http://repositorio.ufc.br/handle/riufc/497 2026-08-02T12:02:20Z 2026-08-02T12:02:20Z Sousa, Wesley Maia de http://repositorio.ufc.br/handle/riufc/87330 2026-07-31T14:03:50Z 2026-01-01T00:00:00Z

Título: Desempenho operacional e custos em obras de redes coletoras de esgoto: uma abordagem com aprendizado de máquina e dados reais em Fortaleza-CE Autor(es): Sousa, Wesley Maia de Abstract: Sewer network construction projects are characterized by high operational variability, strong dependence on local implementation conditions, and recurrent challenges related to physical and financial planning. In this context, this dissertation investigated the operational performance and implementation costs of sewer collector networks and household connections based on real field data collected from a large-scale project carried out in Fortaleza, Ceará, Brazil, between October 2022 and August 2025. The analyzed database comprised approximately 160 km of sewer network, 21,000 household connections, and 35 months of continuous monitoring of productive, geometric, operational, climatic, and economic variables. The research was structured into two complementary pillars. The first evaluated the physical performance of construction crews through machine learning models applied to operational productivity analysis. The second investigated operational costs associated with sewer network implementation, considering labor, equipment, fuel consumption, materials, and execution constraints. The study employed Linear Regression, Ridge Regression, Random Forest, Gradient Boosting, and Support Vector Regression algorithms, combined with a temporal validation protocol based on Nested Cross-Validation and TimeSeriesSplit, aiming to preserve chronological causality and reduce optimistic evaluation bias. The results revealed a significant discrepancy between ex-post performance and forecast-safe prospective performance, highlighting the limitations of predictive generalization in real-world urban construction scenarios. Although high goodness-of-fit metrics were obtained during historical adjustment, the models exhibited substantial degradation under external temporal validation, indicating that operational and construction-related variables present highly unstable dynamics over time. Nevertheless, the modeling process enabled the identification of relevant drivers associated with productivity and costs, including excavation depth, pavement type, groundwater lowering, seasonality, and crew composition. Based on these findings, an operational simulation prototype was developed to support comparative scenario analysis and planning decision-making in sanitation projects, using empirically calibrated parameters derived from the observed historical database. The research contributes to advancing machine learning applications in sanitary engineering by simultaneously demonstrating both the explanatory potential of these techniques and the limitations of their predictive application in real and temporally dynamic operational environments. Tipo: Dissertação

2026-01-01T00:00:00Z Maurício, Ricardo Alves http://repositorio.ufc.br/handle/riufc/87299 2026-07-31T13:58:01Z 2026-01-01T00:00:00Z

Título: Estudo de campo e acoplamento do Cormix2 ao Delft3D para simulação da descarga de efluentes sanitários no rio Guamá, Belém-PA Autor(es): Maurício, Ricardo Alves Abstract: Rapid population growth in several Brazilian cities has occurred along the banks of their main water bodies and has often not been accompanied by adequate sanitation infrastructure, resulting in the indiscriminate discharge of domestic and industrial effluents and cousing environmental pollution. Due to the increasing need to implement new Wastewater Treatment Plants (WWTPs), the use of computational modeling to investigate efluent dilution in receiving water bodies has become an essential management tool for evaluating such projects. Therefore, the study aimed to implement and evaluate the offline coupling between the near-field model CORMIX2 and the far-field model Delft3D-WAQ to simulated treated effluent from the Mártir WWTP discharged through a multiport diffuser into the Guamá River, in Belém, Pará, Brazil. To support the computational simulations, bathymetric surveys, current profile measurements, and water quality assessments were conducted along the river reach considered to be the area influenced by the influence of the effluent discharge. The plumes simulated with CORMIX2 were georeferenced in Civil 3D according to the hydrodynamic fields generated by Delft3D- FLOW. The coupling results showed that, in well-mixed estuaries characterized by low effluent discharge rates and ambient thermotolerant coliform concentrations substantially higher than the effluent concentrations at the end of the Regulatory Mixing Zone (RMZ), the far-field dilution estimates generated by CORMIX2 were consistent with those calculated by Delft3D- WAQ. Under these conditions, CORMIX2 alone was sufficient to estimate concentrations in both the near and far fields. Furthermore, the designed outfall met the applicable regulatory requirements, at the estimated concentrations at the end of the RMZ remained below the limits established by Brazilian environmental legislation. Negatively buoyant discharges were also found to represent the critical condition for near-field dilution. Tipo: Dissertação

