Experimental Investigation of Dual-Path Inline Mixing System for Sprayers in Corn-Soybean Strip Intercropping Mode

Abstract

Corn-soybean strip intercropping, which fully utilizes land resources and has high total yield and soil fertility, has become a modern agricultural cultivation mode that is actively promoted. In order to solve the weed problem in corn-soybean strip intercropping, the agricultural technology requirements cannot be met by traditional pre-mixed spraying machines, so a direct injection dual-path inline mixing system was designed for the corn-soybean strip intercropping mode. The system was integrated to improve its installation convenience and universality, and was capable of fulfilling the requirements for the simultaneous application of two types of pesticides at varying mixing ratios. The system mainly consists of a water solvent injection module, glyphosate (pesticide for corn) inline mixing module, and a fomesafen (soybean pesticide) inline mixing module. First, the detection rules of the mixing ratio of related pesticides based on the electrical conductivity measurement principle were studied. Then, the working characteristics of the designed direct injection dual-path inline mixing system were studied through experiments using different pesticides and mixing ratio adjustment ranges. The mixing uniformity test showed that the designed direct injection dual-path inline mixing system had good mixing uniformity, and the maximum uniformity coefficient of the mixing ratio was 9.7%. The stability test showed that the mixing ratio of the designed dual-path inline mixing system was relatively stable, with the maximum standard deviation of the mixing ratio accounting for about 2.2% of its average value, and the maximum average deviation was less than 1.5%. The precision and response time test showed that the designed dual-path inline mixing system had an average deviation of the mixing ratio of less than 2.7% under the condition of a step signal target mixing ratio, and the response time was a maximum of 3.4 s. The results show that the designed dual-path inline mixing system has good performance, and the research findings provide a reference for the design and optimization of inline mixing systems.

1. Introduction

Corn and soybean are significant food and cash crops globally. As international demand increases, their cultivation practices are continuously evolving and improving [1,2]. Recently, the intercropping mode of corn and soybean has garnered considerable attention due to its efficient use of land resources, the substantial enhancement of overall land productivity, and an improvement in soil fertility [3,4]. For instance, in the high-yield mode of corn-soybean intercropping explored and practiced in Enyang, Sichuan Province, China, this intercropping mode increased the yields of soybean and corn per acre by 18.97% and 39.18%, respectively [5]. Nevertheless, while corn-soybean strip intercropping offers numerous benefits, it also imposes stringent demands on the operational performance of the associated agricultural machinery [6]. According to agronomic requirements, the planting distance between corn and soybeans is relatively narrow, and the types of diseases and pests affecting these two crops differ significantly [7,8,9,10]. Traditional independent application methods for weed control are not only inefficient, but also fail to ensure precise dosage and coverage, leading to issues such as misapplication, overuse, and wastage of pesticides, thereby posing potential risks to crop growth and environmental safety [11].

In recent years, to address the issues of pharmaceutical waste, environmental contamination, and health risks associated with traditional premixing technologies, inline mixing systems for plant protection machinery have garnered significant attention and research. Scholars both domestically and internationally have conducted extensive studies on inline pesticide mixing systems from various perspectives to enhance its functionality. To meet the demands of precision agriculture, Chen et al. devised a direct injection inline pesticide mixing system utilizing fuzzy PID control, engineered a non-contact pesticide mixing ratio detection module based on optical transmission principles, constructed a testing platform, and conducted comprehensive horizontal and longitudinal comparative tests, thereby validating the superior operational performance of the designed inline pesticide mixing system [12]. To achieve variable spraying and width control, Jiang et al. introduced an innovative injection inline mixing control system, incorporating high-precision real-time liquid flow detection technology and a DMC dynamic matrix control algorithm to monitor and regulate the liquid injection volume, while also optimizing the system’s transmission time delay and mixing consistency. These advancements have significantly improved the performance of inline pesticide mixing systems [13]. However, the two aforementioned direct injection inline mixing systems are limited to performing basic mixing and matching for a single pesticide, which fails to satisfy the specific requirements of the corn-soybean strip intercropping mode. Therefore, it is imperative to develop a dual-path inline mixing system for pesticides to cater to their unique plant protection needs including different pesticides and varying mixing ratios. Han et al. devised an inline mixing system for a rod sprayer tailored for corn-soybean strip intercropping mode that incorporated methods such as the calibration of mixing ratio-conductivity and the construction of a control mode algorithm, thereby enabling its practical application in complex planting scenarios. The performance of this system was validated through real-world vehicle testing [14]. Nevertheless, the current inline pesticide mixing system lacks integration, making it challenging to adapt to other pesticide application machinery. Additionally, due to the uneven terrain of field roads, ensuring the operational stability of critical components like the pesticide pump poses significant difficulties, thus affecting the overall work performance. Based on principal component analysis (PCA), Dai et al. integrated image processing and machine learning techniques in the detection of mixing ratios in inline mixing systems [15], significantly enhancing the accuracy of assessing the mixing uniformity and ratio. This approach enabled a comprehensive analysis of the dynamic concentration consistency within the system. However, due to the stringent illumination requirements for high-speed cameras and the time-consuming nature of subsequent analyses, this method is more suitable for studying the mixing characteristics of dynamic flow fields rather than the real-time monitoring of sprayer field operations.

