Design and Test of a Straw-Clearing-Depth Self-Adaptive Control System of a Front-Mounted Seedbed-Preparation Device

Abstract

In northeast China, most seedbed-preparation devices use the ground-wheel profiling method to ensure their operational stability. However, during the wide-width operation of the front-mounted seedbed-preparation device, the poor trafficability characteristics and the low profiling accuracy of the ground-wheel profiling mechanism result in unstable straw clearing depth, poor straw clearing quality, and the low operational efficiency of the seedbed-preparation device. In order to solve the above problems, a straw-clearing-depth self-adaptive control system of a front-mounted seedbed-preparation device was designed. The key structural design of the self-adaptive control system was completed through theoretical analysis. The performance test results of the self-adaptive control system showed that the lifting speed of the front-suspension mechanism was greater than 0.2 m/s in the manual button control mode, and the relative error between the target value and the actual value of the straw clearing depth was 10.8% under the self-adaptive profiling control mode. The three-factor and five-level quadratic regression orthogonal rotation center combination test method was adopted to conduct a parameter combination optimization test, with the machine operation speed, the operation depth of the straw clearing knife, and the straw covering amount as test factors, and the straw clearing rate, the qualified rate of operation depth, and the consistency of straw clearing between rows as evaluation indices. The results indicated that when the machine operation speed was 5~8.8 km/h, the operation depth of the straw clearing knife was 50 mm, the straw covering amount was 0.9~1.44 kg/m², the straw clearing rate was ≥86%, the qualified rate of operation depth was ≥86%, and the consistency of straw clearing between rows was ≥83%. Field tests were carried out on the machine using operation speeds of 5 km/h, 6 km/h, 7 km/h, and 8 km/h under the conditions of an operation depth of the straw clearing knife of 50 mm and a straw covering amount of 1.2 kg/m². The results showed that the straw clearing rate, the qualified rate of operation depth, and the consistency of straw clearing between rows were all within the optimized range under different machine operation speeds, which was basically consistent with the optimized results.

1. Introduction

Conservation tillage is an advanced agricultural tillage technology that facilitates direct sowing on straw-covered land through the methods of no-tillage seeding and reduced-tillage seeding. Conservation tillage has the advantages of reducing wind and water erosion, improving soil structure, increasing soil organic-matter content, and increasing crop yield [1,2]. It is also an important initiative to protect national food security and plays a key role in ecological environmental protection and agricultural sustainable development [3,4]. As a product of the reform and development of the traditional farming system, compared with the conventional ridge-planting model, the large-ridge double-row planting model can improve ventilation conditions, reduce water evaporation, strengthen soil drought resistance and moisture preservation, and increase seedling emergence rate. The combination of conservation tillage technology and the large-ridge double-row planting mode is one of the important methods to ensure the sustainable growth of grain yield in Heilongjiang [5,6,7].

No-tillage seeding technology is the core of conservation tillage promotion and application, and the straw treatment of seedbeds is the premise and basis for no-tillage seeding. The straw covering on the ground can effectively regulate ground temperature, inhibit water evaporation, and improve soil fertility, crop quality, and yield [8,9,10]. The seedbed-preparation device of the 2BMFJ-series straw clearing and covering no-tillage seeder adopts the method of straw lateral throwing to achieve the construction of high-quality seedbeds and even straw coverage after sowing [11,12]. At present, the research on the seedbed-preparation device of the 2BMFJ-series straw clearing and covering no-tillage seeder mainly focuses on the design of key components and parameter combination optimization of small-range operation machines to achieve the purpose of improving the straw clearing rate and reducing power consumption and machine vibration [13,14]. However, the problems of unstable straw clearing depth and different straw clearing effects between rows in wide-range operations have not been investigated. The straw clearing depth of a seedbed-preparation device (the operation depth of the straw clearing knife) is a key index to measure seedbed-preparation quality. If the operation depth of the straw clearing knife is too shallow, crop root stubble will remain on the ground surface, which can easily cause the blockage of the fertilizing and seeding openers; if the operation depth of the straw clearing knife is too deep, the soil disturbance and power consumption of the machine will be increased [15]. The profiling mechanism of the seedbed-preparation device is the key component to ensure the stability of the straw clearing depth. In order to solve the problem of current seeders being unable to sow normally in wide-range operations, Chen et al. [16] designed a front-mounted large-ridge double-row seedbed-preparation device. The sowing and fertilizing device was attached to the rear of a tractor, which greatly improved the utilization rate of existing seeders. However, its profiling mechanism adopted the method of ground-wheel profiling, and the height of the ground wheel could be adjusted by turning the screw, which could not ensure the consistency in the operation depth of the seedbed-preparation device’s straw clearing knife during the wide-width operation. Moreover, the front-mounted ground wheel compacts the soil and makes it difficult to turn the direction, resulting in a poor straw clearing effect and low operation efficiency of existing seedbed-preparation devices.

The profiling mechanism is a key component to ensure the straw clearing depth stability of the seedbed-preparation device. At present, the research on the profiling mechanism of the no-tillage seeder mainly focuses on the control of the furrow opening and seeding depth; however, there is little involvement in the straw clearing depth control of the seedbed-preparation device. The profiling mechanism is mainly mechanical, and it includes the ground wheel, the depth-limiting wheel, the adjusting spring, and the parallel four-linked rod. The profiling power comes from the device’s own gravity and the ground-support force; the profiling accuracy is low and the adaptability to operation conditions is poor. So, the methods of mechanical profiling cannot meet the requirements of a wide-width operation of a seedbed-preparation device.

Electro-hydraulic profiling can solve the existing problems in ground-wheel profiling during the wide-width operation of a seedbed-preparation device. Electro-hydraulic profiling is mainly composed of a ground-profiling mechanism, a hydraulic system, and a control system. The ground-profiling mechanisms can be divided into contact types and remote-sensing types. The remote-sensing profiling mechanism mostly uses ultrasonic sensors and laser distance sensors to obtain surface information. Compared with the mechanical profiling, it has the characteristics of a short response time and high profiling accuracy. However, for an original stubble field with a large amount of straw, the remote-sensing profiling mechanism can be easily affected by straw, temperature, and moisture. The contact profiling mechanism uses a combination of mechanical profiling components and angle sensors or displacement sensors to obtain surface information, which can effectively solve the problems that exist in remote-sensing profiling.

