Intelligent system for diagnosis and proactivity in electrical distribution systems

• Grant number: 11/06394-5
• Support Opportunities: Regular Research Grants
• Start date: August 01, 2011
• End date: July 31, 2013
• Field of knowledge: Engineering - Electrical Engineering - Power Systems
• Principal Investigator: Carlos Roberto Minussi
• Grantee: Carlos Roberto Minussi
• Host Institution: Faculdade de Engenharia (FEIS). Universidade Estadual Paulista (UNESP). Campus de Ilha Solteira. Ilha Solteira , SP, Brazil
• City of the host institution: Ilha Solteira
• Associated researchers: Anna Diva Plasencia Lotufo ; Mara Lúcia Martins Lopes

Abstract

This research intends to develop an intelligent computational system to analyze the abnormalities, as short circuits and voltage disturbances. The objective of the system is to identify and localize possible sources of such operation anomalies, proposing proactive solutions (i.e., acting previously within the concepts of predictive maintenance). The computational system is based on a neuro fuzzy architecture combined with the Dempster-Shafer evidence theory. The neural network used is an ART (Adaptive Resonance Theory) descendent which is stable and plastic, and presents an excellent performance in precision, velocity in answering, and especially providing the knowledge extraction of such events continuously. However, the difficulty in using the ART neural networks concerns to manipulate the input and output information. The emphasis of the project is to develop preventive strategies, investigating and indentifying probable risks to the EPS security that can lead to interruption or perturbations in providing energy to the users. Considering the centralized information of the acquisition data system and sensors, one can detect the faults even so in naive stage, providing the proactivity of the real operational problems that can cause damages to attending the EPS loads, moreover to reduce electrical costs in components. This integrated system must detect eventual operational problems, recognize several faults, and evaluate considering the risks that can compromise the EPS and show preventive actions to the operators. (AU)

Scientific publications

(References retrieved automatically from Web of Science and SciELO through information on FAPESP grants and their corresponding numbers as mentioned in the publications by the authors)

LIMA, FERNANDO P. A.; LOTUFO, ANNA DIVA P.; MINUSSI, CARLOS ROBERTO. Wavelet-artificial immune system algorithm applied to voltage disturbance diagnosis in electrical distribution systems. IET GENERATION TRANSMISSION & DISTRIBUTION, v. 9, n. 11, p. 1104-1111, AUG 6 2015. (11/06394-5)

LIMA, FERNANDO P. A.; LOTUFO, ANNA D. P.; MINUSSI, CARLOS R.. Disturbance detection for optimal database storage in electrical distribution systems using artificial immune systems with negative selection. Electric Power Systems Research, v. 109, p. 54-62, APR 2014. (11/06394-5)