Abstract: Traditional water quality evaluation methods are greatly affected by subjective factors. In order to further understand the water quality of the Huangshui mainstream in the upper reaches of the Yellow River, the interval type Bayesian model was used to evaluate water quality based on the monitoring data of BOD₅, COD, NH₃-N, TP and Cr⁶⁺ of 12 sections in the Huangshui mainstream from January to December 2018. The results show that the interval type Bayesian model can directly and accurately reflect the water quality change of monitoring sections, which basically accords with the trend in water quality of “good water quality-deteriorated water quality-improved water quality” from the upstream to the downstream of the Huangshui River. The sections with poor water quality are mainly located in the urban concentrated reaches with concentrated population and developed industries, and the excessive pollution should be the main reason for the deterioration of water quality. In the future, these reaches should be selected as the priority areas for water pollution control and water quality monitoring in the Huangshui River.