Design and Testing of a Closed Multi-Channel Air-Blowing Seedling Pick-Up Device for an Automatic Vegetable Transplanter

Abstract

In this study, a closed multi-channel air-blowing plug seedling pick-up device and a combined plug tray were designed to address the issues of complex structure, high seedling damage rates and low pick-up efficiency in fully automated vegetable transplanter systems. The device operates by sealing the plug seedlings in a seedling cup, where compressed air is channeled into the sealed cavity through multiple passages during the seedling pick-up process. The upper surface of the seedling plug is subjected to uniform force, overcoming the friction and adhesion between the plug seedlings and the tray. This process presses the seedlings into the guide tube, completing the pick-up operation. A mechanical model for the plug seedlings was developed, and the kinetics of the pick-up process were analyzed. The multi-channel high-pressure airflow was simulated and evaluated, identifying three key parameters affecting seedling pick-up performance: water content of the seedling plug, air pressure during pick-up, and air-blowing duration. Using these factors as variables, and with seedling pick-up rate and substrate loss rate as evaluation indicators, single-factor experiments and a three-factor, three-level orthogonal experiment were conducted. The experiments’ results showed that the best seedling pick-up performance was achieved when the water content of the plug was 20%, the air pressure was 0.3 MPa, and the air-blowing time was 30 ms. Under these conditions, the seedling pick-up success rate was 97.22%, and the substrate loss rate was 10.46%.

1. Introduction

The technology of plug seedling cultivation and transplanting improves the multiple cropping index, as well as vegetable yield and quality, while reducing pests, diseases, and weeds. It has been widely adopted and applied [1,2,3]. China as a country is the largest producer of vegetables in the world. According to data from the National Bureau of Statistics of China, the vegetable planting area in China reached 22,434 km² in 2022, with a production of 799.97 million tons. China primarily uses semi-automatic transplanting methods. In this process, the transplanter completes the automatic planting, but manual labor is still required to pick and place seedlings. Consequently, workers face high labor intensity, making it challenging to further enhance transplanting efficiency. Fully automatic transplanting is a promising solution to these issues and represents the future of vegetable plug seedling transplanting [4,5,6,7].

Developed countries have been studying automatic transplanters for several years now, and have developed various models based on local crops, agronomic practices, and geographical characteristics. For example, in Western Europe and the United States, large-scale wide-width transplanters which integrate machine, electric, hydraulic, and pneumatic systems feature complex structures and high automation, making them suitable for large-scale transplanting on plains. In Japan, the purely mechanical small transplanter is known for its flexible movement and high seedling-picking accuracy, making it ideal for small-scale transplanting in greenhouses or on complex terrains. Although China began researching automatic transplanting later than many developed countries, progress has been rapid [8,9,10]. Several research institutions and companies have developed ejector-type [11], clamp stem type [12,13,14,15], insert-clamp [16,17,18,19,20,21], straight-drop [22], air-assisted [23,24], and other mechanisms for picking seedlings. However, issues such as complex structures, high seedling injury rates, and low seedling-picking efficiency persist [25,26,27].

The straight-drop seedling-picking device allows plug seedlings to fall directly from the plug tray into the planting or branch device, with seedlings cast during the picking process. Compared to other seedling-picking methods, the straight-drop approach offers advantages such as there being no need to adjust the seedlings’ posture, ease of orderly seedling picking, a simple and compact structure, and a low seedling injury rate [22]. There are several types of straight-drop seedling-picking devices, including mechanical downward pressure [28], negative pressure [29], and direct blowing [22]. Mechanical downward pressure devices press the plug seedlings with the end-effector, which directly contacts the delicate seedlings, resulting in a high injury rate. Negative pressure devices have a lower injury rate but require high-power fans to generate negative pressure, which consumes significant energy and demands high machining accuracy and sealing performance. Our research team has developed an air-blow-vibration composite seedling-picking method. Although this achieves non-contact seedling extraction, aiming the nozzle directly at the plug causes uneven force distribution on the plug surface, substrate splashing, high noise, and excessive energy consumption [30].

