Response Prediction and Experimental Validation of Vibration Noise in the Conveyor Trough of a Combine Harvester

Abstract

The noise generated by combine harvesters during operation has drawn growing attention, particularly that of the conveying trough shell, whose noise generation mechanism remains unclear. This study investigated the vibration radiation noise characteristics of conveying troughs by analyzing a chain system with 83 links using numerical simulation and experimental validation. A dynamic model of the conveyor chain system was developed, and the time domain reaction force at the bearing support was used as excitation for the trough shell’s finite element model. Modal and harmonic response analyses were performed to obtain the vibration response, which served as an acoustic boundary input for the LMS Virtual Lab. The indirect boundary element method was used to compute the radiated noise, achieving coupled modeling of chain system vibration and trough shell noise. Simulation results revealed that the maximum radiated noise occurred at approximately 112 Hz, closely matching experimental data. Comparative analysis of transmitted noise at 500 Hz and 700 Hz showed acoustic power levels of 98.4 dB and 109.52 dB, respectively. Results indicate that transmitted noise dominates over structural radiation in energy contribution, highlighting it as the primary noise path. This work offers a validated prediction model and supports noise control design for combine harvester conveying troughs.

1. Introduction

With the accelerating development of agricultural mechanization, combine harvesters have become indispensable equipment in large-scale grain production, offering improved productivity, reduced labor demand, and greater operational efficiency [1]. However, the increasingly sophisticated structure and power configuration of modern combine harvesters have also introduced new engineering challenges, particularly in terms of noise and vibration during operation [2,3]. Excessive noise not only affects the comfort and working efficiency of the operator but may also lead to long-term health issues, including hearing loss, fatigue, and reduced cognitive performance [4]. Long-term exposure to high-intensity mechanical noise can adversely affect the operator’s auditory system, nervous system, and even psychological well-being. Studies have shown that a considerable proportion of agricultural machinery operators suffer from noise-induced hearing loss and chronic fatigue [5]. Moreover, noise can increase stress levels and cause reduced concentration, which may affect operational efficiency and safety [6]. As a result, noise control has become an important objective in the design and refinement of agricultural machinery.

The noise from the conveyor trough is mainly caused by the impact force generated by the meshing of the chain system and the chain wheel and the vibration transmitted to the shell. Therefore, identifying and predicting the noise source location and generation mechanism of conveyor troughs have significant practical application value. In recent years, scholars at home and abroad have conducted extensive research on the noise problems of combine harvesters. Regarding the research on chain system noise, Nakazawa et al. took the chain pins of the continuously variable transmission as the research object and predicted the influence of the difference in cross-sectional area curves on noise [7]; Carletti et al. proposed a simplified pump vibration acoustics model [8], obtained the acceleration spectrum of the measurement points by measuring the vibration of the gear pump shell, solved the noise generated by the gear pump based on modal analysis, and verified the feasibility of the method through experiments; Zhou et al. established a dynamic model of the double-planetary-gear power coupling mechanism, analyzed the vibration response of the shell under the excitation of bearing constraint forces, and, based on the vibration displacement on the shell surface, established a prediction model of the shell radiation noise, and finally optimized the shell through damping structure and reinforcing rib structure [9]; Lee et al. used a perforated thin box in the pressurized chamber window to verify its noise reduction effect on white noise, traffic noise, and construction noise [10]; and Yayin et al. used the acoustic–structural coupling method to study the mutual coupling of noise and structure of the double-screw air compressor and found that at the same frequency, the greater the surface acceleration of the shell, the greater the sound pressure level [11].

The primary mechanism of noise generation in the conveyor trough stems from the internal chain–sprocket system, which operates under complex dynamic conditions. As the chain links engage with the sprocket teeth during continuous motion, periodic impact forces are generated [12,13]. These mechanical shocks are transmitted to the trough shell through the bearings and support structures, inducing structural vibrations that result in radiated noise [14,15]. In addition to these internal excitations, the trough shell is also influenced by external factors such as engine-induced vibrations and road surface irregularities during field travel. This combination of multiple overlapping excitations, compact structural geometry, and complex transmission paths makes the conveyor trough a challenging subsystem to analyze, model, and control from a noise engineering perspective [16].

Despite the significance of this issue, a review of the existing literature reveals a noticeable gap in targeted studies that focus specifically on the noise generation and propagation characteristics of the conveyor trough [17,18,19]. Most previous studies have adopted a macro-level approach, analyzing the overall noise levels of harvesters and attributing noise sources based on generalized measurements and assumptions [20,21,22]. While such approaches offer a broad understanding of noise performance, they fall short in diagnosing localized structural problems or in supporting component-level noise reduction strategies.

Although previous studies have made meaningful progress in the overall noise control of agricultural machinery, especially in the context of combine harvesters, most of these studies primarily focus on general machine-level noise suppression strategies or external sound field control. For instance, some researchers have explored the use of sound insulation materials and vibration dampers to reduce cabin noise exposure [23,24], while others have attempted to optimize the layout of mechanical components to minimize noise generation [25,26]. However, relatively few studies have addressed the specific noise generation mechanisms within the internal structures of key components—particularly the conveyor trough, which plays a critical role in grain transport during harvesting operations [27,28].

