Method for adjusting an adjustable bearing block for a conveyor roller

Abstract

This invention relates to the field of conveyor roller technology, specifically to an adjustment method for an adjustable bearing housing for conveyor rollers. It solves the technical problem that existing solutions have significant limitations in terms of real-time performance, accuracy, and adaptive capability, affecting the overall smoothness of the conveying process. The method includes: acquiring multi-source sensing data of the target cargo during the conveying process; determining the physical attribute vector of the target cargo based on mechanical and visual data; determining the center-of-gravity offset vector of the target cargo during the conveying process based on displacement data; determining adjustment parameters based on the physical attribute vector, the center-of-gravity offset vector, and the historical adjustment records of the target cargo; and adjusting the bearing housing according to the adjustment parameters. This invention is applicable to conveyor roller adjustment scenarios.

Description

Technical Field

[0001] This invention relates to the field of conveyor roller technology, and more specifically to an adjustment method for an adjustable bearing housing for conveyor rollers. Background Technology

[0002] In automated storage and retrieval systems (AS / RS) and intelligent logistics conveying systems, roller conveyors, as core continuous conveying equipment, play a crucial role in the efficient and stable transport of various goods. Their operational accuracy and stability directly affect the throughput efficiency, cargo safety, and reliability of subsequent sorting and scanning processes within the entire warehousing system. Currently, adjustable bearing housings are widely used in conveyor design to adjust the roller axis position to adapt to different loads and operating conditions. These bearing housings typically allow for manual or semi-automatic fine-tuning of height and angle via mechanical means. The adjustment relies heavily on preset fixed parameters or the operator's experience and judgment; some advanced systems may also integrate simple displacement or pressure sensors for feedback control. However, facing complex and ever-changing actual operating environments, especially considering the significant differences in cargo weight, size, and surface characteristics, as well as dynamic factors such as elastic deformation of the frame during transport, existing technologies have significant limitations in terms of real-time adjustment, accuracy, and adaptability. This makes it difficult for the conveyor to maintain optimal roller posture and trajectory when handling different goods, thus affecting the overall smoothness of the transport and increasing the risk of cargo deviation, jamming, or even falling. Summary of the Invention

[0003] To address the significant limitations of existing technologies in terms of real-time performance, accuracy, and adaptability, which affect the overall stability of conveying, this invention aims to provide an adjustment method for an adjustable bearing housing for conveyor rollers. The specific technical solution adopted is as follows:

[0004] In a first aspect, the present invention provides an adjustment method for an adjustable bearing seat for a conveyor roller, the method comprising: acquiring multi-source sensing data of a target cargo during the conveying process; the multi-source sensing data including: mechanical data, displacement data, visual data, and identity data; the mechanical data reflecting the interaction force between the cargo and the conveyor roller; the displacement data reflecting the temporal changes in the position of the conveyor roller bearing seat; the visual data reflecting the surface properties of the cargo; and the identity data determining the historical adjustment records of the target cargo; and determining a physical attribute vector of the target cargo based on the mechanical data and the visual data; the physical attribute vector including: deformation shadow. The system includes: a deformation influence index and a smoothness parameter; a deformation influence index to characterize the degree of influence of cargo weight on the deformation of the conveyor frame; a smoothness parameter to characterize the frictional characteristics of the cargo surface; determination of the center of gravity offset vector of the target cargo during the conveying process based on displacement data; the center of gravity offset vector includes: cumulative offset and offset trend; the cumulative offset to characterize the total amount of center of gravity offset; the offset trend to characterize the rate and direction of change of the center of gravity; determination of adjustment parameters based on physical property vectors, center of gravity offset vectors, and historical adjustment records of the target cargo; adjustment parameters to adjust the bearing seat position and attitude; and adjustment of the bearing seat according to the adjustment parameters.