This paper presents the design of a closed multi-channel air-blowing plug seedling-picking device. The closed air-blowing system can generate the necessary seedling-picking force in a very short air-blowing duration (30 ms), achieving non-contact picking without substrate splashing. It operates at high speed with low energy consumption. The device can pick seedlings either from the entire row or one by one, depending on the automatic transplanting machine’s requirements. The seedling pick-up device was mounted on a planting chassis with an air compressor to enable the fully automatic transplanting of vegetable plug seedlings. Through theoretical and experimental analysis, the optimal design of the seedling-picking device was achieved, and the best combination of seedling-picking parameters was identified.

2. Materials and Methods

2.1. Machine Structure and Working Principle

2.1.1. Overall Design of Automatic Transplanter

The seedling-picking device and the planting chassis were assembled into a fully automatic vegetable transplanting machine (Figure 1). The automatic seedling auto-picking device was the core component of the fully automatic transplanting machine, which directly affects the performance of the whole machine. Therefore, this paper focuses on the closed multi-channel air-blowing type of seedling-picking device. It was mainly composed of combination-type plug trays, a tray feeding device, a seedling-picking unit and a seedling-guiding tube, etc., as shown in Figure 1. The tray feeding device was composed of a seedling feeding bottom plate, a seedling feeding motor, two chains with levers, and two guide strips. The chain lever was covered with a silicone sleeve. The two guide strips were attached to the seedling feeding base plate. The pressure plates on both sides are, respectively, fixed on the left- and right-side plates, and the tension-bearing seat tensions the chain. The tray feeding device drives the plug seedlings to move forward intermittently along the guide strip and aligns the plug seedlings with the seedling-guiding tube row by row. The seedling-picking unit was composed of an “Π” shape bracket, air cylinders, beam, guide rails, seedling-picking actuator, etc. Linear bearings are installed at both ends of the beam. The beams drive the seedling-picking actuators to slide up and down along the guide rails under the drive of the air cylinder.

The combination plug tray of 6 × 6 holes (Shandong University of Technology, Zibo, China) was matched with the seedling pick-up device. It was composed of a plate with holes, a pallet, plug seedlings, and an isolation tube, as shown in Figure 2. The transparent mesh isolation cylinder was an optional accessory, which was used for plug seedlings such as cabbage with large width or poor uprightness, to solve the problems of entanglement between plug seedlings, resulting in difficulty in picking seedlings and damage to seedlings. In the process of raising seedlings, when the plug seedlings have a small width, the isolation tube was sleeved on the seedling tube to limit the lateral extension of the plug seedlings.

2.1.2. Working Principle of the Seedling Selection System

The performance of the seedling-picking actuator (Shandong University of Technology, Zibo, China) directly affects the qualified rate of vegetable seedling transplanting. Due to machining inaccuracies and assembly errors, it was difficult to align the mouths of the seedling cups on the same plane, resulting in an inability to ensure proper airtightness. Therefore, the seedling-picking actuator should have an automatic adjustment function, so that under the condition of deviation between the plug tray and the seedling cup, the mouth of each seedling cup can still closely fit the plug tray and seal the plug seedlings. The upper and lower nuts of the seedling-picking actuator fix the sliding sleeve on the beam, and the seedling cups are fixed under the bottom of the air duct; the air duct is installed with the spring and then passed through the sliding sleeve, and the spring is pre-tightened with the pre-tightening nut. When the seedling cup is under pressure, the spring contracts, and the air guide tube moves up a certain distance along the sliding sleeve relative to the beam, so that each seedling cup is pressed against the plug tray, as shown in Figure 3.

The structure of the seedling cup is shown in Figure 4; this is a nylon 3D printed part. The outer diameter s₁ is 50 mm, the inner diameter s₂ is 43 mm, the height H is 124 mm, the distance h between the air outlet and the cup mouth is 22 mm, and the top of the seedling cup is provided with 3 mounting holes (diameter 3.5 mm) and an air inlet hole (diameter 6 mm), and 6 flow passages (diameter 2 mm), evenly arranged in the cup shell. A silicone sealing ring was installed at the bottom opening of the seedling cup.