In this study, a direct injection multi-path inline mixing system for corn-soybean strip intercropping mode was designed in response to the specific operational requirements of contemporary sprayers. This system ensures uniform mixing as well as rapid, accurate, and stable control of the mixing ratio. By employing real-time control of the electrically regulated valves, the system achieved precise regulation of the injection liquid flow. A direct injection dual-path inline pesticide mixing system was developed, integrating a single-path inline pesticide mixing module. Extensive testing was conducted to ensure the uniformity of pesticide mixing and the precision of the mixing ratio control, thereby validating the superior performance of the designed dual-path inline pesticide mixing system. The findings of this study serve as a valuable reference for the development of inline pesticide mixing systems tailored to the corn-soybean strip intercropping mode. The integrated inline pesticide mixing system, after appropriate modifications and expansion, can be effectively applied to a variety of crops, demonstrating broad applicability and versatility.

2. Materials and Methods

2.1. Agronomic Conditions of the Corn-Soybean Strip Intercropping Mode

The corn-soybean strip intercropping mode significantly enhances land utilization and increases the farmers’ income given a fixed land area. To maximize the use of growing space, the intercropping method typically involves planting two to six rows of soybean and two to four rows of corn in the corn and soybean strip compound planting mode. In this study, we adopted a configuration of four rows of soybean and two rows of corn. Specifically, four rows of soybean and two rows of corn were planted alternately, with a row spacing of 30 cm within each crop belt and a 70 cm spacing between the two kinds of crop belts. Based on the agronomic requirements of the selected corn-soybean strip intercropping mode, modifications were made to the nozzle positioning, and additional application baffles were introduced. Figure 1 illustrates the agronomic requirements and nozzle layout for the corn-soybean strip intercropping mode.

Corn (Zea mays), an annual tall herbaceous plant that originated in Central and South America, is one of the world’s most productive crops. It not only serves as a vital food crop, but is also an essential raw material for industries including healthcare, light manufacturing, and chemical processing. Soybean (Glycine max), an annual herb that originated in China, is now widely cultivated globally. Soybeans are a crucial food crop, with seeds rich in plant-based proteins, making them suitable for producing a variety of soy products, extracting soybean oil, brewing soy sauce, and isolating proteins. The selected corn and soybeans were non-GMO crops. Currently, the staggered timing of pesticide application across two planting zones is a prevalent technique in strip intercropping. However, this approach exhibits low efficiency and hinders agricultural mechanization, thereby impeding the widespread adoption of strip intercropping [16,17]. To enhance operational efficiency, a self-propelled spray boom sprayer was employed as the operational platform to perform simultaneous full-range application spraying. Fomesafen and glyphosate were utilized as herbicides for the soybeans and corn, respectively [18,19]. Fomesafen is primarily applied to manage broadleaf weeds in soybean, cotton, and other crop fields. It contains 95% active ingredient and appears as a grayish-white powder solid [20]. Glyphosate is extensively used for weed control in agricultural fields and is characterized by its white crystalline powder form [21,22]. Both herbicides exhibit ionic dissociation properties in aqueous solutions, leading to an increase in solution conductivity with the rise in mixing ratio. The conductivity values of the concentrated pesticide solution and the final mixed spray solution can be measured using a conductivity sensor, allowing for the indirect determination of the mixing ratio of the two pesticides [23,24,25,26]. This method served to validate the working performance of the designed direct-injection dual-path inline mixing system.

A QIWEI (Hangzhou, China)-produced DDS-11A conductivity transmitter was chosen as the testing apparatus, featuring a conductivity measurement range of 0–2 × 10⁵ μS/cm and an accuracy of ±1% F.S. This device is extensively utilized in the detection of solution conductivity signals and mixing ratios due to its high precision, broad measurement range, and ease of use. For the experiment, a 250 g/L fomesafen solution from JUFENGYUAN (Qingdao, China) and a 30% glyphosate solution from DINGBANG (Nanning, China) were selected. Based on the application guidelines for these pesticides, the typical mixing ratios were approximately 0.2% for the fomesafen solution (with 250 g/L active ingredient) and 2% for the glyphosate solution (with 30% active ingredient). Consequently, a stock solution of fomesafen was prepared at a 1:10 ratio with water, yielding a total volume of 100 mL. The 30% glyphosate solution served as the stock solution for glyphosate. Following the mixing ratios specified in

Table 1. The value of the electrical conductivity of the diluted fomesafen mixed solution according to different mixing ratios.

No.	Fomesafen Stock Solution (mL)	Water (mL)	Mixing Ratio	Mean Conductivity (μS/cm)
1	5	100.0	1:20	568.6
2	5	200.0	1:40	441.3
3	5	300.0	1:60	401.8
4	5	400.0	1:80	378.6
5	5	500.0	1:100	364.5
6	5	600.0	1:120	358.7
7	5	700.0	1:140	353.8
8	5	800.0	1:160	349.5
9	5	900.0	1:180	347.8
10	5	1000.0	1:200	342.2

and

Table 2. The value of the electrical conductivity of the diluted glyphosate mixed solution according to different mixing ratios.