[0005] In conjunction with the first aspect mentioned above, in one possible implementation, the method specifically includes: determining a first height adjustment amount based on the deformation influence index, smoothness parameter, and historical height compensation reference value; the first height adjustment amount is used to synchronously adjust the height of the bearing seats on both sides of the roller; the historical height compensation reference value is determined based on the historical height compensation value in the historical adjustment record; determining a second height adjustment amount based on the cumulative offset, offset trend, smoothness parameter, and historical angle compensation reference value; the second height adjustment amount is used to differentially adjust the height of the bearing seats on both sides of the roller; the historical angle compensation reference value is determined based on the historical angle compensation value in the historical adjustment record.

[0006] In conjunction with the first aspect above, in one possible implementation, the method specifically includes: controlling the adjustment mechanism to perform synchronous equal-height adjustment of the bearing seats on both sides of the drive roller according to a first height adjustment amount; and controlling the adjustment mechanism to perform differential height adjustment of the bearing seats on both sides of the drive roller according to a second height adjustment amount.

[0007] In conjunction with the first aspect mentioned above, in one possible implementation, the method specifically includes: determining the deformation influence index based on the axial pressure component and lateral pressure component in the mechanical data, as well as displacement data; determining the basic friction parameters based on the axial pressure component and lateral pressure component; determining the smoothness parameter based on the basic friction parameter and surface texture parameter; and determining the surface texture parameter based on visual data.

[0008] In conjunction with the first aspect mentioned above, in one possible implementation, the method specifically includes: determining a displacement difference sequence corresponding to the target cargo conveying process based on the height change of the bearing seats on both sides of the roller in the displacement data; performing integral processing on the displacement difference sequence to determine the cumulative offset; and performing trend fitting analysis on the displacement difference sequence to determine the offset trend.

[0009] In conjunction with the first aspect mentioned above, in one possible implementation, the method further includes: querying a preset adjustment database to determine historical adjustment records based on identity data; the historical adjustment records include: historical adjustment data of goods of the same category as the target goods; and determining historical height compensation benchmark values ​​and historical angle compensation benchmark values ​​based on the historical adjustment data.

[0010] In conjunction with the first aspect mentioned above, in one possible implementation, the method specifically includes: acquiring mechanical data through a force sensing component; the mechanical data includes axial pressure components and lateral pressure components acting on the roller shaft; acquiring displacement data through a displacement measurement component; acquiring visual data through a vision acquisition component; the visual data includes at least an image of the target cargo surface; and acquiring identity data by reading the information identifier of the target cargo through an identity recognition component.

[0011] In conjunction with the first aspect above, in one possible implementation, the method further includes: collecting reference displacement data of the roller bearing housing under no-load conditions of the conveyor; and determining the displacement data based on the difference between the displacement data of the roller bearing housing collected under load conditions and the reference displacement data.

[0012] In conjunction with the first aspect mentioned above, in one possible implementation, the method further includes: preprocessing the collected mechanical data, displacement data, and visual data; the data preprocessing includes at least one of the following: filtering and denoising, outlier removal, and data alignment.

[0013] In conjunction with the first aspect mentioned above, in one possible implementation, the method further includes: associating and storing the adjustment parameters with identity data, physical attribute vectors, and centroid offset vectors; and updating the associating and stored data to the adjustment database.

[0014] The present invention has the following beneficial effects:

[0015] This invention, by simultaneously collecting and fusing mechanical, displacement, visual, and identity data, enables the system to comprehensively perceive cargo characteristics and conveyor status. By extracting vectors representing the static physical properties of the cargo and vectors representing its dynamic center of gravity shift, it provides precise and multi-dimensional input for adjustment decisions. Finally, by combining historical experience to generate and execute adjustment parameters, it achieves a shift from passive response to proactive prediction and adaptive adjustment. Overall, this method significantly improves the conveying system's adaptability to different cargoes and operating conditions, ensuring high stability and reliability of the conveying process. This solves the technical problem that existing solutions have significant limitations in real-time performance, accuracy, and adaptability, affecting the overall stability of the conveying process. Attached Figure Description

[0016] To more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0017] Figure 1 This is one of the flowcharts illustrating an adjustment method for an adjustable bearing housing for a conveyor roller according to an embodiment of the present invention.

