Parameter Optimization of Spiral Fertilizer Applicator Based on Artificial Neural Network

To determine the optimal fertilizer discharging performance, a spiral applicator was designed according to orchard agricultural requirements. The influence of different parameter combinations speed, blade diameter, and pitch on coefficient variation (CV) discharge uniformity predicted using neural-network-based model by Box–Behnken design (BBD) test. According extracted results, neural network has good prediction ability, with determination mean relative error reaching 0.99 2.29%, respectively. impact performances examined from both macroscopic microscopic perspectives. During process, openness formed between blades outlet presented periodic changes continuous rotation blade, thus resulting in uneven particles. In addition, there are interacting force chains among particles, which not broken time during procedure, discharge. With comprehensive consideration efficiency, effect, CV uniformity, combination after optimization as follows: rotating speed 47.6 rpm, diameter 90 mm, 60 19.05%. Under this combination, effect efficiency were considered be relatively good. Our work provides references for combination.