In this context, the lack of in-depth research on localized noise sources, especially the interaction between the conveyor chain system and the trough shell, represents a significant research gap [29,30]. The existing literature rarely provides detailed analyses or predictive models that identify how and where such internal noises are generated under different operational conditions [31,32]. Thus, this study proposes a comprehensive research approach that combines theoretical prediction with experimental verification to better understand and control the noise sources within the conveyor trough of combine harvesters [33,34].

The method begins with multi-condition vibration and noise testing, which includes field tests under various load and speed scenarios to collect vibration signals and sound pressure data from different parts of the conveyor trough [35,36]. These tests aim to characterize how the noise levels and vibration patterns change across different operating states, allowing for a more precise identification of dominant noise sources. For example, high-vibration zones at the midsection of the conveyor shell often correlate with weak structural support or resonant frequencies, which are difficult to detect through theoretical modeling alone [37,38,39].

Subsequently, dynamics modeling and simulation analysis are conducted. Based on multi-body dynamics and finite element methods, the internal structure of the conveyor trough—including the chain system, sprockets, and housing—is accurately modeled using tools like ADAMS and ANSYS [40,41,42]. Modal and harmonic response analyses are then performed to predict the vibration response of the system under typical excitation forces generated by the moving chains. These simulations provide detailed maps of displacement, stress, and sound pressure distribution, which serve as a foundation for predicting potential noise hotspots [43,44,45].

Finally, to ensure the validity of the predictive models, the simulation results are compared against the experimental field data. By cross-referencing the predicted vibration and noise distributions with real-world measurements, discrepancies can be identified and model parameters refined accordingly [46,47]. This verification process not only confirms the accuracy of the theoretical approach but also deepens the understanding of how localized structural interactions contribute to overall noise emissions. Furthermore, through this combined method, the mechanisms behind chain–shell interaction-induced noise are clarified, which is essential for developing targeted noise control strategies in future harvester designs [48,49].

Moreover, direct noise measurement of the conveyor trough shell in field environments presents considerable practical challenges. The internal space is often too compact to accommodate sensors without interfering with system operation [50,51]. Furthermore, the simultaneous operation of multiple noise-generating components makes it difficult to isolate the contribution of the trough shell. Given these limitations, traditional experimental techniques alone are insufficient for identifying the precise sources and mechanisms of noise within the conveyor subsystem.

To overcome these limitations, this study proposes a modeling and simulation approach based on dynamic response prediction, combined with experimental validation. This approach allows for the decoupling of various excitation sources, quantification of individual contributions to structural vibration, and prediction of radiated acoustic energy. The response prediction method integrates multi-body dynamic modeling of the chain system and finite element simulation for research.

2. Materials and Methods

2.1. Structure of the Conveying Channel and Chain Engagement

The conveying channel of a combine harvester is a crucial component that connects the cutting system and the threshing system, being mainly used to transport the harvested crops from the header to the threshing device. Its structure generally consists of the following parts: the conveying channel shell, the conveying chain (or belt), the guide rail, the drive mechanism, and the adjustment mechanism. The conveying channel shell is the framework of the entire conveying channel and usually made of high-strength steel plates, which are wear-resistant and impact-resistant. The interior of the shell provides support and a running track for the chain or belt conveying mechanism and also protects the conveyed crops from spilling out. To meet the conveying requirements of different crops, the shell is usually designed with a certain inclination to ensure a smooth flow of the material into the threshing device. The inclination angle plays a critical role in facilitating the gravitational movement of granular materials and preventing blockages or reverse flow. Based on both literature references and our preliminary experiments, the optimal inclination angle varies depending on the type of crop: for wheat, an angle of approximately 45° provides a good balance between flow rate and stability; for rice, which has smaller and smoother grains, a slightly lower angle of around 40° is more suitable to maintain continuous movement without slippage; and for maize, due to its larger kernel size and higher frictional resistance, a steeper angle of about 50° is preferred to ensure consistent feeding into the threshing mechanism. These angles were selected to optimize the conveying efficiency and adapt to the specific flow characteristics of each crop type. The conveying chain or belt is the core component of the conveying channel, responsible for quickly and continuously transporting the cut crops to the threshing device. The conveying chain is typically made of high-strength, wear-resistant materials, with claws or protrusions on its surface to increase the grip on the crops and the stability of the conveying process. For certain special crops, a flexible belt may be used to reduce damage to the crops. The guide rail is an auxiliary support device inside the conveying channel, located along the running path of the chain or belt. Its main function is to reduce the friction of the chain’s operation, improve the conveying efficiency, and prevent the chain from jumping or deviating during high-speed operation. The drive mechanism provides power to the conveying channel and is usually composed of a hydraulic motor or a gear transmission system. The drive mechanism is connected to the conveying chain or belt through gears, sprockets, or tensioning wheels to ensure that the conveying mechanism operates at the predetermined speed. At the same time, the design of the drive mechanism needs to be coordinated with the entire harvester’s power system to ensure that the conveying channel works in synchronization with the cutting and threshing systems. The adjustment mechanism is used to regulate the working state of the conveying channel; for example, the tensioning device can adjust the tightness of the chain or belt to ensure its stability during operation. Some combine harvesters are also equipped with angle adjustment devices to adapt to the harvesting needs of different terrains or crop types [12,33,52,53,54].