No.	Fomesafen Stock Solution (mL)	Water (mL)	Mixing Ratio	Mean Conductivity (μS/cm)
1	5	100.0	1:20	20,258.7
2	5	200.0	1:40	10,537.6
3	5	300.0	1:60	7324.9
4	5	400.0	1:80	5649.2
5	5	500.0	1:100	4687.4
6	5	600.0	1:120	4061.8
7	5	700.0	1:140	3578.3
8	5	800.0	1:160	3238.4
9	5	900.0	1:180	3012.5
10	5	1000.0	1:200	2751.9

, the stock solutions of fomesafen and glyphosate were diluted accordingly. The conductivity values of the diluted mixed solutions were then measured and calibrated using a conductivity transmitter, with each measurement repeated three times to obtain the average value.

As illustrated in Table 1, the electrical conductivity values of the fomesafen mixing solutions, prepared at various dilution ratios, were measured three times for each group, and the mean values were calculated. These data served to calibrate and analyze the conductivity and mixing ratios of the aqueous solutions containing different proportions of fomesafen, allowing for curve fitting of the aforementioned data. Upon fitting, it was evident that the functional relationship between the mixing ratio β and the electrical conductivity EC of the fomesafen mixing solution was as follows:

$$EC=457\times e^{{\displaystyle \frac{-1}{26.5\beta }}}+350$$

(1)

As illustrated in Figure 2, the R² of the fitting curve was 0.99373. Based on this functional relationship, the calculation of the electrical conductivity of the fomesafen mixed solution was converted into a mixing ratio signal, thereby validating the precision of the mixing ratio adjustment for the fomesafen stock solution.

As illustrated in Table 2, the conductivity values of the glyphosate mixed solution, prepared at various dilution ratios, were measured three times for each group, and the mean values were calculated. Based on the collected data, the conductivity of the glyphosate aqueous solutions at different concentrations and the ratio of mixed substances were calibrated and analyzed. These data were then subjected to curve fitting. Following the fitting process, the functional relationship between the mixing ratio β and the conductivity of the glyphosate mixed solution was established as follows:

$$EC=34768\times e^{{\displaystyle \frac{-1}{27.3\beta }}}+3293.7$$

(2)

As illustrated in Figure 3, the R² of the fitting curve was 0.99316. Based on this functional relationship, the calculation of the electrical conductivity of the glyphosate mixed solution was converted into a mixing ratio signal, thereby validating the precision of the mixing ratio adjustment for the glyphosate stock solution.

2.2. Working Principle of Direct Injection Dual-Path Inline Mixing System

To address the issue of weed damage under the intercropping mode of corn and soybean, a direct injection dual-path inline pesticide mixing system was developed to satisfy the specific plant protection requirements for various pesticides and distinct mixing ratios [27]. As illustrated in Figure 4, the designed direct injection dual-path inline pesticide mixing system primarily comprised three components: a water solvent injection module, a glyphosate (for corn) injection mixing module, and a fomesafen (for soybean) injection mixing module.

The water solvent injection module primarily consisted of a DC brushless motor, plunger pump, water tank, bypass reflux regulating valve, water solvent flow transmitter (for corn), and water solvent flow transmitter (for soybean) and was capable of monitoring the flow signals into each pesticide injection mixing module in real-time, facilitating the subsequent calculation of the mixing ratio data. A bypass reflux control valve was installed to regulate the total flow of water and solvent entering the dual-path mixing system. The pesticide injection mixing module comprised a static mixer, pesticide flow transmitter, check valve, electric regulating valve, pesticide pressure transmitter, high-pressure pesticide tank, pesticide pump, DC brushless motor, pesticide tank, electromagnetic relief valve, and accumulator. The pesticide pump initially pumped the pesticide solution into the liquid storage component, which included a high-pressure pesticide tank and accumulator, serving to stabilize the pressure and store energy. Subsequently, the injection flow rate was controlled by precisely adjusting the opening of the electric regulating valve, thereby controlling the mixing ratio. An electromagnetic relief valve was also installed at the end of the liquid injection mixing module to prevent excessive pressure in the liquid storage component and provide pressure relief protection.

The SHUANGLONG (Binzhou, China)-produced VP110-03 fan nozzle was selected for spraying, with a nozzle pressure of 0.3 MPa and a flow rate of 1.1–1.2 L/min per nozzle. A total of 18 nozzles were installed. Based on the 3WPG-300 high-clearance self-propelled sprayer designed by our team, a dual-path inline mixing system was developed. The sprayer’s forward speed was set at 6 km/h. Additionally, it was equipped with a JD-26-3 three-cylinder piston pump to serve as the water pump, which had a flow rate ranging from 12 to 20 L/min, with a maximum of 20 L/min. The mixing ratio β ranged from 1:200 to 1:20, with a maximum of 1:20. The maximum total flow rate of the mixed solution can be obtained by summing the injection flow rate of the pesticide and the water solvent flow rate. The injection flow range of the liquid pesticide can be calculated using Equation (4), while the maximum flow rate of the mixed solution can be determined using Equation (5).

$$Q_{\max }=Q_w+Q_p$$

(3)

$$Q_p=Q_{\mathrm{w}}\times \beta$$

(4)

$$Q_{\max }=Q_{\mathrm{w}}\times (1+\beta )$$

(5)

Among these parameters, Q_max represents the maximum total flow rate of the mixed solution, Q_w denotes the water solvent flow rate, β signifies the ratio of the mixing ratio, and Q_p indicates the injection flow rate of the pesticide. The substituted data calculation revealed that the maximum flow rate Q_max was 21 L/min, with Q_p ranging from 0.06 to 1 L/min.