In order to solve the above problems, based on the seedbed-preparation device of the 2BMFJ-6DL-type straw clearing and covering no-tillage seeder, a straw-clearing-depth self-adaptive control system of a front-mounted seedbed-preparation device, based on fuzzy PID control, was designed. The key parameters affecting the main structure of the self-adaptive control system were determined and the quadratic regression orthogonal rotating center combination test method was used to explore the optimal parameter combination affecting the system’s working performance.

2. Materials and Methods

2.1. Structure and Operation Principle

The straw-clearing-depth self-adaptive control system of a front-mounted seedbed-preparation device (hereinafter referred to as the self-adaptive control system) is designed with the seedbed-preparation device of 2BMFJ-6DL-type straw clearing and covering no-tillage seeder as the carrier. The system overall structure and assembly relationship are shown in Figure 1, which is mainly composed of three parts: seedbed-preparation device, hydraulic system, and self-adaptive control system. The self-adaptive control system is composed of front-suspension mechanism, profiling mechanism, and a control system of three parts. Among them, the hydraulic pump of the hydraulic system is fixed at the rear of the tractor and connected with the power output shaft of the tractor through the coupling, providing power for the front-suspension hydraulic system. The seedbed-preparation device is connected to the front of the tractor through the front-suspension mechanism, and the sowing and fertilizing device is connected with the rear of the tractor, which together constitute the no-tillage and precision-sowing compound operation unit.

During no-tillage sowing operation, the straw and stubble on the surface of the original stubble field are moderately crushed by the seedbed-preparation device and then thrown to the side of the moving direction to create a seedbed without straw cover and stubble residue for the sowing and fertilizing device. The fertilizing and sowing furrow openers open ditches and complete fertilization and sowing operations; then, the soil is covered before moderate repression is carried out to the seeding belt. During the return operation, the seedbed-preparation device then throws the straw onto the surface of the sown seedbed to achieve an even straw covering after sowing [17,18,19]. The operation principle of the self-adaptive control system is shown in Figure 2. The control system is mainly composed of STM32 control unit, solenoid valve, and its drive module and button module. With the help of the hydraulic system and the control system, the profiling mechanism can detect and adjust the straw clearing depth of the seedbed-preparation device in real time; that is, the operation depth of straw clearing knife. Before the operation of the machine, the front-mounted seedbed-preparation device is adjusted through the button module to make the straw clearing knife rise and fall to the target operation depth. In the process of operation, the profiling mechanism converts the operation depth signal of the straw clearing knife into its own rotation angle signal and sends it to the control system. The control system controls the opening and closing of the solenoid valve according to the signal received. When the left side of electromagnetic directional valve is energized, the pressure of no-rod chamber of the hydraulic cylinder increases, pushing the piston rod to the right, and then the seedbed-preparation device drops; when the right side of electromagnetic directional valve is energized, the pressure of rod chamber of the hydraulic cylinder increases, pushing the piston rod to the left, and then the seedbed-preparation device rises, so as to achieve the effect of real-time detecting and adjusting the straw clearing depth of the seedbed-preparation device.

The front-mounted seedbed-preparation device is equipped with a 6.6 m wide precision seeder to complete the compound operation of no-tillage sowing. The operation width of the machine is large, and the operation environment of each seeding row is complicated. In order to ensure the operation quality, as shown in Figure 1, three sets of profiling mechanisms are uniformly arranged on the seedbed-preparation device. If more than two sets of profiling mechanisms detect that the distance between the seedbed and the seedbed-preparation device decreases, that is, the operation depth of straw clearing knife increases, the seedbed-preparation device will be raised upward to prevent the increase in power consumption and avoid structural damage of the device at the same time. If three sets of profiling mechanisms simultaneously detect that the distance between the seedbed and the seedbed-preparation device increases, that is, the operation depth of straw clearing knife decreases, the seedbed-preparation device will be fallen to ensure the same seedbed preparation effect of different sowing rows [20].

2.2. Design of Self-Adaptive Control System

2.2.1. Design of Front-Suspension Mechanism

Based on the New Holland 1104-type tractor, according to relevant requirements in the national standard GB/T 10916-2003 for the front-suspension mechanism design, the front-suspension mechanism structure is designed. The overall structure of the front-suspension mechanism is shown in Figure 3a, which is mainly composed of the front-suspension frame, lower tie rod, upper tie rod, and hydraulic cylinder.

A rectangular coordinate system is established, as shown in Figure 3b, which takes the center O of the tractor’s front wheel as the origin, the forward direction as the positive direction of the x axis, and the vertical direction as the positive direction of the y axis. Two lower tie rods are hinged with the front-suspension frame at point B through the hinge pin. The hydraulic cylinder is hinged with the front-suspension frame and the lower tie rod at points A and C. According to the transportation and operation requirements of the seedbed-preparation device, the piston rod of hydraulic cylinder adopts the method of downward arrangement, and the stroke of hydraulic cylinder is 190 mm. The dimension parameters of the front-suspension mechanism are shown in

Table 1. Dimension parameters of the front-suspension mechanism.

Parameter	Value
L0/mm	578
e/mm	70
S/mm	566–756
LBC/mm	350
LBD/mm	820
α/°	6.91

Note: α is the deflection angle of the upper hinge point A of the front-suspension frame relative to the lower hinge point B of the front-suspension frame (counterclockwise is positive).

.

During the operation of seedbed-preparation device, the straw clearing knife enters a certain depth into the soil to ensure stubble crushing, conveying, and throwing. Its driving force is provided by the machine itself. The distance between the seedbed-preparation device and the ground surface is affected by the height from the hinge point B of lower tie rod to the ground surface. If the height from the hanging point D of lower tie rod to the ground surface is greater than the height from the hinge point B of lower tie rod to the ground surface, the lifting force of the front-suspension mechanism on the seedbed-preparation device will increase, which makes it difficult for the straw clearing knife to enter the soil, which affects the straw clearing effect.