The chain engagement structure of the conveying channel in a combine harvester is the core mechanism for conveying crops, and its design directly affects the efficiency and stability of crop conveying. The chain engagement structure mainly consists of the conveying chain, sprockets, drive shaft, and tensioning device. The conveying chain is the core component and is usually made of high-strength steel, with claws or protrusions on its surface to grip and fix the crops, ensuring that the crops do not fall off during conveying. The chain is arranged in a closed loop, passing around the drive sprocket and the guide sprocket and closely cooperating with the guide rail inside the conveying channel to reduce running friction and improve stability. The sprocket is a key component for power transmission in the chain system and is divided into the drive sprocket and the driven sprocket. The drive sprocket is connected to the drive shaft and is usually powered by hydraulic or mechanical transmission, precisely engaging with the chain to drive its operation through rotation. The engagement between the sprocket teeth and the chain links needs to be precise to avoid tooth skipping or slippage. The tensioning device is used to adjust the tightness of the chain to ensure it does not derail or become loose during operation. It is usually composed of springs or a hydraulic system and can automatically or manually adjust the tension of the chain to extend its service life and improve conveying efficiency [35,37,55,56].

During the actual operation of a combine harvester, the sources of vibration and noise in the conveying channel are extensive. In an unloaded state, they can roughly be classified into chain system vibration noise, bearing vibration noise, and conveying channel shell vibration noise. Through spectral analysis of the noise of the chain and rake-type conveying channel in the literature, it is known that although there are many factors affecting the noise of the conveying channel, the frequency that causes the peak noise of the conveying channel is close to the engagement frequency of the chain system. Therefore, it is determined that the main cause of the vibration and noise in the conveying channel is the engagement impact of the chain system. The driving sprocket and the chain of the conveyor chain system transfer the transportation power through their mutual interaction, as shown in Figure 1. During this process, when the conveyor chain system is subjected to internal and external excitations, it will cause vibration and noise. The internal excitation of the conveyor trough mainly comes from the fluctuation of the chain link speed during meshing, while the external excitation mainly comes from the fluctuation of the input torque and speed, the change in the road surface, and the fluctuation transmitted from other working components of the combine harvester. When the conveyor trough has feed-in, as shown in Figure 2, the external excitation also includes the friction between the grains and the conveyor trough shell, as well as the change in the system mode caused by the feed-in of the material [40,42,43,45].

Internal and external excitations cause a dynamic impact on the conveyor trough during operation, resulting in dynamic vibration responses of the conveyor chain system and the conveyor trough shell. The noise of the conveyor trough can be divided into structural transmission noise and air transmission noise. The former mainly consists of the radiated noise caused by the vibration of the bridge housing due to the meshing impact force of the chain system being transmitted through the feed line drive shaft and bearings, and the radiated noise of the chain system is due to the meshing impact force. The latter is mainly the secondary sound radiation formed by the shell structure due to the vibration of the chain system radiating noise through the structure of the conveyor trough shell, which is the transmitted noise.

2.2. Analysis Method

When an object vibrates, the deviation of each point on its structure from the equilibrium position can usually be represented by a vector, which has a certain proportional relationship. This is the mode of the structure, which is the inherent vibration characteristic of the structure. When solving the structure’s mode, the aim is to obtain the natural frequency and natural mode shape of the structure. If a linear time-invariant system has N degrees of freedom, its motion equation can be expressed as Equation (1),

$$\left[M\right]\left\{\ddot{x}\left(t\right)\right\}+\left[C\right]\left\{\dot{x}\left(t\right)\right\}+\left[K\right]\left\{x\left(t\right)\right\}=\left\{F\left(t\right)\right\}$$

(1)

where [M] is the mass matrix; [C] is the damping matrix; [K] is the stiffness matrix; x(t), ẋ(t), and ẍ(t) are the displacement vector, velocity vector, and acceleration vector, respectively; and F(t) is the excitation force vector. In practical engineering, the excitation force is mainly divided into random load, impact load, and periodic load.

When [C] = 0, it is called undamped free vibration. If the external load is 0, Equation (1) can be written as Equation (2):

$$\left[M\right]\left\{\ddot{x}\right\}+\left[K\right]\left\{x\right\}=0$$

(2)

Substituting x = e^(jωt) into Equation (2) gives the characteristic Equation (3):

$$\left(\left[K\right]-{\omega }_i^2\left[M\right]\right)\left\{X\right\}=\left\{0\right\}$$

(3)

Equation (3) is a homogeneous equation of the displacement amplitude vector [x], ω_i is the natural frequency of the system, and solving this equation results in obtaining the mode shape vector corresponding to ω_i. According to the modal analysis theory, the effective modal mass ratio of the low-order mode is higher, and its influence on vibration is greater. Therefore, the focus can be placed on the low-order modal analysis of the key components of the conveyor trough.

Sound waves are a type of compressional wave. Sound pressure is the difference between the medium pressure and the static pressure when the sound wave causes the air pressure to change. When the measurement point is in the dense part, the sound pressure is positive, and when it is in the rarefied part, the sound pressure is negative. Sound pressure level is a characteristic unit that reflects the human ear’s response to the strength of sound, with the unit of dB. The relationship between sound pressure level and sound pressure is given by Equation (4):

$$L_p=10\lg {\displaystyle \frac{p^2}{{p_0}^2}}$$

(4)

In the equation, L_P is the sound pressure level, dB; p is the effective sound pressure, Pa; and p₀ is the reference sound pressure, which is 2 × 10⁻⁵ Pa.