Considering the operating environment, installation method, valve flow characteristics, and cost-effectiveness of the flow transmitters, the PN-HI2144 Hall flow transmitter manufactured by Wenzhou PENAI Company (Wenzhou, China) was selected for the water solvent flow measurement, while the GICAR-5111 turbine flowmeter produced by Hangzhou YIKONG Technology Company (Hangzhou, China) was chosen for the pesticide injection flow measurement. The GICAR-5111 high-precision turbine flowmeter was installed at the outlet of the electric control valve, with a flow detection range of 2.7 to 124.8 L/h, a working pressure of 0.2 to 1 MPa, and an accuracy of ±2%. This setup enabled the real-time monitoring of the flow rate of the injected pesticide. By collecting the flow signals from both the water solvent and the pesticide liquid, the mixing ratio can be accurately approximated.

2.3. Working Principle of Direct Injection Dual-Path Inline Mixing Control System

The STM32F103VET6 development board serves as the primary controller in the direct injection dual-path inline mixing control system developed by this research institute. At the core of the controller lies a 32-bit processor based on the ARM Cortex-M3 architecture, capable of operating at frequencies up to 72 MHz. The system primarily collects detection signals from pressure transmitters, flow transmitters, conductivity transmitters, and other sensors via analog-to-digital (AD) conversion. To eliminate the interference caused by significant noise and ensure the accuracy of the collected data and the stability of subsequent control processes, a limiting mean filtering algorithm was incorporated into the program.

The ability to achieve accurate, stable, and rapid control of the mixing ratio is directly correlated with the precision of the actuator [28]. For this purpose, the FRSQT11F-16P electric regulating valve manufactured by Jiangsu FREESUN Company (Suzhou, China) was selected as the actuator to ensure precise control. The precise and stable operation of the electric regulating valve was primarily achieved through digital-to-analog (DA) conversion by the main controller, along with external modules such as a signal amplification module and a DC signal attenuation module. Consequently, a direct injection dual-path inline mixing control system was established, as illustrated in Figure 5. The overall workflow of the dual-path inline pesticide mixing system is depicted in Figure 6. The adjustment parameters of the fuzzy PID were K_p = 2.5, K_i = 0.5, and K_d = 0.01. The fuzzy rules are shown in

Table 3. Parameter K_p control rule adjustment.

e (Error)	ec (Rate of Error Change)
	NB	NM	NS	ZO	PS	PM	PB
NB	PB	PB	PM	PM	PS	ZO	ZO
NM	PB	PB	PM	PS	PS	ZO	NS
NS	PM	PM	PM	PS	ZO	NS	NS
ZO	PM	PM	PS	ZO	NS	NM	NM
PS	PS	PS	ZO	NS	NS	NM	NM
PM	PS	ZO	NS	NM	NM	NM	NB
PB	ZO	ZO	NM	NM	NM	NB	NB

,

Table 4. Parameter K_i control rule adjustment.

e (Error)	ec (Rate of Error Change)
	NB	NM	NS	ZO	PS	PM	PB
NB	NB	NB	NM	NM	NS	ZO	ZO
NM	NB	NB	NM	NS	NS	ZO	ZO
NS	NB	NM	NS	NS	ZO	PS	PS
ZO	NM	NM	NS	ZO	PS	PM	PM
PS	NM	NS	ZO	PS	PS	PM	PB
PM	ZO	ZO	PS	PS	PM	PB	PB
PB	ZO	ZO	PS	PM	PM	PB	PB

and

Table 5. Parameter Kd control rule adjustment.

e (Error)	ec (Rate of Error Change)
	NB	NM	NS	ZO	PS	PM	PB
NB	PS	NS	NB	NB	NB	NM	PS
NM	PS	NS	NB	NM	NM	NS	ZO
NS	ZO	NS	NM	NM	NM	NS	ZO
ZO	ZO	NS	NS	NS	NS	NS	ZO
PS	ZO	ZO	ZO	ZO	ZO	ZO	ZO
PM	PB	NS	NS	PS	PS	PS	PB
PB	PB	PM	PM	PM	PS	PS	PB

. Firstly, the target pesticide mixing ratio information of each pesticide at cruising speed was calculated according to the agronomic requirements of the pesticides to be applied, then the target pesticide mixing ratio information was input through the display terminal connected to the controller, and the controller calculated the target pesticide injection flow rate according to the real-time water solvent flow rate. The high precision turbine flowmeter detection signal was used as the feedback signal, and the controller controlled the opening of the electric regulating valve according to the difference between the target value and the feedback value to adjust the mixing ratio.

