Impact of Variable Device Structural Changes on Particle Deposition Distribution in Multi-Rotor UAV

Abstract

The aim of this study was to investigate the effect of structural changes in variable fertilizer application devices on the distribution of particle deposition in UAVs. With the rapid development of drone technology, particularly in particulate spreading, drones have demonstrated significant potential due to their efficiency and precision. This paper evaluates the impact of different variable adjustment modes of the device on particulate deposition distribution through drone spreading experiments and particulate deposition data analysis. In this study, device structure change is the main variable factor, and flight altitude, flight speed and ambient wind speed are single quantitative factors. Experiments were conducted by varying the structure of the device to test the detailed deposition distribution of the device under group a, b, and c structures. Experimental results indicate that by choosing different variable combinations, the spreading device can achieve various fertilizer deposition states to meet regional needs. Among all 27 variable groups, the fertilizer particle deposition data for group b1b2b3 is relatively uniform, with three-quarters of particulate deposition values being 3 g/m² and the maximum value being 4 g/m². However, even with a relatively uniform distribution of fertilizer particles, the coefficient of variation for group b1b2b3 remains high (36.5%), with a range of 4.5% to 41%. Under different group adjustments, the particle distribution shows the smallest variability range in group b1b2b3, with a range of 15.71–26.44% and a variability difference of 10.73%. The particle distribution shows the largest variability range in group a1a2b3, with a range of 0.78–35.06% and a variability difference of 34.28%. These research conclusions provide important guidance for the study and practice of drone spreading systems.

1. Introduction

In recent years, the rapid development of drone technology has brought innovative solutions to various industries [1]. In the field of environmental management and agriculture, the use of drones is gradually increasing, especially in the spreading of particulate matter and pesticide spraying [2,3]. Due to their flexibility and efficiency, agricultural drones have become an important tool in these fields. Drones can rapidly spread particles such as fertilizers, seeds, or environmental remediation materials in areas like rice paddies and large fields [4,5]. However, the effectiveness of spreading is influenced by a variety of factors.

The distribution of particulate deposition is crucial for spreading effectiveness, as it not only affects environmental remediation outcomes but also directly impacts resource utilization efficiency [6]. Research indicates that during drone spreading, various parameters of the variable device (such as spraying speed, particle size and distribution, and dispensing angle) significantly affect particulate deposition patterns [7,8]. By adjusting these variables, it is possible to significantly improve the uniformity of particulate distribution, thereby enhancing operational effectiveness.

Centrifugal spreading UAVs and pneumatic spreading UAVs, using centrifugal discs and pneumatic spreading devices, respectively, achieve high efficiency in UAV spreading over large areas. However, their precise fertilizer supplementation for small-area economic crops is prone to the spreading of fertilizer beyond the limited area and the spreading of fertilizer beyond the crop growth requirements, and lacks a more refined selection of fertilizer variables [9,10]. In order to achieve precise fertilizer supplementation for cash crops, it is important to understand the effects of drones and diffusion devices on particle deposition in order to select the optimal method of operation [11].

Traditional centrifugal discs and pneumatic spreading devices have an overall variable, variable range that is not fine enough, and are only suitable for efficient spreading in large areas. In the face of small areas of economic crops to supplement the fertilizer, it is easy to cause over-spreading of fertilizer and environmental pollution caused by the economic crops to absorb the remaining fertilizer. The device proposed in this study achieves fine adjustment of spreading fertilizer by adjusting the structural changes of the unit, provides a variety of fine fertilizer deposition parts, and can replenish fertilizer according to the fertilizer lacking in the growth of actual crops in small areas, thus effectively reducing the excessive waste of fertilizer and environmental pollution.

Therefore, this study will focus on how the variable devices of spreading drones affect particulate deposition distribution. Through systematic experimental research, this paper will explore the variations in particulate deposition under different device configurations to provide theoretical support for optimizing spreading strategies. This study not only helps to provide a refined solution for fertilizer supplementation in small areas by UAVs but also provides scientific guidance for the application of UAV technology in fertilizer management.

2. Materials and Methods

2.1. Equipment

Figure 1 illustrates the overall structure of the spreading drone and the particle motion principle of the spreading device. As shown in Figure 1.1, the variable spreading device is installed below the drone. The device controls the descent of particulate flow through a flow regulation fan. The device is equipped with three sets of adjustment units to alter the descent distribution of the flow, thereby achieving the same flow state under different distribution conditions. Some of the structural changes in the device and a schematic representation of the particle flow are shown in Figure 1.2.

