Cutting surface roughness prediction model for cutting carbon steel using premixed abrasive water jet

Precision machining using a premixed abrasive water jet is an important problem to solve. The establishment of predictive mathematical models, physical model analyses, experimental research, and experimental verification are relatively mature. However, there has been little research on the cutting surface roughness of premixed abrasive water jets, so comprehensive studies and theoretical systems are scarce. Consequently, in this study, Q235 carbon structural steel, which is widely used in the industry, was selected for preliminary research. First, based on dimensional analysis, four key influencing parameters on the cutting surface roughness (R_a) of the premixed abrasive water jet were determined—that is, the water jet working pressure (P), feed rate (v), target distance (h), and nozzle outlet diameter (d). A full-factor premixed abrasive water jet cutting experiment L₈₁(3⁴) was then designed and conducted for Q235 carbon structural steel based on the four key influencing parameters. Moreover, based on an analysis of polarity variance, significant relationships between the four key influencing parameters were obtained, with d > v > P > h, and the optimal parameter combinations were obtained—that is, P = 28 MPa, v = 50 mm/min, h = 4 mm, and d = 1.14 mm. Under this parameter combination, the R_a was at its smallest (0.760 µm). Finally, based on multiple nonlinear regression analysis, a predictive mathematical model for the cutting surface roughness of Q235 carbon structural steel was developed, with an arithmetic mean error of 3.811% and a standard deviation of 1.499%. The predictive model exhibited good reliability and predictive effect.