Currently, numerous studies have been conducted on the inline detection of mixing ratios, with mainstream methods encompassing the light transmittance method, high-speed camera method, electrical conductivity method, and fiber optic sensor method, among others [29,30]. However, these approaches necessitate the incorporation of various detection components into the inline mixing system, thereby requiring additional installation space and making the detection elements susceptible to external interferences such as field turbulence, leading to inaccurate detection outcomes. Consequently, this study proposed utilizing real-time water solvent flow transmitter signals and high-precision turbine flowmeter signals, after undergoing filtering and processing, to acquire real-time mixing ratio signals. The discrepancy between the obtained mixing ratio signal and the target mixing ratio was incorporated into the control algorithm as an influencing factor to facilitate subsequent adjustments to enhance the control accuracy of the mixing ratio.

To ensure the feasibility of the detection method for the aforementioned mixing ratio information, the glyphosate stock solution containing 30% active ingredient, as previously mentioned, was selected as the test material. The glyphosate mixed solution was prepared under the target mixing ratio conditions according to the test parameters outlined in

Table 6. The actual conductivity results of the mixed glyphosate solution under the condition of the target mixing ratio.

No.	The Inverse of the Target Mixing Ratio	Target Conductivity (μS/cm)	Actual Conductivity (μS/cm)	Value of Error (μS/cm)	Percentage of Error (%)
1	20	20,258.7	19,849.5	409.2	2.0
2	40	10,537.6	9968.1	569.5	5.4
3	60	7324.9	7456.6	131.7	1.8
4	80	5649.2	5361.7	287.5	5.1
5	100	4687.4	4598.3	89.1	1.9
6	120	4061.8	3964.5	97.3	2.4
7	140	3578.3	3688.4	110.1	1
8	160	3238.4	3165.2	73.2	2.3
9	180	3012.5	2948.9	63.6	2.1
10	200	2751.9	2649.8	102.1	3.7

. A conductivity transmitter was employed to measure the target conductivity of the mixed solution. Concurrently, the target mixing ratio for the inline pesticide mixing system was set based on the conditions specified in Table 6. Once the inline pesticide mixing system reached a stable operating state, the pesticide spraying liquid was collected from each nozzle, and the electrical conductivity values were measured using the conductivity transmitter. The average value obtained represents the actual electrical conductivity of the pesticide mixing solution. The actual conductivity results of the glyphosate mixed solution under the target mixing ratio conditions are presented in Table 6.

Based on the data presented in Table 6, it is evident that the maximum error of mixing ratio was 5.4%, a relatively minor discrepancy that satisfied the criteria for mixing ratio detection. Consequently, the aforementioned method was capable of accurately collecting information regarding the mixing ratio.

Due to the expansion of the original single-path inline pesticide mixing system to a dual-path configuration, and considering the distinct types of pesticide dispensed by each nozzle, it was imperative to revise the nozzle connection pipelines. To ensure uniformity in the quantity of dispensed liquid from each nozzle, the spray consistency was validated prior to the installation of the pesticide injection mixing module.

As illustrated in Figure 7, during the steady spraying operation of the sprayer, a measuring cup was positioned beneath each nozzle to collect the liquid volume dispensed within the specified time frame (10 s). The volumetric data from 18 nozzles are presented in Figure 7. The coefficient of variation for the spray quantity of each nozzle head over 10 s was 5.96%, indicating a high level of spray uniformity that satisfied the operational requirements of the plant protection equipment.

2.4. Direct Injection Two-Path Inline Mixing System Integration and Real Sprayer Installation

Based on the initial spray consistency test, it can be concluded that the pipeline component of the designed dual-path inline mixing system for corn-soybean strip intercropping met the fundamental application requirements. However, due to the high number of internal actuators and the complexity of installation, the proposed inline pesticide mixing system could benefit from modular integration to enhance its applicability and portability. Figure 8 illustrates the integrated schematic and physical layout of the pesticide injection mixing module.

Following the integration of the pesticide injection mixing module, it was connected to the dual-path inline pesticide mixing system for corn and soybean strip intercropping, which was part of the 3WPG-300 high-clearance self-propelled sprayer designed by our team. The overall structure and installation position are illustrated in Figure 9. The integrated pesticide injection mixing module only features four ports: the water solvent injection interface, pesticide injection interface, pesticide reflux interface, and the mixture outflow interface. The remaining internal pipelines are integrated internally, eliminating the need for an additional actuator.

3. Results

Based on the overall architecture of the direct injection dual-path inline pesticide mixing system designed in this study, a pesticide injection mixing module was modularized and integrated, and the spray application system of the existing sprayer by our team was enhanced [31,32,33,34,35], thereby enabling it to meet the specific requirements for weed control spraying in the intercropping mode of corn and soybeans. As illustrated in Figure 10, the improved sprayer was equipped with a direct injection dual-path inline mixing system. Given the constraints of the experimental plot, the stability test of the mixing ratio adjustment, the accuracy of the mixing ratio adjustment, and the response time test were conducted by reciprocating the sprayer within the experimental plot. The detection of the mixing ratio information for the dual-channel inline mixing system was performed using a high-precision turbine flow meter, which was largely independent of the experimental plot conditions. Consequently, the selected experimental plot satisfied the test requirements.

