A platform was built up to collect the pressure data near the nozzle. The instantaneous flow data was calculated based on the relationship between pressure and flow by using Kalman filtering algorithm, and the average spray flow rate was obtained by integrating over the instantaneous flow, then an automatic piecewise linear fitting method was designed after considering the relationship between the duty cycle of PWM signal and the average spray flow rate. In order to verify the control effects of the spray flow rate model, the PWM spray flow rate models based on Kalman filtering were established under the spray pressure of 0.2, 0.3 and 0.4MPa, respectively, and the experiments of spray flow rate control accuracy were performed under the same conditions. The results showed that the coefficient of determination (R2) in the models above were all more than 0.995 under the three spray pressures, and the spray flow rate control error was within ±6%. This method provides a reference of building spraying control model for rapid and online use.