检索结果-内蒙古大学图书馆

Solving Nonlinear Filtering Problems Using a Tensor Train Decomposition Method

IEEE TRANSACTIONS ON AUTOMATIC CONTROL 2023年第7期68卷 4405-4412页

作者： Li, Sijing Wang, Zhongjian Yau, Stephen S. -T. Zhang, Zhiwen Univ Hong Kong Dept Math Hong Kong Peoples R China Univ Chicago Dept Stat Chicago IL 60637 USA Tsinghua Univ Dept Math Sci Beijing 100084 Peoples R China

In this article, we propose an efficient numerical method to solve nonlinear filtering (NLF) problems. Specifically, we use the tensor train decomposition method to solve the forward Kolmogorov equation (FKE) arising from the NLF problem. Our method consists of offline and online stages. In the offline stage, we use the finite difference method to discretize the partial differential operators involved in the FKE and extract low-dimensional structures in the solution tensor using the tensor train decomposition method. In the online stage using the precomputed low-rank approximation tensors, we can quickly solve the FKE given new observation data. Therefore, we can solve the NLF problem in a real-time manner. Finally, we present numerical results to show the efficiency and accuracy of the proposed method in solving up to six-dimensional NLF problems.

关键词： Duncan-Mortensen-Zakai (DMZ) equation forward Kolmogorov equations (FKEs) nonlinear filtering (NLF) problems real-time algorithm tensor train (TT) decomposition method

来源：评论

学校读者我要写书评

暂无评论

An Adaptive Interpretable Safe-RL Approach for Addressing Smart Grid Supply-Side Uncertainties 6th

An Adaptive Interpretable Safe-RL Approach for Addressing Sm...

引用

6th International Workshop on Explainable and Transparent AI and Multi-Agent Systems (EXTRAAMAS)

作者： Dey, Sumanta Verma, Praveen Dasgupta, Pallab Dey, Soumyajit Indian Inst Technol Kharagpur Kharagpur W Bengal India Synopsys Inc Sunnyvale CA USA

ISBN: (纸本)9783031700736;9783031700743

Integration of large-scale renewable energy sources is the future of smart grids. However, weather conditions significantly affect the generation of renewable energy sources, which causes supply-side uncertainties. Even with the current state-of-the-art technologies, weather conditions, solar radiance, etc., can be predicted nearly accurately, only a few minutes early. Existing intelligent grid optimization methods used to solve Optimal Power Flow (OPF) consider day-ahead planning for congestion management and generator rescheduling. Current OPF-based methods suffer many drawbacks for the large grid, like longer computation time and local convergence issues, and are computationally expensive. This drawback makes existing OPF methods unsuitable for the congestion management problem when considering supply-side uncertainties, mainly due to the large-scale integration of renewable energy sources. Over the years, Reinforcement Learning (RL) methods have gained popularity for solving constrained optimization tasks at scale. However, even if the existing Deep-RL (DRL) based policies perform best for large and complex tasks, the lack of interpretability due to using deep neural networks at their core makes them not a trustworthy solution when safety constraints like congestion-free operation are considered. Therefore, to ensure that DRL policies are used safely in critical infrastructures like smart grids, this paper proposes a centralized barrier-penalty-based DRL method that utilizes the Adaptive Safety Shield framework as an extra layer of safety. The DRL-based policy is trained to find the economically optimal grid state while considering the barrier penalty value to reduce branch loads. Besides, the Adaptive Safety Shield is trained to learn the congestion criteria and guide the RL in avoiding such scenarios during exploration. We have also empirically shown that the proposed method has demonstrated similar or better performance than the current OPF-based methods.