The effective value of the sound pressure within time T is calculated as Equation (5):

$$\stackrel{-}{p}=\sqrt{{\displaystyle \frac{1}{T}}{\int }_0^T{\left(p_{va\mathrm{r}}\left(t\right)\right)}^2\mathrm{d}t}$$

(5)

p_var is the instantaneous sound pressure. Sound pressure is a function of time and space, and it changes with time and position during propagation. The sound pressure at a certain moment is called the instantaneous sound pressure. In actual measurement, when measuring the noise during the stable operation of a machine within the audible range, the sound pressure will change multiple times. However, due to the limited sensitivity of the human ear to sound pressure changes, it is difficult to accurately perceive such fluctuations. Therefore, although the sound pressure varies, the overall sound seems relatively stable. In practice, the effective value of the sound pressure is usually used to calculate the sound pressure level.

Both sound pressure and sound pressure level are objective physical quantities used to describe sound, but they cannot fully reflect the human ear’s perception of sound. Two noises with the same sound pressure level, such as high-frequency noise, sudden noise, or noise with large fluctuations in sound pressure level, have a greater disturbance to the human ear than low-frequency and stable noise. To consider the subjective perception of humans, many researchers have conducted extensive experiments and obtained the equal-loudness curves that reflect the subjective perception of the human ear, and proposed the concept of loudness. By weighting the objective sound pressure level, the more realistic impact of noise on the human ear can be obtained.

2.3. Numerical Simulation of Vibration of Conveyor Trough Shell

The shell was modeled in 3D using Solidworks Ver.2022 software. To facilitate subsequent analysis, unnecessary holes, bolts, chamfers, threads, roller support parts, and other unnecessary parts of the conveyor trough shell were deleted, and the conveyor trough shell was placed at an angle of 30° with the ground in the assembly. The vibration data of the surface of the conveyor trough device were calculated with Workbench simulation verification software.

The 3D model of the built shell structure was saved in the file with the suffix “.x_t” format for subsequent modal analysis. After the conveyor trough shell model was imported, Workbench automatically generated bound contacts, but the structural mesh was not continuous. In Catia V5 software, the assembly of the shell was combined into a single part. During the operation, the initial positions of the conveyor chain system and the conveyor trough shell did not match, so the position of the conveyor trough shell could be adjusted in Catia for subsequent analysis.

The finite element model of the conveyor trough shell was established through meshing, as shown in Figure 3, with 120,503 nodes and 59,979 elements. The material of the conveyor trough shell was uniformly set as structural steel, with the following performance parameters: elastic modulus E = 2.0 × 10¹¹ Pa, Poisson’s ratio u = 0.3, and density ρ = 7850 kg/m³. According to the actual installation situation of the conveyor trough shell, the constraint condition was as follows: fixed support was applied at the connection between the shell and the hydraulic device, the surface of the introduction pipe bracket base, and the cylindrical surface of the bolt connection of the introduction pipe bracket. Modal analysis was conducted on the conveyor trough shell, and the first fifty natural frequencies are shown in

Table 1. The first 50 order modal frequencies of the conveying trough shell.

Order	Frequency/Hz	Order	Frequency/Hz	Order	Frequency/Hz
1	45.177	18	180.06	35	293.76
2	51.964	19	189.5	36	298.13
3	62.351	20	190.04	37	311.7
4	66.732	21	201.46	38	317.75
5	68.365	22	208.37	39	322.63
6	78.48	23	209.26	40	334.9
7	92.104	24	221.66	41	347.79
8	100.55	25	224.49	42	349.43
9	109.46	26	227.8	43	357.
10	115.87	27	237.26	44	358.29
11	123.18	28	238.13	45	365.26
12	127.92	29	242.79	46	367.35
13	136.99	30	251.6	47	375.2
14	144.86	31	265.21	48	377.87
15	155.1	32	270.59	49	382.73
16	159.68	33	278.94	50	387.93
17	174.49	34	286.93

.

From Table 1, it can be seen that the modal frequencies of the conveyor trough shell were mostly concentrated in the low-frequency range. From the vibration mode shapes of the conveyor trough shell, it can be observed that there was almost no vibration displacement at the constrained parts. The vibration displacement of the shell appeared on multiple surfaces of the shell. During the operation of the combine harvester, by avoiding the natural frequencies and vibration mode directions, the noise of the conveyor trough shell could be reduced, which is of great significance for the design improvement of the conveyor trough.

The transient dynamic analysis of the conveyor trough shell was conducted based on the full method using Ansys Workbench (ANSYS 2023 R1). Since the magnitude of the bearing load cannot be defined in the form of a table in Workbench, a surface load was applied at the connection between the conveyor trough shell and the bearing to simulate the varying load of the bearing. At the connection between the header and the conveyor trough, a force of 1500 N was added to simulate the weight of the header. The addition of boundary conditions is shown in Figure 4. The end time of the step was 0.76 s, large deformation and automatic time step were turned off, and the number of sub-steps was set to 400.

The vibration of the shell can be evaluated by the vibration acceleration on its surface. Four measurement points were selected on the surface of the shell, including two points on the upper surface of the shell and one point on each of the left- and right-side surfaces, as shown in Figure 5. Figure 6 shows the vibration displacement of the four measurement points under the action of the bearing load. It can be seen that the vibration amplitude of measurement point 4 was greater than that of measurement point 3, indicating that this area is more affected by the chain transmitted from the header.