2.1.3. Working Principle of the Seedling Pick-Up

Put the seedling tray on the seedling-feeding bottom plate; the seedling-feeding motor drives the left and right chains to rotate clockwise, and the lever of the chain moves the seedling tray forward intermittently, so that the plug seedlings are aligned with the seedling-guiding tube row by row (Figure 1). The plug seedlings are affected by the friction and adhesive force between the plug seedlings and the plug tray, and the plug seedlings remain in the plug tray. The air cylinder drives the beam down, and the whole row of seedling cups cover the seedlings to form a sealed cavity. The PLC touch screen control system controls the solenoid valve to pass high-pressure air into the sealed cavity for a certain period of time, and the air pressure in the seedling cups rapidly rises, under the action of gravity and air pressure; the plug seedlings overcome the friction and adhesive force and fall into the seedling-guiding tube to complete the seedling extraction. After the plug seedlings were taken out, the air cylinder drives the beam to rise, the tray feeding device sends the next row of plug seedlings to the seedling-guiding tube, and the air cylinder presses the beam down again to start the next seedling-picking cycle. The principle of picking seedlings is shown in Figure 5.

2.2. Mechanical Model of Plug Seedlings in the Process of Picking Seedlings

The force analysis of the plug seedlings when picking the seedlings was carried out. The force of the plug seedlings is shown in Figure 6, where N₁ and N₂ are the support forces of the tray tube on the plug body, F is the air blowing force, mg is the gravity of the plug seedlings, f is the friction between the plug body and the tray tube, and F_nj is the adhesive force between the plug body and the tray tube. During the process of picking seedlings, the adhesive force disappears gradually after the plug body slips along the tray tube. With the downward movement of the plug seedlings, the contact area between the plug body and the tray tube gradually decreases, and the friction force between the plug body and the tray tube gradually decreases. Therefore, the falling process of the plug body is a variable acceleration motion, which can be divided into the plug seedling acceleration stage, the plug seedling deceleration stage, and the plug seedling re-acceleration stage.

(1)

Pour compressed air into the seedling cup, F₁(t) + mg > f₁(t) + F_nj, this is the accelerated falling process.

Seedling cup air pressure P_b

$$P_b={\displaystyle \frac{PQt}{V_b}}$$

(1)

Air blowing force F₁

$$F_1\left(t\right)=P_{b}A$$

(2)

Friction force f

$$f\left(t\right)={\displaystyle \frac{\left(l-S_1\left(t\right)\right)N}{l}}\mu$$

(3)

According to Newton’s second law:

$$F_1\left(t\right)+mg-f_1\left(t\right)-F_{nj}=m{a}_1\left(t\right)$$

(4)

Acceleration a₁

$$a_1\left(t\right)={\displaystyle \frac{F_1\left(t\right)+mg-f_1\left(t\right)-F_{nj}}{m}}$$

(5)

Speed v₁

$$v_1\left(t\right)={{\displaystyle \int }}_0^{t1}{a}_1\left(t\right)dt$$

(6)

Displacement S₁

$$S_1\left(t\right)={{\displaystyle \int }}_0^{t1}{v}_1\left(t\right)dt$$

(7)

where p represents air pressure (Pa); Q is the flow rate (m³/s); t is the air blowing duration (s); V_b is the volume of the seedling cup (m³); A is the cross-sectional area of the plug body (m²); l is the length of the tray tube (m); N is the contact force between the plug seedling and the plug tray (N); μ is the coefficient of friction between the plug seedling and the tray tube; and S is the displacement of the falling plug seedling (m).

(2)

Stop aeration to the seedling cup, F₂(t) + mg < f₂(t); this is the plug seedling slow down and stop moving process.

According to Newton’s second law

$$mg+F_2\left(t\right)-f_2\left(t\right)=m{a}_2\left(t\right)$$

(8)

Acceleration a₂

$$a_2\left(t\right)={\displaystyle \frac{mg+F_2\left(t\right)-f_2\left(t\right)}{m}}$$

(9)

Speed v₂

$$v_2\left(t\right)={{\displaystyle \int }}_0^{t1}{a}_1\left(t\right)dt+{{\displaystyle \int }}_{t1}^{t2}{a}_2\left(t\right)dt$$

(10)

Displacement S₂

$$S_2\left(t\right)={{\displaystyle \int }}_{t1}^{t2}{v}_2\left(t\right)dt$$

(11)

To ensure successful seedling picking, the speed of pouring of compressed air should be controlled to avoid this stage, which may cause the seedlings to be stationary in the tray tube, and fail to be picked up.

(3)

The plug seedlings fall along the seedling tube for a certain distance but do not break away from the seedling tube. At this time, F₃(t) + mg > f₃(t) is the process of acceleration of the falling of the plug.

