Dynamic Thermal Rating (DTR) enhances grid flexibility by adapting line capabilities to weather conditions. For this purpose, DTR-based technologies require reliable and continuous measurement of the conductor temperature along the line route, which could hinder their wide-scale deployment due to the prohibitively high number of required sensors. Existing machine learning-based DTR methods infer conductor temperature from weather variables avoiding using complex and expensive measurement techniques, but their estimation accuracy greatly relies on the availability of a comprehensive set of measured data. To face these issues, this paper proposes the usage of transfer learning, a data-driven technique allowing the reduction of the number of sensors by transferring knowledge from a single calibrated source sensor to many target sensors. To the best of the author's knowledge, at the time of writing, the proposed approach is the first application of Transfer Learning in the domain of DTR which is validated on real transmission lines data. Experimental results from several real transmission lines equipped with self-organizing sensors-based DTR architecture show that transfer learning enhances the conductor temperature estimation reliability and accuracy of machine learning-based DTR techniques, suggesting the potential for practical applications, and reducing costs without losing accuracy for practitioners and system operators.

@en

The rapid growth of internet-connected devices has led to a significant increase in the number of cyber attacks, resulting in security challenges related to IoT. Researchers have discovered a new attack technique that can be used for launching large DDoS attacks, which involves TCP reflective amplification by abusing middleboxes and IoT devices. In order to assist Internet Service Providers (ISP's) in mitigating this vulnerability present at their customers, a deeper understanding of this novel attack technique is needed.
The thesis primarily focuses on exploring vulnerable devices and their end-users within the consumer network of a Dutch ISP, KPN. The ultimate goal is to gather more information on the types of vulnerable devices and actors involved to eventually assist an ISP in making informed decisions to remediate the vulnerability in their network.
The study found that the problem can be described in two different issues: vulnerable middleboxes and vulnerable consumer IoT devices with broken TCP protocols. The problem of vulnerable middleboxes has been solved in the network of the Dutch ISP as manufacturers have released updates remediating the vulnerability. This is not the case for vulnerable consumer IoT, as updating consumer IoT devices does not necessarily address the vulnerability present in the devices that have been identified. However, vulnerability notifications can potentially be useful for end-users to encourage them to update their vulnerable devices.
The study highlights the presence of vulnerable devices in the ISP network that cannot be remediated by updating the device due to the unavailability of a fix. This calls for the exploration of alternative notification methods like walled garden notifications for ISP's to address the issue as mail notifications seem not feasible at the moment of writing. While updating devices is a suggested solution, it may not be feasible for end-users with vulnerable consumer IoT devices, making it crucial for manufacturers to ensure their products have secure TCP protocols. While end-users are motivated and capable to keep their vulnerable devices up to date, whether or not they receive a vulnerability notification from their ISP, this action alone will not fully address the vulnerability as long as manufacturers remain unaware of the issue or fail to provide updates to remedy it.

In this paper, we present ALAMBIC, an open-source dockerized web-based platform for annotating text data through active learning for classification tasks. Active learning is known to reduce the need of labelling, a time-consuming task, by selecting the most informative instances among the unlabelled instances, reaching an optimal accuracy faster than by just randomly labelling data. ALAMBIC integrates all the steps from data import to customization of the (active) learning process and annotation of the data, with indications of the progress of the trained model that can be downloaded and used in downstream tasks. Its architecture also allows the easy integration of other types of models, features and active learning strategies. The code is available on https://trusted-ai-labs.github/ALAMBIC/ and a video demonstration is available on https://youtu.be/4oh8UADfEmY.@en

This master thesis aimed to develop an optimized operating strategy for an electricity grid incorporating hydrogen storage, while considering three key performance indicators: financial costs, reliability, and sustainability. The central research question guiding this study was: "How can the electricity demand be reliably met by a system combining solar power, conventional power supply, and hydrogen systems, optimized for sustainability, cost-efficiency, and reliability through adjustment of the operating strategy?"

