the ground penetrating radar (GPR) is one of the most recommended methods for tunnel lining inspection, yet the interpreting the GPR data requires significant time and expertise. Recent attempts to automate this proce...
详细信息
European energy and ancillary service markets have recently undergone many changes. Some were structural, such as the go-live of balancing energy trading platforms MARI and PICASSO, while others, like price spikes, we...
详细信息
ISBN:
(纸本)9798350312584
European energy and ancillary service markets have recently undergone many changes. Some were structural, such as the go-live of balancing energy trading platforms MARI and PICASSO, while others, like price spikes, were consequences of world events. these changes may negatively impact tools that rely on data stationarity, especially machine learning-based models. We use the energy and ancillary service market data for Croatia, France and Germany to determine prerequisites for predicting manual frequency restoration reserve (mFRR) capacity and balancing energy prices. We present a statistical analysis of the data, and draw conclusions about the steps necessary for training machine learning models on this data.
this research describes the development of a cost-effective ultrasonic wire bonding system, which was later integrated into an industrial robot to enhance welding automation. Initially, the team developed an innovativ...
详细信息
ISBN:
(纸本)9780791888117
this research describes the development of a cost-effective ultrasonic wire bonding system, which was later integrated into an industrial robot to enhance welding automation. Initially, the team developed an innovative 3D printed adapter to attach the ultrasonic transducer to the tool tip, along with a 3D Gantry platform for manual wire bonding to determine the optimal process conditions. the study examined the influence of different variables, such as wire type and size, ultrasonic power and duration, and mechanical pressure. In the follow-up phase, the system was further enhanced by integrating the ultrasonic tool with a robotic arm having six degrees of freedom. A control law was developed using the calculated arm stiffness. through experimental testing, which varied compression strength and pulse duration, the team attempted to identify effective operating conditions via destructive bond strength analysis. this integration significantly improved the system's functionality, allowing for more precise automated wire bonding and better data collection for process control. Practical applications demonstrated include automated wire bonding and cutting on aluminum plates and printed circuit boards with small features, using aluminum and gold wires of various diameters. the system's ability to cut wires in-process allows continuous bonding at different locations without halting the process. Future objectives include implementing real-time monitoring of bonding failures using data-driven methods.
the paper focuses on predicting the stateof health of Lithium Ion batteries used in Electric Vehicles to make predictive analytics on Remaining Useful Life as well as understand the battery dynamics for recycling enab...
详细信息
ISBN:
(纸本)9798350360875;9798350360868
the paper focuses on predicting the stateof health of Lithium Ion batteries used in Electric Vehicles to make predictive analytics on Remaining Useful Life as well as understand the battery dynamics for recycling enabling sustainable development. the work makes use of three different datasets, namely the Calce dataset, the lithium-ion battery model dataset, and the Oxford Battery Degradation dataset. Multiple machine learning and deep learning algorithms are adopted to predict the state of the health of the batteries in the datasets and are compared with respect to various metrics.
Body measurement data are inherently inaccurate and quite error-prone due to manual measurement and data collection. In this study, professionally collected and self-collected body measurement data were used to invest...
详细信息
ISBN:
(纸本)9798350359718;9788396960160
Body measurement data are inherently inaccurate and quite error-prone due to manual measurement and data collection. In this study, professionally collected and self-collected body measurement data were used to investigate to what extent potentially erroneous data can be identified during collection by utilizing the anthropologically given correlation of body measurements. the study specifically uses a dataset created within the framework of a project for made-to-measure pattern creation, consisting of data from 2053 female individuals with up to 52 recorded body measurements. Using linear regression, a method for validating the collected data is defined, wherein potentially inconsistent data are identified based on tolerance intervals. the tolerance intervals calculated within the study are specific to the particular application and the personal data used in the study. the outlined method is applicable to almost any set of manually collected body data in at least the triple-digit range, enabling the identification of probable data errors already during their collection.
Citizen science has emerged as a valuable resource for scientific research, providing large volumes of data for training deep learning models. However, the quality and accuracy of crowd-sourced data pose significant c...
