Modeling is one of the most effective methods in river water quality management, planning to reduce the entry of pollutants and control environmental damage. In this article, the water quality of Babolrood river using the Qual2kw model and meta-heuristic model of support vector machine and SVM-GA genetic algorithm with the approach of simulating six water quality parameters including temperature, alkalinity, electrical conductivity, total nitrogen, dissolved oxygen and biochemical oxygen demand in four time periods from 2018 was investigated. Genetic algorithm is used for simulation and optimization, and the hybrid of this model with support vector machine has improved the simulation performance. In order to compare the observed and simulated data, the explanation was used coefficient R2 of the average absolute error MAE and the root mean square of the normalized error NRMSE. The results showed that the best SVM-GA simulation for total nitrogen parameter in February is with MAE 0.15 and NRMSE 1.97. Also, the results of Qual2kw numerical model showed that the best simulation for May is with MAE 0.14 and NRMSE 1.2. The water quality of Babolrood River was evaluated using the NSFWQI index. Based on this, the best water quality is in the month of Bahman with an index of 56 in the medium to good range and the worst quality is in the month of May with an index of 49 in the bad range.