[0018] Figure 2 This is a second schematic flowchart illustrating an adjustment method for an adjustable bearing housing for a conveyor roller, provided in one embodiment of the present invention.

[0019] Figure 3 This is a third schematic flowchart illustrating an adjustment method for an adjustable bearing housing for a conveyor roller, provided as an embodiment of the present invention.

[0020] Figure 4 This is a fourth schematic flowchart illustrating an adjustment method for an adjustable bearing housing for a conveyor roller, provided as an embodiment of the present invention.

[0021] Figure 5 This is the fifth flowchart illustrating an adjustment method for an adjustable bearing housing for a conveyor roller, provided in one embodiment of the present invention.

[0022] Figure 6 This is a schematic diagram of the sixth step in the process of providing an adjustment method for an adjustable bearing housing for a conveyor roller according to an embodiment of the present invention.

[0023] Figure 7 This is a schematic flowchart (seventh) of an adjustment method for an adjustable bearing housing for a conveyor roller provided in an embodiment of the present invention.

[0024] Figure 8 This is the eighth flowchart illustrating an adjustment method for an adjustable bearing housing for a conveyor roller, provided as an embodiment of the present invention.

[0025] Figure 9 This is a schematic flowchart (nine) of an adjustment method for an adjustable bearing housing for a conveyor roller provided in an embodiment of the present invention.

[0026] Figure 10 This is a schematic flowchart of an adjustment method for an adjustable bearing housing for a conveyor roller, provided as an embodiment of the present invention. Detailed Implementation

[0027] To further illustrate the technical means and effects adopted by the present invention to achieve its intended purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effects of an adjustable bearing seat for a conveyor roller according to the present invention. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0028] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0029] The following describes in detail, with reference to the accompanying drawings, a specific scheme for adjusting an adjustable bearing seat for a conveyor roller provided by the present invention.

[0030] Please see Figure 1 The diagram shows a flowchart of an adjustment method for an adjustable bearing housing for a conveyor roller according to an embodiment of the present invention. The method includes the following steps S101-S105, which will be described in detail below.

[0031] S101. Acquire multi-source sensing data of the target cargo during the transportation process.

[0032] The multi-source sensing data includes: mechanical data, displacement data, visual data, and identity data; mechanical data is used to reflect the interaction force between the cargo and the conveyor rollers; displacement data is used to reflect the temporal changes in the position of the conveyor roller bearing seats; visual data is used to reflect the surface properties of the cargo; and identity data is used to determine the historical adjustment records of the target cargo.

[0033] In one possible implementation, a multi-source sensing system is constructed and operates collaboratively to simultaneously collect heterogeneous data streams reflecting the physical characteristics, transport status, and identification of the target cargo. Specifically, when the target cargo enters the preset transport monitoring section, piezoelectric force sensing units deployed on the bearing seats collect mechanical data in real time. The mechanical data includes pressure components along the roller axis and lateral pressure components perpendicular to the axis. Displacement data is collected by a laser displacement sensor set on a stable measurement reference. The displacement data includes the absolute displacement of the bearing seats on both sides of the roller in the height direction. Visual data is collected by a vision sensor installed upstream of the transport path. The visual data is an image sequence containing the surface texture features of the target cargo. At the same time, an RFID device reads the electronic tag attached to the cargo to obtain its unique identification data.

[0034] S102. Based on mechanical and visual data, determine the physical attribute vector of the target cargo.

[0035] The physical attribute vector includes: deformation influence index and smoothness parameter; the deformation influence index is used to characterize the degree of influence of cargo weight on the deformation of the conveyor frame; the smoothness parameter is used to characterize the friction characteristics of the cargo surface.

