FEATURE EXTRACTION AND ESTABLISHMENT BASED ON PUMPING UNIT WORKING CONDITIONS AND GLOBAL FAULT IDENTIFICATION by REN Taizhu, FAN Jun, JIANG Xiaxin

“…Then, two methods of obtaining valve opening and closing points and load variation characteristics were proposed, and 54 new features of global faults of pumping units were extracted, and the characteristic database of working conditions of the pumping unit was established.Finally, the algorithm of decision tree, logistic regression and support vector machine was used to verify that the feature database has good classification effect under different working conditions. …”
Get full text

Load balancing multipath routing protocol for mass remote sensing data downlink in LEO satellite network by Pei-long LIU, Hong-yu CHEN, Song-jie WEI, Hao CHENG, Shuai LI, Jun-yong WANG

“…LEO satellite networks can provide seamless real-time data communication for all kinds of users,which developed rapidly in recent years.At the mean time,the massive payload data down-link system of space data sources,such as remote sensing satellites,still make use of traditional storage and forward mode.The real-time performance of space mission data will be improved effectively,if such satellites are connected to LEO satellite networks equipped with inter satellite links.However,it is necessary to design a specialized satellite network load balancing routing algorithm.Satellite parallel edge-disjoint multipath routing protocol (SPEMR) was designed for remote sensing satellite real-time down-link applications.OPNET simulation results indicate that the performance degradation index(DI) of the multipath scheme implemented by SPEMR is 0.32,which is only 32% of the TLR and 21% of the traditional DSP scheme.It is demonstrated that SPEMR has the better capability of transmitting massive data in real time.…”
Get full text

Detection of opening motion characteristics in DC circuit breakers based on machine vision. by Zhaoyu Ku, Jinjin Li, Dongheng Li, Huajun Dong

“…The captured images are preprocessed, with coarse edges extracted using the Laplacian algorithm. Franklin moment convolution calculations are then applied to determine sub-pixel coordinates of the image edges based on these coarse edges. …”
Get full text

Cooperative Coevolution with Two-Stage Decomposition for Large-Scale Global Optimization Problems by H. D. Yue, Y. Sun

“…Experiments show that CC with TSD is a competitive algorithm for solving LSGO problems.…”
Get full text

Optimizing PID Controller Design for Rotor Systems Suspended by Active Magnetic Bearings by Ibrahim Abubakar, Atte Putkonen, Marek Rehtla, Tuomo Lindh, Stijn Derammelaere, Niko Nevaranta

“…Therefore, a multiobjective genetic algorithm optimization is proposed in this paper to seek an optimal configuration for the controller parameters. …”
Get full text

Recurrences Reveal Shared Causal Drivers of Complex Time Series by William Gilpin

“…We introduce a physics-based unsupervised learning algorithm that reconstructs causal drivers by iteratively building a recurrence graph with glasslike structure. …”
Get full text

Local System Matrix Compression for Efficient Reconstruction in Magnetic Particle Imaging by T. Knopp, A. Weber

“…In the present work, it is shown that the image quality of the algorithm can be considerably improved by using a local threshold for each matrix row instead of a global threshold for the entire system matrix.…”
Get full text

Stochastic Portfolio Selection Problem with Reliability Criteria by Xiangsong Meng, Lixing Yang

“…The equivalent linear (mixed integer) programming models are also deduced for different evaluation strategies. A genetic algorithm with a polishing strategy is designed to search for the approximate optimal solutions of the proposed models. …”
Get full text

Modeling the Interaction Networks about the Climate Change on Twitter: A Characterization of its Network Structure by Mary Luz Mouronte-López, Marta Subirán

“…We also find that the process of emergence of new links in the interaction networks can be properly modeled (with high accuracy) through a Support Vector Machine model using the embeddings provided by the Node2Vec algorithm. A Random Forest model using certain similarity measures as explanatory variables between nodes also provides high accuracy. …”
Get full text

Dynamic Path Optimization in Sharing Mode to Relieve Urban Traffic Congestion by Hongmei Chen, Yuanhang Zhang

“…The results show that both the model and algorithm are reliable and effective; thus, travel efficiency can be effectively improved and traffic jams can be alleviated.…”
Get full text

Ear Recognition Based on Gabor Features and KFDA by Li Yuan, Zhichun Mu

“…The ear detection approach based on improved Adaboost algorithm detects the ear part under complex background using two steps: offline cascaded classifier training and online ear detection. …”
Get full text

Gearbox Fault Diagnosis Based on ALIF-PE-GOLSSVM by Huang Ying, Li Ximei, Ye Renhu, Wang Rui

“…Then the PE value with component (C-PE value) is calculated to obtain the multi-scale characteristics of vibration signals. Then the genetic algorithm is used to optimize the least squares support vector machine (LSSVM). …”
Get full text

PREDICTION AND DAMAGE ANALYSIS OF THE POSTBUCKLING FOR MULTI-SPAR COMPOSITE BOX by YUAN Fei, CHAI YaNan, ZHANG AYing

“…In this paper,a failure algorithm based upon Puck’s failure theory has been developed. …”
Get full text

MP-Start:bandwidth measurement based and phase-divided TCP slow start mechanism by DENG Xiao-heng1, CHEN Zhi-gang1, ZHANG Lian-ming1, ZENG Bi-qing1

