Entity

Time filter

Source Type

Hagen, Germany

Bortfeldt A.,University of Hagen | Homberger J.,Stuttgart University of Applied Sciences
Computers and Operations Research | Year: 2013

The Vehicle Routing and Loading Problem (VRLP) results by combining vehicle routing, possibly with time windows, and three-dimensional loading. Some packing constraints of high practical relevance, among them an unloading sequence constraint and a support constraint, are also part of the VRLP. Different formulations of the VRLP are considered and the issue is discussed under which circumstances routing and packing should be tackled as a combined task. A two-stage heuristic is presented following a packing first, routing second approach, i.e. the packing of goods and the routing of vehicles is done in two strictly separated stages. High quality results are achieved in short computation times for the 46 VRLP instances recently introduced by Moura and Oliveira. Moreover 120 new large benchmark instances including up to 1000 customers and 50,000 boxes are introduced and results for these instances are also reported. © 2012 Elsevier Ltd. Source


Bortfeldt A.,University of Hagen
Computers and Operations Research | Year: 2012

The capacitated vehicle routing problem with three-dimensional loading constraints combines capacitated vehicle routing and three-dimensional loading with additional packing constraints concerning, for example, unloading operations. An efficient hybrid algorithm including a tabu search algorithm for routing and a tree search algorithm for loading is introduced. Computational results are presented for all publicly available test instances. Most of the best solutions previously reported in literature have been improved while the computational effort is drastically reduced compared to other methods. © 2011 Elsevier Ltd. Source


Huba M.,Slovak University of Technology in Bratislava | Huba M.,University of Hagen
Journal of Process Control | Year: 2013

This paper considers PI controller tuning for the Integral Plus Dead Time (IPDT) plant subject to constraints on tolerable deviations from ideal shapes and guaranteeing minimal combined IAE (Integral of Absolute Error) measure composed of weighted IAE values of the setpoint and disturbance step responses. As the ideal responses at the plant output, monotonic (MO) transients are chosen. This implies one-pulse (1P) responses consisting of two monotonic intervals at the plant input. Optimal nominal tunings for the most important situations regarding the servo/regulation trade-off are determined by the newly developed Matlab/Simulink tool based on the performance portrait (PP) method. Generated results confirm Pareto-like limits of performance achievable under PI control and are used in classifying traditional tuning approaches into the setpoint and disturbance oriented ones. Principles of robust PP based tuning are briefly illustrated and conditions of invariance of the closed loop performance against the dead time uncertainty are formulated and checked by simulation.© 2013 Elsevier Ltd. All rights reserved. Source


Huba M.,Slovak University of Technology in Bratislava | Huba M.,University of Hagen
Journal of Process Control | Year: 2013

This paper analyses the optimal nominal tuning of a new modification of predictive disturbance observer (PDO) based filtered PI control (PDO FPI) applied to a first order plus dead time (FOPDT) plant with exactly known parameters. The impacts of applied filters on optimal controller tuning and on achievable closed loop performance are evaluated first of all. The limits of achievable performance are compared with those of traditional two degree of freedom (2DOF) PI control, with both controllers tuned by the multiple real dominant pole method. This comparison shows the potential of PDO FPI control to improve tracking and regulatory dynamics significantly, permitting the Pareto-like servo/regulator trade-off of 2DOF PI control to be removed. Two PDO FPI tuning approaches are proposed, allowing optimal filter degree and time constants to be evaluated. The first tuning scenario considers optimization of closed loop performance expressed in terms of the Integral of the Absolute Error (IAE) weighted alternatively by the relative total variance TV1 of the control signal. This is carried out by changing the filter order n under the constraint that a constant position of the dominant closed loop pole is maintained. This keeps the dynamics of the setpoint step responses almost unchanged. In the second tuning scenario the optimization is carried out under a constraint on constant speed of disturbance step responses. All the main results are then numerically checked for the integral first order plant with dead time by the performance portrait (PP) method. The analysis presented here shows that the new PDO FPI structure substantially enriches the spectrum of controllers applicable to simple control tasks. © 2013 Elsevier Ltd. Source


Patent
Imec and University of Hagen | Date: 2010-01-12

The present invention is related to a photovoltaic device, the device comprising a first layer of a first semiconductor material of a first conductivity type, a second layer of a second semiconductor material of the opposite conductivity type of the first layer, and a third layer of a third porous semiconductor material situated between the first layer and the second layer. The present invention also provides a method for producing the photovoltaic device.

Discover hidden collaborations