This paper presents the recent status of a PhD thesis research, with an innovative proposal for the development of Digital Systems Laboratory classes in engineering Schools. The described system uses IT and Communicat...
详细信息
ISBN:
(纸本)9781932415841
This paper presents the recent status of a PhD thesis research, with an innovative proposal for the development of Digital Systems Laboratory classes in engineering Schools. The described system uses IT and Communication over Internet in order to improve the professor's interaction, multiplying his presence and his support for the students. Joining the flexibility of distance learning with the interactivity and the intensive support of the traditional presential learning, this system permits to enlarge the delivered contents and learning efficiency of a basic and conceptual discipline. Further, a new evaluation and approval procedure is developed and applied. This methodology permits to ensure that the approved student actually knows the entire transmitted knowledge and by using a new concept evaluation grade, it is possible to measure the difficulties of each student in passing through this learning process. This system could be applied with small changes in other knowledge areas.
Electricity price forecasting plays a vital role inthe strategy decision making for almost all power marketparticipants. This article investigates statistical background andpotential relations between different power ...
详细信息
Electricity price forecasting plays a vital role inthe strategy decision making for almost all power marketparticipants. This article investigates statistical background andpotential relations between different power market products (***-ahead prices, intraday prices, etc.). Danish and Croatianpower markets are used for the purpose of the case studyto present the methods used in this article. First, Danish andCroatian power market structures are shortly explained to clarifythe context of the problem. The data collection and preprocessingmethods are described, followed by the core focus of the study:statistical analysis. In addition to the presented histograms ofrespective power market components, we examine interrelationships through statistical analysis, demonstrating significantcorrelations both numerically and graphically. Furthermore,price spreads are investigated as a logical next step of the noticedcorrelations. Our comparative analysis of Danish and Croatianmarket peculiarities reveals three key findings: i) statisticallysignificant relationships between specific market components,ii) distinct behavioral patterns among observed factors, and iii) anopen-access analytical tool with accompanying dataset for futureresearch. Finally, the findings of this article present to marketparticipants an efficient tool to adjust business strategies andincrease profit.
At the beginning of the academic year 2011-2012 the Central Library of Faculty of electricalengineering and computing (FER) set foot in the world of Web 2.0 expanding to social networks: Facebook, Twitter, YouTube an...
详细信息
The main characteristics of computer system developed for support of teaching the Fundamentals of electricalengineering are described, with emphasis on computer use in knowledge evaluation. The ways of actual compute...
详细信息
This paper deals with reduction of losses in electric power distribution system through a dynamic reconfiguration case study of a grid in the city of Mostar,Bosnia and *** proposed solution is based on a nonlinear mod...
详细信息
This paper deals with reduction of losses in electric power distribution system through a dynamic reconfiguration case study of a grid in the city of Mostar,Bosnia and *** proposed solution is based on a nonlinear model predictive control algorithm which determines the optimal switching operations of the distribution *** goal of the control algorithm is to find the optimal radial network topology which minimizes cumulative active power losses and maximizes voltages across the network while simultaneously satisfying all system *** optimization results are validated through multiple simulations(using real power demand data collected for a few characteristic days during winter and summer)which demonstrate the efficiency and usefulness of the developed control algorithm in reducing the grid losses by up to 14%.
Building energy planning is significantly challenged by climate change, particularly the increasing frequency of heat waves impacting heating and cooling demands. Current planning methodologies neglect the impacts of ...
详细信息
Federated recommender systems(FedRecs) have garnered increasing attention recently, thanks to their privacypreserving benefits. However, the decentralized and open characteristics of current FedRecs present at least t...
详细信息
Federated recommender systems(FedRecs) have garnered increasing attention recently, thanks to their privacypreserving benefits. However, the decentralized and open characteristics of current FedRecs present at least two ***, the performance of FedRecs is compromised due to highly sparse on-device data for each client. Second, the system's robustness is undermined by the vulnerability to model poisoning attacks launched by malicious users. In this paper, we introduce a novel contrastive learning framework designed to fully leverage the client's sparse data through embedding augmentation, referred to as CL4FedRec. Unlike previous contrastive learning approaches in FedRecs that necessitate clients to share their private parameters, our CL4FedRec aligns with the basic FedRec learning protocol, ensuring compatibility with most existing FedRec implementations. We then evaluate the robustness of FedRecs equipped with CL4FedRec by subjecting it to several state-of-the-art model poisoning attacks. Surprisingly, our observations reveal that contrastive learning tends to exacerbate the vulnerability of FedRecs to these attacks. This is attributed to the enhanced embedding uniformity, making the polluted target item embedding easily proximate to popular items. Based on this insight, we propose an enhanced and robust version of CL4FedRec(rCL4FedRec) by introducing a regularizer to maintain the distance among item embeddings with different popularity levels. Extensive experiments conducted on four commonly used recommendation datasets demonstrate that rCL4FedRec significantly enhances both the model's performance and the robustness of FedRecs.
This study presents a novel battery-aware approach for connectivity maintenance for multi-agent systems. It extends a classical connectivity controller by incorporating battery states into the computation of control i...
详细信息
Recent advancements in action recognition leverage both skeleton and video modalities to achieve state-of-the-art performance. However, due to the challenges of early fusion, which tends to underutilize the strengths ...
详细信息
暂无评论