“…To set appropriate threshold value（ssthresh） of slow start and eliminate network oscillation caused by bursty traffic,a new measurement-based and phase-divided slow start mechanism was proposed,which was called MP-Start.MP-Start employs effective online bandwidth measurement technology to update the ssthresh with appropriate value dy-namically,and adjusts the increment of congestion window（cwnd） in different phase of the startup.MP-Start increases cwnd with half of the difference between ssthesh and cwnd,iterates and gradually closes up ssthresh,of which the incre-ment is small at start phase and end phase and comparatively large at the middle phase of the startup of a connection,the largest increment is no more than one quarter of ssthresh.In order to improve the performance of this algorithm,a flexi-ble parameterized model and an adaptive parameter setting solution are designed.And series simulation experiments in-dicate that MP-Start significantly decreases dropped packets,data transforming time,packet queue length of shared link,and also improves network performance and smoothness of connections.…”
Get full text

Survey of FPGA based recurrent neural network accelerator by Chen GAO, Fan ZHANG

“…Recurrent neural network(RNN) has been used wildly used in machine learning field in recent years,especially in dealing with sequential learning tasks compared with other neural network like CNN.However,RNN and its variants,such as LSTM,GRU and other fully connected networks,have high computational and storage complexity,which makes its inference calculation slow and difficult to be applied in products.On the one hand,traditional computing platforms such as CPU are not suitable for large-scale matrix operation of RNN.On the other hand,the shared memory and global memory of hardware acceleration platform GPU make the power consumption of GPU-based RNN accelerator higher.More and more research has been done on the RNN accelerator of the FPGA in recent years because of its parallel computing and low power consumption performance.An overview of the researches on RNN accelerator based on FPGA in recent years is given.The optimization algorithm of software level and the architecture design of hardware level used in these accelerator are summarized and some future research directions are proposed.…”
Get full text

Extensible hierarchical codec semantic communication system by Yuyuan ZHANG, Haitao ZHAO, Jibo WEI, Kuo CAO, Yichi ZHANG, Peng LUO, Yueling LIU, Kai MEI

“…Aiming at the problem that most current researches on text semantic communication mainly rely on simulation system for theoretical verification, an extensible hierarchical semantic communication system was proposed by taking advantage of the separation of hierarchical encoding and decoding architecture in semantic level and grammatical level.The system was compatible with the reliable communication technology under the framework of Shannon information through the mode of semantic and syntactic separation, and realized the nested combination of semantic communication and traditional communication.Furthermore, a universal and extensible verification system was built based on the software radio platform to verify the proposed semantic communication system architecture.The verification system took semantic communication software platform as the driving core of hardware drive and algorithm call, integrated the whole process of signal generation, information transmission, data acquisition, decoding and evaluation at the receiving end, and could be further extended for semantic and syntactic level.Finally, the text semantic communication was tested based on this verification system, which verified that it had higher validity and reliability than the traditional communication mode.…”
Get full text

Multi-version consolidation method for block-level backup data by WANG Shu-peng 1, WU Guang-jun1, WU Zhi-gang3, YUN Xiao-chun1

“…Historical version backup data query efficiency was dependent on the version chain length in a multi-version backup data storage system.A novel block-level backup data multi-version consolidation method was presented for this problem.Any version of a historical full system snapshot could be generated within the incremental backup data version sequence.Theoretical analysis showed that it could increase the target snapshot version query efficiency significantly in the version chain.A variable-length block version merging algorithm was presented to decrease the computing complex-ity in the process of snapshot version merging,according to the block-level backup data distribution features.Finally,the concrete applications and performance analysis for block-level version consolidation method was also introduced.The simulated experiments results expose that the version consolidation method can eliminate redundant version of backup data with the same version chain length,and it can increase full system recovery efficiency than traditional methods sig-nificantly.…”
Get full text

Current financial control automation in a holding by O. D. Gaisha, V. A. Dzhioev, V. L. Okorokov

“…Based on the study results, an approach to current financial control automation in a holding company has been proposed: the algorithm of data collection and the format of its processing have been described, as well as approaches to automation of reporting forms for stakeholders in order to present the main trends and factors of deviations necessary for further management decisions. …”
Get full text

Target Torque Control of Two-speed DCT Clutch based on GA-LQR by Zhang Zhigang, Chen Jinjing, Yao Lianhong, Zheng Yanjie

“…The global search ability of the genetic algorithm (GA) is used to optimally select the weighting matrix in the LQR. …”
Get full text

Malicious domain name detection method based on associated information extraction by Bin ZHANG, Renjie LIAO

“…To improve the accuracy of malicious domain name detection based on the associated information, a detection method combining resolution information and query time was proposed.Firstly, the resolution information was mapped to nodes and edges in a heterogeneous information network, which improved the utilization rate.Secondly, considering the problem of high computational complexity in extracting associated information with matrix multiplication, an efficiency breadth-first network traversal algorithm based on meta-path was proposed.Then, the query time was used to detect the domain names lacking meta-path information, which improved the coverage rate.Finally, domain names were vectorized by representation learning with adaptive weight.The Euclidean distance between domain name feature vectors was used to quantify the correlation between domain names.Based on the vectors learned above, a supervised classifier was constructed to detect malicious domain names.Theoretical analysis and experimental results show that the proposed method preforms well in extraction domain name associated information.The coverage rate and F1 score are 97.7% and 0.951 respectively.…”
Get full text