In order to ensure that the seedbed-preparation device maintains a horizontal state during operation and meets the requirements that the height from the hanging point D of lower tie rod to the ground surface is lower than that from the hinged point B of lower tie rod to the ground surface, as shown in Figure 4a, the dimension relation of the front-suspension mechanism meeting operation requirements is:

$$\{\begin{array}{l}{H}_1\ge R-H_2-H_3\\ L_{\mathrm{AE}}\ge \sqrt{{\left(L_{\mathrm{ED}}-L_{\mathrm{BD}}\sin \tau -L_0\right)}^2+{\left(L_{\mathrm{BD}}\cos \tau -e\right)}^2}\end{array}$$

(1)

where τ is the angle between lower tie rod and horizontal plane, °; H₁ is the height from hinged point of lower tie rod to ground surface, mm; H₂ is height from rotation center of straw clearing knife to hanging point of lower tie rod, mm; H₃ is operation depth of straw clearing knife, mm; R is rotation radius of straw clearing knife, mm; L_AE is the length of upper tie rod, mm; and L_ED is vertical height of upper and lower hanging points of seedbed-preparation device, mm.

Referring to the existing seedbed-preparation device [18], the operation depth of straw clearing knife is designed to be 50 mm. The rotation radius of the straw clearing knife is 380 mm. The rotation center of the straw clearing knife is 100 mm higher than the hanging point of the lower tie rod. By substituting the known parameter values into Equation (1), the height from the hinge point B of lower tie rod to the ground surface is not less than 238 mm, and the maximum adjustment length of the upper tie rod is more than 765 mm, meeting the design requirements.

In addition, during the transportation of the seedbed-preparation device, the height from straw clearing knife to ground surface should meet the actual transportation requirements, so as to avoid bumping when passing through uneven road surface. The transportation state of the seedbed-preparation device is shown in Figure 4b. At this time, the hydraulic cylinder retracts to the shortest length, and the lower tie rod raises to the limit position. Additionally, the dimension relation of the front-suspension structure meeting the transportation requirements is:

$$\{\begin{array}{l}{\tau }_1+{\tau }_2+{\tau }_3={90}^{\circ }+{\tau }_0\\ H_4=H_1+L_{\mathrm{BD}}\sin {\tau }_0+L_1\sin {\tau }_1-\left(R-H_2\right)\cos {\tau }_1\end{array}$$

(2)

where τ₀ is the angle between lifting limit position of lower tie rod and horizontal plane, °; τ₁ is the angle between upper and lower hanging point of seedbed-preparation device and the vertical direction, °; τ₂ is the angle between the lower tie rod and the line connecting point A and point D, °; τ₃ is the angle between upper and lower hanging point of seedbed-preparation device and the line connecting point A and point D, °; L₁ is the distance from the nearest straw clearing knife to the hanging point of lower tie rod, mm; and H₄ is the height between the straw clearing knife and ground surface, mm.

The distance of the nearest straw clearing knife to the hanging point of lower tie rod is 90 mm. When the height of the straw clearing knife to ground surface is 350 mm, the height of hinged point B of lower tie rod to the ground surface is greater than 435 mm, meeting the design requirements.

To sum up, when the height from the hinge point B of lower tie rod to the ground surface is 550 mm, the distance of the straw clearing knife to the ground surface during transportation is 465 mm, which meets the actual transportation conditions.

2.2.2. Design of Profiling Mechanism

The profiling mechanism is a key component of the self-adaptive control system to detect the changes of the straw clearing depth in real time. As shown in Figure 5, it is mainly composed of a profiling plate, fixed shaft, angle sensor, and connecting piece. The profiling plate is firmly connected with the fixed shaft, and a certain torque is applied at the fixed shaft to make the profiling plate attach to the ground surface and rotate with the fluctuation of the ground surface. The straw clearing depth of the seedbed-preparation device can be obtained by detecting the rotation angle of the fixed shaft.

During machine operation, the self-adaptive control system adjusts the straw clearing depth in real time according to the angle change of the profiling plate, and it stabilizes the straw clearing depth at 50 mm. According to the structure dimension relationship of the profiling mechanism, the angle between the profiling plate and ground surface is designed to be 45° during normal operation. The fluctuant range of the seedbed-preparation device is set to be 20 mm, that is, the rotation angle of the profiling plate around the fixed axis varies from 36° to 53°. Due to the complex operating environment, the contactless HT-CE100 switch Hall-integrated sensor was selected to obtain real-time information of the angle change of profiling plate to avoid damage to the sensor caused by the impact of profiling mechanism during operation. For the purpose of ensuring profiling effect and avoiding profiling plate falling into the soil due to its own gravity, the length of profiling plate was designed to be 180 mm, while the cutting-edge angle is 30°, the shape is circular, and the material is PVC plastic.

The operation effect of the profiling plate is related to the torque at the fixed shaft and the straw on the ground. Therefore, the force analysis of the profiling plate and corn stalk was carried out, respectively. A rectangular coordinate system is established, as shown in Figure 6, with the center O_f of the fixed shaft as origin, the moving direction of profiling plate as x-axis, the vertical direction as y-axis, and the direction perpendiculars to x-axis and y-axis as z-axis. According to the force analysis of the profiling plate in the xo_fy plane, which is shown in Figure 6a, the equilibrium equation can be obtained as follows.

$$F_{\mathrm{n}0}{L}_{\mathrm{f}}\sin {\alpha }_{\mathrm{f}}+F_{\mathrm{fy}}{H}_{\mathrm{y}}\tan {\alpha }_{\mathrm{y}}+F_{\mathrm{f}0}{L}_{\mathrm{f}}\cos {\alpha }_{\mathrm{f}}+F_{\mathrm{fx}}{H}_{\mathrm{y}}=M_{\mathrm{f}}+\frac{1}{2}{G}_{\mathrm{f}}{L}_{\mathrm{f}}\sin {\alpha }_{\mathrm{f}}$$

(3)

where L_f is the length of profiling plate, mm; α_f is the angle between the profiling plate and the vertical direction in normal operation, °; α_y is the angle between the contact point of profiling plate and corn straw and the vertical direction, °; F_n1 is supporting force of fixed shaft against the profiling plate, N; F_n0 is supporting force of ground surface against the profiling plate, N; F_fy is component force of corn straw on the profiling plate along y-axis direction, N; F_fx is component force of corn straw on the profiling plate along x-axis direction, N; G_f is weight of profiling plate, N; H_y is the height of contact point between profiling plate and corn straw from fixed shaft, mm; F_f0 is the friction of ground surface against profiling plate, N; and M_f is the torque applied at fixed shaft, N·mm.

