Fachbereich Ingenieur- und Naturwissenschaften
Refine
Document Type
- Article (253)
- Conference Proceeding (120)
- Article in a Periodical of the TH Wildau (51)
- Part of a Book (14)
- Preprint (11)
- Doctoral Thesis (10)
- Master's Thesis (8)
- Bachelor Thesis (5)
- Book (4)
- Contribution to a Periodical (2)
Year of publication
Institute
Keywords
- machine learning (15)
- C60 fullerene (9)
- Bildverarbeitung (7)
- Künstliche Intelligenz (6)
- apoptosis (6)
- periodic timetabling (6)
- DNA (5)
- bone density (5)
- deep learning (5)
- leukemic cell (5)
At present and in future the necessity of generating electricity using renewable resources globally is most important and clearly evident. Depending on how the scenarios are designed and disregarding fossil resources the contributions of renewable energy sources have to be increased by a factor of four to six as far as it concerns the installed capacity and the amount of electricity generated. The corresponding annual investment costs amount some hundred billion US-$ and the CO₂ emission reduction equals one Gt. In any case heat to electricity conversion systems based on biomass, geothermal and process heat recovery are expected to provide significant contributions to this objective. Besides several others the conversion technology of a Rankine Cycle Process is most suitable over a wide range of capacities notwithstanding the possibilities to applicate working fluids of non-organic or organic character. Particularly for the heat recovery from the exhaust gas of biogas operated internal combustion engines the type of a newly developed friction expander was designed, constructed, built and evaluated. Based on preceding theoretical and numerical calculations as well as constructional and experimental studies using hexane and hexamethoxydisiloxane in the project reported here saturated water steam was used as working fluid. During the experiments carried out in a special test facility the electric power of the system amounted 1.4 kW representing two thirds of the conceptual design.
From human experts to AI agents: Agentic AI for knowledge-intense problem-solving in logistics?
(2025)
Logistics planning is a complex and complicated problem solving process requiring a wide range of knowledge and competence. A planning person or team represents a particular set of knowledge, skills, experience and personal attitudes. His, her or their level of competence decides about success or failure of the planning and about the quality of its results. With this, logistics planning is knowledge-based and knowledge-creating. Because of this, Knowledge Management can provide tremendous support to the planning person. Even though, there is a wide variety of (software) tools applicable to solve specific sub-problems throughout the planning process, human expertise is the key to develop creative planning solutions so far. Latest developments with regard to Artificial Intelligence (AI) might change this picture. AI is one of the most promising technological advancements, but also one of the most ambivalently discussed topics these days. It has surpassed humans at a number of tasks already and continues in doing so at new tasks in an increasing rate. Because of this, AI is capable to assist humans in many different ways. The latest boost of development moved AI application from those sophisticated communicators or task-oriented AI to intelligent agents or agentic AI allowing for autonomous, goal-driven decision-making. Hybrid approaches combine those advanced AI capabilities with human oversight, offering context-aware adaptability without full autonomy. In this way, artificial intelligence and human intelligence join forces in teams, for example, to solve complex problems creatively. Against this background, the paper analyses agentic AI with regard to its potential to support creative problem solving in a knowledge-intense field like logistics planning. Will AI agents become co-workers in logistics planning teams or can agentic AI design logistics solutions autonomously? With this, the paper focusses on chances and challenges coming with the highly dynamic evolution of AI within a complex, knowledge-intense application area where human experience, problem-solving capability and creativity is decisive. It contributes to discussions about necessary adjustments in how organizations manage their knowledge considering artificial and human intelligence.
The aim of this paper is to provide information on the general characterization of multimodal transport, intermodal and combined transport. There is also mentioned the distribution of basic transport processes and types of transhipment. Finally, this paper describes the practical examples of multimodal transport.
This paper deals with the size of the forces affecting the vehicle during the transportation on a flatbed truck, railway wagon or container. It is necessary to secure the vehicle against these forces properly in order to prevent vehicles replacement. The laboratory measurements were performed in order to find out the size of coefficient of friction between the tire and the vehicles loading area. During the measurements, vehicle had different gears in order to determine to what extent the size of mechanical resistances affects the fixing and securing the vehicle against its replacement on the loading area. However, this paper does not propose the solution for securing the passenger vehicle on the loading area of a truck, railway wagon or container.