2.4. Modal Analysis and Vibration Response Analysis of Conveyor Shell

A conveyor chain system with 83 links on one side was selected as the analysis object. The meshing impact force in the frequency domain was used as the excitation to conduct harmonic response analysis on the conveyor chain system, and the bearing reaction force in the time domain was used as the excitation to solve the vibration response of the conveyor trough shell. The boundary element method was used to solve the radiated noise of the conveyor chain system and the conveyor trough shell based on the vibration response. In LMS Virtual Lab 12 software, the acoustic boundary element models of the conveyor chain system and the conveyor trough shell were established to solve the radiated noise generated by the conveyor chain system and the conveyor trough shell, respectively. Finally, based on the modal analysis of the conveyor trough shell and the vibration response data of the conveyor chain system, a transmission noise simulation model was established to determine the transmission noise of the conveyor trough. Due to the small gap structure of the chain system, when the grid size is set to be large, the envelope grid will directly ignore these structures. Therefore, a smaller grid size envelope grid was generated first, and then coarsened. Subsequently, the envelope processing was performed on the surface grid to obtain the envelope grid, which had 44,270 elements and 22,124 nodes.

2.4.1. Modal Analysis

After the envelope grid was completed, modal analysis was conducted on it using the Workbench software. Modal analysis was performed on the key components of the conveyor chain system to solve the natural frequencies and their corresponding mode shapes. The material properties were assigned, as shown in

Table 2. Material properties.

Components	Chain Wheel	Chain Harrow	Roller	Pin Shaft	Sleeve	Passive Drum
Material	45 Cr	Q235-B	20 CrMo	20 CrMo	20 CrMnMo	Structural steel
Density kg/m3	7820	7840	7840	7840	7870	7850
Poisson’s ratio	0.290	0.278	0.278	0.278	0.254	0.3
$\mathrm{Elastic} \mathrm{modulus} \mathrm{Pa}$	2.06 × 1011	2.10 × 1011	2.10 × 1011	2.10 × 1011	2.07 × 1011	2.0 × 1011

.

Since the low-order modes have a greater impact on the dynamic characteristics of the structure, with longer vibration wavelengths, their mode shapes have a significant influence on the stability of the structure. The high-order modes correspond to higher frequencies and cause smaller amplitudes. Due to space limitations, only the natural frequencies and mode shapes of the low-order modes of the conveyor trough system parts are listed here.

2.4.2. Vibration Response Analysis

Modal analysis of the conveyor chain system was conducted to obtain the natural frequencies and mode shapes of the chain system. Based on this, harmonic response analysis could be conducted on the conveyor chain system. During the operation of the chain system, the meshing impact force is transmitted through the sprocket, chain plate, and feed line drive shaft, causing overall vibration of the chain system. The meshing contact force results obtained from the multi-body dynamics simulation model of the conveyor chain system described in the previous chapter were used as the boundary load conditions for the harmonic response analysis. The obtained time domain contact force load was transformed into a frequency domain load through Fourier transform and used as the excitation. The frequency range for the harmonic response analysis was selected as 20 to 1500 Hz, with 74 solutions and an interval of 20 Hz. The vibration results corresponding to different frequencies under the excitation were calculated.

It can be known from the cloud graph of the deformation amount at the corresponding frequency obtained by simulation, at 500 Hz, the deformation mainly occurred near the left side of the conveyor chain system, with significant deformation in the chain rake and sprocket. At 680 Hz, the deformation mainly occurred in the upper chain near the passive roller. At 700 Hz, the deformation mainly occurred in the lower chain of the chain system. The chain of the conveyor chain system responds violently under the action of excitation.

2.5. Equipment and Conditions for Vibration Testing of Conveyor Troughs

Although there are certain differences in the structure of different models of combine harvesters, their working principles are basically the same. To analyze the vibration of the working components of the combine harvester, we selected a certain domestic crawler combine harvester for the vibration test of the working components. Vibration can be measured by analyzing data such as acceleration, velocity, and displacement. In this test, the acceleration value was used to analyze the working components of the combine harvester. The Donghua dynamic signal acquisition and analysis instrument and the three-axis acceleration sensor were used as data sampling tools, as shown in Figure 7. The test equipment used is the DH5902 signal acquisition system of Donghua Test, which is produced by Jiangsu Donghua Test Technology Co., Ltd., Jingjiang, China. The sampling frequency was set to 6 kHz. The parameters and specifications of the equipment used in the test are shown in

Table 3. Equipment for vibration test.

Instrument name	356A16-type accelerometer
Performance indicators	Range/g	Frequency response/kHz	Sensitivity/(mV·g−1)	Lateral sensitivity/%
Technical parameters	±50	0.3~6	100	<5
Instrument name	DH5902N dynamic signal acquisition instrument
Performance indicators	Number of channels	Maximum sampling frequency/kHz	Distortion degree	Signal input method
Technical parameters	32	100	<0.5	IEPE

.

The test site was located in a certain experimental field in Xinfeng Town, Jiangsu Province, as shown in Figure 8. The combine harvester used in the test is the Reilong combine harvester, which is produced by Jiangsu World Group Co., Ltd. of China. Since the combine harvester was in actual use, the staff would adjust the working parameters of the machine according to the actual situation in the field when driving the combine harvester, causing the overall situation of the combine harvester to constantly change. Due to the complex environment in the field, the external excitation on the combine harvester was also different. Considering the above reasons, this paper mainly considers the vibration of the combine harvester under specific working conditions during the test. The set test conditions are shown in

Table 4. Test conditions of combine harvester in the field.