The experiment used a self-developed 6-rotor drone and spreading device, with the drone controlled by the DJI A3 system. The overall structure and control system of the drone are shown in Figure 2.

2.2. Work Principles

Figure 3 is the variable adjustment control process of the normal spreading device. In this study, the control module is divided into a ground signal control module and a device control module, aiming to achieve precise control of the descent speed and amount of fertilizer by regulating the speed of the stepper motor. The ground signal control module is configured with a laptop, microcontroller, and wireless module. The laptop is responsible for program modifications and command execution settings for the microcontroller. The microcontroller, in turn, sends signals to open the device and adjust the speed through the wireless module. The device control module consists of a wireless module, microcontroller, stepper drive module, stepper motor, and execution device. The wireless module receives instructions from the ground signal control module. The microcontroller processes the received instructions and transmits the processed control signals to the stepper drive module. The stepper drive module converts electrical signals into pulse signals, thereby controlling the rotation and speed adjustment of the stepper motor. This, in turn, drives the flow-regulating fan to achieve the desired speed for specific requirements. Through this precise control process, it is possible to effectively adjust the speed of the stepper motor, thus accurately controlling the rotation speed of the flow-regulating fan and achieving fine adjustment of fertilizer discharge.

2.3. Experiment Design

The experiment was conducted from August 29 to 31 at Shandong University of Science and Technology in Zibo. The test site is equipped with a ground-based meteorological station that detects wind speeds, as well as environmental changes. The experimental environment had a temperature of 25 °C, humidity of 30%, and wind speed not exceeding 2 m/s. The fertilizer particles used were urea (medium granules, Grain size diameter 1.18–3.35 mm, N total ≥ 46%), and the flow regulation fan speed was set to 110 r/min. The fertilization drone flew at an altitude of 3 m and a speed of 2.5 m/s, flying from west to east under wind speeds of less than 3 m/s. The test had a take-off buffer and a landing buffer, and the device was already running the spreading when the UAV flight took off. Before the drone arrived at the collection device, the drone had already been broadcasting for a period of time, and the same was performed in the area where the drone landed. The total starting amount of fertilizer was around 10 kg, with around 6 kg remaining at the end of a single flight operation.

A schematic of the test site is shown in Figure 4. On the test site, 3 rows of fertilizer collection boxes (500 × 500 × 200 mm) were arranged parallel to the drone’s flight direction, with a row spacing of 0.5 m, a total length of 6.5 m, and 7 collection boxes per row. The spacing between the collection boxes extended from the center of the flight path (0 m) to both sides, with a spacing of 0.5 m.

Figure 5 illustrates the structure of some of the device test groups. According to the position corresponding to the center line of the regulating modules (modules 1, 2 and 3), they are divided into three main categories, a, b, and c. The modules of group 1 are all at a, so they are named a1a2a3. Module 1 of group 21 is at c, module 2 is at a, and module 3 is at c, so the group is named c1a2c3. At the same time, the test was carried out to test the effect of the change in the structure of the regulating unit on the distribution of the fertilizer deposition for the changes in the regulating unit as the main factor. A total of 27 experimental groups were set up, with specific names listed in

Table 1. Experimental group settings.

Test Group	Combination Name	Test Group	Combination Name	Test Group	Combination Name
1	a1a2a3	10	b1a2a3	19	c1a2a3
2	a1a2c3	11	b1a2b3	20	c1a2b3
3	a1a2b3	12	b1a2c3	21	c1a2c3
4	a1b2a3	13	b1b2a3	22	c1b2a3
5	a1b2b3	14	b1b2b3	23	c1b2b3
6	a1b2c3	15	b1b2c3	24	c1b2c3
7	a1c2a3	16	b1c2a3	25	c1c2a3
8	a1c2b3	17	b1c2b3	26	c1c2b3
9	a1c2c3	18	b1c2c3	27	c1c2c3

.

2.4. Data Processing and Analysis

Figure 6 illustrates the process of collecting and analyzing particulate data from the experiment.

Figure 6.1 shows the fertilizer particles falling into the collection boxes after the drone’s operation. The staff then collected these particles, placed them into bags, weighed them, and recorded the data. Fertilizer spread into the collection device during the test is collected sequentially by means of transparent collection bags made of PE material, according to the number of the bags. The collected fertilizers were weighed and counted in sequence on a precision electronic scale.