During machine operation, the profiling effect of the profiling plate is affected by the straw on the ground surface. If the torque applied at fixed shaft is not enough to make the profiling plate attach to the ground surface, the accuracy of control system information detection will be affected. In order to eliminate the influence of the straw on the profiling effect, the component force of profiling plate on the corn straw along z axis should be greater than the friction force of ground surface against straw. When the contact between the profiling plate and the ground surface is in a critical position status due to the action of straw, as shown in Figure 6b, Equation (4) can be obtained by the force analysis of straw in the yo_fz plane.

$$\{\begin{array}{l}{F}_{\mathrm{n}2}=G_{\mathrm{y}}+F_{\mathrm{fy}}^{\prime }\\ F_{\mathrm{fz}}^{\prime }=\frac{F_{\mathrm{fy}}^{\prime }}{\tan {\alpha }_{\mathrm{j}}}\\ F_{\mathrm{f}2}=F_{\mathrm{n}2}{\mu }_0\end{array}$$

(4)

where F_n2 is the supporting force of ground surface against corn straw, N; F_f2 is the friction force of ground surface against corn straw, N; G_f is the weight of corn straw, N; F′_fy is the component force of profiling plate on corn straw along y-axis direction, N; F′_fz is the component force of profiling plate on corn straw along z-axis direction, N; α_j is the cutting edge angle of profiling plate, °; and μ₀ is the friction coefficient between corn straw and ground surface.

Combining Equations (3) and (4), the torque value applied at the fixed shaft can be obtained through Equation (5).

$$M_{\mathrm{f}}=\frac{G_{\mathrm{y}}{\mu }_0{H}_{\mathrm{y}}\left(\tan {\alpha }_{\mathrm{y}}+{\mu }_0\right)}{\left(\frac{1}{\tan {\alpha }_{\mathrm{j}}}-{\mu }_0\right)}-\frac{1}{2}{G}_{\mathrm{f}}{L}_{\mathrm{f}}\sin {\alpha }_{\mathrm{f}}+H_{\mathrm{y}}{\mu }_0{G}_{\mathrm{y}}$$

(5)

According to the actual measurement results in the field and referring to relevant literature [21], the Dannuo No. 6 corn straw crushed by the combined harvester is taken as the research object. The average length of corn straw is 210 mm, the straw density is 0.24 g/cm³, and the friction coefficient between corn straw and ground surface is 0.49, and we substituted them into Equation (5). When the torque applied at the fixed shaft is 10 N·mm, the profiling mechanism meets the design requirements.

2.2.3. Design of Control System

The fuzzy PID control system is a combination of fixed-parameter PID controller and the fuzzy controller. The parameter value of fixed-parameter PID control is fixed, and it can hardly meet the requirements of control accuracy, response speed, and anti-interference ability under the situation of changeable operating environment, large linear error, and strong time variability [22]. The control algorithm of fixed parameter PID control is shown in Equation (6).

$$U\left(t\right)=K_{\mathrm{p}}e\left(t\right)+K_{\mathrm{i}}{\displaystyle {\int }_0^{t}e\left(t\right)}dt+K_{\mathrm{d}}\frac{de\left(t\right)}{dt}$$

(6)

where U(t) is system output; e(t) is system input; K_p is proportionality coefficient; K_i is integration coefficient; and K_d is differential coefficient.

The effect of fixed-parameter PID control depends on the accuracy of the mathematical model. The front-mounted seedbed-preparation device is used for wide-range operation. the operating environment is complex, and the undulating changes of the ground surface have no specific laws. So, it is impossible to establish an accurate model of the action relationship between the straw clearing depth and the profiling mechanism sensor. Additionally, the problems of overshoot or lag are easy to occur in the control process, which means the control system cannot receive and process the signals from the sensors in time, affecting the stability of system and the operation reliability of machine. Therefore, the fuzzy idea is added on the basis of fixed-parameter PID control. First, the control system obtains the current actual value of the straw clearing depth through the sensor and compares it with the set target value. Then, the system error and error change rate of the straw clearing depth obtained by the comparison would be fuzzy processed. Later, the PID parameters will be adjusted according to fuzzy rule reasoning [23]. Finally, the accurate acquisition and rapid response of straw clearing depth of the front-mounted seedbed-preparation device will be achieved.

The diagram of the fuzzy PID control system for straw clearing depth of front-mounted seedbed-preparation device is shown in Figure 7. The straw clearing depth h_f of front-mounted seedbed-preparation device can be calculated by the rotation angle of profiling plate around the fixed axis. Let the h_f be the straw clearing depth actual value of fuzzy PID control system and h be the set target value of straw clearing depth. We can obtain the error of straw clearing depth e = h_f − h and the error change rate ∆e = de/dt.

The error e and the error change rate de of the seedbed-preparation device straw clearing depth are taken as input variables of the fuzzy PID control system, and the on-line correction values ∆K_P, ∆K_I, and ∆K_D of the proportionality coefficient K_P, integration coefficient K_I, and differential coefficient K_D, respectively, are taken as output variables of the fuzzy controller. In this paper, the fluctuant range of the operation depth of straw clearing knife during normal operation is designed to be 20 mm. Therefore, the basic theoretical domains of error e and the error change rate de of straw clearing depth are set as [−20, 20] and [−100, 100], respectively. The domain of output variables K_P is [−300, 300], and the domains of K_I and K_D are both [−3, 3]. The domains of input language variables E and EC and output language variables are all [−3, 3]. Therefore, the quantization factors of input variables e and de are K_E = 3/20 = 0.15 and K_EC = 3/100 = 0.03, respectively. The proportional factors of output variables ΔK_P, ΔK_I, and ΔK_D are U_P = 100 and U_I = U_D = 1, respectively.

