This work presents a methodology that combines the use of artificial neural networks and fuzzy logic for alarm processing and identification of faulted components in electrical power systems. Fuzzy relations are established and form a database employed to train artificial neural networks. The artificial neural networks inputs are alarm patterns, while each output neuron is responsible for estimating the degree of membership of a specific system component into the class of faulted components. The proposed method allows good interpretation of the results, even in the presence of difficult corrupted alarm patterns. Tests are performed with a test system and with part of a real Brazilian system.Index Terms-Alarm processing, fuzzy logic, neural networks, pattern recognition, power system protection.

No abstract

Goal: to propose a hybrid method, combining AHP and Fuzzy, for project portfolio management. Originality/value: this paper meets the positive characteristics of both methods, adequately weighing the criteria and contemplating process subjectivity. Limitations that exist in both methods, such as the maximum number of alternatives and the difficulty of inserting new alternatives at the end of the process, are overcome. Design/methodology/approach: the AHP is applied for determining the criteria weights and fuzzy is used to compare the alternatives for each criterion. Results: the results show that the proposed hybrid method allows the ranking of many alternatives and provides higher reliability for decision makers. It must be noted that the system is fed by performance indicators, which minimize the subjectivity in decision-making—an important characteristic in the technology management. Moreover, the results provide the possibility of changing the number of projects at any time without influencing the outcome. Practical implications: the proposed hybrid method can be used in different problems that have many alternatives or subjectivity.

This work investigates the construction of fuzzy relations for alarm processing and fault location in electrical power systems. Several data aggregation classes are tested and compared. Fuzzy relations are established with the aid of the knowledge on protection devices operation for faults involving different system components. Tests are performed with a 7-bus test system and with part of a real brazilian system.

Network parameter errors may come from many different sources, such as: imprecise data provided by manufacturers, poor estimation of transmission line lengths, changes in the transmission network design which are not adequately updated in the corresponding database, etc. Network parameter data are used by almost all power system analysis tools, from real time monitoring to long term planning. Parameter errors may contaminate the obtained results and compromise decision making processes. This work proposes a methodology that combines genetic algorithms and power system state estimation to correct single or multiple network parameter errors. Simulations with the IEEE 14-bus test system are performed to illustrate the proposed method.