Experimental Conditions	1	2	3	4	5
Testing environment	Rice fields	Rice fields	Rice fields	Rice fields	Rice fields
Harvesting operation status	Only engine operation	Engine and threshing cleaning work	The whole machine starts, remains stationary, and has no load	The whole machine starts only for walking	Complete machine startup, walking, and harvesting
Speed km/h	0	0	0	8	8

.

3. Results

3.1. Numerical Prediction of Vibration Radiation Noise of Conveyor Trough Shell

Through the acoustic boundary element simulation of the conveyor trough shell, the sound pressure level cloud diagrams at the field point grids were obtained, as shown in Figure 9. These are the sound pressure level distribution cloud diagrams at 0.1 s, 0.2 s, 0.3 s, and 0.4 s. It can be seen that the locations with higher sound pressure levels were at the feed end, discharge end, and two side plates of the conveyor trough shell. It should be noted that although the sound pressure levels shown at the field point grids are relatively high, they are not A-weighted sound pressure levels and are not suitable as indicators of the impact on human ears. To better reflect human auditory sensitivity, the A-weighted sound pressure level at detection point 16 was calculated, and the results are shown in Figure 10. While the overall A-weighted sound pressure level appears low, indicating limited impact on human hearing under the excitation of the bearing reaction force, a closer inspection of the frequency spectrum at this point reveals distinct peaks at 500 Hz, 700 Hz, and 1400 Hz. These frequencies correspond to the meshing frequency and impact characteristics of the conveyor chain system, suggesting that localized periodic excitations still contribute to notable acoustic energy in specific bands. Although these peaks may not significantly affect the A-weighted total level, they can be important for evaluating mechanical resonance and potential vibration-related fatigue. This suggests that the conveyor system exhibits dominant acoustic response characteristics at these frequencies during operation.

The radiation noise characteristics of the conveyor chain system were analyzed using LMS Virtual Lab, with the noise calculation frequency range set from 40 to 1500 Hz and the same solution step as the harmonic response analysis. By combining modal analysis with sound pressure distribution results, we found that the middle section of the conveyor trough, due to insufficient rigid support, was more prone to local resonance under chain movement excitation, leading to higher sound pressure response in that area. Further analysis of the chain system’s noise distribution allowed us to identify the main frequency components and the key locations of the noise sources.

Figure 11 shows the sound power curve of the conveyor chain system. It can be seen that the first noise peak at 500 Hz reached 78.04 dB. As the frequency continued to increase, the noise decreased and then increased again, reaching 75.34 dB at 700 Hz. Finally, a small peak appeared at 1400 Hz, with a noise estimate of 64.82 dB. Figure 12 shows the sound pressure distribution cloud map corresponding to the above frequencies. From the cloud map, it can be seen that the peaks were mainly distributed on the chain system, and the maximum sound pressure values often occurred on the chain rake. This corresponds to the main deformation positions in the modal analysis: the vibration deformation mainly occurs in the chain part of the conveyor chain system, and the sound radiation area of the chain rake is larger than that of other parts, so this part radiates more noise energy. When the frequency is higher, the impact force of meshing is transmitted to the passive roller, which also corresponds with the results of the modal analysis.

The noise distribution of the conveyor chain system was observed at plane field point 3, resulting in Figure 13 and Figure 14. The results show that at 500 Hz, the highest sound pressure levels appeared at the active sprocket and the middle section of the chain system, with the largest red areas. Peak1 corresponds to the middle of the conveyor chain system, while Peak2 corresponds to the sprocket and the feed shaft section. It is evident that the highest sound pressure was generated in the middle of the conveyor chain system, which differs from traditional chain drive noise analysis results—where the maximum sound pressure typically occurs on the surface of the sprocket. In this study, the high sound pressure area was located in the middle of the line connecting the contact point between the chain and the sprocket and the contact point between the chain and the driven roller. This area lacks structural support, resulting in larger local deformation and, consequently, stronger sound radiation.

At 700 Hz, the position with the highest sound pressure level on the plane field point 3 was at the feed shaft, and the second highest was near the passive roller. It can be seen that at this time, the vibration was transmitted to the passive roller and caused significant vibration noise. It is worth noting that at 500 Hz, the position with the highest noise on the chain system surface was at the chain rake, but by observing the plane field point 3, it can be seen that the sound pressure level of the noise generated by the chain rake decreased significantly after being transmitted to this plane field point. It can be seen that although the noise generated by the chain rake was large, the transmission efficiency in this direction was not high.

3.2. Verification of Conveyor Groove Vibration Test

To comprehensively understand the vibration characteristics of the conveyor of the combine harvester, during the test, for conditions 1, 2, and 3, a total of eight measurement points were selected at the main vibration source positions of the combine harvester and on the outer shell of the conveyor. The measurement points of the conveyor of the combine harvester were used as the response, and the other measurement points were used as the vibration sources. The vibration was measured using a three-axis acceleration sensor. The layout of each measurement point is shown in

Table 5. Location of vibration measuring points.

