Stochastic approach to determination of suspended sediment concentration in tidal rivers by artificial neural network and genetic algorithm
Bibliographic record
Abstract
Because of the interaction between tidal and fluvial flows in tidal rivers, sampling and measurement of suspended sediment concentration is very complex. Determination of suspended sediment concentration in tidal rivers is a very important problem in some countries such as Canada and United Kingdom (UK) (for example Bay of Fundy in Canada and Bristol Channel in UK). A numerical model cannot show suspended sediment concentration in tidal river accurately. Fluvial flows bring sand and gravel particles from the watershed, while tidal flow brings silt particles from the sea in flood time and returns them to the sea in ebb time. Interaction between tidal and fluvial flows, relation between suspended sediment concentration and return periods of them, correction of suspended sediment distribution coefficient for use in tidal limit of rivers, finding the best method for determination of suspended sediment concentration in tidal limit of rivers and optimization of it are major difficulties and challenges for determination of suspended sediment concentration. For overcoming these challenges in this research, a perceptron artificial neural network is trained and validated by observed data. For training of the artificial neural network (ANN), Levenberg–Marquardt training method is applied. For decreasing of the mean square error (MSE) and increasing of efficiency coefficient, parameters of ANN are optimized by genetic algorithm (GA) method. The GA method optimizes the number of nodes of hidden layers of ANN that is trained by Levenberg–Marquardt training method. Two sets of data are introduced into a network. Inputs of first network are distance from upstream of river, flood return period, and tide return period. These return periods are determined by observed data and governing stochastic distribution on them. Inputs of second network are distance from upstream of river, flood discharge, and ebb height. Output of these networks is suspended sediment concentration. Observed data show that maximum suspended sediment concentration is concerned with ebb that tidal flow and fluvial flow are in one direction. Because of a shortage of observed data especially in extreme conditions, a numerical model was developed. This model was calibrated by observed data. Results of numerical model convert to two regression relations. These relations are functions of distance from the upstream of river, discharge of flood (or flood return period) at upstream, and ebb height (or ebb return period) at downstream. Then the artificial neural network is tested with the remainder of observed data and results of the numerical model. Sensitive analysis shows that distance from the upstream of river and flood discharge are the most effective governing factors on suspended sediment concentration in first and second network, respectively. For the case study, the Karun River in south west of Iran is considered. This river is the most important tidal river in Iran.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.002 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".