Figure 6.2 shows the directions for analyzing particulate data, including particle deposition analysis and effective spreading width analysis of the drone. For a single route there is the option of adjusting the device for different fertilizer deposition states. The data from the three collection strips were also calculated, and when the variability in the three data points was small, the mean of the three data points was taken for data analysis. When the variability was large, the data set was rounded off and a re-test was performed. Calculations of fertilizer particle deposition mass and coefficient of variation were carried out for the collected experimental data according to Equations (1) and (2).

The experimental data are presented by calculating the deposition amount. Deposition amount refers to the unit area deposition calculated based on the quantity of fertilizer in the collection boxes after executing a single flight line. The calculation formula is as follows.

$$Q=M/S$$

(1)

where Q is the actual deposition amount in g/m²; M is the amount of particles in the collector in g; S is the bottom area of the collector, which is a constant of 0.25 m². Three fertilizer deposition curves can be obtained from the three sets of transverse deposition data collected under a single flight line. The particle deposition width is calculated using the simulation overlap method. The specific formula is as follows.

$$\overline{M}={\displaystyle \frac{\sum M_i}{n}},SD={\displaystyle \frac{\sqrt{{\sum }_1^n{(M_i-\overline{M})}^2}}{n-1}},CV={\displaystyle \frac{SD}{\overline{M}}}\times 100\%$$

(2)

where M is the deposition amount between the center lines of the first and third deposition curves after overlapping in g/m². $\overline{M}$ is the average value, in g/m². n is the number of samples of the overlapping deposition curves.

3. Results

3.1. Spreading Deposition Quality

Figure 7 shows the distribution changes in fertilizer particle mass for the spreading drone under different device groups. The figure includes 27 groups of different adjustment units, divided into three groups according to arrangements a, b, and c. Each group contains nine adjustment units. In group a, the data show that the maximum deviation for group a1a2b3 is 1.32 g, and the maximum deviation for group a1b2a3 is 2.41 g. In group b, the data show that the maximum deviation for group b1b2b3 is 0.41 g, and the maximum deviation for group b1c2a3 is 1.63 g. In group c, the data show that the maximum deviation for group c1c2a3 is 1.11 g, and the maximum deviation for group c1b2c3 is 2.39 g. Fine-grained differences in the fertilizer distribution of the UAV spreading devices with different adjusted unit structures were observed.

Figure 8 shows the deposition of fertilizer particles under 27 variable spreading device combination modes.

Figure 8a displays the fertilizer particle deposition for the main a1 group. Overall, group a1b2a3 has the highest fertilizer particle deposition, reaching 9.8 g/m². The minimum deposition value for all groups is 0, occurring in the edge drift areas of the drone operation. Group a1b2b3 shows the most significant fertilizer deposition data, with a three-quarter median value of 4.8 g/m² and a maximum value of 6.2 g/m². In contrast, group a1b2c3 has more uniform fertilizer particle deposition, with a three-quarter median value of 2.4 g/m² and a maximum value of 6.1 g/m².

Figure 8b shows the deposition of fertilizer particles under the main b1 group. Overall, group b1c2a3 has the highest fertilizer particle deposition, at 7 g/m². The minimum deposition value for all groups is 0, mainly occurring in the edge drift areas of the drone operation. Group b1a2b3 shows the most significant fertilizer deposition data, with a three-quarter value of 4.7 g/m² and a maximum value of 6.5 g/m². In contrast, group b1b2b3 has more uniform fertilizer particle deposition, with a three-quarter median value of 3 g/m² and a maximum value of 4 g/m².

Figure 8c shows the deposition of fertilizer particles under the main c1 group condition. Overall, group c1b2c3 exhibits the highest fertilizer particle deposition, at 2.4 g/m². The minimum deposition value across all groups is 0, which occurs in the edge drift areas during drone operation. Group c1b2b3 shows the most significant fertilizer deposition data, with a three-quarter quantile value of 1.2 g/m² and a maximum value of 1.5 g/m². In contrast, group c1b2a3 displays more uniform fertilizer particle deposition, with a three-quarter quantile median value of 0.8 g/m² and a maximum value of 1.7 g/m².

Data on particulate deposition indicate that the deposition state of the fertilizer varies when the sowing device selects different variable combinations. Among all variable groups, the deposition data for fertilizer particles in group b1b2b3 is the most uniform, with 75% of the particulate deposition values being 3 g/m² and the maximum value reaching 4 g/m².

