Estimation of river flow by using the artificial intelligence and trend analysis methods

In this study, artificial intelligence and statistical methods were used to develop estimation models for the determination of current data which could not be measured by current observation stations. Artificial intelligence methods; they are generally developed in programming languages and produce data by means of learning algorithms of computer systems. Statistical methods; It is expressed as methods based on mathematical expressions and mathematical modeling. 10 flow monitoring stations in the Porsuk basin, which is the lower basin of Sakarya basin and rainfall observation station data in Eskişehir province were used. Estimation models have been developed with the obtained data. These models; The forward feed back propagation method of artificial neural networks, which is an artificial intelligence method, is formed by the gene expression programming method of another artificial intelligence method and the multiple linear regression method which is a parametric trend analysis method. Forward feed back propagation network; is an artificial neural network method. The results obtained from the models prepared with different values and different educational inputs; coefficients of determination, mean square error and square root of mean square error were compared. When the results of the study were examined, it was seen that feed back propagation neural network produced the best data prediction models. Gene expression programming method failed to generate prediction data more than feed back feed propagation method. The most unsuccessful results were obtained by multiple linear regression method. The most unsuccessful results were obtained by multiple linear regression method. As a result; forward-feed back propagation neural network; It was able to successfully estimate the current values that could not be measured or wanted to be estimated and showed that it could be used in hydraulic works.