Abstract: Wheat blending is one of the widely used technical means in flour production, especially special flour production, which is closely related to product quality, processing cost, and economic efficiency, etc. A systematic study of the changes in the physicochemical characteristics of flour milling from different quality types of wheat varieties, and clarification of their matching effects can more accurately predict the quality traits of matching flour, and provide scientific guidance for the production and processing of special wheat flour. In this study, Xinong 979, Weilong 169, Xinong 511, and Xiaoyan 22 were used as raw materials, and combined in two-by-two combinations according to the mass ratios of 0∶100, 20∶80, 40∶60, 60∶40, 80∶20, and 100∶0 to prepare wheat flour. The protein properties, starch pasting properties, and dough rheological properties of the flours were determined. Subsequently, the physicochemical properties and the wheat-matching effect of different wheat varieties were analyzed by regression curve fitting. In order to clarify the variation pattern and the effect of wheat blending, a regression curve fitting was performed, “ monomial” or “ binomial” equation was adopted to reveal the variation of flour quality characteristics with the proportion of wheat blending combinations. The results showed that the protein content, wet gluten content, pasting temperature, stability time, and maximum tensile resistance of the flour exhibited a “ monomial” variation, showing an average effect, while the peak viscosity exhibited a “ binomial” variation, showing a negative effect, when Weilong 169, Xinong 979 and Xinong 511 were blended with Xiaoyan 22, respectively. The protein content, wet gluten content, water absorption and extensibility of the flour exhibited a “ binomial” variation, when Xinong 979 and Weilong 169 were blended with Xinnong 511, respectively. The protein content, wet gluten content, extensibility of the flour exhibited a “ binomial” variation, showing a negative effect, when Xinong 979 were blended with Weilong 169.The physicochemical properties of wheat flour can be predicted effectively by regression curve fitting, which can provide a reference for scientific wheat blending process design and special flour production.