The ongoing industrialization of material extrusion additive manufacturing requires tools for the acquisition of condition data for the formalized evaluation of process qualities. However, current approaches for process monitoring are not sufficient regarding the needed flexibilities and capabilities for interpreting process data. Therefore, in this thesis it is investigated to what extent quality indicators for material extrusion can be derived based on self-learning image processing methods for a layer-by-layer anomaly detection and an evaluation of the digital part information.
For this purpose, the optical properties of layer surfaces are analyzed and an image processing system is developed on this basis. To include digital part information in the image processing steps, the NC program for the machine control is analyzed. Furthermore, different layer areas are detected in segmentation steps, characteristic feature vectors for layer surfaces are extracted and unsupervised machine learning methods are used to detect imperfections. Finally, quality indicators for the area and the number of layer imperfections are determined. The evaluation of the image processing system based on a realistic data set and shows that influences of an illumination are significant and can be partially represented in a model. Additionally, uncertainties of digital part information occurring in manufacturing practice are quantified. Furthermore, a proposal for quality classes is derived from an empirical field study.
additive manufacturing, material extrusion, quality indicators, quality classes, image processing
Die forging enables components of very good quality with excellent technical properties to be produced economically in large quantities. However, design freedom is limited because die forging cannot produce components with undercuts. The simple applicability of undercut forging could further increase the variety of forged parts. In particular, the process can reduce the amount of machining required, thus increasing process and resource efficiency. In this context, improved technical component properties, such as increased fatigue strength, can also be expected. Conversely, these improved properties allow a reduction in weight without a reduction in technical properties. Currently, no forging processes with undercut insertion for complex forging part geometries are known. Therefore, an undercut forging process for a complex component was researched in this work. A steel piston with an undercut piston pin bore is chosen as an example component. The process was developed and optimized virtually before finally being tested experimentally.
Forging, die forging, undercuts, FEM, steel pistons
Tool-integrated process variable monitoring during cross-wedge rolling of hybrid semi-finished products in the context of tailored forming process chains
Components made from several materials offer the potential to improve properties and increase functional integration. For example, it is possible to produce lighter and application-optimised components by combining steel and aluminium. The combination of different iron alloys can improve the mechanical properties of a component and thus increase the service life of rolling contacts, for example. Within the framework of the Collaborative Research Centre 1153 "Process Chain for the Production of Hybrid High-Performance Components by Tailored Forming" at Leibniz Universit¨at Hannover, the production of hybrid components is being researched. In particular, high demands are placed on the production of suitable semi-finished products by means of various joining processes, targeted heating and subsequent forming. Process monitoring of the forming of hybrid semi-finished products by means of cross-wedge rolling is necessary to ensure consistent component qualities and to detect component defects at an early stage. Within the scope of this dissertation, approaches for tool- and machineintegrated process monitoring are investigated. It is shown that by means of different sensor types a reproducible assessment of the process parameters can be carried out and the compliance with process limits can be monitored. Process parameters such as semi-finished product temperature and material as well as tool temperature, speed and position can be used automatically to determine process deviations.
Cross-Wedge Rolling, Process Monitoring, Tailored Forming, Hybrid Components
Globalization and the associated increase in competitive pressure are presenting companies with new challenges. Customer demands for fast delivery times, a wide range of variants and favorable prices lead to constant pressure for improvement in companies. In this context, the focus is also on optimizing or increasing the efficiency of logistics activities. To carry out logistical activities in the warehouse, transport systems are required that are selected individually for each company. This selection is very time-consuming for companies and requires a high professional competence of the employees. To support this, a method was developed in this thesis, with which the transport system selection in the warehouse can be carried out in a supported manner, taking into account the necessary degree of changeability and automation. During the development of the method, both the given and the necessary degree of automation and changeability of transport systems were determined individually. In addition, costs were taken into account, the decision components (costs, automation changeability) were considered in combination and the application of the method in companies was prepared. Through an exemplary application it could be confirmed that the developed method can accelerate a selection project, reduce the commissioning of additional consulting services and that the selection of employees can be carried out without high expertise. In addition to the resulting increase in efficiency of the selection project, the user-friendliness could be increased while maintaining the same quality of results. Furthermore, the individual sub-methods can also be used independently of each other and a transfer of the developed method to other systems is possible.
