Modelo multicritério de avaliação de risco de segurança de barragens baseado em lógica fuzzy

Abstract

Dams are hydraulic structures that contribute to the socioeconomic development of a region. However, the absence of adequate safety procedures increases the risks of failures with the potential to cause significant impacts to downstream areas. This work resulted in the developed a fuzzy methodology for safety risk assessment of water accumulation dams, framed within the National Dam Safety Policy (PNSB). Simplified modeling of dam breach floodplains was performed to identify features of interest within the flood surface boundary and estimate the socioeconomic and environmental impacts in the downstream valley. The Fuzzy Analytic Hierarchy Process (FAHP) was used to obtain the weights of indicators and the corresponding aggregated scores for vulnerability and potential impact criteria. In the fuzzy inference phase, the system was divided into two stages to obtain a more representative model. In the first stage, vulnerability was estimated using conditional "If-Then" statements to estimate the combined effect of technical characteristics, conservation status, and safety plans. Then, defuzzified vulnerability values were combined with potential damage to estimate the resulting potential risk. When comparing the results obtained by the model with those of the methodology for classifying dams by Risk Category (CRI), it was observed that the studied dams were categorized in equivalent classes. Regarding the proposed periodicity, the deadlines for the entrepreneur to carry out Regular Safety Inspections (ISR) were more rigorous than those determined based on Resolution nº. 236 from the ANA. This allowed for establishing priorities for safety procedures, such as monitoring and sensing, suitable for field conditions. Therefore, the proposed model allowed for incorporating uncertainties of the experts’ subjective judgments into the decision-making process and can be considered an improvement in normative assessment to estimate the risk of dams with similar characteristics.