Medição de vazão em rios utilizando o método de Chiu e velocimetria por imagem de partículas
River flow results from the complex interaction between hydrological processes and hydrodynamic phenomena. The uncertainty associated with this flow presents a central challenge in the fields of hydraulics and hydrology, particularly regarding discharge estimation and the characterization of the spa...
| Autor: | |
|---|---|
| Formato: | tesis doctoral |
| Estado: | Versión publicada |
| Fecha de publicación: | 2025 |
| País: | Brasil |
| Recursos: | Universidade Federal do Ceará (UFC) |
| Repositorio: | Repositório Institucional da Universidade Federal do Ceará (UFC) |
| Idioma: | portugués |
| OAI Identifier: | oai:repositorio.ufc.br:riufc/82647 |
| Acesso em linha: | http://repositorio.ufc.br/handle/riufc/82647 |
| Access Level: | acceso abierto |
| Palavra-chave: | Hidráulica Hidrometria Canais abertos Velocidade de fluxo Monitoramento em rios Hydraulics Hydrometry Open channels Flow velocity River monitoring CNPQ::CIENCIAS AGRARIAS::ENGENHARIA AGRICOLA |
| Resumo: | River flow results from the complex interaction between hydrological processes and hydrodynamic phenomena. The uncertainty associated with this flow presents a central challenge in the fields of hydraulics and hydrology, particularly regarding discharge estimation and the characterization of the spatiotemporal distribution of water velocity. The Chiu method, based on a probabilistic approach, has been employed to reduce these uncertainties by enabling discharge estimation from surface velocity data.This research aims to estimate streamflow in rivers of the Brazilian semi-arid region through the combined use of the Chiu method and Large-Scale Particle Image Velocimetry (LSPIV). The study was divided into two stages: (I) analysis of the variability of Chiu’s entropy coefficient across three spatial scales (cross-section, river reach, and watershed); and (II) evaluation of the effectiveness of combining the Chiu method with LSPIV for streamflow estimation.In the first stage, hydrometric data from COGERH (2017–2022) were used, encompassing 667 discharge measurements across twelve cross-sections of the Acaraú River. The estimation of the entropy coefficient ϕ(M) and maximum velocity was performed using root mean square error (RMSE) minimization. Statistical analysis included descriptive statistics, ANOVA, confidence intervals, and cluster analysis. Calibration and validation of ϕ(M) were carried out using the Nash-Sutcliffe Efficiency (NSE) coefficient. Results showed that ϕ(M) had a global mean of 0.69 with a coefficient of variation (CV) of 12%, indicating relative stability across scales, whereas the M parameter had a global mean of 2.85 and a CV of 60%, indicating higher variability. The model performed well at all spatial scales, with NSE values above 0.6. The best performance occurred at the watershed scale, with NSE values of 0.95 for calibration and 0.92 for validation.In the second stage, the efficiency of combining the Chiu method with LSPIV for discharge estimation was evaluated. Three approaches were tested: (i) reference method using ADV and floaters; (ii) LSPIV combined with the Chiu method using a calibrated ϕ coefficient; and (iii) LSPIV with the Chiu method and elliptical distribution refinement. Measurements were conducted at six cross-sections of the Acaraú River. Method (ii) yielded the best results, with an average absolute error of 20.7%, outperforming method (iii), which had an error of 23.27%. The accuracy of the estimates was influenced by factors such as ϕ coefficient calibration, image quality, camera positioning, and the number of ground control points (GCPs). The study concluded that, despite operational limitations, integrating LSPIV with the Chiu method is a promising, low-cost, and non-intrusive solution for monitoring discharge in small to medium-sized rivers, especially in semi-arid regions. |
|---|