Transport system, system selection, automation, ability to change
Absolutely measuring angle of rotation and measuring torque are of great importance in industry. So far, the acquisition of both variables is done separately. Within the scope of this thesis, a non-contact measurement method for the determination of the absolute rotation angle and torque is investigated. In order to detect the torque of a mechanical shaft contactless, two coded markings in the circumferential direction and corresponding distance are applied with a laser, which are red by means of an optical system. Both markings encode the absolute angle of rotation within 360°, so that when the two angles of rotation are detected, an angle difference and thus torque are inferred. Further development of existing codings as well as the detectability of the absolute rotation angles and the torque calculated therefrom are described in this work. In addition, the detection of the circularity and the effect of this on the angle measurement in the context of the scientific question is examined. For this purpose, both the model presentation and experimental measurement results are described in the work. The simultaneous detection of the circularity offers the possibility of determining the angle-dependent change in distance, so that an application in condition-based maintenance (e.g. of bearings) is conceivable. Furthermore, preliminary considerations for industrial applicability are included in order to enable a compact design and further optimize data processing. In experiments with an engine test bench, the implemented prototypical measuring system is investigated and discussed in industrial practice. At a speed of 500 min-1, the general feasibility of optical torque measurement was demonstrated with an error of up to 11 %.
optical torque measuring system, rotation angle, coding, circularity, angle difference method
The assembly of XXL-products is often organized according to the job-site principle. With this principle, the assembly objects are arranged stationary and the necessary resources and materials are brought there. If several job-site assemblies are arranged simultaneously at the assembly area, an area-allocation-planning is necessary to prevent the available area from becoming a bottleneck. The area allocation planning can follow different design variants, which differ regarding the degree of freedom in the area arrangement, their organizational framework conditions as well as the tolerance against external influences. The possible general design variants and how they differ in terms of performance, planning reliability and resulting costs for
different framework conditions has not been researched yet.
This study shows that companies can optimally align their area-allocation-planning regarding the performance indicators productivity, planning reliability and expenditure by the implementation of an appropriate design variant. For this purpose, the special characteristics of job-site organized assemblies are analyzed, relevant influencing factors identified and the possible variants for the design of the area-allocation-planning are derived. In order to evaluate and compare them with each other, a simulation model is applied which maps the area allocation planning for all design variants depending on the influencing factors. The results of a simulation study carried out with this model confirm the different characteristics of the design variants and their individual suitability for specific conditions.
assembly, job-site-principle, area allocation planning, xxl-products
Manufacturing companies are trying to meet individual customer requirements by increasing the number of product variants. The customization of products often takes place in the assembly. The assembly is also the last step in the industrial value-added process and is often affected by disruptions.
Disruptions in the assembly processes have a negative impact on the manufacturing company's logistical objectives and therefore their competitiveness. Due to the high complexity of assembly, a disturbance often leads to further disturbances. In addition, manufacturing companies are forced to reduce time, capacity and inventory reserves due to global competitive pressure and therefore can react less flexibly to disruptions.
For this reason, an approach to the production planning and control of the assembly is developed which describes the interdependencies of actions for dealing with disruptions to the logistical objectives.
production planning and control, assembly, possibility of disturbance, sequencing, simulation
The dismantling of wind turbines is gaining in importance. The reasons for this are the expiry of feed-in tariffs for a large part of the German plant stock as well as the technical and economic end-of-life of wind farms. Operators and dismantling companies are faced with the challenge of organizing the dismantling process efficiently. There is a conflict of objectives between the positioning of the complex and expensive dismantling at the site of operation and the expensive transport of components of a wind turbine that are not further dismantled to a dismantling factory where a more cost-effective dismantling is possible. In the research work, the most important factors influencing disassembly are summarized in an impact model and transferred into a mathematical model. The mathematical model is based on a site planning and allocation problem for wind turbine disassembly. By applying the model to a framework scenario and conducting a parameter study, recommended actions for the individual components in the dismantling of a wind turbine are developed. In addition to the allocation of dismantling tasks to dismantling sites, these also included recommendations for action with regard to dismantling planning and documentation.
Wind energy, dismantling, site planning and allocation problem, dismantling networks
Increasing quality requirements combined with high cost pressure make efficient quality management indispensable, especially for manufacturing companies. As part of quality management, complaints management ensures that complaints are received and processed and that the causes of defects are identified and eliminated. The 8D method is a tool for the profitable documentation and processing of complaints. Parallel to the execution of the 8D method, all steps performed are recorded in an 8D report. There are no objective internal inspection or control bodies that ensure the integrity and correctness of an 8D report.
Therefore, this thesis aims to support the verification of 8D-reports by an automated evaluation system. Thereby, 8D-reports should be evaluated both formally and in terms of content.
8D-Report, computational linguistics, quality assessment, complaints
As a result of the energy turnaround and the expansion of the grid, energy-intensive manufacturing companies are confronted with further increases in energy costs and are urged to reduce them through their own measures. In this context, energy costs can be positively influenced by organisational measures, whereby the actual power requirement remains unchanged. This can be done within the machine allocation planning.