[0036] In one possible implementation, the physical attribute vector consists of at least two dimensions: First, a deformation influence index, which quantifies the degree to which cargo weight drives the deformation of the conveyor frame structure. Its determination relies on a correlation analysis between the pressure component in the mechanical data and the height change in the displacement data. The index value is positively correlated with the cargo weight and the resulting elastic deformation of the frame. Second, a smoothness parameter, which characterizes the frictional properties of the interaction between the cargo surface and the rollers. This parameter is determined by fusing basic friction parameters from mechanical data and surface texture parameters from visual data. The basic friction parameters are estimated based on the relationship between axial and lateral pressure components, while the surface texture parameters are obtained through texture feature analysis of cargo surface images. The final smoothness parameter value comprehensively reflects the smoothness of the cargo surface; a higher value indicates a smoother surface and lower frictional resistance. Combining the calculated deformation influence index and smoothness parameter constitutes the physical attribute vector of the target cargo.

[0037] S103. Determine the center of gravity offset vector of the target cargo during the transportation process based on displacement data.

[0038] The center of gravity offset vector includes: cumulative offset and offset trend; the cumulative offset is used to characterize the total amount of the center of gravity that has been offset; the offset trend is used to characterize the rate and direction of change of the center of gravity.

[0039] In one possible implementation, the process of determining the center of gravity offset vector includes dynamically tracking and analyzing the temporal changes in the bearing seat position reflected by displacement data to construct a vectorized index characterizing the dynamic behavior of the cargo's center of gravity. The center of gravity offset vector consists of at least two dimensions: first, the cumulative offset, which quantifies the total deviation of the cargo's center of gravity from the ideal centerline within the completed transport segment. The cumulative offset is determined by integrating or accumulating the temporal signal of the height displacement difference between the bearing seats on both sides of the target cargo; second, the offset trend, which characterizes the instantaneous offset direction and rate of change of the cargo's center of gravity at the current and nearby times. The offset trend is extracted by trend fitting and analysis of the displacement difference temporal signal, for example, by differentiating or calculating the slope of the signal to obtain its instantaneous rate of change and direction. The combination of the cumulative offset and the offset trend constitutes the dynamic center of gravity offset vector of the target cargo.

[0040] S104. Determine the adjustment parameters based on the physical attribute vector, the center of gravity offset vector, and the historical adjustment records of the target cargo.

[0041] The adjustment parameters are used to adjust the position and orientation of the bearing housing.

[0042] In one possible implementation, the determination of the adjustment parameters includes at least two related calculation processes: First, based on the deformation influence index and smoothness parameter in the physical property vector, and combined with the historical height compensation benchmark value in the historical adjustment record, a first height adjustment amount is calculated. This adjustment amount is used to instruct the adjustment mechanism to perform synchronous equal height adjustment on the bearing seats on both sides of the roller to compensate for the systematic deformation of the frame caused by the weight of the cargo. Second, based on the cumulative offset and offset trend in the center of gravity offset vector, and combined with the smoothness parameter and the historical angle compensation benchmark value in the historical adjustment record, a second height adjustment amount is calculated. This adjustment amount is used to instruct the adjustment mechanism to perform differential height adjustment on the bearing seats on both sides of the roller to correct the roller axis angle deviation caused by the offset of the center of gravity of the cargo.

[0043] S105. Adjust the bearing housing according to the adjustment parameters.

[0044] In one possible implementation, the adjustment process includes two related and coordinated actions: First, based on a first height adjustment amount, the drive actuator (e.g., a servo electric cylinder) in the adjustment mechanism responsible for vertical lifting is controlled to raise or lower the bearing seats on both sides of the roller by an equal amount and synchronously, so as to complete the height compensation for the overall deformation of the frame; Second, based on a second height adjustment amount, the drive unit (e.g., a micro servo actuator) in the adjustment mechanism capable of independent or differential control is controlled to adjust the bearing seats on both sides of the roller by unequal amounts (including possibly opposite directions), thereby changing the tilt angle of the roller axis to actively correct the trajectory of the cargo's center of gravity offset.

