Abstract

Abstract
Wind power instability and inconsistency involve the reliability of renewable power energy, the safety of the transmission system, the electrical grid stability and the rapid developments of energy market. The study on wind power forecasting is quite important at this stage in order to facilitate maximum wind energy growth as well as better efficiency of electrical power systems. In this work, we propose a novel hybrid data driven model based on the concepts of deep learning-based convolutional-long short term memory (CLSTM), mutual information, evolutionary algorithm, neural architectural search procedure, and ensemble-based deep reinforcement learning (RL) strategies. We name this hybrid model as DOCREL. In the first step, the mutual information extracts the most effective characteristics from raw wind power time series datasets. Second, we develop an improved version of the evolutionary whale optimization algorithm in order to effectively optimize the architecture of the deep CLSTM models by performing the neural architectural search procedure. At the end, our proposed deep RL-based ensemble algorithm integrates the optimized deep learning models to achieve the lowest possible wind power forecasting errors for two wind power datasets. In comparison with fourteen state-of-the-art deep learning models, our proposed DOCREL algorithm represents an excellent performance seasonally for two different case studies.

Abstract
Robotics is a multidisciplinary subject that typically involves mechanics, electronics, and computer science concepts. For this reason, robotic projects are particularly well suited to the European Project Semester framework since they allow students with different backgrounds to contribute to the overall team objective in their specific knowledge areas. This chapter briefly presents illustrative examples of robotic projects that have been developed by teams of students participating in the European Project Semester at the School of Engineering of the Polytechnic Institute of Porto. It concludes by presenting and discussing student feedback, namely on the program and the projects developed.

Abstract
In this paper, a decision support application, for the air cargo planning and distribution, is proposed. The freight forwarding sector has been working to be assertive and efficient in responding to the market through an efficient approach to planning and allocation problems. The main goal is to minimize costs and improve performance. A real air cargo distribution problem for a freight forwarder was addressed. This project emerged from the need to efficiently plan and minimize costs for the distribution of thousands of m(3) (cubic meters) of air cargo, while considering the market restrictions, such as aircraft availability and transportation fees. Through the GRG algorithm adaptation to the real problem, it was possible to respond to the main goal of this paper. The development of an easy-to-use application ensures a quick response in the air distribution planning, focusing on cost reduction in transportation. With the application development it is possible to obtain real earnings with immediate effect.

Abstract

Abstract
-This work presents a study on how an immersive virtual environment's level of interaction and fidelity can affect the quality of experience (QOE) in a real estate context. Four versions of the virtual space were created with the level of interaction and the level of fidelity varying between them. The QoE dimensions considered in this work are user satisfaction, lighting quality, interior space quality, and interaction features. The sample comprises 28 participants, of which 21 are men and 7 are women, aged between 18 and 29 years. Results show that, overall, the level of fidelity is more relevant when the level of interaction is low, assuming the movement around the apartment is statistically higher in high-fidelity experiences.