The study quantified all key performance indicators in terms of costs, with sustainability measured by the price of carbon credits, providing a quantifiable measure of CO2 emissions. A discrete event model was utilized to simulate an electricity system on an hourly basis, integrating solar energy, a hydrogen system, and non-sustainable energy sources. The particle swarm optimization algorithm, enhanced with linear decay, was employed to identify the most optimal operating strategies within this model. Four distinct scenarios were formulated to analyze the optimal operating strategy under varying circumstances.
The findings revealed that when the price of hydrogen was significantly higher than that of conventional resources, the optimal strategy favored the avoidance of hydrogen electricity due to its high costs. Electrolysis costs and fuel cell expenses were identified as the primary cost drivers. Conversely, in scenarios where hydrogen was competitive with conventional resources, the optimal operating strategy incorporated a small portion of non-sustainable electricity while maximizing the use of hydrogen-based systems.

In conclusion, this master thesis addressed the research question by optimizing the operating strategy of an electricity grid integrating hydrogen technology. Presently, the financial advantages of non-sustainable electricity outweigh the sustainability benefits of hydrogen electricity. The study highlighted the need for substantial cost reductions in hydrogen system components and an increased price of carbon credits to realize a future scenario where hydrogen electricity competes with natural gas.

The pursuit of sustainability is at the forefront of international conversation in an era marked by urgent environmental challenges. This research aims to shed a light on tools, indicators, and methodologies used to assess sustainability, paving the way for informed decisions, actionable insights, and a brighter tomorrow.

The Paris Climate Accords urges the importance for sustainability and the reduction of environmental impact. At the same time, the demand for animal protein is growing due to an increase in world population and changing diets, resulting in growing challenges regarding the environmental impact of the agricultural and livestock sector. Firms aim to provide technological solutions that contribute to a sustainable way of farming. One of the technological solutions that lack scientific research is the livestock feeding robot. This knowledge gap offers the opportunity to investigate measuring the sustainable performance of livestock feeding robots. Additionally, scientific literature describes Life Cycle Assessment as a common and fitting methodology to measure the environmental impact of products. Furthermore, firms want to become more sustainable for competitive advantage and to be able to meet future regulations. Measuring sustainability is essential in the transition towards sustainability because the measurement can provide insights for decisions about strategic planning, product design, and supply chain design. Therefore, this research examines how the Life Cycle Assessment methodology measures the environmental impact of a livestock feeding robot life cycle by executing a case study.

A literature review was conducted to gain more detailed information on the Life Cycle Assessment methodology and environmental impact. But also, to identify multiple types of Life Cycle Assessment methodologies. The Fast Track Life Cycle Assessment methodology was selected based on data availability and compatibility with the goal of the research t0 measure the environmental impact of a livestock feeding robot. Carbon Footprint and Eco-costs were selected as impact indicators due to their practical characteristics for firms. The case study resulted in a Carbon Footprint of 34944 kg CO2 equivalents, equal to the offset of 1588 mature trees existing for one year. Additionally, the Eco-costs results in 9192 euro, representing the required investment to lower the environmental impact to a sustainable level by selecting the best available alternative technology which is needed to meet the required level of emission allowances.

To conclude, this study provides a case study about measuring the environmental impact of a product life cycle. Although this study focuses on the assessment of a livestock feeding robot, the same case study design can be used to measure the environmental impact of similar products. Therefore, other firms can replicate the case study design to measure environmental impact and meet future regulations, as well as maintain or improve their competitive advantage.

Modeling the relationship between rainfall and runoff is a longstanding challenge in hydrology and is crucial for informed water management decisions. Recently, Deep Learning models, particularly Long short-term memory (LSTM), have shown promising results in simulating this relationship. The Transformer, a newly proposed deep learning architecture, has also demonstrated the ability to outperform LSTM in machine translation, text classification, etc. However, there has been limited research on applying Transformers for rainfall-runoff modeling.