The fuzzy PID control system needs to find out the fuzzy relationship between three parameters and the error and the error change rate, and constantly detect the error and the error change rate according to the fuzzy control principle during operation, so as to modify three parameters online [24,25]. Based on a large number of tests, the input and output of the self-adaptive control system were tested, and the fuzzy control rules were established. The domain of input language variables E and EC and the output language variables ΔK_P, ΔK_I, and ΔK_D are divided into seven fuzzy subsets. The fuzzy subset corresponding to the input language variable E is taken as {NB, NM, NS, ZO, PS, PM, PB}, indicating that the actual straw clearing depth relative to the set value is “extremely high”, “very high”, “high”, “just right”, “low”, “very low”, and “extremely low”, respectively. The fuzzy subset corresponding to the input language variable EC is taken as {NB, NM, NS, ZO, PS, PM, PB}, representing the variation trend of the straw clearing depth errors “extremely rapid increasing”, “rapid increasing”, “increasing”, “unchanged”, “decreasing”, “rapid decreasing”, and “extremely rapid decreasing”, respectively. The fuzzy subset corresponding to the output language variables is taken as {NB, NM, NS, ZO, PS, PM, PB}, representing “positive large”, “positive medium”, “positive small”, “zero”, “negative small”, “negative medium”, and “negative large”, respectively. The fuzzy rules were established [26,27], as shown in

Table 2. Fuzzy PID control rules.

E	EC
	NB	NM	NS	ZO	PS	PM	PB
NB	PB/NB/PS	PB/NB/NS	PM/NM/NB	PM/NM/NB	PS/NS/NS	ZO/ZO/NM	ZO/ZO/PS
NM	PB/NB/PS	PB/NB/NS	PM/NM/NB	PS/NS/NM	PS/NS/NM	ZO/ZO/NS	NS/ZO/ZO
NS	PM/NB/ZO	PM/NM/NS	PM/NS/NM	PS/NS/NM	ZO/ZO/NS	NS/PS/NS	NS/PS/ZO
ZO	PM/NM/ZO	PM/NM/NS	PS/NS/NS	ZO/ZO/NS	NS/PS/NS	NM/PM/NS	NM/PM/ZO
PS	PS/NM/ZO	PS/NS/ZO	ZO/ZO/ZO	NS/PS/ZO	NS/PS/ZO	NM/PM/ZO	NM/PB/ZO
PM	PS/ZO/PB	ZO/ZO/PS	NS/PS/PS	NM/PS/PS	NM/PM/PS	NM/PB/PS	NB/PB/PB
PB	ZO/ZO/PB	ZO/ZO/PM	NM/PS/PM	NM/PM/PM	NM/PM/PS	NB/PB/PS	NB/PB/PB

.

2.3. Test Scheme Design

The field test was carried out in the Xiangyang Experimental Practice Base of Northeast Agricultural University in September 2022, as shown in Figure 8. The previous crop in the test field was corn Dannuo No. 6. The average soil firmness of 0~5 cm soil layer was 28.3 kg/m², the average soil moisture content of 0~5 cm soil layer was 21.5%, and the straw covering amount was 1.25 kg/m². The test equipment included New Holland 1104 tractor (Shanghai New Holland Agricultural Machinery Co., Ltd., Shanghai, China), TZS-1 soil moisture meter (Zhejiang Topunnong Technology Co., Ltd., Zhejiang, China), SZ-3 soil hardness meter (Zhejiang Topunnong Technology Co., Ltd., Zhejiang, China), ACS-30 electronic scale (Guangdong Juheng Precision Measurement & Control Co., Ltd., Guangdong, China), etc. The test was divided into two parts, including the performance test of self-adaptive control system and the field test of whole machine.

2.3.1. Performance Test of the Self-Adaptive Control System

(1)

Manual button control performance test of the front-suspension mechanism

The self-adaptive control system was installed in the specified position. The lifting sensitivity test of the device was carried out through the button module. The stopwatch was used to record the time used by the seedbed-preparation device to lift the specified height. Manual button control performance test of the front-suspension mechanism was carried out with the lifting distance as test factor and the lifting time as the system response index.

(2)

Profiling performance test of the self-adaptive control system

The function of the self-adaptive control system is to stabilize the straw clearing depth of straw clearing knife at about 50 mm. In order to test the profiling performance of the seedbed-preparation device in stable operation, the profiling performance test of self-adaptive control system was carried out. The operation depth of straw clearing knife was manually adjusted through button module before operation. Then, the actual operation depth of knife after stable operation was measured and compared with the target value. The relative error between the actual depth and the target depth was used as the evaluation index of the self-adaptive control system.

2.3.2. Whole-Machine Field Test

The front-mounted seedbed-preparation device was suitable for wide-range operation, and the environmental factors, such as straw amount, varied greatly between different rows. In order to verify the overall performance of self-adaptive control system in actual operation, the three-factor and five-level quadratic regression orthogonal rotation center combination test was adopted to conduct parameter combination optimization test, with the machine operation speed, operation depth of straw clearing knife, and straw covering amount as test factors, and the straw clearing rate, qualified rate of operation depth, and consistency of straw clearing between rows as evaluation indices. According to the 2BMFJ-series straw clearing and covering no-tillage seeder and actual-production experience, the machine operation speed was selected to be 5~10 km/h, and the operation depth of straw clearing knife was 20~80 mm. The straw covering amount in the field is related to various factors, such as corn varieties. Based on the literature [28] and the actual measurement, the range of the straw covering amount was determined to be 0.8~1.6 kg/m². Test factor levels were coded as shown in

Table 3. Test-factor level coding table.

Test Factors		Coded Value
		−1.68	−1	0	1	1.68
X1	Operation speed/(km·h−1)	5	6	7.5	9	10
X2	Operation depth of straw clearing knife/(mm)	20	32	50	68	80
X3	Straw covering amount/(kg·m−2)	0.8	0.96	1.2	1.44	1.6

.