关键词： Smart Grid Congestion Management Renewable Energy real-time algorithm Safe Reinforcement Learning

来源：评论

学校读者我要写书评

暂无评论

An Online Unsupervised Dynamic Window Method to Track Repeating Patterns From Sensor Data

引用

IEEE TRANSACTIONS ON CYBERNETICS 2022年第6期52卷 5148-5160页

作者： Kowsar, Yousef Moshtaghi, Masud Velloso, Eduardo Leckie, Christopher Kulik, Lars Univ Melbourne Sch Comp & Informat Syst Interact Design Lab Parkville Vic 3010 Australia Univ Melbourne Sch Comp & Informat Syst Parkville Vic 3010 Australia Alexa AI Amazon Manhattan CA USA

Short bursts of repeating patterns [intervals of recurrence (IoR)] manifest themselves in many applications, such as in the time-series data captured from an athlete's movements using a wearable sensor while performing exercises. We present an efficient, online, one-pass, and real-time algorithm for finding and tracking IoR in a time-series data stream. We provide a detailed theoretical analysis of the behavior of any IoR and derive fundamental properties that can be used on real-world data streams. We show that why our method, unlike current state-of-the-art techniques, is robust to variations in repeats of the same pattern adjacent to each other. To evaluate our algorithm, we build a wearable device that runs our algorithm to conduct a user study. Our results show that our algorithm can detect intervals of repeating activities on edge devices with high accuracy (over 70% F1-Score) and in a real-time environment with only a 1.5-s lag. Our experimental results from real-world datasets demonstrate that our approach outperforms state-of-the-art algorithms in both accuracy and robustness to variations of the signal of recurrence.

关键词： Activity monitoring online algorithm pattern mining real-time algorithm time-series analysis unsupervised learning wearable technology

来源：评论

学校读者我要写书评

暂无评论

Flexible Scenario for Background Suppression in Heavy Element Research

引用

PHYSICS OF ATOMIC NUCLEI 2022年第12期85卷 1981-1987页

作者： Ibadullayev, D. Tsyganov, Yu. S. Polyakov, A. N. Voinov, A. A. Shumeiko, M. V. Joint Inst Nucl Res Flerov Lab Nucl React Dubna 141980 Russia Inst Nucl Phys Alma Ata 050032 Kazakhstan Eurasian Natl Univ Astana 010000 Kazakhstan

New algorithms to operate with new analog spectrometer of the DGFRS-2 installed at DC-280 cyclotron setup are presented. The main goal of application of these algorithms is to search an optimal time correlation recoil-alpha parameter directly during the acquisition C++ code execution. A new real-time flexible algorithm is presented in addition to the conventional ER-alpha one which is in use for a few years at the DGFRS-1 setup installed at the U-400 FLNR cyclotron. Note that the spectrometer operates together with the 48 x 128 strip DSSD (Double Side Strip Detector;48 x 226 mm(2)) detector and low pressure pentane-filled gaseous detector (1.2 Torr;80 x 230 mm(2)) are presented schematically. First beam test results in Ca-48 induced nuclear reactions are presented too.

关键词： alpha decay real-time algorithm detection system cyclotron heavy ion C plus plus gaseous detector silicon radiation detector

来源：评论

学校读者我要写书评

暂无评论

Enhancing Railway Earthquake Early Warning Systems with a Low Computational Cost STA/LTA-Based S-Wave Detection Method

引用

SENSORS 2024年第23期24卷 7452页

作者： Katakami, Satoshi Iwata, Naoyasu Railway Tech Res Inst 2-8-38 Hikari cho Kokubunji Tokyo 1858540 Japan

To enhance real-time S-wave detection in the railway earthquake early warning (EEW) system, we improved the existing short-term average/long-term average (STA/LTA) algorithm. This enhancement focused on developing a more robust and computationally efficient method. Specifically, we introduced noise reflecting P-wave amplitude information before the P-wave to better distinguish between P- and S-waves. By applying this modified STA/LTA method, we achieved a significant improvement in S-wave detection accuracy. For seismic waveforms from stations located within 100 km of the epicenter of each earthquake, with magnitude of M5.5-6.5 and depths <= 100 km, the detection accuracy within 1.5 s of the correct time (manual picking) was 81.0%, compared to the 49.0% accuracy of the currently operational railway EEW system. Importantly, despite the improved accuracy, the computational cost of the new method remains comparable to the existing system, allowing for easy integration into the operational EEW system. This development is crucial for preventing false alarms, especially moderate earthquakes (similar to M6) because issuing warn-ings in unnecessary areas can have a significant social impact. Future plans involve implementing this method into the current system to further improve early warning capabilities and minimize false alarms.