The research examined the performance of using Transformer architecture, including its time series forecasting variants, to develop rainfall-runoff models using the CAMELS (US) data set. These models were compared to the LSTM regional rainfall-runoff models, with a particular focus on snow-driven basins as the attention mechanism in Transformer is believed to allow it to attend to the earlier precipitation events in the meteorological forcing. Additionally, the Transformer's potential as a global rainfall-runoff model was also tested using the global Caravan data to determine if it could learn and generalize a wide range of rainfall-runoff behaviors, allowing it to potentially be applied in ungauged basins.

The results suggest that while Transformer and its variants may not be able to fully replace LSTM for rainfall-runoff modeling, the variant called Reformer has shown promise for daily discharge forecasting in snow-driven basins, particularly in terms of peak flow and low flow prediction. However, using the global Caravan data for building a global rainfall-runoff model was not successful due to uncertainty in the forcing data, particularly precipitation. The code for Transformer-based rainfall-runoff modeling is available publicly at https://github.com/Numpy-Panda/neuralhydrology_Transformer.

This thesis studies companies which developed open source software but adopted cloud protection licenses, a novel type of software licenses which protect their business from competitors but potentially affect their open source communities. It addresses the practical problem of executives of the companies who miss information about the effects of the license change on the health of their community. From a scientific perspective, there is no existing research about cloud protection licenses. The thesis presents the required background information, including software licenses and commercial open source software. It considers community activity and community structure as indicators of community health and presents corresponding research. Five propositions about the impact of the license change are presented and examined using data from projects of MongoDB, Elasticsearch and Redis by creating visualizations using CHAOSS and Python scripts. First, we expect that changing the licensing terms from an open source license to a cloud protection license leads to reduced community activity. We evaluated both major types of community activity, technical activity and social activity, and could not find support for the proposition but contradicting developments in all projects. Second, we expect less individuals joining the community. Here, the decomposed time series data indicates contradicting developments in the projects of MongoDB and Redis. Third, we expect more community members leaving the community. When assessing this proposition, we found that the decomposed time series data is congruent with the proposition in the project of MongoDB. However, the data of the projects of Elasticsearch and Redis indicates no support for it. Fourth, we expect increased knowledge concentration among individuals in the community. To evaluate the knowledge concentration among community members, the onion model is applied which associates each member with a role for each quarter. The development of the shares of the onion roles within the community is analyzed, and the results contradict the proposition. Fifth, we expect increased knowledge concentration considering organizations in the community. To examine the last proposition, the proportion of contributions authored by employees of the respective company is determined for each month and the development evaluated. There is no contradicting evidence in the data and the developments in the projects are congruent with the proposition. In summary, the data indicates that the impact of a license change to a cloud protection license on community health is rather small and constrained to the concentration of knowledge in respect to organizations. Therefore, we can recommend executives to adopt a cloud protection license if it addresses the strategic needs of their companies. This thesis continued the research on the impact of choice and change of software licenses on open source communities by including cloud protection licenses and previously disregarded aspects of community structure. It relates to the Management of Technology study program by taking the perspective of companies to analyze the effects of their strategic choices on their open innovation processes. By studying the effects of a license change, it examines a problem that is located at the intersection of technology, organizations, and strategy.

The limitation of transmission lines thermal capacity plays a crucial role in the safety and reliability of power systems. Dynamic thermal line rating approaches aim to estimate the transmission line’s temperature and assess its compliance with the limitations above. Existing physics-based standards estimate the temperature based on environment and line conditions measured by several sensors. This manuscript shows that estimation accuracy can be improved by adopting a data-driven Digital Twin approach. The proposed method exploits machine learning by learning the input–output relation between the physical sensors data and the actual conductor temperature, serving as a digital equivalent to physics-based standards. An experimental assessment on real data, comparing the proposed approach with the IEEE 738 standard, shows a reduction of 60% of the Root Mean Squared Error and a decrease in the maximum estimation error from above 10 °C to below 7 °C. These preliminary results suggest that the Digital Twin provides more accurate and robust estimations, serving as a complement, or a potential alternative, to traditional methods.@en