Modeling based on computational intelligence in river suspended load estimation (Baqorabad Station of the Qamroud River)

Estimating the sediment load of rivers is one of the important and practical issues in the studies and design of water engineering projects, such as the design and development of irrigation and drainage networks, water extraction from rivers, etc. Sediment concentration can be calculated by direct or indirect methods, which are usually expensive and time-consuming direct methods. Various factors affect this phenomenon, which makes their analysis difficult. Statistical and regression models are the most common methods of analysis, which often provide erroneous results due to the linear solution of these phenomena. Therefore, they cannot model the sedimentation phenomenon with acceptable accuracy. Hydraulic models cannot always be trusted due to the need for a lot of data and sometimes the unavailability of the required data and the inaccuracy of the data due to human error for simulating sediments. Nowadays, intelligent fuzzy and neural conductor systems, due to their ability to solve complex and nonlinear phenomena, have found many applications in various water engineering problems, including sedimentation. The purpose of this research is to evaluate and compare adaptive neural fuzzy models (ANFIS), machine Support vector (SVM) and GEP gene expression programming in estimating the sediment load of selected rivers in Central Province. For this purpose, the performance of three types of support vector machine (SVM), adaptive neural system (ANFIS) and GEP gene expression programming in the simulation of sediment load of rivers was studied.

In this research, first, the long-term daily statistics of temperature, rainfall, average flow rate and sediment concentration of Hasan Abad hydrometric and sediment measuring station located on the main branch of the Tirah River were collected. Then, data sufficiency test for analysis, checking the correlation between parameters of river discharge, precipitation, temperature with sediment discharge, and determining the long-term average of suspended sediment in the studied stations were performed. In the next step, a suitable combination of input variables was selected. The design of input parameter pattern can be based on the relationship between flow and sediment flow parameters, rainfall, temperature, flow flow and sediment flow. Determining the most appropriate time delay of input parameters in flow and sediment modeling. Estimation of sediment discharge using support vector machine (Svm), gene expression programming (GEP) and fuzzy-adaptive neural system (Anfis), comparing three data mining methods with each other as well as with gauge curve and observational data. The appropriate design of the structure of soft calculation models is used, in this research the number of data required for training (usually more than 70 percent), research data as training and also determining the required data (usually between 20 and 30 percent) is used for validation and testing.

In this research, The best performance of the GEP model has been obtained for pattern number 13. In this model, the R2 explanatory coefficient and the RMSE error obtained from the model are 0.97 and 0.033, respectively. The coefficient of explanation R2 and the RMSE error of the models in predicting suspended sediment values in the test phase were obtained as 0.53, 3.18 for the ANFIS model, and 0.70, 15.16 for the SVR model, respectively. Comparing the results of ANFIS and GEP models with the SVR model indicates the superiority of the GEP model in predicting the amount of suspended sediment according to Verdi model number 13. According to the obtained results, it can be seen that the performance of GEP model was better compared to other models. SVR and ANFIS models are ranked second and third. According to the obtained results, it can be said that the GEP model as a powerful and high-speed model can be used to model the suspended sediment in the Qomroud catchment area of Bagherabad station.

The results show the acceptable performance of the models compared to the gauge curve. Also, the results showed the superiority of the GEP model with the highest coefficient of determination R2 with a value of 0.99 and the lowest root mean square error RMSE in terms of tons per day with a value of 0.010. In this regard, the efficiency of the SVR model was somewhat better than the ANFIS model. The obtained results showed that all three investigated data mining methods provide far better results than the sediment gauge curve.