2.3.3. Measurement of Evaluation Indices

(1)

Straw clearing rate

Straw clearing rate (Y₁) refers to the ratio of total straw amount before and after operation in the operation area of the unit group of seedbed-preparation device. A five-point sampling method is used to determine straw sampling before and after the operation. The sampling point area is 1 × 0.5 m. The straw clearing rate can be expressed as:

$$Y_1=\frac{1}{3}{\displaystyle \sum _{\mathrm{i}=1}^3\frac{m_0-m_1}{m_0}}\times 100\%$$

(7)

where Y₁ is straw clearing rate, %; m₀ is straw amount before operation in the test area, kg/m²; and m₁ is straw amount after operation in the test area, kg/m².

(2)

Qualified rate of operation depth

The qualified rate of operation depth (Y₂) refers to the ratio of the operation depth of straw clearing knife reaching the set range within the operation range of machine. Adjust the operation depth of straw clearing knife before the machine operates, then measure the operation depth of straw clearing knife every 1.5 m after machine operates steadily.

(3)

Consistency of straw clearing between rows

Consistency of straw clearing between rows (Y₃) is used to describe the difference of straw clearing rate of each row in machine operation range. There are differences in straw covering amount and terrain variation between adjacent rows, which resulted in different amounts of straw clearing per unit time and affected the quality of seedbed preparation. The consistency of straw clearing between rows is measured after the completion of seedbed preparation (the ground surface with length × width of 180 × 500 mm is randomly selected to measure the straw covering amount, and the average value of 5 measurements is as the final result). The consistency of straw clearing between rows could be expressed as:

$$Y_3=1-\sqrt{{\displaystyle \sum _{i=1}^6{\left(6W_{\mathrm{i}}-{\displaystyle \sum _{i=1}^6{W}_{\mathrm{i}}}\right)}^2}/6{\left({\displaystyle \sum _{i=1}^6{W}_{\mathrm{i}}}\right)}^2}$$

(8)

where,

$$W_{\mathrm{i}}=\frac{{\displaystyle \sum _{\mathrm{j}=1}^5{W}_{\mathrm{ij}}}}{5},\left(\mathrm{i}=1,2,3,4,5,6\right)$$

where, Y₃ is consistency of straw clearing between rows, %; W_i is covering amount of corn straw after clearing in the corresponding test area of row i, kg; and W_ij is covering amount of corn straw sampled at the jth time in the corresponding test area of row i, kg.

3. Results

3.1. Performance Test of Self-Adaptive Control System

The manual button control performance test of the front-suspension mechanism was carried out with the lifting distance as the test factor and the lifting time as the system response index. A total of eight groups of tests were carried out. Each group of tests was repeated three times, and the average value was used as the final test result. The test results are shown in

Table 4. Manual control test results of front-suspension mechanism.

Test Numbers	Lifting Height/mm	Response Time/s	Lifting Speed/(m/s)
1	50	0.17	0.29
2	−50	0.19	0.26
3	80	0.29	0.27
4	−80	0.35	0.23
5	110	0.42	0.26
6	−110	0.47	0.23
7	140	0.58	0.24
8	−140	0.64	0.22

Where “−” represented the descending height of the seedbed-preparation device.

.

It could be seen from the test results that the time of lifting the same distance was basically the same, and the lifting speed was higher than 0.2 m/s, which met the requirements of machine operation.

According to the 2BMFJ-series straw clearing and covering no-tillage seeder and actual-production experience, the operation depth of the straw clearing knife was 20~80 mm. The profiling performance test results of the self-adaptive control system are shown in

Table 5. Profiling performance test results of self-adaptive control test.

Test Numbers	Operation Depth/mm	Target Depth/mm	Actual Depth/mm	Relative Error/%
1	20	50	41.2	17.6
2	30	50	43.1	13.8
3	40	50	45.8	8.4
4	50	50	47.9	4.2
5	60	50	52.7	5.4
6	70	50	55.8	11.6
7	80	50	57.3	14.6

. From the test results, we could see that when the target depth of the straw clearing knife was 50 mm, the relative error between the actual depth and the target depth was 10.8% under different operation depth conditions after stable operation, which indicated that the self-adaptive control system had a great sensitivity and high stability, meeting the requirements of field operation.

3.2. Whole-Machine Field Test Results

According to the three-factor and five-level quadratic regression orthogonal rotation center combination test, a total of 23 groups of tests were carried out. Each group of tests was repeated three times, and the average value was the final test result, as shown in

Table 6. Test scheme and results.

Number	Test Factors			Evaluation Indexes
	X1/ (km/h)	X2/ (mm)	X3/ (kg/m2)	Y1/(%)	Y2/(%)	Y3/(%)
1	−1	−1	−1	85.9	86.3	81.8
2	1	−1	−1	85.2	87.5	80.3
3	−1	1	−1	87.1	87.6	83.4
4	1	1	−1	88.2	87.2	82.6
5	−1	−1	1	87.1	88.1	81.9
6	1	−1	1	84.6	87.1	80.7
7	−1	1	1	90.2	87.4	82.6
8	1	1	1	88.4	85.4	81.8
9	−1.68	0	0	86.9	88.1	83.1
10	1.68	0	0	85.7	87.2	81.5
11	0	−1.68	0	84.9	86.4	80.2
12	0	1.68	0	89.1	85.7	82.7
13	0	0	−1.68	87.2	88.9	82.4
14	0	0	1.68	88.2	88.1	81.2
15	0	0	0	89.1	89.4	83.9
16	0	0	0	90.1	89.4	84.3
17	0	0	0	89.5	89.1	84.6
18	0	0	0	89.4	89.2	84.1
19	0	0	0	89.3	89.5	83.7
20	0	0	0	89.3	88.7	83.9
21	0	0	0	89.1	89.1	83.3
22	0	0	0	89.2	89.4	83.6
23	0	0	0	89.5	89.3	84.5

.

Design-expert 8.0.6 was applied to conduct variance analysis on the test results, as shown in

Table 7. Variance analysis.

