a tied structure with consideration of transfer (movement) of parameters. A large volume of training data was used to measure transfer of each parameter during training, and tying relationships regarding the transfer vectors were organized between parameters performing statistically similar movements. In particular, in the current study, the authors concentrated on mean vectors of fundamental distributions and followed movements of these mean vectors during training of initial models (speaker-independent models) by acoustic data from different speakers. The structure was defined by identifying the mean vectors characterized by strong correlation of movements during training, and tying their corresponding transfer vectors. Speaker adaptation tests confirmed high training efficiency of the model obtained as a result of this tying. © 2000 Scripta Technica, Syst Comp Jpn, 31(14): 74–82, 2000

海技の教育訓練方法に関する研究-II : 乗船実習と操船シミュレータ訓練の比較検討

Recently, the ship handling simulator has been introduced into various marine technical training for improvement of each techniques. The training effectiveness using ship maneuvering simulator was already certified, however, there are few reports about quantitative comparison of the training efficiency between the onboard training and simulator training. In this paper, authors analyzed the characteristic of learning process of ship handling techniques by experiments for comparison of the training efficiency between the training ship and ship maneuvering simulator. As the result, it was clarified that there is the disparity of training efficiency based on the concept of elemental technique and compound ability between the both training methods. Consequently, it was confirmed that the higher training efficiency is obtained by the compound training using the training ship and ship maneuvering simulator.

Biomechanics of the Snatch: Toward a Higher Training Efficiency

Biomechanics of the Snatch: Toward a Higher Training Efficiency