**1. Introduction**

Single-seed precision seeding refers to a type of seeding technology that separates crop seeds one by one and controls their distribution position precisely on the seedbed. This technology encounters difficulties in the separation of seeds and in preventing the formed uniform seed flow from being re-dispersed on the inner wall of the seed pipe and seedbed [1–3]. To solve the above problems, a seeding-wheel-type pneumatic seeder that works via pneumatic single-grain separation and orderly and stable casting in a low position was designed. However, its internal airway structure is complex with many branches, has a dense arrangement, variable cross-sectional areas, and displays long and narrow bending, which will inevitably cause pressure losses, and lead to further problems, such as low pneumatic utilization efficiency and energy waste [4].

Scholars have conducted extensive research on the airflow field associated with the pneumatic seed-metering mechanism [5–8]. For example, Yazgi et al. [9,10] used the CFD method to study the effect of a vacuum plate with different numbers of holes on the uniformity of seed spacing. The highest performance was reported when 26 and 36 holes were used for cotton and corn, respectively. Ghafori et al. [11] analyzed the pressure drop function during the suction and transport of corn and barley. The lowest pressure drop for corn and barley occurred at the air velocity of 20 <sup>m</sup> · <sup>s</sup>−<sup>1</sup> and 15 <sup>m</sup> · <sup>s</sup><sup>−</sup>1, respectively. Yang Shandong et al. [12–14] used the large eddy simulation method with Fluent software to analyze the flow field of the positive-pressure air chamber inside a seed-metering device, and studied the variation law of the internal flow field of the seedmetering device under different parameters. It was shown that the seeding performance

**Citation:** Zhang, X.; Wen, Z.; Wang, Q.; Li, H.; Zhang, Z.; Liu, J. Research on Characteristics of Airway Pressure Loss in Seeding-Wheel-Type Pneumatic Seeder. *Agriculture* **2022**, *12*, 2021. https://doi.org/ 10.3390/agriculture12122021

Academic Editors: Vadim Bolshev, Vladimir Panchenko and Alexey Sibirev

Received: 1 November 2022 Accepted: 23 November 2022 Published: 26 November 2022

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2022 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

was better when the plate speed was lower than 18 <sup>r</sup> · min<sup>−</sup>1. Zhao Zhan et al. [15–17] used Fluent software to calculate the force of seeds in the real seed suction airflow field and highspeed photography technology to obtain the seed transient adsorption and falling trajectory. The results showed that the average seed-spacing interval error reached its minimum value at 1.5 kPa pressure and a −5◦ angle and that the error increased almost linearly with increasing cylinder rotational speed. Xiaolong Lei et al. [18,19] carried out a numerical simulation of seed motion in the distribution head and tracked the seed migration trajectory and distribution behavior. The simulation results showed that, when the streamlined angle increased from 10◦ to 50◦, the variation coefficient of seed distribution decreased initially and then increased for an inlet diameter of 20 mm and an airflow velocity of 20 <sup>m</sup> · <sup>s</sup><sup>−</sup>1. Song shi et al. [20] calculated the gas–solid two-phase flow of a seed-metering device by introducing the pressure gradient force model into the coupled analysis of EDEM-CFD. The results were given as follows: the curvature coefficient of the seed guide groove curve was 0.265, the depth of the seed guide groove was 2.57 mm, and the slope angle of the seed guide groove was 15.33◦.

The above scholars all used the computational fluid dynamics method to simulate the internal flow field that is difficult to measure, and the simulation results were similar to the experimental results, which shows that it is an effective method to study the pneumatic seed-metering mechanism. However, the above research mainly focuses on the largediameter pneumatic and intensive conveying air duct or the cylindrical chamber inside the seed-metering device, which is characterized by a spacious air duct and simple structure. There is no relevant research on long, narrow and complex air ducts with variable diameters.

Therefore, we focused on the fluid domain of seeding-wheel-type pneumatic seeders and used the computational fluid dynamics method to explore the types and principles of the pressure loss in each area. The key parameters that affect the pressure loss were identified and the airflow field distribution and the airflow motion law in the seed-placing device were defined to determine the best structure and working parameters. The results show that it is possible to weaken the obstruction of the complex airway structure to the airflow, reduce the pressure loss, and improve pneumatic utilization and the effect of sowing. The results will also provide theoretical guidance for pressure drop analysis of complex airways.