Source Variance	Y1/(%)				Y2/(%)				Y3/(%)
	Sum of Squares	Freedom	F	p	Sum of Squares	Freedom	F	p	Sum of Squares	Freedom	F	p
Model	65.20	9	69.65	<0.0001	34.65	9	69.27	<0.0001	37.37	9	30.42	<0.0001
X1	2.57	1	24.69	0.0003	1.01	1	18.17	0.0009	3.58	1	26.24	0.0002
X2	24.17	1	232.35	<0.0001	0.49	1	8.74	0.0111	7.18	1	52.63	<0.0001
X3	2.28	1	21.93	0.0004	0.28	1	4.99	0.0437	0.71	1	5.22	0.0398
X1X2	0.78	1	7.51	0.0168	0.84	1	15.21	0.0018	0.15	1	1.11	0.3116
X1X3	2.76	1	26.55	0.0002	1.81	1	32.48	<0.0001	0.011	1	0.082	0.7785
X2X3	0.91	1	8.76	0.0111	1.44	1	26.00	0.0002	0.55	1	4.04	0.0657
X12	17.19	1	165.22	<0.0001	5.73	1	103.07	<0.0001	4.96	1	36.32	<0.0001
X22	9.96	1	95.71	<0.0001	21.69	1	390.23	<0.0001	11.76	1	86.15	<0.0001
X32	4.67	1	44.88	<0.0001	1.40	1	25.21	0.0002	8.56	1	62.71	<0.0001
Residual	1.35	13			0.72	13			1.77	13
Lack of fit	0.6	5	1.29	0.3562	0.24	5	0.81	0.5748	0.31	5	0.33	0.8799
Corrected	0.75	8			0.48	8		3.04	1.47
Total	66.55	22			35.37	22			39.14	22

Note: (0.01 < p < 0.05) indicated a significant effect. (p < 0.01) indicated a highly significant effect.

. The model was significant, and the lack of fit was not significant. For straw clearing rate, the machine operation speed, operation depth of the straw clearing knife, and straw covering amount all had a highly significant effect. The primary and secondary relation of the influence of each test factor on straw clearing rate was the operation depth of the straw clearing knife, machine operation speed, and straw covering amount. For the qualified rate of operation depth, the machine operation speed had a highly significant effect. The operation depth of the straw clearing knife and the straw covering amount had a significant effect. The primary and secondary relation of each test factor on the qualified rate of operation depth was machine operation speed, the operation depth of the straw clearing knife, and straw covering amount. For consistency of straw clearing between rows, the machine operation speed and the operation depth of the straw clearing knife had a highly significant effect. The straw covering amount had a significant effect. The primary and secondary relation of the influence of each test factor on the consistency of straw clearing between rows was the operation depth of the straw clearing knife, machine operation speed, and straw covering amount. The interaction between machine operation speed and straw covering amount had a highly significant effect on straw clearing rate. The interaction between machine operation speed and the operation depth of the straw clearing knife had a significant effect on straw clearing rate. The interactions between the operation depth of the straw clearing knife and the straw covering amount had a significant effect on straw clearing rate. The interaction of all test factors had highly significant effects on the qualified rate of operation depth. The interaction of all test factors had no significant effect on the consistency of straw clearing between rows.

Based on the principle of efficiency maximization, parameter combination optimization was carried out under the conditions of the machine operation speed of 5~10 km/h, the operation depth of the straw clearing knife of 20~80 mm, and the straw covering amount of 0.8~1.6 kg/m². The objective function and constraint conditions were as follows:

$$\{\begin{array}{l}86\%\le Y_1\le 100\%\\ 86\%\le Y_2\le 100\%\\ 83\%\le Y_3\le 100\%\\ s.t.\{\begin{array}{l}5 \text{km/h}\le X_1\le 10 \text{km/h}\\ 20 \mathrm{mm}\le X_2\le 80 \mathrm{mm}\\ 0.8 {\text{kg/m}}^2\le X_3\le 1.6 {\text{kg/m}}^2\end{array}\end{array}$$

(9)

The optimization module in Design-Expert 8.0.6 software was applied for multi-objective optimization. The optimization interval of machine operation speed and straw covering amount is shown in Figure 9. When the machine operation speed was 5~8.8 km/h, the operation depth of the straw clearing knife was 50 mm, the straw covering amount was 0.9~1.44 kg/m², the straw clearing rate was ≥86%, the qualified rate of operation depth was ≥86%, and the consistency of straw clearing between rows was ≥83%.

In order to verify the accuracy of the parameter combination optimization results, the field tests were carried out on machine operation speeds of 5 km/h, 6 km/h, 7 km/h, and 8 km/h under the conditions of the operation depth of the straw clearing knife of 50 mm and the straw covering amount of 1.2 kg/m². The test results are shown in

Table 8. Field validation test results.

X1/ (km/h)	Y1/(%)	Y2/(%)	Y3/(%)
5	87.1	88.7	83.8
6	87.5	87.9	83.4
7	86.2	87.1	84.1
8	86.4	86.8	82.9

. Under different machine operation speed conditions, the straw clearing rate, the qualified rate of operation depth and the consistency of straw clearing between rows were all within the optimized range, which proved that the optimization results were reliable.

4. Discussion

4.1. Analysis of the Influence Law of Each Factor on Straw Clearing Rate

Figure 10a shows the interaction between machine operation speed and straw covering amount on the straw clearing rate when the operation depth of the straw clearing knife is at the design center point (50 mm). When the straw covering amount is constant, the straw clearing rate tends to increase first and then decrease with the increase in machine operation speed. This is mainly because the amount of straw going into the seedbed-preparation device per unit time increases with the increase in the machine operation speed, making the straw migration ability decrease. When the machine operation speed is constant, the straw clearing rate increases gradually with the increase in the straw covering amount, but the corresponding speed of the extreme point of straw clearing rate decreases gradually with the increase in the straw covering amount and the machine operation speed. This is mainly because the straw amount entering into seedbed-preparation device per unit time increases with the increase in the straw covering amount when the machine operates at a low speed, leading to improvements in straw migration ability and straw clearing rate. Additionally, with the increase in machine operation speed, the cutting pitches of the straw clearing knife become larger, and the cutting and migration times of straw decrease in unit time, further resulting in a reduction in straw clearing rate. Figure 10b shows the interaction between the operation depth of the straw clearing knife and straw covering amount on the straw clearing rate when the machine operation speed is at the design center point (7.5 km/h). When the straw covering amount is constant, the straw clearing rate increases first and then decreases with the increase in the operation depth of the straw clearing knife. This is mainly because the vertical striking surface of the knife on the seedbed is circular. When the operation depth of the straw clearing knife is small, the contact area between the rotation circumference of the straw clearing knife and ground surface increases with the increase in the operation depth, and the straw amount going into the seedbed-preparation device per unit time increases. When the operation depth is larger, the rotation radius of the straw along the knife decreases, which leads to the increase in the slip time of the straw along the knife, resulting in the decrease in straw migration per unit time and straw clearing rate. Figure 10c shows the interaction between machine operation speed and the operation depth of the straw clearing knife on the straw clearing rate when the straw covering amount is at the design center point (1.2 kg/m²). The interaction between machine operation speed and the operation depth of the straw clearing knife on the straw clearing rate is significant.