3.1. Mixing Uniformity Test of the Dual-Path Inline Mixing System

Following the installation of the integrated pesticide mixing injection module and the configuration of the sprayer mixing system based on the structure of the direct injection dual-path inline mixing system depicted in Figure 4, the mixing uniformity of the dual-path inline mixing system was evaluated. For this test, a glyphosate solution containing 30% active ingredient was used as the glyphosate stock solution, and the stock solution of fomesafen was prepared at a 1:10 ratio with water. The detailed testing procedure is outlined below:

(1)

Set the electric regulating valve to an opening degree of 33% and maintain this setting; operate the water pump at its standard gear, ensuring that all other operational components function normally.

(2)

After allowing the dual-path inline mixing system to reach a stable operating state, collect the spray liquid from nozzles installed on the sprayer’s spray rod.

(3)

Measure the electrical conductivity of the collected spray liquid from each nozzle after settling and calculate the mixing ratio.

The spray solution emitted from 18 nozzles mounted on the spray rod was collected, and its conductivity was measured after a two-hour settling period. Following multiple trials, the mixing ratio of the spray liquid from the 18 nozzles was determined, as detailed in

Table 7. The electrical conductivity value of each nozzle spraying liquid.

Nozzle Serial Number	Electrical Conductivity Value of Pesticide Solution			Mixing Ratio of Test 1	Mixing Ratio of Test 2	Mixing Ratio of Test 3
	Test 1	Test 2	Test 3
1	355.9	353.2	353.3	0.00868	0.00761	0.00764
2	353.0	357.3	352.3	0.00753	0.00912	0.00716
3	359.2	354.5	352.7	0.00966	0.00815	0.00725
4	356.8	351.9	352.6	0.00897	0.00688	0.00729
5	355.4	354.7	351.9	0.00850	0.00824	0.00688
6	358.1	351.8	351.7	0.00936	0.00683	0.00677
7	353.7	353.4	354.4	0.00782	0.00770	0.00813
8	357.9	353.2	353.6	0.00930	0.00759	0.00779
9	351.8	353.9	350.5	0.00682	0.00792	0.00553
10	357.4	355.0	353.1	0.00915	0.00835	0.00756
11	356.3	352.0	351.5	0.00881	0.00697	0.00658
12	354.5	352.8	354.5	0.00817	0.00743	0.00815
13	3386.8	3471.5	3425.6	0.00618	0.00694	0.00657
14	3474.9	3519.7	3426.4	0.00697	0.00727	0.00658
15	3357.4	3468.6	3580.7	0.00581	0.00692	0.00764
16	3465.7	3576.5	3528.5	0.00690	0.00761	0.00733
17	3523.3	3418.9	3467.4	0.00730	0.00651	0.00691
18	3469.1	3497.3	3386.5	0.00693	0.00713	0.00618

. Specifically, the solutions dispensed by nozzles numbered 1–12 correspond to the dilution of the fomesafen stock solution, while those from nozzles 13–18 represent the dilution of the glyphosate stock solution.

According to the results in Table 7, it can be concluded that the uniformity coefficient of variation of the mixing ratio of the fomesafen stock solution and glyphosate stock solution in test 1 was 9.4% and 7.7%, respectively. In test 2, the uniformity coefficient of variation of the mixing ratio of the fomesafen stock solution and glyphosate stock solution was 8.4% and 4.8%, respectively. In test 3, the uniformity coefficient of variation of the mixing ratio of the fomesafen stock solution and glyphosate stock solution was 9.7% and 7.2%, respectively.

The results obtained are presented in Figure 11. The maximum coefficient of variation of the mixing uniformity across the three tests was 9.7%, with the overall coefficient of variation remaining below 10%. Consequently, this finding validates the high level of mixing uniformity achieved by the direct injection dual-path inline mixing system developed in this study, confirming its suitability for pesticide application in corn-soybean strip intercropping mode.

3.2. Stability Test of Mixing Ratio Adjustment for Two-Path Inline Mixing System

The stability test of the mixing ratio adjustment was conducted on a sprayer equipped with a direct injection dual-path inline mixing system. The test materials were fomesafen and glyphosate stock solutions prepared as described in Section 3.1, with the pump maintained at its normal operating gear. The specific testing procedure involved setting the target mixing ratio of the two pesticide formulations to 1:50 and subsequently reducing it to 1:170; the denominator of the mixing ratio was tested at intervals of 40. When only the pump is active, the mixing ratio approaches zero. Upon activating the pesticide pump, the liquid storage component becomes pressurized, and the inline mixing control system adjusts the electric regulating valve’s aperture to achieve the target mixing ratio. Real-time monitoring of water solvent flow and single-path liquid flow was achieved using a water solvent flow transmitter and a high-precision turbine flow meter, allowing for the calculation of the real-time mixing ratio. Based on the deviation between the real-time and target mixing ratios, the electric regulating valve’s aperture is dynamically adjusted to maintain the real-time mixing ratio at the desired value. Here, β represents the mixing ratio of mixed pesticide. A total of five tests were conducted, each lasting approximately 80 s. The results are presented in Figure 12.