#### **2. Structure and Fluid Analysis**

#### *2.1. The Structure and Working Principle of Seeding-Wheel-Type Pneumatic Seeder*

Seeding wheels are the main carriers used for picking seeds, transporting seeds, and casting seeds. Their structure and installation are shown in Figure 1 [21]. Several seedsucking holes are evenly distributed along the edge of the seeding wheel and are connected to the internal pneumatic distribution center through a long and narrow airway with variable diameters. The upper-end face of the pneumatic distribution center is connected to a rubber sealing ring fixed with a pneumatic distribution cover, and they are press-fitted and sealed. The inner air cavity of the rubber sealing ring is connected to the negative pressure aperture to provide negative pressure force for the seed-sucking hole. During operation, the pneumatic distribution cover remains stationary, the seeding wheel rotates with the ground wheel, and the two are connected by thin-walled bearings. A spring in the locking mechanism is used to press the seed wheel and air distribution cover together. The outer wall of the seeding wheel, which is installed on both sides of the edge of the seeding wheel, flattens the soil with a certain force, which makes a difference to the depth limit and level of flattening. The seed-sucking boss can extend to the lower soil layer to complete the casting of seeds at a suitable position.

**Figure 1.** Schematic diagram and three-dimensional modeling of seeding wheel structure. (**a**) Schematic diagram of seeding wheel structure: (1) seed-sucking hole; (2) airway; (3) thin-walled bearing; (4) negative pressure aperture; (5) rubber sealing ring; (6) seeding axle; (7) locking mechanism; (8) positive pressure aperture; (9) outer wall; (10) seed-sucking boss; (**b**) three-dimensional modeling of seeding wheel structure.

#### *2.2. Fluid Domain Modeling*

The function of the airway inside the seeding wheel is to transport the fan's pneumatic supply to the seed-sucking hole, and try to reduce the pressure loss during conveying [16,17,22], so that the seed-sucking hole has enough airflow velocity and negative pressure to perform its function. In addition, the airway should demonstrate good pneumatic distribution, making the difference between negative pressure and airflow velocity in each seed-sucking hole small, which is not significantly related to their location.

The negative pressure fluid domain inside the seeding device is the inner space enclosed by the seeding wheel, rubber sealing ring, and pneumatic distribution cover. The seeding wheel and the pneumatic distribution cover will rotate during operation, thereby dividing the entire negative pressure fluid domain into two parts, the static part and the moving part. The static part is the upper fluid domain surrounded by the pneumatic distribution cover, which is the green part shown in Figure 1b. The moving part is the underfluid domain surrounded by the seeding wheel, which is the blue part shown in Figure 1b. The seeding wheel has 16 sub-airways that are not connected, as shown in Figure 1a. Because the inner pneumatic distribution cover is divided into areas with air and areas without air, the seed-sucking hole at the end of the sub-airway has negative pressure only when the sub-airway is connected to areas with air. These two parts will rotate during operation, which contributes to the airflow's on–off control by switching the corresponding position. Using Space Claim software to extract the fluid domain, a simplified 3D model of the seeding device was imported into the workbench of ANSYS, as shown in the central part of Figure 2.

**Figure 2.** Fluid domain division.

According to Figure 2, there are 12 lower sub-air channels connected to the upper air chamber, and there is a lower sub-air channel at the edge of the domain (only about half of this is connected to the upper air chamber at the top). The analysis below focuses on the airway in the connected position only. The structure of the negative pressure fluid domain is complex, involving various forms of airway structural changes, and it is difficult to conduct an overall analysis. Therefore, the negative pressure fluid domain is divided into the following four areas according to the different types of aerodynamic changes: the variable-section panhandle area, the bending area of the air pipe, the air chamber confluence area, and the connection area of the negative pressure aperture and air chamber. These are shown in Figure 2.

#### *2.3. Analysis of Pressure Loss in Airway*

#### 2.3.1. Aerodynamic Analysis of Variable-Section Panhandle Area

The energy loss of the fluid can be divided into the along-path loss and the local loss [23] according to the difference in the airway geometrical boundary that causes the fluid energy loss. The airflow velocity inside the airway is much less than the sound velocity, so air can be regarded as an incompressible fluid. The pneumatic input is negative pressure. The entire pneumatic system's inlet is the 12 seed-sucking holes, and the outlet is a negative pressure aperture. The analyses of the four areas were conducted using the airflow inlet. The cross-sectional view of the variable-section panhandle area is shown in Figure 3.