4.2. Analysis of the Influence Law of Each Factor on Qualified Rate of Operation Depth

Figure 11a shows the interaction between machine operation speed and straw covering amount on the qualified rate of operation depth when the operation depth of the straw clearing knife is at the design center point (50 mm). The impact trend of the machine operation speed on the qualified rate of operation depth is affected by the straw covering amount. When the machine operation speed is constant, the qualified rate of operation depth increases with the increase in straw covering amount. However, with the increase in the machine operation speed, the qualified rate of operation depth gradually decreases with the increase in straw covering amount, mainly because when the straw covering amount is small, the profiling plate can always stick to the ground surface in its moving direction and move the straw to both sides of its moving direction under the action of its own torque. With the increase in straw covering amount, when the torque applied to the profiling plate is unable to move straws to both sides of its moving direction, the profiling plate will pass over the straw, and the rotation angle around the fixed shaft will be increased. The self-adaptive control system will adjust the operation depth of the straw clearing knife according to the wrong signal, resulting in the decrease in the qualified rate of operation depth. Figure 11b shows the interaction between operation depth of the straw clearing knife and straw covering amount on the qualified rate of operation depth when the machine operation speed is at the design center point (7.5 km/h). The effect trend of the operation depth of the straw clearing knife on the qualified rate of the operation depth is affected by the straw covering amount. When the straw covering amount is constant, the qualified rate of operation depth increases first and then decreases with the increase in the operation depth of the straw clearing knife. Figure 11c shows the interaction between machine operation speed and the operation depth of the straw clearing knife on the qualified rate of operation depth when the straw covering amount is at the design center point (1.2 kg/m²). The interaction between machine operation speed and the operation depth of the straw clearing knife on the qualified rate of operation depth is significant.

4.3. Analysis of the Influence Law of Each Factor on Consistency of Straw Clearing between Rows

Figure 12a shows the influence law of the straw covering amount and operation speed on the consistency of straw clearing between rows when the operation depth of the straw clearing knife is at the design center point (50 mm). When the straw covering amount is constant, the consistency of straw clearing between rows tends to increase and then decrease with the increase in operation speed. This is mainly because when the operation speed is low, the straw transportation amount of the seedbed-preparation device increases slowly with the increase in operation speed. The straw transport condition of the seedbed-preparation device is stable, and the consistency of straw removal between rows is large. When the operation speed is high, the straw will collide with the seedbed-preparation device transport shell along its moving direction after it is detached from the straw clearing knife, and the collision will increase the straw conveying time, resulting in different straw conveying amounts for different rows of straw clearing knives, decreasing the consistency of straw clearing between rows. When the operation speed is constant, the consistency of straw clearing between rows tends to increase and then decrease with the increase in straw covering amount. This was mainly because the straw covering amount can affect the operation effect of the profiling mechanism. When the straw covering amount is large, the profiling plate cannot continue to adhere to the ground movement under the action of its own torque; instead, it passes over the straw. At this time, the rotation angle of the profiling plate increases, and the operation depth of the straw clearing knife decreases, resulting in a decrease in the amount of straw transported in this area and a decrease in the consistency of straw clearing between rows. Figure 12b shows the influence law of the operation depth and operation speed on the consistency of straw clearing between rows when the straw covering amount is at the design center point (1.2 kg/m²). When the operation speed is constant, the consistency of straw clearing between rows tends to increase and then decrease with the increase in operation depth. This is mainly because when the operation depth is small, the contact area between the straw clearing knife and the ground surface increases with the increase in operation depth. When the operation depth is large, the radius of straw movement with the straw clearing knife decreases, the sliding time along the knife increases, and then straw accumulates in the seedbed-preparation device, resulting in a decrease in the consistency of straw clearing between rows.

5. Conclusions

(1) A straw-clearing-depth self-adaptive control system for a front-mounted seedbed-preparation device was designed. The poor trafficability characteristic and the low profiling accuracy of the ground-wheel profiling mechanism, which resulted in the unstable straw clearing depth, poor straw clearing quality, and low operation efficiency of seedbed-preparation devices were solved;

(2) The performance test results of the self-adaptive control system showed that the lifting speed of the front-suspension mechanism was greater than 0.2 m/s in the manual button control mode, and the relative error between the target value and actual value of straw clearing depth was 10.8% under the self-adaptive profiling control mode;

(3) The primary and secondary relation of the influence of each test factor on the straw clearing rate was the operation depth of the straw clearing knife, machine operation speed, and straw covering amount. The primary and secondary relation of each test factor on the qualified rate of operation depth was machine operation speed, the operation depth of the straw clearing knife, and straw covering amount. The primary and secondary relation of the influence of each test factor on the consistency of straw clearing between rows was the operation depth of the straw clearing knife, machine operation speed, and straw covering amount;

(4) When the machine operation speed was 5~8.8 km/h, the operation depth of the straw clearing knife was 50 mm and the straw covering amount was 0.9~1.44 kg/m², the straw clearing rate was ≥86%, the qualified rate of the operation depth was ≥86%, and the consistency of straw clearing between rows was ≥83%. Field tests were carried out using machine operation speeds of 5 km/h, 6 km/h, 7 km/h, and 8 km/h with a straw clearing knife operation depth of 50 mm, and straw covering amount of 1.2 kg/m². The test results indicated that under different machine operation speed conditions, the straw clearing rate, the qualified rate of operation depth, and the consistency of straw clearing between rows were all within the optimized range, meeting the requirements of agronomy production technology.