The Experts below are selected from a list of 44370 Experts worldwide ranked by ideXlab platform
Alexander L. Julian - One of the best experts on this subject based on the ideXlab platform.
-
Hybrid Energy Storage Control in a Remote Military Microgrid With Improved Supercapacitor Utilization and Sensitivity Analysis
IEEE Transactions on Industry Applications, 2019Co-Authors: Giovanna Oriti, Norma Anglani, Alexander L. JulianAbstract:This paper presents a novel power flow control system for a remote military microgrid with hybrid energy storage. A combination of batteries and supercapacitors (SCs) is managed by the novel control system to increase the battery life by redirecting the higher frequency current that would have to flow in the battery if SCs were not present. This paper offers a practical solution to manage the SC current and ensure that the SCs are never overcharged or commanded to support the system when they are discharged to the lower operating limit chosen. The new controller allows the independent selection of the low-pass Filter Parameter and the number of SCs. By making the most out of these two degrees of freedom, we investigate different configurations, identifying the one achieving the highest cash flow for the overall system. Modeling, simulations, and experimental verification are presented and linked to the sensitivity analysis of the economics of the military microgrid.
-
Hybrid Energy Storage Control in a Remote Military Microgrid with Improved Supercapacitor Utilization and Sensitivity Analysis
2018 IEEE Energy Conversion Congress and Exposition (ECCE), 2018Co-Authors: Giovanna Oriti, Norma Anglani, Alexander L. JulianAbstract:This paper presents an improved power flow control system for a remote military microgrid with hybrid energy storage. A combination of batteries and supercapacitors (SCs) is managed by the novel control system to increase the battery life by redirecting the higher frequency current that would have to flow in the battery if SCs weren't present. This work offers a practical solution to manage the SC current and ensure that the SCs are never overcharged or commanded to support the system when they are discharged to the lower operating limit chosen. The new controller allows the independent selection of the low pass Filter Parameter and the number of SC S . By making the most out of these two degrees of freedom, we investigate different configurations, identifying the one achieving the best economics of the system. Modeling, simulations and experimental verification are presented and linked to the sensitivity analysis of the economics of the military microgrid.
Junli Deng - One of the best experts on this subject based on the ideXlab platform.
-
Switching harmonic suppression design based on multi-objective optimization algorithm with constraint processing
Compel-the International Journal for Computation and Mathematics in Electrical and Electronic Engineering, 2020Co-Authors: Songtao Huang, Haozhe Wang, Anwen Shen, Junli DengAbstract:Traditional switching harmonic suppressor design methods require domain experts to adjust design Parameters due to various complex performance requirements and practical limitations in switching ripple suppressor designs. The purpose of this paper is to present a method for Filter Parameter design.,An improved non-dominated sorting genetic algorithm II (NSGA II) was used in the inductor-capacitor-inductor (LCL) Filter design to find the optimal design Parameters, and a method was proposed to handle the constraints by transforming the them into decision variables.,The performance of the proposed algorithm in Parameter designing was verified by simulation on MATLAB and experimental results on hardware-in-the-loop plat-form with StarSim software. The results indicate that the optimization algorithm has a better effect than the traditional expert Parameters on each optimization index, especially on the switching harmonic suppression.,The paper presents an improved multi-objective optimization algorithm with ingenious constraints handing to obtain better Filter Parameters and reduces switching harmonics.
-
Switching harmonic suppression design based on multi-objective optimization algorithm with constraint processing
COMPEL - The international journal for computation and mathematics in electrical and electronic engineering, 2020Co-Authors: Songtao Huang, Haozhe Wang, Anwen Shen, Junli DengAbstract:Purpose Traditional switching harmonic suppressor design methods require domain experts to adjust design Parameters due to various complex performance requirements and practical limitations in switching ripple suppressor designs. The purpose of this paper is to present a method for Filter Parameter design. Design/methodology/approach An improved non-dominated sorting genetic algorithm II (NSGA II) was used in the inductor-capacitor-inductor (LCL) Filter design to find the optimal design Parameters, and a method was proposed to handle the constraints by transforming the them into decision variables. Findings The performance of the proposed algorithm in Parameter designing was verified by simulation on MATLAB and experimental results on hardware-in-the-loop plat-form with StarSim software. The results indicate that the optimization algorithm has a better effect than the traditional expert Parameters on each optimization index, especially on the switching harmonic suppression. Originality/value The paper presents an improved multi-objective optimization algorithm with ingenious constraints handing to obtain better Filter Parameters and reduces switching harmonics.
Hong Zhang - One of the best experts on this subject based on the ideXlab platform.