However, logistical target values, such as a high delivery reliability, must not be neglected. Therefore, in this thesis a method for energy- and logistics cost oriented machine allocation planning was developed, which considers energetic and logistic target values simultaneously during planning. In the developed method, the planned peak load was used as an energetic target value and valued monetarily via the performance price. The planned schedule deviation in order issue was used as the logistical target value as the target value in the method and was valuated in monetary terms via the stock of manufacturing orders with early planned completion.
The developed method was converted into a mathematical model and a problem-specific algorithm was developed to solve the model. By applying the developed method it could be confirmed that with the additional consideration of the energetic target figure planned load peak in the machine allocation planning, the total costs resulting from the implementation of the planning can be significantly reduced compared to a sole consideration of the logistic target figure. Furthermore, the qualitative effects of the variation of the considered input variables on the total costs resulting from the implementation of the machine allocation plan could be shown.
Machine occupancy planning, energy costs, performance price, schedule deviation
To determine the influence of the process combination of cross wedge rolling with multidirectional forming on the burr formation in single-cylinder crankshaft preforms, cross wedge rolling and multidirectional forming geometries were determined. For this purpose, FEM simulations were first used to narrow down the value range of the five varied parameters for subsequent experimental investigations.
The parameters were the shoulder angle, the reduction in cross-sectional area, the bearing offset, the forming speed and the workpiece temperature. After the evaluation of both investigations, there was potential for adjustments to the FEM simulations. In the course of the search for suitable parameters for adaptation, friction was identified as significant in the FEM simulations.
On the one hand, the values were varied within the friction coefficient friction factor model used; on the other hand, the suitability of another friction model, the IFUM friction model, was examined.The latter influenced the results of the FEM simulations in such a way that a sufficient approximation to the experimental investigations was achieved and a mathematical model could be formed on this basis.
cross wedge rolling, multidirectional forming, crankshaft, friction
Ever-stricter requirements for the quality of the finished product and the energy consumption pose difficult challenges for industry, especially the automotive industry. The market demands high-quality components at a reasonable cost.
Modern technologies and innovative methods help to meet this challenge. Process monitoring has become a key technology and enables the improvement of safety, quality, and efficiency in various fields. Until recently, production ? from the design of the aluminum melting furnace to the daily process ? relied largely on tra-ditional methods and experience.
This thesis investigates a new method for monitoring a melt-ing process and shape changes in the furnace by means of optical sensors for the first time. To this end, this thesis deals with an innovative analysis algorithm for using an optical meas-uring system that is able to monitor the melting block despite the red-hot furnace walls.
melting process, analysis algorithm, light-field camera, process monitoring
In the burr-free forging of aluminium, material can unintentionally penetrate into the gap areas between the tool elements, thus creating so-called tinsel burr.
Burr makes handling and positioning of the components in subsequent forging or machining operations more difficult and thus leads to position and tolerance errors on the component. In this thesis, two different methods for the first numerical prediction of the formation of flake burr were developed.
For specific areas of finished parts, it was possible to qualitatively predict the formation of flake burr in all areas considered. Furthermore, a method for the reduction of flake burr by geometrical optimization of preforms using evolutionary algorithms was developed. The optimized preforms met or exceeded the established quality requirements compared to conventionally developed preforms with a significantly reduced optimization time of less than ten minutes.
Solid forming, burr-free forging, flash, forecast, optimization
When forging aluminum, the main cause of tool failure is adhesion-related wear. The high affinity for the adhesion of aluminum alloys to the steel materials of the forging dies results in so-called "stickers", which have a negative influence on the forging result. As a result, the engravings on the tools must be regularly overhauled. The overhaul interval depends largely on the workpiece geometry and temperature as well as on the tool temperature, material and surface.
This paper develops a model that predicts the adhesive wear and thus determines the tool life or the time of potential tool failure. This model is based on the correlation of simulative and experimental data. After evaluating the sensitivity of the varied parameters, influencing variables such as forming or die temperature are combined in a data mining model. With the help of various algorithms, the surface quality and therefore the tool life can be predicted after simulation of any aluminum forming process.
forging, adhesion, aluminum, wear
Automated guided vehicle systems (AGVS) have become indispensable in advanced production facilities. Due to significant progress in the field of AGVS and the increased automation within production plants, the potential applications for AGVS increase. So far the roadmaps for the vehicles are mostly generated manually, which leads to long and laborious planning phases. This thesis examines how system planners' knowledge can be integrated into a pathfinding algorithm in order to combine human logic with mathematical optimization to generate roadmaps for AGVS that are both efficient and applicable.