3.2. Effective Width of Deposition

Figure 9 shows the effective width calculated from particulate deposition data. Figure 7.a illustrates the effective width of particulate deposition for group a. Overall, the highest coefficient of variation (CV) for group a1b2b3 under the same effective width conditions is 78%. Additionally, the lowest coefficient of variation (CV) for the particulates is 0, corresponding to the state where the original route remains unshifted. For group a1c2a3, when the effective width reaches 2 m, the coefficient of variation (CV) is 35%, with a range from 18% to 35%.

Figure 9b shows the effective width of particulate deposition for group b. Overall, the highest coefficient of variation (CV) for group b1b2a3 under the same effective width conditions is 76%. Additionally, the lowest coefficient of variation (CV) for the particulates is 0, corresponding to the state where the original route remains unshifted. For group b1a2a3, when the effective width reaches 2 m, the coefficient of variation (CV) is 31%, with a range from 18% to 31%. For group b1b2b3, when the effective width is 2 m, the coefficient of variation (CV) is 41%, with a range from 4.5% to 41%.

Figure 9c shows the effective width of particulate deposition for group c. Overall, the coefficient of variation (CV) for group c1c2a3 can reach up to 40%, while the lowest CV for particulates is 0, corresponding to the state where the original route remains unshifted. For group c1b2b3, when the effective width reaches 2 m, the coefficient of variation (CV) is 27%, with an initial minimum CV of 19% and a range from 19% to 27%. In group c1b2a3, when the effective width reaches 2 m, the coefficient of variation (CV) is 40%, with an initial minimum CV of 11% and a range from 11% to 40%.

Table 2. Range of particle variables in group a.

Group	Variance	Variable Scope/%	Group	Variance	Variable Scope/%	Group	Variance	Variable Scope/%
a1a2a3	0.46	3.92–43.87	a1b2a3	0.80	0.25–60.5	a1c2a3	0.39	3.64–1.45
a1a2c3	0.44	1.27–49.36	a1b2b3	0.50	0.75–39.10	a1c2b3	0.52	3.64–45.87
a1a2b3	0.49	0.78–35.06	a1b2c3	0.64	0.29–63.66	a1c2c3	0.50	3.64–45.87

presents the relevant data on the particle variability range for group a. In group a1b2a3, the particles have the highest variance of 0.8, with a variability range of 0.25–60.5%. In group a1c2a3, the particle fertilizer has the smallest variance of 0.39, with a variability range of 1.45–3.64%. In group a1b2c3, the particles have the largest variability range, from 0.29% to 63.66%, with a variability difference of 63.37%. In group a1a2b3, the particles have the smallest variability range, from 0.78% to 35.06%, with a variability difference of 34.28%.

Table 3. Range of particle variables in group b.

Group	Variance	Variable Scope/%	Group	Variance	Variable Scope/%	Group	Variance	Variable Scope/%
b1a2a3	0.37	3.07–41.43	b1b2a3	0.32	6.74–40.41	b1c2a3	0.44	2.73–43.18
b1a2b3	0.51	0.98–39.71	b1b2b3	0.16	15.71–26.44	b1c2b3	0.42	2.99–42.64
b1a2c3	0.48	2.43–41.85	b1b2c3	0.36	6.95–41.19	b1c2c3	0.44	1.01–41.92

presents the data on the particle variability range for group b. In group b1a2b3, the particles have the highest variance of 0.51, with a variability range of 0.98–39.71%. In group b1b2b3, the particle fertilizer has the smallest variance of 0.16, with a variability range of 15.71–26.44%. In group b1c2c3, the particles have the largest variability range, from 1.01% to 41.92%, with a variability difference of 40.91%. In group b1b2b3, the particles have the smallest variability range, from 15.71% to 26.44%, with a variability difference of 10.73%.

Table 4. Range of particle variables in group c.

Group	Variance	Variable Scope/%	Group	Variance	Variable Scope/%	Group	Variance	Variable Scope/%
c1a2a3	0.41	1.86–39.36	c1b2a3	0.38	4.86–43.48	c1c2a3	0.47	1.26–37.28
c1a2b3	0.55	0.24–41.95	c1b2b3	0.80	0.25–37.59	c1c2b3	0.48	0.5–36.07
c1a2c3	0.47	0.25–41.73	c1b2c3	0.51	0.24–58.68	c1c2c3	0.43	1.77–43.43

presents the data on the particle variability range for group c. In group c1b2b3, the particles have the highest variance of 0.8, with a variability range of 0.25–37.59%. In group c1a2a3, the particle fertilizer has the smallest variance of 0.41, with a variability range of 1.86–39.36%. In group c1b2c3, the particles have the largest variability range, from 0.24% to 58.68%, with a variability difference of 58.43%. In group c1c2b3, the particles have the smallest variability range, from 0.5% to 36.07%, with a variability difference of 35.57%.

