Análise da confiabilidade e modelagem cinética de lagoas de estabilização no nordeste no Brasil

• Ano de defesa: 2020
• Autor(a) principal: Alves, Matheus Sales
• Orientador(a): Pereira, Erlon Lopes
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Palavras-chave em Português: Saneamento; Esgotos; Lagoas de estabilização; Monte Carlo, Método de; Effluent quality; Monte Carlo simulation
• Link de acesso: http://www.repositorio.ufc.br/handle/riufc/52083

The present work is based on monitoring data of 10 Waste Stabilization Pond systems in operation in the state of Rio Grande do Norte, northeastern Brazil. All systems treat predominantly domestic effluents and consist of a facultative pond followed by two maturation ponds. The parameters considered were Biochemical Oxygen Demand in its total fraction (BODt) and soluble (BODs), Chemical Oxygen Demand in its total fraction (CODt) and soluble (CODs), Total Suspended Solids (TSS), Total Ammonia (TAM), Thermotolerant Coliforms (TTC), Chlorophyll-a, Dissolved Oxygen (DO) and pH. The systems performance was evaluated based on compliance with different concentration limits, according to the level of restriction of the effluent discharge standards. Based on the data collected, low percentages of compliance with restrictive standards were observed, especially with respect to Thermotolerant Coliforms and Ammonia. The methodology developed by Niku et al. (1979) was used to propose an alternative approach to the design of Waste Stabilization Ponds based on the coefficient of reliability (COR) calculated for an acceptable measure of non-compliance. The COR was obtained from the coefficient of variation (CV) of effluent concentrations and the standardized normal variable associated with a 95% probability of not exceeding. The large dispersion in coliform counts resulted in the lowest COR values among the evaluated parameters. The mean effluent concentrations proved to be much higher than those required to comply with the quality standards set over 95% of the time. Seeking to provide an overview of the occurrence of failure and the inherent risks of the treatment process, Monte Carlo simulation was used as a tool for analyzing the percentage of exceedance of the standards required for the launch. The Kolmogorov - Smirnov (KS) adherence test followed by the Akaike Information Criterion (AIC) allowed the selection of the most representative Probability Distribution Function (PDF) of each sample set. Simulations with 3000 iterations allowed to construct synthetically the selected distributions and, thus, to calculate the percentages of non-attendance in relation to the launching standards of interest. Among the simulated quality parameters, BOD and COD presented the highest risk of failure, with percentages of exceedances above 80% in all ponds in the series. The percentages of failure of the Thermotolerant Coliforms and pH parameters were correlated with physical and operational variables, such as hydraulic retention time, depth and organic loading rates. The resulting simulations provide an overview of the occurrence of failures and the risks inherent in waste stabilization pond technology. The kinetics of organic matter removal in the stabilization ponds operating in the typical conditions of northeastern Brazil were also investigated. First order removal rates (k) for BOD and COD were achieved assuming optimal hydraulic patterns of complete mixed and plug flow. The k-values were correlated with design factors and geometric characteristics of the ponds. The increase in the organic loading rate has led to an increase in the first order removal rate. In contrast, the increased ratio of hydraulic retention time (TRH) to depth (H) led to a reduction in removal rates. The regression analysis showed that the organic loading rate and the TRH/H ratio can be used as explanatory variables in the computation of kinetic coefficients more adequate to the local reality.