The results derived from the stability test of the mixing ratio were analyzed and processed, yielding the mean value, standard deviation, and mean variation of the data once the real-time mixing ratio had stabilized near the target mixing ratio. These results are presented in

Table 8. Results of the stability analysis of the mixing ratio.

No.	Target Mixing Ratio 1:β	Fomesafen			Glyphosate
		Mean Value of Mixing Ratio	Standard Deviation of Mixing Ratio	Mean Variation (%)	Mean Value of Mixing Ratio	Standard Deviation of Mixing Ratio	Mean Variation (%)
1	50	49.3	3.02	1.45%	49.6	3.10	0.64%
2	90	90.2	3.42	0.25%	90.2	2.77	0.26%
3	130	130.2	3.37	0.13%	130.1	3.05	0.07%
4	170	169.3	3.63	0.43%	170.5	3.62	0.27%

. The mean value of mixing ratio deviated from the target value by no more than 1.5%, with a maximum average deviation of 1.45%. The overall trend indicated a decline. This phenomenon can be attributed to the utilization of a fuzzy PID control algorithm to regulate the 1:β signal; as the mixing ratio increases, the fuzzy PID adjustment parameters remain unaltered, leading to a reduction in the average deviation as the mixing ratio β decreases. The maximum standard deviation was 3.63, representing approximately 2.2% of the mean value. The stability of the mixing ratio was satisfactory, fulfilling the operational performance criteria for inline mixing in the context of sprayer applications.

3.3. Tests on Accuracy and Response Time of Mixing Ratio Adjustment of Two-Path Inline Mixing System

The accuracy of the mixing ratio adjustment and response time were evaluated for a sprayer equipped with a direct injection dual-path inline mixing system. The test materials comprised fomesafen and glyphosate stock solutions prepared as described in Section 3.1, with the pump operating at its standard working gear. The testing procedure was as follows: the target mixing ratios of the two pesticides were set incrementally by a denominator step of 40 within the range of 1:40 to 1:160, conducting a total of three trials. Initially, the inline mixing control system initiated operation 25 s after the water pump started. The real-time mixing ratio was calculated based on data from the water solvent flow transmitter and high-precision turbine flow meter, which monitored the single-path water solvent flow and pesticide flow rates. According to the real-time mixing ratio signal, the electric regulating valve’s aperture was dynamically adjusted to ensure that the actual mixing ratio matched the target mixing ratio in real-time. Each of the three tests, where β represented the mixing ratio of pesticide, lasted approximately 42 s. The results are presented in Figure 13.

The accuracy of the mixing ratio and response time test results were analyzed and processed. The mean value, standard deviation, average deviation and response time of the real-time mixing ratio and target mixing ratio were obtained after the difference of the denominator was less than 5. The analysis results are presented in

Table 9. Results of the accuracy and response time analysis of the mixing ratio.

No.	Mixing Ratio 1:β	Fomesafen			Response Time (s)	Glyphosate			Response Time (s)
		Mean Value of Mixing Ratio	Standard Deviation	Mean Variation (%)		Mean Value of Mixing Ratio	Standard Deviation	Mean Variation (%)
1	40–80	82.1	2.85	2.66%	3.4 s	82.0	2.81	2.49%	1.9 s
2	80–120	121.4	2.19	1.18%	1.3 s	121.9	3.09	1.55%	2.1 s
3	120–160	160.6	3.18	0.35%	1.5 s	159.5	2.96	0.34%	2.2 s

. The mean value of the mixing ratio deviated by less than 2.7% from the target value, indicating a high level of adjustment accuracy. The maximum standard deviation was 3.18, which accounted for approximately 2% of the mean value, suggesting good stability. Despite the application of a limiting average filter, some noise remained due to the MCU collecting the output voltage signal of the mixing ratio detection module via AD conversion. The response time for all three tests was less than 3.4 s, which was sufficiently short to meet the performance requirements of the inline pesticide mixing system used in the corn-soybean strip intercropping mode.

4. Discussion

Due to the distinct pesticide requirements of the corn-soybean strip intercropping mode, traditional spraying methods that alternate between the two crop strips over time present several challenges including reduced efficiency, adverse effects on crop growth, and wastage of pesticide solution. This study aimed to develop a direct injection dual-path inline mixing system tailored for the intercropping mode of corn and soybean, ensuring simultaneous and comprehensive pesticide application. By leveraging high-precision sensor data acquisition, the real-time mixing ratio was determined, and fuzzy PID control was employed to achieve real-time regulation of the electric control valve, thereby ensuring precise control of the injection flow rate. The designed direct injection dual-path inline mixing system was integrated and installed, and field tests were conducted using our team’s existing self-propelled sprayer as the test platform, validating the operational performance of the developed system.

Key findings of this study include the following:

(1)

Utilizing the DDS-11A conductivity transmitter, preliminary experiments were conducted to derive a mathematical mode for the conductivity values of the fomesafen and glyphosate mixtures.