关键词： real-time algorithm earthquake early warning STA/LTA S-wave picking K-net

来源：评论

学校读者我要写书评

暂无评论

CRAR: Accelerating Stereo Matching with Cascaded Residual Regression and Adaptive Refinement

引用

ACM TRANSACTIONS ON MULtimeDIA COMPUTING COMMUNICATIONS AND APPLICATIONS 2022年第3期18卷 74-74页

作者： Zeng, Linghua Tian, Xinmei Univ Sci & Technol China Hefei 230027 Peoples R China

Dense stereo matching estimates the depth for each pixel of the referenced images. Recently, deep learning algorithms have dramatically promoted the development of stereo matching. The state-of-the-art result is achieved by models adopting deep convolutional neural networks. However, a considerable computational burden is also introduced, which slows the inference. To solve this problem, previous works down-sampled the input images to decrease the spatial size. However, down-sampling increases the error rate and its lower bound. In this article, we accelerate stereo matching algorithms through the improvement of network structure. Inspired by network compression, we conduct decomposition and sparsification to squeeze the computationally expensive cost optimization network. It is sparsified and then decomposed into smaller networks, which are designed and trained in a cascaded manner to reach the nearest possible performance of the larger network. Previous methods have utilized numerous refinement methods to adjust the coarse disparity. We integrate refinement methods to create an unified algorithm to utilize parallelism for running devices to further accelerate the inference. The extensive experiments on Kitti2015, Kitti2012, and Middlebury datasets demonstrate the efficiency of our method.

关键词： Stereo matching depth estimation real-time algorithm

来源：评论

学校读者我要写书评

暂无评论

Proper Orthogonal Decomposition Method to Nonlinear Filtering Problems in Medium-High Dimension

引用

IEEE TRANSACTIONS ON AUTOMATIC CONTROL 2020年第4期65卷 1613-1624页

作者： Wang, Zhongjian Luo, Xue Yau, Stephen S-T Zhang, Zhiwen Univ Hong Kong Dept Math Hong Kong Peoples R China Beihang Univ Sch Math Sci Shahe Campus Beijing 102206 Peoples R China Tsinghua Univ Dept Math Sci Beijing 100084 Peoples R China

In this paper, we investigate the proper orthogonal decomposition (POD) method to numerically solve the forward Kolmogorov equation (FKE). Our method aims to explore the low-dimensional structures in the solution space of the FKE and to develop efficient numerical methods. As an important application and our primary motivation to study the POD method to FKE, we solve the nonlinear filtering (NLF) problems with a real-time algorithm proposed by Yau and Yau combined with the POD method. This algorithm is referred as POD algorithm in this paper. Our POD algorithm consists of offline and online stages. In the offline stage, we construct a small number of POD basis functions that capture the dynamics of the system and compute propagation of the POD basis functions under the FKE operator. In the online stage, we synchronize the coming observations in a real-time manner. Its convergence analysis has also been discussed. Some numerical experiments of the NLF problems are performed to illustrate the feasibility of our algorithm and to verify the convergence rate. Our numerical results show that the POD algorithm provides considerable computational savings over existing numerical methods.

关键词： Heuristic algorithms Mathematical model real-time systems Convergence Stochastic processes Discrete wavelet transforms Duncan-Mortensen-Zakai equation nonlinear filtering (NLF) problems proper orthogonal decomposition (POD) real-time algorithm

来源：评论

学校读者我要写书评

暂无评论

Fast parallel blur detection on GPU

引用

JOURNAL OF real-time IMAGE PROCESSING 2020年第4期17卷 903-913页

作者： Giang Son Tran Thi Phuong Nghiem Burie, Jean-Christophe Univ Sci & Technol Hanoi Vietnam Acad Sci & Technol ICTLab 18 Hoang Quoc Viet Hanoi Vietnam Sorbonne Univ IRD UMMISCO Unite Modelisat Math & Informat Syst Complexes F-93143 Bondy France Univ La Rochelle L3i Lab Av M Crepeau F-17042 La Rochelle 1 France

Blur detection, a task to determine whether an image is blurred or not, is very helpful in various applications of image processing and computer vision. In this paper, we propose a novel method to accelerate blur detection algorithms based on Haar wavelet transform. The method decouples data dependency to gain fast 3-level Haar wavelet transform. With the obtained independence, the blur detection steps can be performed in parallel using native GPU thread blocks. We evaluated our proposed method on embedded devices, desktop and server. Our experiments show that on desktop and server, the proposed method obtains a huge performance speedup. On embedded devices, our GPU-based 3-level Haar wavelet transform is up to 4.9 times better performance and 4.3 times better power efficiency than CPU-based blur detection algorithms.

