Optimization of Spraying Parameters and Efficacy Evaluation for Maize Disease and Pest Control Using Plant Protection UAVs

To optimize the spraying parameters of plant protection unmanned aerial vehicles (UAVs) for controlling pests and diseases in the mid-to-late growth stages of maize and to enhance droplet deposition uniformity and field control efficacy, this study systematically evaluated the effects of different flight speeds, operating altitudes, and application rates on droplet deposition patterns, as well as on the control efficacy against spider mites, corn borers, and ear rot by a multi-factor orthogonal experiment. Based on field deposition data, three models for predicting spraying quality were constructed and compared: Support Vector Machine Regression (SVR), Random Forest Regression (RFR), and a Back-Propagation Neural Network (MLP). The Differential Evolution (DE) algorithm was employed for parameter optimization. The results demonstrated that the MLP model exhibited the best performance in predicting spraying quality (R2 = 0.860). Through model-based optimization and field efficacy assessment, the optimal operational parameter combination was determined as follows: flight speed of 7 m/s, operating altitude of 4 m, and application rate of 2.5 L per hectare. Compared to the parameters based on farmer experience, the optimal parameter combination improved the relative control efficacy against spider mites, Asian corn borer, and ear rot by approximately 10.2%, 18.4%, and 28.9%, respectively. The findings of this study provide direct technical support for UAV-based pest and disease control during the mid-to-late growth stages of maize in Jilin Province, and the established methodological framework can serve as a reference for parameter optimization in other regions.