- 标题
- 摘要
- 关键词
- 实验方案
- 产品
-
[IEEE 2019 IEEE CHILEAN Conference on Electrical, Electronics Engineering, Information and Communication Technologies (CHILECON) - Valparaiso, Chile (2019.11.13-2019.11.27)] 2019 IEEE CHILEAN Conference on Electrical, Electronics Engineering, Information and Communication Technologies (CHILECON) - All-optical Routers Modeled through the Matrix Method with NVidia CUDA Development Framework
摘要： In this letter, we present a novel efficient automated tracing algorithm, called Compound Ray Recorder (CRR), to measure landscape heterogeneity efficiently without any supporting data sets. The main advantages of this method are: 1) the definition of a unified calculation framework for landscape heterogeneity is proposed and 2) no ancillary data are required, and the whole procedure can be automatically performed without any expert support or subjective evaluation. The results of tests using the proposed CRR method with actual satellite data show that it can accurately quantify the level of heterogeneity of a variety of landscapes. By normalizing the image size, the method constructs a unified framework for comparison of different regions or image extents. Meanwhile, the CRR method has been applied to time-series tracing of urban expansion and seasonal changes in the Poyang Lake area, thereby providing a new approach for monitoring landscape changes. Furthermore, heterogeneity changes mapping, and quantitative comparisons between the proposed method and existing methods are also performed.
关键词： landscape heterogeneity,index,Comparisons,heterogeneity changes mapping,time series
更新于2025-09-19 17:13:59
-
[IEEE 2019 IEEE 46th Photovoltaic Specialists Conference (PVSC) - Chicago, IL, USA (2019.6.16-2019.6.21)] 2019 IEEE 46th Photovoltaic Specialists Conference (PVSC) - Field-Aging Test Bed for Behind-the-Meter PV + Energy Storage
摘要： Data missing in collections of time series occurs frequently in practical applications and turns out to be a major menace to precise data analysis. However, most of the existing methods either might be infeasible or could be inefficient to predict the missing values in large-scale coevolving time series. Also, the evolving of time series needs to be handled properly to adapt to the temporal characteristic. Furthermore, more massive volume of data is generated in many areas than ever before. In this paper, we have taken up the challenge of missing data prediction in coevolving time series by employing temporal dynamic matrix factorization techniques. First, our approaches are optimally designed to largely utilize both the interior patterns of each time series and the information of time series across multiple sources to build an initial model. Based on the idea, we have imposed hybrid regularization terms to constrain the objective functions of matrix factorization. Then, temporal dynamic matrix factorization is proposed to effectively update the initial already trained models. In the process of dynamic matrix factorization, batch updating and fine-tuning strategies are also employed to build an effective and efficient model. Extensive experiments on real-world data sets and synthetic data set demonstrate that the proposed approaches can effectively improve the performance of missing data prediction. Even when the missing ratio reaches as high as 90%, our proposed methods still show low prediction errors. Dynamic performance demonstrates that the methods can obtain satisfactory effectiveness and efficiency. Furthermore, we have also demonstrated how to take advantage of the high processing power of Apache Spark to perform missing data prediction in large-scale coevolving time series.
关键词： missing data prediction,time series,Apache Spark,Matrix factorization
更新于2025-09-19 17:13:59
-
Parallel information processing using a reservoir computing system based on mutually coupled semiconductor lasers
摘要： Via the nonlinear channel equalization and the Santa-Fe time series prediction, the parallel processing capability of a reservoir computing (RC) system based on two mutually coupled semiconductor lasers is demonstrated numerically. The results show that, for parallel processing the prediction tasks of two Santa-Fe time series with rates of 0.25 GSa/s, the minimum prediction errors are 3.8 × 10?5 and 4.4 × 10?5, respectively. For parallel processing two nonlinear channel equalization tasks, the minimum symbol error rates (SERs) are 3.3 × 10?4 for both tasks. For parallel processing a nonlinear channel equalization and a Santa-Fe time series prediction, the minimum SER is 6.7 × 10?4 for nonlinear channel equalization, and the minimum prediction error is 4.6 × 10?5 for Santa-Fe time series prediction.
关键词： Santa-Fe time series prediction,parallel processing,reservoir computing,nonlinear channel equalization,semiconductor lasers
更新于2025-09-19 17:13:59
-
[IEEE 2019 Big Data, Knowledge and Control Systems Engineering (BdKCSE) - Sofia, Bulgaria (2019.11.21-2019.11.22)] 2019 Big Data, Knowledge and Control Systems Engineering (BdKCSE) - UV- Vis Spectroscopy and Chemometrics Analysis in Distinguishing Different Types of Bulgarian Honey
摘要： In this paper, we formulate a novel time series representation framework that captures the inherent data dependency of time series and that can be easily incorporated into existing statistical classification algorithms. The impact of the proposed data representation stage in the solution to the generic underlying problem of time series classification is investigated. The proposed framework, which we call structural generative descriptions moves the structural time series representation to the probability domain, and hence is able to combine statistical and structural pattern recognition paradigms in a novel fashion. Two algorithm instantiations based on the proposed framework are developed. The algorithms are tested and compared using different publicly available real-world benchmark data. Results reported in this paper show the potential of the proposed representation framework, which in the experiments investigated, performs better or comparable to state-of-the-art time series description techniques.