关键词： Image blur detection Haar wavelet transform GPU parallelism real-time algorithm Embedded device

来源：评论

学校读者我要写书评

暂无评论

real-time agent-based control of plug-in electric vehicles for voltage and thermal management of LV networks: formulation and HIL validation

引用

IET GENERATION TRANSMISSION & DISTRIBUTION 2020年第11期14卷 2169-2180页

作者： Veloso, Cesar Garcia Rauma, Kalle Orjuela, Julian Fernandez Rehtanz, Christian KTH Royal Inst Technol Stockholm Sweden Tech Univ Catalonia Barcelona Spain TU Dortmund Univ Inst Energy Syst Energy Efficiency & Energy Ie3 Dortmund Germany Univ Victoria Inst Integrated Energy Syst Victoria BC Canada

Ensuring a stable and reliable operation of current and future distribution networks represents a major challenge for system operators aggravated by the global proliferation of plug-in electric vehicles (PEVs). While the introduction of a controlled charging process would be advantageous to minimise the impacts PEVs cause in the system, a suitable, efficient and ready to be implemented solution is still missing. The present work addresses this issue by proposing a smart charging management solution capable to simultaneously combat the main network impacts derived from the energy needs of the vehicles. This is done by means of an agent-based hierarchical real-time algorithm which combines a local decentralised nodal voltage management with a centralised thermal control conceived to minimise the impact upon participating users. The effectiveness of the proposed system is tested both using a simulation environment considering multiple PEV penetration levels and employing commercially available charging stations and cars through hardware-in-the-loop simulations. The results reveal how all network violations are successfully attenuated by peak shaving the total aggregated charging demand and ensuring a correct system operation for all penetration scenarios while inflicting no impact on the participating users.

关键词： distribution networks distributed power generation electric vehicles hardware-in-the loop simulation real-time systems temperature control plug-in electric vehicles thermal management LV networks distribution networks global proliferation controlled charging process smart charging management real-time algorithm local decentralised nodal voltage management centralised thermal control multiple PEV penetration levels charging stations aggregated charging demand real-time agent-based control hardware-in-the-loop simulations

来源：评论

学校读者我要写书评

暂无评论

A real-time Detection algorithm for Freezing of Gait in Parkinson's Disease

A Real-Time Detection Algorithm for Freezing of Gait in Park...

引用

IEEE International Symposium on Circuits and Systems (ISCAS)

作者： Chang, Yi-Fan Ding, Jian-Jiun Hu, Hao Yang, Wen-Chieh Lin, Kwan-Hwa Wu, Po-Hung Natl Taiwan Univ Grad Inst Commun Engn Taipei Taiwan Natl Taiwan Univ Grad Inst Phys Therapy Taipei Taiwan Tzu Chi Univ Dept Phys Therapy Taipei Taiwan

ISBN: (纸本)9781479934324

Freezing of gait (FOG) is a symptom of Parkinson's disease (PD). In this paper, we develop a real-time FOG detection algorithm. To achieve the goal of real time, we adopt the generalized spectrogram with asymmetric windows and the asymmetric smooth filter. Furthermore, instead of using only the frequency domain information, we apply the features in both the time domain and the frequency domain. Simulations show that the proposed FOG detection algorithm has higher accuracy than other methods. Moreover, the time delay of the proposed algorithm is only 1 second, which is 2.25-3.75 times less than that of existing methods.

关键词： 3-D accelerometer signal analysis freezing of gait time-frequency analysis generalized spectrogram with asymmetric windows real-time algorithm

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：