Neural-Fuzzy system applied for determine the harmonic contribution on power distribution feeders

• Grant number: 15/12510-9
• Support Opportunities: Scholarships in Brazil - Scientific Initiation
• Start date: September 01, 2015
• End date: August 31, 2017
• Field of knowledge: Engineering - Electrical Engineering - Power Systems
• Principal Investigator: Ricardo Augusto Souza Fernandes
• Grantee: Israel Wilson Agostinho
• Host Institution: Centro de Ciências Exatas e de Tecnologia (CCET). Universidade Federal de São Carlos (UFSCAR). São Carlos , SP, Brazil

Abstract

This research project has as main objective the development of a method for determining the percentage of responsibility of both the consumer and the distribution utility on harmonic distortion measured at Points of Common Coupling (PCC). It is worth mentioning that the proposed method will use a capacitor bank to be allocated at the PCC, generating interference in the electrical signals and changing the power flow. So, after the measurements of voltage and current at the PCC, the signals will be pre-processed, i.e., some calculations will be performed, such as RMS values of voltages and currents, telephone interference factor, crest factor, form factor, harmonics of voltage and current, total harmonic distortion of current and voltage, skewness, kurtosis, Rényi entropy, Shannon entropy, among other. Based on this pre-processing step, it can be obtained a set of scalar values, which are considered as attributes for the generated database. This way, these attributes will be analyzed by means of correlation in order to be better to select the most relevant attributes. Consequently, the Neural-Fuzzy System will be automatically adjusted to estimate the harmonic contributions of both the utility and consumer connected at the PCC. Finally, it is important to comment that the proposed methodology will be evaluated tested in relation to the PRODIST standard.