The airway structure in this area is shown in Figure 3a. There are three cylindrical cavities arranged from small to large. The first cavity is the seed-sucking hole, which is also the executed structure for the terminal, and its cross-sectional area is mainly related to the performance of seed absorption. Its structure is not studied in this paper. The second cavity is the airway section I, which is contained in the seed-sucking boss. Its cross-sectional area is limited by the size of the seed-sucking boss and has a maximum size of 10 mm. The third cavity is the airway section II, and its cross-sectional area is adjustable. When the diameters of airway section I and airway section II are equal, the airway structure shown in Figure 3b can be observed.

The variable-section panhandle area mainly involves two forms of pressure loss. When the air passes from the seed-sucking hole to the airway section I and from the airway section I to the airway section II (unequal diameter), the type of aerodynamic change that occurs is the sudden expansion of the pipe. It is mainly caused by the local pressure loss; while in the airway section I and II, it is caused by the pressure loss along the path.

#### 2.3.2. Aerodynamic Analysis of Variable Diameter Transition

The type of aerodynamic change in the variable diameter is caused by diffusion flow [24], which can be divided into the following two forms: sudden and gradual expansion. The space with a variable diameter is narrow, and it is difficult to make a tapered airway with a small angle and a long lead. Therefore, this is regarded as the sudden expansion type. After the cross-sectional area of the sudden expansion pipe changes, the fluid will move away from the wall and form a vortex, resulting in the loss of airflow energy. This part of the local energy loss can be calculated as follows [25]:

$$p\_{\uparrow 1} = (1 - \frac{A\_1}{A\_2})\frac{\rho v^2}{2} \tag{1}$$

where *pj*<sup>1</sup> is the energy loss per unit volume of the gas in the form of sudden expansion (pressure loss), Pa; *<sup>ρ</sup>* is the medium density, kg · <sup>m</sup><sup>−</sup>3; *<sup>A</sup>*<sup>1</sup> is the inlet cross-sectional area, m2; *<sup>A</sup>*<sup>2</sup> is the outlet cross-sectional area, m2; *<sup>v</sup>* is the inlet airflow velocity, m · <sup>s</sup><sup>−</sup>1.

It can be observed from Equation (1) that the pressure loss here is mainly related to the cross-section ratio and the velocity in the inlet. Under the condition of a given inlet with negative pressure, the velocity in the inlet can be regarded as constant. Moreover, because the diameter of the seed-sucking hole and the cross-sectional area of the airway section I are not variable, *A*<sup>1</sup> can be regarded as invariable. Therefore, the pressure loss of this part is mainly related to the outlet cross-sectional area *A*<sup>2</sup> (the cross-sectional area of the airway section II). When the cross-sectional areas of airway section I and section II are equal, local pressure loss occurs once. Otherwise, it will occur twice.

#### 2.3.3. Aerodynamic Analysis of Long and Narrow Air Pipe

The size, shape, and direction of the flow cross-sections inside airway section I and section II remain unchanged, so the fluid encounters only the frictional resistance provided by the airway wall, which is the loss that occurs along the path [26,27]. The resistance loss coefficient along the path (*λ*) is related to the fluid flow state and often is calculated using the Reynolds number (*Re*). The Reynolds number calculation formula [28] is as follows:

$$\text{Re} = \frac{\rho \overline{v} d\_0}{\mu} \tag{2}$$

where *μ* is the dynamic viscosity of the fluid, Pa·s; *d*<sup>0</sup> is the characteristic length of the airflow section, m; *<sup>v</sup>* is the average flow velocity of the airway section, m · <sup>s</sup><sup>−</sup>1.

