In this research, polymeric ferric sulfate (PFS) was used as a coagulant to treat wastewater containing Cd(II), Pb(II), and Cu(II). Based on a large number of single-factor experiments, response surface methodology (RSM) with Box-Behnken design (BBD) was employed to investigate the effects of PFS dosage, pH value, and precipitation time on the removal efficiencies of these heavy metals, and a quadratic polynomial response surface model was established. The result for the analysis of variance (ANOVA) shows that the significant order of these factors was as follows: pH > PFS dosage > precipitation time. The optimum PFS dosage, pH value, and precipitation time were found to be 59.6 mg/L, 8.04, 45.4 min for coagulation experiment in the combined heavy metals wastewater with the concentration of Cd(II), Pb(II) and Cu(II) of 5, 50, and 25 mg/L, 50 times of the requirement of the "Integrated Wastewater Discharge Standard of China" (GB 8978-1996), respectively. The remaining concentrations of combined heavy metals in both pure water and real wastewater could all reach the "Integrated Wastewater Discharge Standard in China" (GB 8978-1996). Meanwhile, the zeta potential of the supernatant and the fractal dimension (D-f) of the flocs after precipitation under the optimal coagulation conditions were measured to illustrate the consistency of the removal efficiency of heavy metals with the trend of zeta potential and fractal dimension. The electric neutralization and purge coagulation play a relatively dominant role in the coagulation of heavy metals in aqueous solution.