The effective width of particle variable deposition data show that different sowing device combinations exhibit different fertilizer coefficients of variation (CVs) at the same width. In group a1c2a3, when the effective width is 2 m, the coefficient of variation (CV) is 35%, with a range from 18% to 35%. In group b1b2b3, when the effective width is 2 m, the coefficient of variation (CV) is 41%, with a range from 4.5% to 41%. In group c1b2a3, when the effective width is 2 m, the coefficient of variation (CV) is 40%, with a range from 11% to 40%. In cases where the fertilizer particle distribution is relatively uniform (group b1b2b3), the difference in the coefficient of variation is significant, at 36.5%. Under different group adjustments, the particle distribution shows the smallest variability range in group b1b2b3, with a range of 15.71–26.44% and a variability difference of 10.73%. The particle distribution shows the smallest variability range in group a1a2b3, with a range of 0.78–35.06% and a variability difference of 34.28%.

4. Discussion

4.1. The Effect of Rotor Wind Fields on Sediment Distribution

The rotor-induced wind field generated by drones significantly affects the movement and deposition patterns of particulate matter. It alters the trajectory and deposition location of particulate matter by driving air flow [12]. The rotor-induced wind field causes the dispersion of particulate matter, with its intensity and direction affecting the extent of dispersion. Strong wind fields result in a broader distribution of particulate matter, increasing deposition unevenness, while weaker wind fields lead to more concentrated deposition. Therefore, variations in the wind field significantly influence the deposition patterns of the particulate matter.

The aim of this study was to investigate the effect of structural changes in the conditioning unit on fertilizer deposition and to understand the more refined distribution of fertilizer deposition in this state. Therefore, the experiments were designed with the conditioning unit as the main factor, while keeping the flight speed of 2.5 m/s, flight altitude of 3 m, and external wind field of <3 m/s as invariant factors so as to maintain the consistency of the rotor wind field during the operation, reducing the impact from the rotor wind field on the evaluation. The Flunt rotor wind field simulation is performed on this UAV, and the wind field velocity cloud map changes in the integrated direction, as shown in Figure 10. The rotor wind field has an overall change in wind field velocity in the XZ plane, converging from the outside towards the middle of the UAV. The spreading device is installed at the position of the UAV’s proper center, which can effectively reduce the influence brought by the rotor wind field of the UAV.

Additionally, the wind field can cause stronger elastic collisions during the fall of particulate matter, leading to some particles escaping the collection area. To collect as much particulate matter as possible, this study used larger collection devices and set up multiple collection routes. This ensures that after a single drone operation, particulate deposition data from different routes remain consistent, thereby reducing the impact of the wind field on deposition. Subsequent tests could further reduce the effect of the wind field on particle deposition.

4.2. Multifactorial Effects on Sediment Distribution

Wind speed, flight altitude, flight speed, and flow rate collectively influence deposition distribution [13].

Smith et al. tested the effect of three wind speeds, 1.3, 2.7, and 4.0 m/s, on fertilizer deposition. The results of the tests showed that higher wind speeds increased the spreading range and suspension time of the particles, which could lead to uneven deposition [14]. Chojnacki et al. set the flight heights to 0.5 m and 1 m during the analytical tests of the disc spreading device. The results showed that the spreading uniformity was lower when the flight height was 1 m than when the flight height was 0.5 m. Flight altitude affects particle settling time and distribution range; higher altitude may cause particles to settle over a larger area [15]. Li et al. determined through simulation tests that the wind field has a significant effect on the landing of particles when the flight speed is 5 m/s. And with the increase in flight speed, the influence area of downwash will move further along the downstream direction accordingly. Flight speed determines the transport distance and settling pattern of particles; faster speeds may result in deposition at greater distances [16]. Parish et al. conducted tests for application rate changes and showed that the offset of particle deposition range increased at lower application rates. At higher application rates, the particle deposition profile was cheaper and the deposition distribution was distorted. Flow rate determines the particle load and deposition range; higher flow rates result in a broader distribution of deposits [17].