关键词： structural generative descriptions (SGDs),time series representation,time series classification,Statistical-structural pattern recognition
更新于2025-09-16 10:30:52
-
[IEEE 2019 IEEE 4th International Future Energy Electronics Conference (IFEEC) - Singapore, Singapore (2019.11.25-2019.11.28)] 2019 IEEE 4th International Future Energy Electronics Conference (IFEEC) - A Comparative Study of Flexible Power Point Tracking Algorithms in Photovoltaic Systems
摘要： Land-cover datasets are crucial for research on eco-hydrological processes and earth system modeling. Many land-cover datasets have been derived from remote-sensing data. However, their spatial resolutions are usually low and their classification accuracy is not high enough, which are not well suited to the needs of land surface modeling. Consequently, a comprehensive method for monthly land-cover classification in the Heihe river basin (HRB) with high spatial resolution is developed. Moreover, the major crops in the HRB are also distinguished. The proposed method integrates multiple classifiers and multisource data. Three types of data including MODIS, HJ-1/CCD, and Landsat/TM and Google Earth images are used. Compared to single classifier, multiple classifiers including thresholding, support vector machine (SVM), object-based method, and time-series analysis are integrated to improve the accuracy of classification. All the data and classifiers are organized using a decision tree. Monthly land-cover maps of the HRB in 2013 with 30-m spatial resolution are made. A comprehensive validation shows great improvement in the accuracy. First, a visual comparison of the land-cover maps using the proposed method and standard SVM method shows the classification differences and the advantages of the proposed method. The confusion matrix is used to evaluate the classification accuracy, showing an overall classification accuracy of over 90% in the HRB, which is quite higher than previous approaches. Furthermore, a ground campaign was performed to evaluate the accuracy of crop classification and an overall accuracy of 84.09% for the crop classification was achieved.
关键词： HJ-1/CCD,multiple classifiers,phenology,river basin,multiple scales,time-series analysis,Crop classification,land cover,multisource remotely sensed data
更新于2025-09-16 10:30:52
-
Multi-Temporal Cliff Erosion Analysis Using Airborne Laser Scanning Surveys
摘要： Rock cli?s are a signi?cant component of world coastal zones. However, rocky coasts and factors contributing to their erosion have not received as much attention as soft cli?s. In this study, two rocky-cli? systems in the southern Baltic Sea were analyzed with Airborne Laser Scanners (ALS) to track changes in cli? morphology. The present contribution aimed to study the volumetric changes in cli? pro?les, spatial distribution of erosion, and rate of cli? retreat corresponding to the cli? exposure and rock resistance of the Jasmund National Park chalk cli?s in Rugen, Germany. The study combined multi-temporal Light Detection and Ranging (LiDAR) data analyses, rock sampling, laboratory analyses of chemical and mechanical resistance, and along-shore wave power ?ux estimation. The spatial distribution of the active erosion areas appear to follow the cli? exposure variations; however, that trend is weaker for the sections of the coastline in which structural changes occurred. The rate of retreat for each cli?–beach pro?le, including the cli? crest, vertical cli? base, and cli? base with talus material, indicates that wave action is the dominant erosive force in areas in which the cli? was eroded quickly at equal rates along the cli? pro?le. However, the erosion proceeded with di?erent rates in favor of cli? toe erosion. The e?ects of chemical and mechanical rock resistance are shown to be less prominent than the wave action owing to very small di?erences in the measured values, which proves the homogeneous structure of the cli?. The rock resistance did not follow the trends of cli? erosion revealed by volume changes during the period of analysis.