The combination of mathematical planning and human planning was achieved by combining a fuzzy inference system with a traditional pathfinding algorithm - the A* algorithm. The fuzzy inference system stores the knowledge of the system planners in the form of fuzzy rules and the output of the rules directly influence the path planning of the A* algorithm.
pathfinding, expert system, fuzzy logic, automated guided vehicle system, automated guided vehicle
Rising procurement costs for electrical energy and an increase in electricity price volatility due to the increased feed-in of renewable energies endanger the international competitiveness of manufacturing companies in Germany. To meet this challenge, the demand for energy must be adapted to the energy supply in the medium term. Demand-side energy management requires companies to consume more energy when energy is cheap. A starting point for this is the production control, since this has the task to implement the production plan created by production planning due to frequent unavoidable disturbances. It thus determines when which order of production is received and processed. In this thesis the influence of the production control on the energy costs is described and integrated into an existing active model of the production control. Based on the active model, a sequential rule is developed as a decision model, which takes into account not only the energy prices, but also the time-related urgency of the individual orders and thus enables energy cost savings while taking into account the logistical target values. Finally, the simulation-based validation of this sequence of rules proves part of the model of action. Based on the validated impact model, the application prerequisites and the influences on the logistic target values ??of an energy price-oriented order sequence rule are described.
Production control, sequencing, energy prices, energy costs, electricity price volatility
The introduction of continuous condition monitoring of machinery and equipment is associated with obstacles, especially for retrofits in the maritime sector. Sensor networks in conjunction with wireless, energy-autonomous sensor nodes completely avoid the cabling effort for energy supply and data transmission, so that the sensors for measuring value acquisition can be integrated with a single installation effort. Energy Harvesting systems allow the sensor nodes to be supplied with energy from the immediate environment.
The aim of the present work is therefore to develop a model that enables the energetic design of wireless, energy self-sufficient sensor nodes for condition monitoring of machines and plants.
wireless sensor network, energy harvesting, energy-independent sensors, condition monitoring
Automated guided vehicles (AGVs) are unsteady conveyors and suitable for logistics. To reduce installation costs and to make AGVs economically more feasible for small- and medium-sized enterprises (SMEs), autonomously acting AGVs are used. These AGVs can be ordered decentrally via natural communication channels of humans. Most frequently, people interact via speech. The aim of this thesis is to examine how well a decentralized speech control for autonomously operating AGVs works in intralogistics and which impact the operator's cognitive information processing has. For this purpose, a speech control system for AGVs and an electroencephalography (EEG)-based measurement system were developed to investigate cognitive information processing. Four independent studies were conducted to evaluate the interaction with the AGVs. The results confirm the assumption that a speech control for AGVs in intralogistics can be used speaker-independently. However, noise has a significant influence on the recognition error of the speech control. In addition, it was found that the interaction is realizable in real time and that inattention and relaxation of the operator have a significant effect on the work result.
cognition, speech control, automated guided vehicles (AGVs), human-machine-interaction (HMI)
The preform design of forging processes mainly influences the economics of a multistage forging process and is very time consuming. This thesis presents an automated approach for preform design, aiming to reduce the development effort for multistage forging processes. Therefore analytical equations are derived to describe the quality of preforms. The equations take into account all main quality (formfilling, folds) and economic parameters (amount of flash, forming force, manufacturing expenses) for preforms.
To solve the multicriteria optimization problem an evolutionary algorithm is used. The results show that it is possible to obtain a suitable preform by using the algorithm in less than 60 seconds. Furthermore a method to reduce the amount of flash for cross wedge rolled preforms is developed in this thesis. Based on an equation describing the minimum required amount of flash in each area of the final part an optimized mass pre-distribution is achieved. Compared to conventional preforms a flash reduction of 66 % is reached.
bulk forming, evolutionary algorithm, preform optimization, flash reduction
The most common method of bulk forming processes is closed die forging with flash. In these processes a surplus of material is used to avoid forging defects such as a missing filling of the cavities. The surplus material is driven through the flash gap and forms the flash. Therefore, the flash gap design has a distinct influence on the material flow within the die and thereby on the cavity filling, the material usage and other process parameters. In indus-try the flash gap is fixed once the dies are manufactured and cannot be changed without additional manufacturing.
A moveable flash gap, which can be changed vertically during the forging operation, was developed within this thesis. It can be used to influence the material flow within the forming process. Thus, the filling of the cavity can be improved. By the use of a variable flash gap in the forging process of a trial part, a maximum increase in the cavity filling of 4.6 mm (17.2 %) was achieved. Furthermore, the influences of different process properties, such as the billet temperature, the trigger force or the height of the moveable flash gap, were evaluated.
To allow the application of a variable flash gap to different forging parts, a guideline was designed. For that purpose, forging parts were classified into characteristic zones. Within the zones the influence of differently designed flash gaps were evaluated using FEA. The result of this evaluation is a recommendation on possible designs of a variable flash gap for differ-ent forging parts, to allow an improvement of the cavity filling.
bulk forming, finite element analysis (FEA), flash gap, material flow, die design