Condition	1, 2, 3				4, 5
Detection point	1	2	3	4	1	2	3
Installation position	On the conveyor trough panel	Left side of conveyor trough panel	Conveyor trough panel on the right	Conveyor trough drive shaft bearing seat	On the conveyor trough panel	Left side of conveyor trough panel	Conveyor trough panel on the right
Detection point	5	6	7	8	4	5	6
Installation position	Vibration screen drive shaft bearing seat	Fan drive shaft bearing seat	Threshing drive shaft bearing seat	header	Conveyor trough drive shaft bearing seat	Threshing drive shaft bearing seat	Header

. During the test, the combine harvester was started in the rice field in sequence: the engine, the threshing and cleaning device, the header, and the conveyor. After each start operation, it was ensured that each working condition lasted for more than 15 s. Conditions 4 and 5 were the process of starting the entire combine harvester in the field and completing a period of walking before harvesting the rice in the field. And during the tests of conditions 4 and 5, the noise test of some positions of the combine harvester was also completed.

The noise test was conducted and the Fourier transform was performed on some operating conditions to obtain the corresponding sound pressure frequency domain data. The frequency and sound pressure measurements are shown in

Table 6. Spectrum diagram of sound pressure under different working conditions.

(a) Small Throttle of the Engine		(b) Big Throttle of the Engine		(c) Threshing and Cleaning Start		(d) Whole Machine Startup
Frequency	Sound Passure /Pa	Frequency	Sound Passure /Pa	Frequency	Sound Passure /Pa	Frequency	Sound Passure /Pa
58.594	0.597	83.008	0.782	117.188	1.158	95.215	1.127
117.188	0.440	473.633	0.439	95.215	0.926	112.305	0.570
29.297	0.431	385.742	0.376	190.43	0.564	102.539	0.707
146.484	0.162	114.746	0.286	85.449	0.561	119.629	0.679
131.836	0.111	122.07	0.267	168.457	0.560	168.457	0.648

.

Table 6a shows the frequency and sound pressure measurements of the engine at a small throttle. By converting the main frequencies, it can be known that the engine speed at this time was 1770 rpm, 58.594 Hz was the main frequency of the engine, and 29.297 Hz was the half frequency of the engine. It can be known that the noise at the measurement point above the conveying trough at a small throttle was mainly caused by the engine. Similarly, when the engine was at a large throttle, the peak frequency of the sound pressure was the main frequency of the engine. After the combine harvester started the threshing and cleaning device, it can be seen from Table 6c that the sound pressure value had a significant change, and in the frequency and sound pressure measurements, the peak sound pressure was no longer the ignition frequency of the engine. After the whole machine started, the main peak frequency of the sound pressure appeared near 112.305 Hz, close to the meshing frequency of the conveying chain in the conveying trough, indicating that the start of the conveying trough and the header had a significant impact on this sound pressure measurement point. The summary is shown in

Table 7. Summary of results.

Parameter	Value/Result	Notes
Frequency range (Hz)	40–1500 Hz	For noise calculation of conveyor chain system
Maximum sound pressure level (dB)	87.2 dB	At the active sprocket
Peak sound pressure location	Middle of conveyor chain system	Location of highest noise at 500 Hz
Sound pressure distribution	Highest at feed and discharge ends	High SPL at both ends of the conveyor trough
Transmission noise impact	Greater than structural noise	Transmission noise contributes more energy

.

4. Discussion

4.1. Transmission Analysis of Conveyor Trough Shell

The transmission noise of conveyors is closely associated with the vibration noise generated by the internal conveyor chain system. It results from the acoustic–vibration coupling between the chain system and the conveyor housing. Prior to analyzing the transmission noise characteristics of the conveyor housing, it is necessary to import its constrained mode shapes and define the coupling relationships between the structural and acoustic meshes. In this study, the structural meshes of the conveyor housing (excluding support components) and the internal conveyor chain system were processed using HyperMesh and applied as the acoustic boundary meshes. The acoustic mesh is illustrated in Figure 15. Additionally, the mesh of the conveyor housing was designated as the field point mesh to capture the sound pressure distribution. It should be noted that the analysis of the transmission noise characteristics of the conveyor housing in this study was primarily based on vibro–acoustic coupling simulations using a combination of finite element and boundary element methods and has not yet been experimentally validated. Although the simulation model was developed to closely approximate real operating conditions, using reasonable boundary constraints and material parameters, the results remain theoretical to some extent. Due to the complexity of building an experimental setup and the challenges of accurate testing, physical experiments have not been conducted at this stage. In future work, we plan to perform synchronized measurements of structural responses (such as acceleration and displacement) and the surrounding sound field in order to validate the simulation results and enhance the reliability and engineering applicability of the model.

After importing the modal analysis file and the acoustic mesh into the acoustic analysis module of LMS Virtual Lab, the material and properties of the fluid medium were defined and the preprocessing of the acoustic mesh was completed. Then, the two parts that constitute the acoustic mesh as the housing element group and the chain system element group were defined. After the vibration data of the conveyor chain system were imported, its vibration displacement data were used as the boundary condition of the chain system element group. Considering the actual installation position of the conveyor, a baffle was set below the conveyor model to simulate a fully reflective ground, and the coupled boundary element model was obtained, as shown in Figure 16.

Figure 17 shows the sound power curve of the conveyor housing after the acoustic–vibration coupling. At 700 Hz, the sound power reached the maximum value of 109.52 dB; at 500 Hz, it reached 98.4 dB. Compared with the sound power of the conveyor chain system, although only the noise of the chain system exists in the model, the simulation results show that the frequency corresponding to the maximum sound power after coupling changed.