Air is the fluid object studied in this paper and by taking the parameters under normal pressure at 20 ◦C, *<sup>ρ</sup>* can be calculated as 1.205 kg · <sup>m</sup>−<sup>3</sup> and *<sup>μ</sup>* as 1.79 × <sup>10</sup>−<sup>5</sup> Pa·s. The characteristic length of the airflow section, which is also the diameter of the air pipe, is 0.02 m. The average velocity of the airway section II after stabilization is calculated by pre-simulation to be about 2 <sup>m</sup> · <sup>s</sup><sup>−</sup>1. The Reynolds number (*Re*) is solved at 2692, which is greater than the critical value of 2300. Therefore, at this time, the fluid flow state in the airway is turbulent, but the value is less than 4000, indicating that the fluid in the airway is in the

transition state between laminar flow and turbulent flow [29,30]. Therefore, the resistance loss coefficient (*λ*) along the path in the area increases with the increase in Reynolds number (*Re*) and has a weak relationship with the relative roughness of the wall [31].

$$p\_{f1} = \frac{l\_{\mathcal{S}} \rho^{\frac{3}{4}} \mathcal{Q}^{\frac{7}{4}}}{800 d\_{\mathcal{S}}^3 \mu^{\frac{1}{3}}} \tag{3}$$

where *lg* is the length of the pipe, mm; Q is the flow rate, <sup>m</sup><sup>3</sup> · <sup>s</sup><sup>−</sup>1; *dg* is the diameter of the pipe, mm. *pf1* is the pressure loss along the path of the long and narrow airway part.

It can be observed from the formula that the pressure loss (*pf1*) in the airway part is mainly related to the length of the pipe (*lg*), the diameter of the pipe (*dg*), and the flow rate (*Q*). The length of the pipe (*lg*) is the pneumatic conveying distance, which is determined by the size of the seeding wheel. The flow rate (*Q*) is mainly related to the given negative pressure. Therefore, in order to reduce air resistance in design, one must adjust the diameter of the pipe (*dg*). The larger the diameter of the pipe, the smaller the pressure loss along the path in the long and narrow airway.

Therefore, according to the above research, there are three kinds of design schemes in this area, which are as follows:


The direction of the airway shifts from horizontal to longitudinal in the bending area of the air pipe, as shown in Figure 4.

**Figure 4.** Bending area of the air pipe.

When the air goes flows through the connection, there is a velocity difference between the inner and outer airflows, which generates eddy currents, resulting in a large local energy loss. Due to space constraints, this is in the form of a sharp bend. The pressure loss in the bending area of the air pipe [32] can be calculated as follows:

$$p\_{\rm w} = [0.946 \sin^2(\frac{\theta}{2}) + 2.05 \sin^4(\frac{\theta}{2})] \frac{\rho \overline{v}^2}{2} \tag{4}$$

where *pw* is the local pressure loss in the bending area of the air pipe, Pa; *θ* is the angle between the two air pipes, (◦).

It can be observed from the formula that the pressure loss in this part is related to the angle (*θ*) between the two air pipes. The larger the angle, the smaller the pressure loss. However, when the position of the air-passing aperture is fixed, the larger the angle is and the larger the horizontal length of the inclined section is. The relationship of the angle (*θ*) between the horizontal length and the two air pipes is as follows:

$$\theta = 180^\circ - \arctan(\frac{h\_f}{l\_c - R}) \tag{5}$$

where *hf* is the vertical distance from the center of the outermost air pipe to the plane where the air-passing aperture is located, mm; *lc* is the horizontal length of the distribution center, mm. *R* is the distance between the center of the longitudinal airway and the center of the sowing wheel, mm.

Restricted by the structure and thin-walled bearings, the horizontal length of the distribution center should not be greater than 150 mm. The vertical distance should be less than the width of the outer wall of the seeding wheel (100 mm). It can be solved by formula 5 that the angle (*θ*) between the two air pipes should be less than 122.62◦, meaning that the horizontal value range of the air pipes' angle is 90–120◦.

#### 2.3.5. Aerodynamic Analysis of Air Chamber Confluence Area

The air chamber confluence area is shown in Figure 5.

**Figure 5.** Air chamber confluence area.

This area is the air chamber space formed by the air afflux from the longitudinal air pipes of each lower sub-airway to the rubber sealing ring in the pneumatic distribution cover [33], which is essentially the flow form of multiple small-section pipes that transform into large-section cavities. The energy loss that occurs is mainly local pressure loss, which mainly depends on the cross-sectional area on both sides, but it is difficult to change, and therefore optimize.