(2)

Based on the traditional single-path inline pesticide mixing system, the direct injection dual-path inline pesticide mixing system was extended to cover the entire spectrum of pesticide spraying applications. Leveraging a fuzzy PID control algorithm, we designed the control system and principles for the direct injection dual-path inline mixing system including its physical connection diagram and control flowchart. After developing a method to calculate the real-time mixing ratio data using flow signals, we conducted a spray consistency test to verify the feasibility of the direct injection dual-path inline mixing system.

(3)

The designed direct injection dual-path inline mixing module was integrated into the 3WPG-300 high-clearance self-propelled sprayer, which served as the test platform for field trials. Multiple groups of tests were conducted to ensure the uniformity of mixing and the precision control of the mixing ratio. The maximum coefficient of variation for the mixing ratio was 9.7%, indicating good mixing uniformity. The maximum standard deviation of the mixing ratio was approximately 2.2%, with a maximum average deviation of less than 1.5%, demonstrating excellent stability in adjusting the mixing ratio. When the target mixing ratio was set as a step signal, the maximum average deviation remained below 2.7%, and the maximum response time was 3.4 s, confirming that the mixing ratio can be adjusted accurately and promptly.

Due to the necessity of accurately controlling the electric control valve based on the discrepancy between the target mixing ratio and the real-time mixing ratio, it is imperative to conduct real-time detection of the mixing ratio information and ensure the adjustment speed and accuracy of the electric control valve. The difference in the mixing ratio served as the input for the control system, and a fuzzy PID control method was employed to achieve real-time control of the mixing ratio. Given that the pressure-flow-opening characteristic of the electric regulating valve is nonlinear, the fuzzy PID controller can flexibly adjust the control strategy, thereby better adapting to the dynamic behavior changes of the system and enhancing the precision and stability of the mixing ratio regulation.

5. Conclusions

(1)

A direct injection dual-path inline pesticide mixing system was designed to fulfill the requirements of corn-soybean strip intercropping mode. This system enables the simultaneous pesticide application for both crops, thereby enhancing the efficiency and convenience of weed control operations while significantly reducing the pesticide waste. Through calibration, we established the relationship between the mixing ratio and conductivity of herbicides commonly used in corn and soybean fields. Additionally, we designed the control system and control flowchart for the direct injection dual-path inline mixing system. Experimental validation confirmed the feasibility of calculating the real-time mixing ratio of the water solvent flow and pesticide injection flow data collected by high-precision flow transmitters.

(2)

In order to enhance installation convenience and minimize space occupancy, the pesticide mixing injection module within the designed direct injection dual-path inline pesticide mixing system was integrated, and an optimized overall pipeline layout was implemented. Post-integration, the pesticide mixing injection module retained only essential outlet and inlet ports. Using the 3WPG-300 high-clearance self-propelled sprayer developed by our team as a test platform, the designed direct injection dual-path inline mixing system was installed, and the original pipeline configuration and nozzle positions were adjusted according to the agronomic requirements of the corn-soybean strip intercropping mode. Spray consistency tests revealed that the maximum coefficient of variation in spray volume uniformity across 18 nozzles was 5.96%, indicating excellent spray consistency. Mixing uniformity tests confirmed that the maximum coefficient of variation of mixing uniformity in pesticide injection mixing ratio was 9.7%, demonstrating satisfactory mixing uniformity. Therefore, the designed direct injection dual-path inline pesticide mixing system meets the fundamental requirements for pesticide application in agricultural machinery.

(3)

A control strategy for the two-path inline pesticide mixing system was developed, and its stability, accuracy, and response time were evaluated using a sprayer equipped with a direct injection dual-path inline mixing system. The stability test results indicated that the maximum standard deviation of the mixing ratio was approximately 2.2%, while the maximum average deviation was less than 1.5%. These findings confirm that the mixing ratio stability is satisfactory and meets the operational requirements of the sprayer’s inline mixing system in field applications. From the tests on the adjustment accuracy and response time, it was observed that when the target mixing ratio followed a step signal, the maximum average deviation of the mixing ratio was less than 2.7%, and the maximum response time was 3.4 s. Both the mean deviation and response time decreased as the target mixing ratio decreased. Consequently, these results validate that the designed dual-path inline pesticide mixing system can fulfill the demands for stable, accurate, and rapid adjustment of the pesticide mixing ratio required by plant protection machinery in the corn-soybean strip intercropping mode, demonstrating excellent working performance.

(4)

Based on the aforementioned conclusions, future research will concentrate on enhancing the performance of the inline pesticide mixing system utilized in the corn-soybean strip intercropping mode. Specifically, efforts will be directed toward designing a high-performance inline pesticide mixing system, optimizing the actuator selection, refining the system’s operational principles, and advancing the detection technology for pesticide mixing ratios. Potential approaches include employing control algorithms integrated with artificial neural networks or utilizing image processing techniques for the real-time monitoring of mixing ratios and uniformity. These improvements aim to significantly enhance the stability, accuracy, and speed of mixing ratio adjustments.