4.2. The Difference Between Actual Situation and Simulation

Several field point sound pressure level cloud diagrams with higher sound pressure levels are shown in Figure 18 and Figure 19. At 700 Hz, the position with the highest sound pressure level appeared on the lower plate of the conveyor housing; at 500 Hz, the position with the highest sound pressure level was at the location closest to the feed line drive shaft of the conveyor, and the other was near the active sprocket on the upper panel of the conveyor housing. By comparing the left- and right-side plates of the conveyor housing, it can be found that the noise generated near the header drive chain of the conveyor housing was greater. The areas in red on the cloud diagram are mostly planes perpendicular to the chain rake, indicating that these are the positions that need to be optimized.

In this study, structural noise (caused by the vibration of the conveyor housing) and transmission noise (generated by the internal conveyor chain system and radiated through the housing) are considered approximately incoherent sound sources. In practical engineering applications, perfectly coherent sound sources are almost non-existent; therefore, treating these two sources as incoherent is a reasonable assumption for engineering analysis. Based on this assumption, the superposition of their sound pressure levels can be estimated using the principle of noise superposition. When the sound pressure levels of two incoherent sources differ significantly, the source with the lower sound pressure contributes minimally to the overall level and can be considered secondary in noise control strategies.

Figure 20 shows the transmission noise. Simulation results show that at the same detection points, the A-weighted sound pressure levels of transmission noise were generally higher than those of structural noise, indicating that transmission noise radiates more energy and has a greater acoustic impact. Therefore, in noise reduction design, priority can be given to mitigating transmission noise, for example, by optimizing chain pitch, tensioning mechanisms, or improving the acoustic insulation of the housing.

It should be noted that this conclusion is based solely on simulation analysis without experimental validation. Since complex coupling may exist between the conveyor chain system and the housing, their coherence characteristics require further confirmation through experimental measurements, such as synchronized structural response and acoustic field testing. The assumption of incoherence adopted in this study serves as a simplification for modeling purposes and does not imply complete physical independence between the two noise sources.

In this study, we conducted a detailed analysis of the noise characteristics of a conveyor trough system and proposed the noise propagation features based on acoustic–vibration coupling. Compared with other similar studies, our findings show certain differences. For example, some previous studies have indicated that noise in traditional chain drive systems is mainly concentrated at the sprocket surface. However, our results reveal that the central area of the conveyor trough exhibits higher noise levels. This is primarily due to the lack of sufficient structural support in this region, which makes it more prone to local resonance. Additionally, in terms of noise frequency distribution, our analysis suggests that high-frequency components contribute more significantly to the overall noise level.

The comparison with other research demonstrates that differences in experimental conditions, structural configurations, and analytical methods can lead to different noise propagation patterns and characteristics. Therefore, this study not only provides new insights into the noise characteristics of a conveyor trough system but also establishes a reference framework for future research. In particular, the results contribute to the localization of noise sources and understanding of noise transmission mechanisms, offering a theoretical basis for effective noise control strategies.

In conclusion, this study reveals the complexity and diversity of a conveyor trough system’s noise, highlights the concentration of noise in the central region due to insufficient support, and compares the findings with other similar studies, underscoring the originality of our work. Future research will focus on experimentally validating the simulation results and addressing practical noise control issues in real operating environments to further enhance the noise reduction performance of agricultural machinery.

5. Conclusions

This paper uses finite element, boundary element, and experimental methods to calculate and analyze the vibration radiation noise of the conveyor housing and obtains the following conclusions:

(1)

The simulation results of the vibration radiation noise of the conveyor housing are consistent with the experimental results; the sound pressure levels at the fundamental frequency of 112 Hz are basically consistent; the established noise numerical prediction model is basically accurate, and the prediction method is feasible.

(2)

The noise generated by the engine and the conveyor chain system structure on the sound pressure level of the conveyor housing was compared and analyzed. When the combine harvester is operating with the conveyor and header running, the average A-weighted sound pressure level decreases significantly, indicating that the noise generated during harvesting is more intense than in other operating conditions.

(3)

However, the transmission noise of the conveyor housing at 500 Hz and 700 Hz was compared and analyzed, with sound power levels of 98.4 dB and 109.52 dB, respectively. The research results show that when the sound pressure levels produced by two sound sources differ significantly, the sound source with the lower sound pressure level has almost no effect on the superimposed result. Compared with the sound pressure at the same detection point, the influence of the transmitted noise at this point is greater than that of the structural noise, and the transmitted noise transmits more energy.

The findings of this study provide a theoretical basis and design reference for noise control in the development of combine harvesters and similar agricultural machinery. By applying the proposed noise–vibration coupling simulation method and optimization strategies, manufacturers can reduce noise transmission into the operator cabin, thereby improving the working environment for tractor operators, reducing noise exposure, and enhancing operational comfort and efficiency. Moreover, this method is versatile and applicable to structural noise analysis in other agricultural and industrial machinery, enabling virtual simulation and early-stage noise prediction to identify potential acoustic issues, reduce modification costs, and enhance overall machine performance. These outcomes support the advancement of agricultural equipment toward greener, quieter, and more efficient systems, offering valuable guidance for intelligent and human-centered machinery design.