According to Newton’s second law

$$F_3\left(t\right) mg-f_3\left(t\right)=m{a}_3\left(t\right)$$

(12)

Acceleration a₃

$$a_3\left(t\right)={\displaystyle \frac{F_3\left(t\right)+mg-f_3\left(t\right)}{m}}$$

(13)

Speed v₃

$$v_3\left(t\right)={{\displaystyle \int }}_0^{t1}{a}_1\left(t\right)dt+{{\displaystyle \int }}_{t1}^{t2}{a}_2\left(t\right)dt+{{\displaystyle \int }}_{t2}^{t3}{a}_3\left(t\right)dt$$

(14)

Displacement S₃

$$S_3\left(t\right)={{\displaystyle \int }}_{t2}^{t3}{v}_3\left(t\right)dt$$

(15)

The conditions for successful seedling picking are v₃ > 0; reducing the speed of the plug seedling leaving the seedling tray can reduce the impact of the plug seedlings and other devices.

According to the analysis of the process of picking seedlings, the factors affecting the performance of picking seedlings are air-blowing force, adhesive force, friction, and gravity. The air-blowing force is positively correlated with the air pressure and air-blowing time for picking seedlings. The adhesive force and friction force are related to the ratio of the substrate, the moisture content, the friction coefficient of the seedling tube, etc. The gravity of plug seedlings is related to the ratio of substrate and moisture content. The ratio of substrate composition of the plug body has been determined according to the type of crops. Therefore, under given conditions, the main factors affecting the qualified rate of seedling picking are the water content of plug body, the air pressure of picking seedlings and air-blowing time.

2.3. Airflow Simulation Analysis of Air-Blown Seedling Picking

The computational fluid dynamics (CFD) analysis software Flow Simulation of Solidworks 2022 was used to simulate the flow of direct-blowing seedling picking and closed multi-channel air-blowing seedling picking [31], and the simulation results were compared and analyzed. Solidworks 2022 software was closely integrated with Flow Simulation software, which can realize the seamless connection between CAD and CFD, build a 3D model in Solidworks software, divide mesh in Flow Simulation software, set calculation domain, boundary conditions, target and mesh, etc., and perform fluid simulation analysis.

Simulation analysis of the traditional direct-blowing method and the closed multi-channel air-blowing method described in this paper are carried out, respectively.

2.4. Experiment Conditions

Through the analysis of the plug body mechanics model in the process of picking seedlings, the performance of picking seedlings was mainly related to the air pressure, air-blowing time, and water content of the plug body. Taking the air pressure, the water content of the plug, and air-blowing time as the influencing factors of this experiment, the qualified rate of the seedlings and the loss rate of the substrate were investigated as indicators.

The seedling-picking experiment was carried out in the Agricultural Robot Laboratory of Shandong University of Technology. The test objects were pepper seedlings (Zhonghe 5605), using 6 × 6 holes self-made combined seedling trays, and the seedlings were raised in the solar greenhouse of Beijing Zhongnong Futong Horticulture Co., Ltd., Beijing, China. The seedlings were 56 days old, the plant height was 80.3 mm, the stem diameter was 2.6 mm, and the width was 95.1 mm, the diameter of the plug was 28 mm, the height of the plug was 40 mm, and the substrate was peat, vermiculite, and perlite (volume ratio 3:1:1). The equipment used for the experiment was as follows: closed multi-channel air-blown seedling picking device (Figure 7), experimental plug seedlings, plate for catching plug seedlings (material: foam pastic), air compressor (power: 800 W, air volume: 60 L/min, Shengyuan, Taizhou, China), vernier caliper (range: 150 mm, resolution: 0.01 mm, Delixi, Leqing, China), moisture meter (model: JHY-600, load range: 120 g, accuracy ± 0.1%, Xinxiongfa, Xiamen, China), electronic scale (range: 2 kg, accuracy: 0.01 g, Dayang Sensor, Bengbu, China).

2.5. Experiment Method and Index

The qualified rate of picking seedlings and the substrate loss rate were used as the evaluation indexes for picking seedlings. The qualified rate of picking seedlings refers to the ratio of the number of plug seedlings in a single test minus the number of plug seedlings not taken out, the number of scattered plug seedlings, the number of damaged seedlings, and the number of plug seedlings in a single test. The substrate loss rate was the ratio of the weight of the scattered substrate to the total weight of the plug seedlings.

