This paper deals with the application of second order come convex optimization to the real time management of the wastewater treatment plants and sewer network of Paris and of its suburbs. It presents preliminary resu...
详细信息
This paper deals with the application of second order come convex optimization to the real time management of the wastewater treatment plants and sewer network of Paris and of its suburbs. It presents preliminary results applied on a simple case study composed of (simple) validated models of three wastewater treatment plants (Seine Aval (SAV), Seine Centre (SEC) and Seine Gresillons (SEG)) and their (inter)connections to the sewer network modelled by transport delays. Copyright (c) 2024 The Authors. This is an open access article under the CC BY-NC-ND license (https://***/licenses/by-nc-nd/4.0/)
Print circuit board is one of the important components of the electronics industry. The Gerber format is a common format for printing circuit boards. Aiming at the lack of software to analyze the network analysis of G...
详细信息
This paper proposes a self-adaptive volt-var optimal control strategy to address the challenges that traditional control faces in adapting to changing distribution network operating conditions and the difficulty that ...
详细信息
In order to solve the problems of incomplete representation of video frame features in video summarization tasks, most methods can easily ignore timing information, while data noise, incomplete or varying quality vide...
详细信息
ISBN:
(纸本)9798400718212
In order to solve the problems of incomplete representation of video frame features in video summarization tasks, most methods can easily ignore timing information, while data noise, incomplete or varying quality video clips may be encountered, which tend to produce undesirable summaries under suboptimal conditions, we propose a video summarization network based on space-time attention and genetic algorithm optimization (VSSTGA). First, an improved Transformer network is used instead of the original backbone network, which is able to model the overall structure, long-distance dependency, and contextual relationships;then, a separate space-time attention mechanism is used, which reduces the computational effort and allows for the construction of temporal features;finally, mutation and crossover in the Genetic Algorithm is used to optimize the extracted features, which achieves the effect of evolving the model through data augmentation, and improves the robustness of the model. Finally, using mutation and crossover in genetic algorithm to optimize the extracted features can achieve the effect of evolving the model through data enhancement and improve the robustness and generalization ability of the model. The F-scores on the SumMe dataset and TVSum dataset are 53.04% and 63.07%, which are 2.8% and 1.0% higher than the existing methods, respectively. In conclusion, our method achieves accurate and comprehensive extraction of video frame features and generates more comprehensive video summarization results by introducing separate space-time attention mechanisms and genetic algorithms.
In HRM(Human resource management), it has become an inevitable trend to integrate human resources, improve management methods and accelerate the overall development of human resources through information technology. I...
详细信息
Distributed power sources can lead to problems such as increased active network loss and voltage quality degradation of the system when they are integrated into the grid. In this paper, we propose a reactive power opt...
详细信息
Generally, power flow optimization aims to minimize active power loss or generation cost in active distribution network. However, the control variables include continuous and discrete types. Since both discrete variab...
详细信息
In the present era, the integration of renewable energy into the power grid, along with the utilization of large-scale power electronic converters, has made power flow calculations more challenging. To ensure accurate...
详细信息
Satellite image denoising is of utmost importance in various applications weather monitoring, flood control and crop monitoring focusing on enhancing the visual quality of affected images. Denoising is a contemporary ...
详细信息
The proceedings contain 267 papers. The topics discussed include: deep learning prediction for multiple source-load via VMD-RIME-LSTM;a hybrid GRU-LightGBM model for day-ahead electricity price forecasting;remaining-c...
ISBN:
(纸本)9798350375138
The proceedings contain 267 papers. The topics discussed include: deep learning prediction for multiple source-load via VMD-RIME-LSTM;a hybrid GRU-LightGBM model for day-ahead electricity price forecasting;remaining-capacity-based distributed control for harmonic sharing in islanded microgrids;networking mode and evaluation method of low voltage direct current charging and discharging system in residential communities;the three-stage model and analysis of cable terminal boot failure development;electromagnetic performance analysis and optimization of asynchronous motors based on finite element analysis;research on normal force of linear induction motor based on Ansoft;and reliability evaluation method of distribution network under extreme disasters based on index reuse.
暂无评论