-
An Improved Speech Synthesis Algorithm with Post Filter Parameters Based on Deep Neural Network
Communications Signal Processing and Systems, 2020Co-Authors: Shunjie Dong, Chunyang Li, Hong ZhangAbstract:Statistical Parameters speech synthesis typically relies on context-dependent Hidden Markov Model (HMM) that is based on decision tree clustering. However, the shortcomings of clustering decision tree, restricted to a feature rigid subdivision model space, results in smooth speech Parameters generated from HMM. In this paper, Deep Neural Network (DNN) is put forward to replace clustering decision tree, and we propose a post Filter-Parameter-based speech synthesis improvement algorithm. This method enhances the formant region of synthesized speech spectrum by selecting the most optimized Filter Parameter according to the flatness of spectrum. The experimental results show that DNN effectively can modify the deficiency of two smooth Parameters. Furthermore, the improved post Filter algorithm increases the naturalness of synthesized speech.
-
CSPS (3) - An Improved Speech Synthesis Algorithm with Post Filter Parameters Based on Deep Neural Network
Lecture Notes in Electrical Engineering, 2019Co-Authors: Shunjie Dong, Hong ZhangAbstract:Statistical Parameters speech synthesis typically relies on context-dependent Hidden Markov Model (HMM) that is based on decision tree clustering. However, the shortcomings of clustering decision tree, restricted to a feature rigid subdivision model space, results in smooth speech Parameters generated from HMM. In this paper, Deep Neural Network (DNN) is put forward to replace clustering decision tree, and we propose a post Filter-Parameter-based speech synthesis improvement algorithm. This method enhances the formant region of synthesized speech spectrum by selecting the most optimized Filter Parameter according to the flatness of spectrum. The experimental results show that DNN effectively can modify the deficiency of two smooth Parameters. Furthermore, the improved post Filter algorithm increases the naturalness of synthesized speech.
Frede Blaabjerg - One of the best experts on this subject based on the ideXlab platform.
-
an active trap Filter for switching harmonic attenuation of low pulse ratio inverters
IEEE Transactions on Power Electronics, 2017Co-Authors: Haofeng Bai, Xiongfei Wang, Poh Chiang Loh, Frede BlaabjergAbstract:Switching harmonic attenuation has always been challenging for inverters used in high-power conversion applications, where ratio of switching to fundamental frequency is low. Addition of multiple LC -trap Filters is no doubt a feasible cost-effective method, which has increasingly been used, but generally susceptible to Filter Parameter variations and harmonic resonances. This paper hence presents an alternative active trap Filter (ATF), based on a series- LC -Filtered inverter, for attenuating switching harmonics in a flexible, while yet not cost burdensome, approach. A direct impedance synthesis method has also been proposed for the ATF to better enforce its active switching harmonic bypassing ability. Compared with conventional schemes for controlling active power Filters, the proposed method is more readily implemented, since it requires neither current reference generation nor high-bandwidth current control loop. Moreover, the use of a series LC Filter at its ac side helps the ATF to reduce its inverter voltage and power ratings. Compensated frequency range of the ATF can hence be enlarged by using a comparably higher switching frequency and a proper step-by-step design procedure to be presented in this paper. Simulation and experimental results have confirmed the design procedures, and hence expected performance of the ATF.
Jianhuang Lai - One of the best experts on this subject based on the ideXlab platform.
-
Filter-in-Filter: Low Cost CNN Improvement by Sub-Filter Parameter Sharing
Pattern Recognition, 2019Co-Authors: Guotian Xie, Kuiyuan Yang, Jianhuang LaiAbstract:Abstract Increasing the number of Parameters seems to have improved convolutional neural networks, e.g. increasing the depth or width of the networks. In this paper, we propose a scheme to improve CNNs by deriving the six sub-Filters from a Filter, which share Parameters among them and enhance the expressibility of the Filter. We first defined the sub-Filters of a Filter, and by visualizing a well-trained CNN, we verified that these sub-Filters could recognize multiple meaningful patterns with different visual characteristics, even when the Filter containing them was not activated. These findings revealed that the Filter has the potential to recognize multiple patterns. Inspired by these findings, we proposed the Filter-in-Filter (FIF) scheme to enhance the expressibility of a Filter, by making full use of its sub-Filters to recognize multiple meaningful sub-patterns. We verified the effectiveness of FIF on three image classification benchmark datasets, namely Tiny ImageNet, CIFAR-100 and ImageNet. Our experimental results showed that our models achieved consistent improvement over the base CNNs on the benchmark datasets, e.g. AlexNet and VGG16 using FIF achieved approximately 1% improvement on ImageNet. The sub-Filters share the Parameters and most of the computational cost with the Filter containing them; therefore, FIF does not increase the number of Parameters and increases the computational cost only slightly.