2026-01-01T00:00:00Z Braga, Lucas Silva http://repositorio.ufc.br/handle/riufc/87286 2026-07-28T15:22:51Z 2026-01-01T00:00:00Z

Título: Estudo da influência da condição de Courant na análise de vazamento em redes de distribuição de água por meio do método transiente inverso e do algoritmo genético Autor(es): Braga, Lucas Silva Abstract: One of the major problems affecting water distribution systems is the occurrence of leaks. In fact, scientists and researchers continuously seek ways to prevent this problem in the most cost-effective manner possible. With technological advancements, computational techniques have become increasingly suitable for the rapid and economical detection of leaks. This study aims to apply the Inverse Transient Method in conjunction with the Genetic Algorithm Optimization Method to identify leak locations in two selected water distribution networks, while analyzing the influence of the Courant condition on the stability of the calculations. To achieve this objective, specialized software tools were employed to evaluate the effects of different Courant condition cases and to verify whether their stability criteria remain valid for the analyzed situations. Initially, data for both steady-state and transient flow conditions were obtained. Subsequently, data simulated by the Genetic Algorithm were generated for the identification of leak locations within the networks. Based on these results, a comparison was performed among the different flow rate and hydraulic head values obtained under the various Courant condition cases. Tipo: Dissertação

2026-01-01T00:00:00Z Silva, Ana Carolina Vasconcelos http://repositorio.ufc.br/handle/riufc/87283 2026-07-28T18:53:53Z 2026-01-01T00:00:00Z

Título: Aplicação do FMEA na avaliação de risco de piping em fundações sedimentares: estudo de caso da Barragem Jaburu i (CE) Autor(es): Silva, Ana Carolina Vasconcelos Abstract: The safety of earth dams on sedimentary foundations represents a critical geotechnical challenge in the Brazilian semiarid region, given the susceptibility of friable sandstones and siltstones to internal erosion (piping) and the history of accidents and incidents within this geotechnic-geological context. This master’s dissertation proposes an adaptation of the Failure Mode and Effects Analysis (FMEA) methodology for risk assessment in such structures, using the Jaburu I Dam (CE) as a case study. The research is justified by the progressive and rapid nature of erosive processes which, in sedimentar media, can evolve into failure scenarios regardless of reservoir level fluctuations or external actions. The developed methodology is based on the Baldi Method, featuring Severity (S), Occurrence (O), and Detection (D) scales calibrated to the regional geotechnical contexto, incorporating instrumentation control levels (Normal, Attention, and Alert) into the detection criteria. Based on Taveira et al. (2025), which describes the Potential Failure Mode Analysis (PFMA) conducted by COGERH in 2024, and updated data from the 2025 Regular Inspection, 11 erosive failure modes were selected and analyzed. The technical evidence examined included the history of anomalies, such as sediment transport in flow meter MV-04, past interventions, visual inspections, and instrumentation data. The results showed that risk classification is strongly influenced by detection capacity (D). Failure modes in regions with adequate instrumentation, such as the foundation and left abutment (MF1, MF2, MF3 and MF7), presented tolerable risk, while those in regions with insuficiente monitoring, such as the right abutment, spillway, and intake (MF8, MF10, MF11, MF5, and MF6) reached intolerable risk levels. The risk matrix (S x O) positioned all analyzed failure modes as severe failures. It is concluded that the adapted FMEA is na effective tool for the strategic priorization of risks, allowing for the hierarchy of interventions in dams on sedimentay foundations based on objective and update evidence. Tipo: Dissertação

2026-01-01T00:00:00Z