详细信息
ISBN:
(纸本)9798350359718;9788396960160
Citizen science has emerged as a valuable resource for scientific research, providing large volumes of data for training deep learning models. However, the quality and accuracy of crowd-sourced data pose significant challenges for supervised learning tasks such as plant trait detection. this study investigates the application of AI techniques to address these issues within natural science. We explore the potential of multi-modal data analysis and ensemble methods to improve the accuracy of plant trait classification using citizen science data. Additionally, we examine the effectiveness of transfer learning from authoritative datasets like PlantVillage to enhance model performance on open-access platforms such as iNaturalist. By analysing the strengths and limitations of AI-driven approaches in this context, we aim to contribute to developing robust and reliable methods for utilising citizen science data in natural science.
this work presents a component-based data acquisition system for reading multi-sensor modules in the context of active mobility and light vehicles. Leveraging technologies coming from electric mobility in a light vehi...
详细信息
ISBN:
(纸本)9798350386301;9798350386318
this work presents a component-based data acquisition system for reading multi-sensor modules in the context of active mobility and light vehicles. Leveraging technologies coming from electric mobility in a light vehicle scenario means dealing with low power requirements and limited computational resources. To collect data simultaneously from different sensor platforms, an ad-hoc multi-process application has been developed for a single-board computer. this data acquisition module supports multi-protocol communication interfaces and it can aggregate data about the environment, the status of the vehicle and the user. Moreover, it can support the integration with an IoT communication infrastructure. A prototype of the system has been implemented to prove the feasibility of the architecture.
there is a growing concern about the dynamic landscape of cyber security threats escalating, and the need for improvement in defence capabilities against emerging sophisticated incidents. In response, this paper prese...
详细信息
ISBN:
(纸本)9798400717185
there is a growing concern about the dynamic landscape of cyber security threats escalating, and the need for improvement in defence capabilities against emerging sophisticated incidents. In response, this paper presents a solution called the Cyber Incident Simulation System, which enables system security engineers to simulate cyber-physical attacks and incidents without the requirement to affect or disrupt the ongoing business operation of the system. Leveraging graph-based threat modelling and AI-generated incident data, the system empowers professionals to predict the effect of the incident within the system under study. the synthetic data is used by anomaly-based Intrusion Detection Systems (IDSs) and other additional security controls to improve their detection algorithms to enhance their accuracy and effectiveness. the Cyber Incident Simulation System is designed to enhance the cyber security measures through the simulation of various incident scenarios.
Fully Homomorphic Encryption (FHE) is a key technological enabler for secure computations as it allows a third-party to perform arbitrary computations on encrypted data learning neither the input nor the results of a ...
详细信息
ISBN:
(纸本)9798400717185
Fully Homomorphic Encryption (FHE) is a key technological enabler for secure computations as it allows a third-party to perform arbitrary computations on encrypted data learning neither the input nor the results of a computation. Notwithstanding the recent theoretical breakthroughs in FHE, building a secure and efficient FHE-based application is still a challenging engineering task where optimal choices are heavily application-dependent. Taking linear regression as a case-study, we investigate the programming and configuration solutions to implement FHE-based applications. We show that, although obviously slower than the non-homomorphic version, the implementation of linear regression on homomorphically encrypted data is viable provided the programmer adopts appropriate programming expedients and parameters selection.
the proceedings contain 222 papers. the topics discussed include: DRL-deploy: adaptive service function chains deployment with deep reinforcement learning;accuracy vs. efficiency: achieving boththrough hardware-aware...
ISBN:
(纸本)9781665435741
the proceedings contain 222 papers. the topics discussed include: DRL-deploy: adaptive service function chains deployment with deep reinforcement learning;accuracy vs. efficiency: achieving boththrough hardware-aware quantization and reconfigurable architecture with mixed precision;cmss: collaborative modeling of safety and security requirements for network protocols;FGPA: fine-grained pipelined acceleration for depthwise separable CNN in resource constraint scenarios;Dyacon: JointCloud dynamic access control model of data security based on verifiable credentials;understanding the runtime overheads of deep learning inference on edge devices;and alleviating imbalance in synchronous distributed training of deep neural networks.
暂无评论