This paper presents the impact of air resistance (air resistance) on fuel consumption of the particular vehicle under real traffic conditions. In the introductory and subsequent part of the paper, the individual resistances influencing the moving vehicle are characterized. In the following chapter, the specific factors such as air resistance, driving speed, air temperature and air resistance coefficient in the context of their impact on fuel consumption of the particular vehicle are identified and described; as well as the comparisons of fuel consumption when the vehicle is moving on the highway with closed and open windows under different conditions of road traffic are performed.
A semiconductor based hydrogen sensor system was optimized by various modifications, which allow an improved detection of very low concentrations of hydrogen in air. The foundation for new investigations on the sensor structure are modifications of substrate and gate structures. Establishment of reference structures is a major aim. A possible drift compensation could be the use of aluminum or alloys for sensor system in order to stabilize signal in Metal Oxide Semiconductor (MOS) respectively Metal Electrolyte Insulator Semiconductor (MEIS) structures. Gold is more likely not capable to function for drift compensation as a pure metal. Nafion™ treatment for cover up the palladium gate seems not to be suitable as a reference, besides could be an option to stabilize sensor signal responses and protect sensors from environmental influences.
The advancement of 3-dimensional printing technology over the past ten years has raised interest in and accessibility to these devices. Due to its consistent growth and demand, 3D printing is becoming a consumer-friendly, reasonably priced craft. Due to technological advancements, it is becoming increasingly integrated into broader fields of science and research, as well as the manufacturing sector. The need for customized solutions is constantly growing across several industries. The 3D extrusion of hydrogels is advancing in the field of biomaterials with a broad spectrum of biomedical applications. Hydrogel extrusion prints heads often use stepper motors or pneumatic pressure systems to push the substrate onto a surface. These techniques are well-suited for materials with high viscosity. While these systems are usually bound to the syringe volume, a refillable reservoir enables them to print above the syringe’s limitations. We developed a low-cost standalone heating system for flexible tubes to control the temperature, hence avoiding jellification and clogging of the tubing system leading to the nozzle of the print head. The system connects to an in-house made peristaltic pump, which forces the e.g. gelatin through the nozzle with low pulsation, enabling us to extrude multiple layers of precise tempered gelatin. The heating system is based on easily available materials and electronic components and does not require expensive tools.
This study presents a new mathematical model for determining the specific growth rate of biomass in biotechnological production processes, which aims to optimize the production of biotechnological products such as the advanced material polyhydroxyalkanoates. The specific growth rate is classified by the FDA as a critical process parameter that affects product quality and quantity, but is difficult for laboratory personnel to determine. Therefore, a simple and robust method for real-time monitoring and control is crucial. According to the current state of the art, the established Luedeking-Piret model for determining the specific growth rate requires the determination of the biomass as an absolute value to initialize the model and to determine two further model parameters. However, determining the biomass is time-consuming and error-prone. The new relative model replaces this value with the relative change in biomass, which can be easily recorded using standard laboratory methods such as optical density measurement. This eliminates the need for time-consuming and resource-intensive preliminary work. Despite this simplification, simulation tests have shown that the new model delivers identical results to the established model. It represents an independent, precise alternative and offers advantages in terms of handling. The results underline the model’s potential to make bioprocesses more sustainable and efficient. Especially in research, material consumption, laboratory time and costs can be reduced compared to the established model. Future experiments will further investigate the performance of the new approach compared to the established model.
The ability to read and write is an important requirement for an active participation in social life – maybe even the most important one. It is a prerequisite for education and in most cases also for professional success. So, it is important, that children learn how to read. In this paper, a case study is presented which investigates the question how humanoid robots can support the motivation of children to read. For this, we considered the following questions: Can humanoid robots, like the NAO robot from Aldebaran Robotics, motivate children to read more and with pleasure? Is there a setup which is technically stable enough to work outside a lab environment and without developer support? The underlying project uses children’s curiosity about humanoid robots to create an exciting learning experience. The idea can be summarized as follows: Children read to a NAO robot and communicate to the robot about the content of the book. What started as a research project is now an application that is readily used in several public libraries in Germany. Currently, our so-called Reading-NAO can interact in German and in English. The paper also introduces the idea, its realization, practical experiences and future plans for the project.