关键词： cli? retreat,time-series analysis,cli? coastlines,airborne laser scanner
更新于2025-09-16 10:30:52
-
[IEEE 2019 IEEE PES Asia-Pacific Power and Energy Engineering Conference (APPEEC) - Macao, Macao (2019.12.1-2019.12.4)] 2019 IEEE PES Asia-Pacific Power and Energy Engineering Conference (APPEEC) - An Adaptive Ramp-Rate Control for Photovoltaic System to Mitigate Output Fluctuation
摘要： Data missing in collections of time series occurs frequently in practical applications and turns out to be a major menace to precise data analysis. However, most of the existing methods either might be infeasible or could be inefficient to predict the missing values in large-scale coevolving time series. Also, the evolving of time series needs to be handled properly to adapt to the temporal characteristic. Furthermore, more massive volume of data is generated in many areas than ever before. In this paper, we have taken up the challenge of missing data prediction in coevolving time series by employing temporal dynamic matrix factorization techniques. First, our approaches are optimally designed to largely utilize both the interior patterns of each time series and the information of time series across multiple sources to build an initial model. Based on the idea, we have imposed hybrid regularization terms to constrain the objective functions of matrix factorization. Then, temporal dynamic matrix factorization is proposed to effectively update the initial already trained models. In the process of dynamic matrix factorization, batch updating and fine-tuning strategies are also employed to build an effective and efficient model. Extensive experiments on real-world data sets and synthetic data set demonstrate that the proposed approaches can effectively improve the performance of missing data prediction. Even when the missing ratio reaches as high as 90%, our proposed methods still show low prediction errors. Dynamic performance demonstrates that the methods can obtain satisfactory effectiveness and efficiency. Furthermore, we have also demonstrated how to take advantage of the high processing power of Apache Spark to perform missing data prediction in large-scale coevolving time series.
关键词： time series,missing data prediction,Apache Spark,Matrix factorization
更新于2025-09-16 10:30:52
-
Time series forecasting of solar power generation for large-scale photovoltaic plants
摘要： Accurate solar power forecasting is essential for grid-connected photovoltaic (PV) systems especially in case of fluctuating environmental conditions. The prediction of PV power output is critical to secure grid operation, scheduling and grid energy management. One of the key elements in PV output prediction is time series analysis especially in locations where the historical solar radiation measurements or other weather parameters have not been recorded. In this work, several time series prediction methods including the statistical methods and those based on artificial intelligence are introduced and compared rigorously for PV power output prediction. Moreover, the effect of prediction time horizon variation for all the algorithms is investigated. Hourly solar power forecasting is carried out to verify the effectiveness of different models. The data utilized in the current work comprises 3640 hours of operation data taken from a 20 MW grid-connected PV station in China.
关键词： neural network,statistical methods,PV power forecasting,time series analysis,deep learning,grid-connected PV plant
更新于2025-09-12 10:27:22
-
Location for fault string of photovoltaic array based on current time series change detection
摘要： Various faults inevitably occur in photovoltaic (PV) array due to the harsh external working environment. Therefore, detecting the faults and theirs locations is essential for the PV array. In this paper we propose a method for detecting the faults and their locations based on time series of PV string current. A time series sliding window (TSSW) is adopted. The local outlier factor (LOF) of each current point in the TSSW is calculated. Once a number of LOFs are continuously detected to exceed the threshold value, the PV string can be judged as fault. The experiment results show that the proposed method can detect short circuit fault, open circuit fault and shadow fault for PV string under different irradiance.
关键词： Photovoltaic array,local outlier factor,fault detection,time series
更新于2025-09-12 10:27:22
-
[IEEE 2019 Conference on Lasers and Electro-Optics Europe & European Quantum Electronics Conference (CLEO/Europe-EQEC) - Munich, Germany (2019.6.23-2019.6.27)] 2019 Conference on Lasers and Electro-Optics Europe & European Quantum Electronics Conference (CLEO/Europe-EQEC) - Optical Cavity-Less 40-GHz Picosecond Pulse Generator in the Visible Wavelength Range
摘要： A method combined ensemble empirical mode decomposition, Volterra model and decision acyclic graph support vector machine was proposed to improve adaptability, feature resolution, and identification accuracy when diagnosing mechanical faults in an on-load tap changer of a transformer. In detail, the ensemble empirical mode decomposition algorithm was applied to decompose the multi-channel vibration signals in the switchover process of the on-load tap changer. Then, a Volterra model for the mechanical state of the on-load tap changer was established based on time-frequency characteristics obtained through the use of the ensemble empirical mode decomposition algorithm. Moreover, a matrix of coefficient vectors was also used in the Volterra model. This method will not only reduce the aliasing effect of empirical mode decomposition but also obtain high-resolution characteristics of nonstationary vibration signals. Furthermore, taking the singular values of the Volterra coefficient matrix as the fault characteristic, the data states of the model for diagnosing the on-load tap changer were automatically classified and identified by establishing a rapid, multi-classification decision acyclic graph support vector machine model with a low misjudgment rate. Finally, based on a certain on-load tap changer, the test platform for simulating mechanical faults was built. On this basis, by using the proposed method, the vibration signals generated due to typical mechanical faults, such as loosening of moving contacts, lessening of transition contact, and motor jam were acquired and analyzed, thus validating the effectiveness of the method through case studies. Compared with other methods, the new method could overcome many defects in existing methods and it has higher fault identification accuracy.
关键词： signal processing algorithms,Mechanical variables measurement,power transformers,fault diagnosis,electromechanical devices,time series analysis,support vector machines,switches
更新于2025-09-11 14:15:04