This study only conducted experimental analysis on single wind speeds, flight altitudes, and flight speeds, involving 27 adjustment units. Overall, group a1b2a3 had the highest fertilizer particle deposition of 9.8 g/m² when the flight speed, flight altitude, and wind speed remained stable. It can be chosen in case of high fertilizer deposition for economic crop demand. Group c1b2b3 had the maximum value of 1.5 g/m², which is a lower fertilizer deposition and can be selected in the case of lower crop demand for fertilizer.

Future research should investigate the combined effects of varying wind speeds, flight altitudes, and flight speeds on particulate deposition distribution. This will allow for more accurate prediction and management of deposition distribution. The follow-up study should address the multifactorial (wind speed, flight altitude, flight speed) effects on fertilizer deposition distribution.

4.3. Variable Regulation and Particulate Deposition Characterisation

Previous research has focused on the application of UAV technology and the fundamentals of particulate spreading. Many papers have explored the application of UAVs in agriculture, such as the application of precision fertilizer application and remote sensing-based spraying [18,19]. The environmental pollution and fertilizer waste caused by the fertilizers and pesticides used can be further reduced through precision fertilizer and medicine application. However, there are relatively few studies on the specific effects of variable devices on the distribution of particulate matter deposition. And, there is a lack of more refined fertilizer variable selection. A number of studies have focused on the effects of factors such as spreading height and wind speed on particle deposition, but have often neglected a more fine-grained and comprehensive evaluation of fertilizer deposition [20]. Moreover, existing variable spreading tests usually use a single device structure test design, with few systematic analyses of multiple combinations.

In the combination a series, nine refined variable spreading selection modes are available. Regionalized fertilizer replenishment can be carried out in response to the amount of fertilizer shortages in economic crops by selecting the combination of regulating units in the combination a series that is most suitable for the corresponding shortage of fertilizer so that fertilizer wastage and pollution of the environment are reduced to a minimum. When the economic crop single regional fertilizer demand is the largest, the combination a1b2c3 can be selected; at this time, the device is spreading fertilizer in the widest variable range of 0.29% to 63.66%, and the variability difference is 63.37%. When the demand for fertilizer in a single region of the economic crop is small, it can be chosen in combination a1a2b3, at which time the narrowest variable range of fertilizer is 0.78% to 35.06%, and the variability difference is 34.28%. If the fertilizer is to be more adapted to a small area of economic crops with a shortage of fertilizer, a more refined selection can be made from combination b series or c series. The combination b1b2b3 can be chosen when the demand for fertilizer from economic crops in a single region is more homogeneous, when the range of variability is 15.71–26.44% and the difference in variability is 10.73%. In this way, fertilizer spreading can be minimized and at the same time crop fertilizer requirements can be met.

This study provides guidance in practical applications, especially in optimizing fertilizer distribution, which can effectively meet regional needs and enhance agricultural productivity. These features make this study cutting edge and practical in the field of UAV spreading.

5. Conclusions

This study analyzed the deposition data of granular fertilizers dispersed by drones through broadcasting deposition experiments. It also investigated the effective deposition width of the dispersed fertilizer particles to explore the impact of variable mechanisms of broadcasting drones on the distribution of particulate matter deposition. Based on the research findings, the following conclusions were drawn:

(1)

The particulate deposition data indicates that the fertilizer deposition varies with different variable combinations of the broadcasting device. Among the 27 variable combinations, the b1b2b3 group exhibited the deposition, with three-quarters of the particulate deposition values being 3 g/m² and the maximum value reaching 4 g/m².

(2)

Under relatively uniform distribution of fertilizer particles, the b1b2b3 group had a large coefficient of variation difference (36.5%), with the coefficient of variation ranging from 4.5% to 41%.

(3)

Under different group adjustments, the particle distribution shows the smallest variability range in group b1b2b3, with a range of 15.71–26.44% and a variability difference of 10.73%. The particle distribution shows the largest variability range in group a1a2b3, with a range of 0.78–35.06% and a variability difference of 34.28%.

In conclusion, by adjusting the structure of the variable spreading device, the fertilizer flow down state is changed so as to meet the needs of different regions with different fertilizer distribution states. The device can realize different particle distribution states, and the uniform spreading state is applicable to the required uniform fertilizer replenishment area. Fertilizer distribution under different device states can be selected according to the needs of different regions.