INESCC-DL - Vários
Permanent URI for this collection
Browse
Recent Submissions
- Network reconfiguration to improve reliability and efficiency in distribution systemsPublication . Vitorino, Romeu; Pires Neves, Luís; Jorge, H. M.This paper presents a new method to improve reliability and also minimize active power losses in radial distribution systems (RDS) through a process of network reconfiguration. The methodology adopted to enhance reliability, uses the Monte Carlo (MC) simulation and an historical data of the network such as the severity of the potential contingencies in each branch. Due to the greater number of possible configurations and the need of an efficient search, is also used an improved genetic algorithm (IGA), with adaptive crossover and mutation probabilities and with other new features. The method analyses the RDS in a perspective of optimization considering no investment, and a perspective of optimization where is given the possibility to place a limited number of tie-switches, defined by a decision agent, in certain branches. The effectiveness of the proposed method is demonstrated through the analysis of a 69 bus RDS.
- Load forecasting based on neural networks and load profilingPublication . Sousa, João; Pires Neves, Luís; H. M. JorgeThis work presents a novel perspective of load forecasting based on neural networks and load profiling. In addition to the variables that are typically used to predict future load demand, such as past load values, meteorological variables, seasonal effects or macroeconomic indexes, it is expected that the use of load profiles and detailed information of individual consumers could favor the forecasting process. The methodology can be extended to different temporal horizons being predicted and the eventual threat of overparametrization is attenuated by the use of neural networks since the complexity of the model does not necessarily depends on the number of its weights and biases, as some of these parameters might be found irrelevant in the process. Another way to reduce the risk of overparametrization and overfitting is through the use of a considerable number of data points (whenever historical data is available) to train the network.
- Impact of load and generation price uncertainties in spot pricesPublication . Gomes, Bruno A.; Saraiva, João T.; Pires Neves, LuísIn this paper it is presented a formulation for the DC Optimal Power Flow problem considering load and generation cost uncertainties and the corresponding solution algorithms. The paper also details the algorithms implemented to allow the integration of losses on the results as well the algorithm developed to compute the nodal marginal price in the presence of such uncertainties. Since loads and generation costs are represented by fuzzy numbers, nodal marginal prices are no longer represented by deterministic values, but instead, by membership functions. To illustrate the application of the proposed algorithms, this paper also includes results based on a small 3 bus system and on the IEEE 24 bus/38 branch test system.
- Network reconfiguration using a genetic approach for loss and reliability optimization in distribution systemsPublication . Vitorino, Romeu; Jorge, Humberto M. M.; Pires Neves, LuísThis paper presents a new method to improve reliability and also minimize losses in radial distribution systems (RDS), trough a process of network reconfiguration, using a genetic algorithm approach. The methodology adopted to enhance reliability, uses the Monte Carlo simulation and an historical data of the network such as the level of reliability and the severity of potential contingencies in each branch. The method analyses the RDS in two perspectives. A first perspective of optimization considering no investment, therefore using only the switches presented in the network, and a second perspective of optimization where is given the possibility to place a limited number of tie-switches and thus get better results. Here, the number of tie-switches and the branches that can receive them are defined by a decision agent. The effectiveness of the proposed method is demonstrated through the analysis of a 69 bus RDS.
- Short-term load forecasting using information obtained from low voltage load profilesPublication . Sousa, João; Pires Neves, Luís; Humberto M.M. JorgeRecent researches in load forecasting are quite often based on the use of neural networks in order to predict a specific variable (maximum demand, active electric power or hourly consumption) using past values of the same variable and other exogenous factors proved to influence the value being predicted. This work aims to explore different input patterns in neural networks incorporating information derived from load profiles of different consumers' classes.