Abstract
This paper deals with a very common problem in the home-textile industry. Given a set of orders of small rectangles of fabric the problem consists of determining the lengths and widths of a set of large rectangles of fabric to be produced and the corresponding cutting patterns. The objective is to minimize the total quantity of fabric necessary to satisfy all orders. The approach proposed uses a biased random-key genetic algorithm for generating sets of cutting patterns which are the input to a sequential heuristic procedure which generates a solution. Experimental tests based on a set of 100 random generated problems with known optimal solution validate quality of the approach.

Abstract

Abstract

Abstract
The development of wireless sensor networks for control and monitoring functions has created a vibrant investigation scenario, where many critical topics, such as communication efficiency and energy consumption, have been investigated in the past few years. However, when sensors are endowed with low-power cameras for visual monitoring, a new scope of challenges is raised, demanding new research efforts. In this context, the resource-constrained nature of sensor nodes has demanded the use of prioritization approaches as a practical mechanism to lower the transmission burden of visual data over wireless sensor networks. Many works in recent years have considered local-level prioritization parameters to enhance the overall performance of those networks, but global-level policies can potentially achieve better results in terms of visual monitoring efficiency. In this paper, we make a broad review of some recent works on priority-based optimizations in wireless visual sensor networks. Moreover, we envisage some research trends when exploiting prioritization, potentially fostering the development of promising optimizations for wireless sensor networks composed of visual sensors.

Abstract
This study presents a methodology for siting and sizing static synchronous compensator (STATCOM) devices in the Portuguese transmission system in order to improve system security following severe grid faults. Security issues arise since the Portuguese transmission system incorporates significant levels of wind generation without fault ride through and reactive current injection capabilities during grid faults. As the transmission system operator (TSO) is responsible for assuring system security, the goal of the study is to take advantage of the proved STATCOM ability for injecting reactive current in order to mitigate the disconnection of large amounts of wind farms in case of severe grid faults. The proposed methodology was developed and tested in coordination with the Portuguese TSO and it is based on the formulation of an optimisation problem in order to minimise the installed STATCOM power while ensuring its compliance with the current grid code requirements, namely in what concerns to the system stability and security. Given the discrete and complex nature of the problem, a hybrid approach, combining both a heuristic method and an evolutionary particle swarm optimisation (EPSO) algorithm was developed. Results show the effectiveness of the proposed methodology as well as its robustness regarding the validity of the obtained solutions while facing the most severe operational scenarios.

Abstract
In open-ended domains, robots must continuously learn new object categories. When the training sets are created offline, it is not possible to ensure their representativeness with respect to the object categories and features the system will find when operating online. In the Bag of Words model, visual codebooks are usually constructed from training sets created offline. This might lead to non-discriminative visual words and, as a consequence, to poor recognition performance. This paper proposes a visual object recognition system which concurrently learns in an incremental and online fashion both the visual object category representations as well as the codebook words used to encode them. The codebook is defined using Gaussian Mixture Models which are updated using new object views. The approach contains similarities with the human visual object recognition system: evidence suggests that the development of recognition capabilities occurs on multiple levels and is sustained over large periods of time. Results show that the proposed system with concurrent learning of object categories and codebooks is capable of learning more categories, requiring less examples, and with similar accuracies, when compared to the classical Bag of Words approach using codebooks constructed offline.