The qualified rate of picking seedlings

$${\eta }_1={\displaystyle \frac{N-U-R-T}{N}}\times 100\%$$

(16)

The substrate loss rate

$${\eta }_2={\displaystyle \frac{M}{P+M}}\times 100\%$$

(17)

where N is the number of plug seedlings in a single test; U is the number of plug seedlings that have not been taken out; R is the number of plug seedlings with stem and leaf damage; T is the number of scattered plug seedlings; M is the weight of the scattered substrate, g; p is the weight of the plug seedlings taken out, g.

2.6. Orthogonal Experiment

To verify the seedling-picking performance of the seedling-picking device, the effects of different combination parameters on the seedling picking performance were studied, and the optimal parameter combination was obtained. According to the single-factor test data, the test factors and levels were selected, and a three-factor and three-level orthogonal test was carried out. Orthogonal table L9 (34) was used for experimental design and analysis, and the levels of experimental factors are shown in

Table 1. Factors and levels of orthogonal experiment.

Level	Factor
	Moisture Content/% A	Seedling Air Pressure/MPa B	Blow Time/ms C
1	20	0.2	30
2	30	0.25	35
3	40	0.3	40

. According to the orthogonal test factor level table, 9 groups of experiments were carried out, and each group was repeated 36 times for a total of 324 experiments.

3. Results and Discussion

3.1. Airflow Simulation Result Analysis

3.1.1. Simulation Analysis of Direct-Blowing Seedling Picking

The direct-blowing nozzle for picking seedlings directly blows air to the plug body, and the continuous high-pressure airflow provides the maintaining air-blowing force. The simulation parameters are set as follows: the inner diameter of the nozzle was 6 mm, the inner diameter of the tray tube (the diameter of the plug body) was 28 mm, The outer diameter of the tray tube was 32 mm, the tray tube was 2 mm higher than the plug body, the diameter of the plug seedling stalk was 2.5 mm, the distance from the nozzle to the plug body was 15 mm, the distance from the nozzle to the seedling stem was 6 mm, and the air pressure for picking the seedling was 0.2 MPa. The simulation results are shown in Figure 8.

It can be seen from Figure 8a,b that the airflow velocity on the surface of the plug on one side of the nozzle was about 100 m/s, and the velocity on the other side was about 40 m/s. The high-speed air blows the plug, causing the substrate to splash. The vertical air- blowing force on the plug body was 5.65 N, but the force on the plug body was not uniform. The force on the plug body on one side of the nozzle was large, and the force on the other side was small (Figure 8c), which will cause the plug seedlings to get stuck in the tray tube. In addition, the open high-speed airflow interferes with the surrounding plug seedlings, causing the non-target plug seedlings to fall off in advance, or the plug seedlings to fall or be damaged. The streamline diagram of the flow field is shown in Figure 8d.

3.1.2. Simulation Analysis of Closed Multi-Channel Air-Blown Seedling Picking

The closed multi-channel air-blowing type of seedling-picking process was simulated. The parameters of the seedling cup are as described in Section 3.1.1. The size of the seedling tube and the plug seedling are the same as those of the direct-blowing seedling-picking simulation. The pressure of the seedling picking at the inlet was 0.2 MPa. The ejection force measurement experiment was performed on plug seedlings with different moisture contents, and the maximum ejection force was measured to be 20 N, so the average air pressure on the upper surface of the plug was set to 0.0328 MPa. The simulation results are shown in Figure 9.

When the compressed air enters the flow channel of the seedling cup, the flow velocity was greater than 85 m/s, and the flow velocity of the air reaching the inside of the cup decreases rapidly. The flow velocity at the outlet of the channel was about 70 m/s, the maximum airflow velocity on the surface of the plug seedling was 50 m/s, and the velocity at the bottom was higher than that at the top (Figure 9a), and the force and airflow velocity on the surface of the plug was relatively uniform. There was turbulent flow in the seedling cup, and the turbulent flow causes certain fluctuations in the stems and leaves, and the force on the upper surface of the plug was evenly distributed (as shown in Figure 9b).

Comparing and analyzing the simulation results of the two types of air-blown seedling picking, the airflow of the closed multi-channel seedling picking was limited in the sealed cavity, the air-blowing force was larger, energy consumption was lower and the airflow did not interfere with the surrounding plug seedlings; the air pressure in the sealed cavity was relatively uniform, and the plug’s upper surface was subjected to uniform air-blowing force; the airflow rate was low, so it was not easy to damage the seedlings.

3.2. Single Factor Test

3.2.1. Effects of Plug Moisture Content and Seedling Air Pressure on Substrate Loss Rate

Increasing the air pressure for seedling-picking can increase the air-blowing force, but the airflow speed in the seedling cup and the moving speed of the plug seedlings during the falling process also increases, which may cause damage to the plug seedlings and substrate damage. To study the relationship between the air pressure with different moisture contents and the substrate loss rate, the single-factor seedling-picking experiments with different moisture contents and the substrate loss rate were carried out. The plug seedlings were uniformly watered and placed in a ventilated place, and the moisture content of the plug body was regularly detected with a moisture meter to obtain the plug seedlings with the required plug body moisture content. The seedling-picking air pressure was adjusted by a precision pressure-regulating valve.

When the water content of the plug body was less than 20%, the plug seedlings were prone to dehydration, which affects the survival rate after planting. The water content of the plug body after watering and draining the water was 70%. The test seedlings were divided into 6 groups according to the moisture content of the plug (20–70%). Each group of test seedlings was subjected to the seedling-picking test under 6 kinds of air pressure conditions, and 12 repeated tests were carried out for each level, with a total of 432 tests. The test results are shown in Figure 10.

The analysis of the test results shows that the substrate loss rate of the plug body increases with the increase in the air pressure. When the air pressure was 0.35 MPa, the substrate loss rate increased significantly. The substrate loss rate with a plug body moisture content of 20% was lower than that of other water content plug seedlings. Therefore, the seedling-picking effect was better when the seedling air pressure was 0.3 MPa and the plug body moisture content was 20%.

3.2.2. The Relationship between the Qualified Rate of Picking Seedlings and the Time of Air Blowing

When the moisture content of the plug body and the air pressure for picking seedlings had been determined, the air blowing time directly affected the efficiency and the qualified rate of picking seedlings. Extending the air blowing time can maintain a certain air-blowing force, but if the air-blowing time was too long, the energy consumption was large, which affected the efficiency of seedling picking, and the plug body would be scattered after the plug seedlings were accelerated and collided with the seedling-receiving device. To study the effect of air blowing time on the performance of picking seedlings, a single-factor seedling picking experiment was performed on the air blowing time. The control parameters are set through the touch screen, and the PLC controls the solenoid valve to adjust the air blowing time. The moisture content of the test seedling plug was 20%, the air pressure of the seedlings was fixed at 0.3 MPa, 8 air blowing times were set, and 12 repeated tests were performed at each level. The test results are shown in Figure 11.

It can be seen from the test results that when the air-blowing time was less than 35 ms, the qualified rate of picking seedlings increased with the increase of air-blowing time; when the air-blowing time was more than 25 ms, the qualified rate of picking seedlings was higher. With the increase of air-blowing time, the qualified rate of picking seedlings did not increase significantly. When the air-blowing time was 30 ms, the requirements for picking seedlings were met, and the qualified rate of picking seedlings was 91.67%.

3.3. Orthogonal Experiment Results and Analysis

The test results are shown in

Table 2. Plan and results of orthogonal experiment.

Test Number	Influencing Factors			Evaluation Indicators
	A	B	C	Qualified Rate η1/%	Loss Rate η2/%
1	1	1	1	97.22	12.33
2	1	2	2	91.67	15.74
3	1	3	3	94.44	17.56
4	2	1	2	91.67	14.23
5	2	2	3	86.11	26.78
6	2	3	1	94.44	18.87
7	3	1	3	88.89	22.33
8	3	2	1	91.67	24.34
9	3	3	2	94.44	19.80

, and the range analysis method and the variance analysis method are used to analyze the test results, as shown in

Table 3. Range analysis.

Index	Factor	K1	K2	K3	R	Better Solution
η1	A	283.33	272.22	275	11.11	A1
	B	277.78	269.45	283.32	13.87	B3
	C	283.33	277.78	269.44	13.89	C1
	Influence	C > B > A
η2	A	45.63	59.88	66.47	20.84	A1
	B	48.89	66.86	56.23	17.97	B1
	C	55.54	49.77	66.67	16.90	C2
	Influence	A > B > C

and

Table 4. Variance analysis.

Index	Source	Sum of Square	Degrees of Freedom	Mean Square	F	p
η1	A	22.28	2	11.14	13.02	0.071
	B	32.50	2	16.25	18.99	0.05
	C	32.59	2	16.29	19.04	0.049
	Error	1.71	2	0.86
	sum	89.08	8
η2	A	75.64	2	37.82	55.51	0.0177
	B	54.42	2	27.21	39.93	0.0244
	C	49.20	2	24.60	36.10	0.027
	Error	1.36	2	0.68
	sum	180.63	8

Note: p ≤ 0.01 means extremely significant, 0.01 < p ≤ 0.05 means very significant, 0.05 < p ≤ 0.1 means significant, and p > 0.1 means not significant.

, respectively. It can be seen from Table 3 and Table 4 that the order of significance of the influence of each factor on each index was consistent. The comprehensive balance method was used to comprehensively analyze the influence of each factor on each index, and a better seedling-picking scheme was obtained as A₁B₃C₁. That is, the water content of the plug was 20%, the air pressure for taking the seedlings was 0.3 MPa, and the air blowing time was 30 ms. The verification test of the optimal scheme was carried out, and the results showed that the qualified rate of seedling picking was 97.22%, and the substrate loss rate was 10.46%. The performance of the optimal scheme for picking seedlings was higher than that of the orthogonal experiments, and its performance meets the requirements for picking seedlings for transplanting vegetable plug seedlings in the field.

3.4. Discussion

The main factors affecting the qualified rate of seedling picking were the failure of picking seedlings and the damage to seedlings. Because the shape of the plug is a cylinder, the mechanical properties of the plug are different, and the bonding force between the plug and the tray is different. There is still a certain air leakage phenomenon in the process of picking seedlings, causing some plug seedlings to fail to pick up seedlings. The main forms of damaged seedlings are loose lumps and leaf damage. When the speed of the plug seedling-picking is high, the plug seedlings collide with the seedling collection tray when they come into contact, resulting in loose lumps. Optimizing the parameters for taking seedlings, reducing the speed at which plug seedlings are separated from the plug tray, and promoting the root development of plug seedlings can reduce the rate of scattered lumps. When the plug seedlings widen greatly, the seedling cup presses some of the bottom leaves of the plug seedlings, causing some leaves to be missing. This can be solved by the following solutions: (1) Improving the sowing accuracy, sowing the seeds to the center of the hole, and controlling the spread of the plug seedlings. (2) Before the true leaves of the plug seedlings are unfolded, installing a transparent grid-like isolation cylinder on the plug tray to constrain the stems and leaves of the plug seedlings, which can avoid damage to the stems and leaves of the plug seedlings when taking the seedlings. Figure 12a shows the seedlings without the use of the isolation tube, and Figure 12b with the use of the isolation tube.

4. Conclusions

• A closed multi-channel air-blowing seedling-picking device was designed, consisting mainly of a combined plug tray, tray feeding device, seedling-picking mechanism, and a seedling-guiding tube. Controlled by a PLC touch screen system, these components work together to achieve low seedling injury, reduced energy consumption, and high-speed seedling picking and placement.
• A dynamic analysis model of plug seedlings was developed to study their movement during the seedling-picking process. Key parameters affecting seedling performance were identified, including the moisture content of the plug, the air pressure for seedling extraction, and the air-blowing duration.
• CFD simulation was conducted for both the closed multi-channel air-blown seedling picking and direct-blowing methods. The results showed that the maximum air velocity on the surface of the closed-type seedling picking plug was 50 m/s, significantly lower than the 100 m/s in the direct-blowing method, resulting in more uniform force distribution on the plug.
• Single-factor test results indicated that when the moisture content of the plug was 20%, air pressure for seedling picking was 0.3 MPa, and air-blowing time exceeded 30 ms, the qualified seedling-picking rate was higher. The orthogonal test results revealed that air-blowing time had a very significant effect on the qualified seedling-picking rate, while the moisture content of the plug and seedling picking pressure had a significant effect. The optimal parameter combination for seedling picking was a plug moisture content of 20%, air pressure of 0.3 MPa, and an air-blowing time of 30 ms. Under these conditions, the qualified seedling picking rate reached 97.22%, and the substrate loss rate was 10.46%, meeting the requirements for pepper seedling picking.