[0045] In one possible implementation, the adjustment of the bearing housing adopts a feedforward control mode. When the system identifies the target cargo upstream of the conveying path (e.g., at the (k-1)th roller) via the radio frequency identification unit, it initiates the calculation and pre-adjustment process of adjustment parameters for the next roller (kth roller). This adjustment process includes two related and coordinated actions: First, based on a first height adjustment amount calculated based on upstream data, the drive actuator (e.g., a servo electric cylinder) responsible for vertical lifting in the adjustment mechanism is controlled to raise or lower the bearing housings on both sides of the kth roller by an equal amount and synchronously. This action aims to pre-compensate for predicted frame deformation before the cargo arrives. Second, based on a second height adjustment amount calculated based on upstream data, the drive unit (e.g., a micro servo actuator) in the adjustment mechanism, capable of independent or differential control, is controlled to adjust the bearing housings on both sides of the kth roller by unequal amounts, thereby pre-setting a expected roller axis tilt angle to actively counteract the predicted center of gravity shift trend. All input data used to calculate the adjustment parameters originates from the results of the target cargo being collected and processed on the upstream roller (k-1). For example, the texture roughness coefficient required to calculate the smoothness parameter S depends on the severity of the longitudinal displacement. This is calculated based on the degree of drastic change in the displacement signal of the upstream roller bearing housing as the goods pass over it. The value characterizes the typical vibration mode caused by this type of cargo on the upstream roller. Based on this, the system predicts the similar effects that it may cause on the downstream roller and compensates in advance.

[0046] The technical solution provided by the above embodiments can bring at least the following beneficial effects: By synchronously collecting and fusing mechanical, displacement, visual, and identity data, the system can comprehensively perceive the characteristics of the cargo and the state of the conveyor; by extracting vectors representing the static physical properties of the cargo and vectors representing its dynamic center of gravity shift, precise and multi-dimensional inputs are provided for adjustment decisions; finally, adjustment parameters are generated and executed by combining historical experience, realizing the transformation from passive response to active prediction and adaptive adjustment. Overall, this method significantly improves the adaptive capability of the conveying system to different cargoes and different working conditions, ensuring high stability and high reliability of the conveying process, thereby solving the technical problem that existing technical solutions have obvious limitations in terms of real-time performance, accuracy, and adaptive capability of adjustment, which affects the overall stability of the conveying.

[0047] In one possible implementation, please refer to Figure 2 The diagram shows a flowchart of an adjustable bearing housing for a conveyor roller according to an embodiment of the present invention. The method includes the following steps S201-S202, which will be described in detail below.

[0048] S201. Based on the deformation influence index, smoothness parameters, and historical height compensation benchmark values, determine the first height adjustment amount.

[0049] The first height adjustment amount is used to synchronously adjust the height of the bearing seats on both sides of the roller; the historical height compensation benchmark value is determined based on the historical height compensation value in the historical adjustment record.

[0050] In one possible implementation, the historical height compensation benchmark serves as the basis for the adjustment amount. Its value is extracted from historical adjustment records of goods of the same category (identified by identity data) as the current target cargo and is statistically verified as a typical height compensation amount. The deformation influence index, as a core scaling factor, directly reflects the actual driving force of the current cargo weight on the frame deformation; the larger the index value, the larger the required height compensation benchmark. The smoothness parameter, as a refined correction factor, is involved in the calculation, reflecting the frictional characteristics of the cargo surface. For cargo with smooth surfaces, it is more sensitive to the flatness of the roller support surface; therefore, a fine-tuning component aimed at pursuing higher flatness needs to be introduced on the basis of weight deformation-based compensation. The final first height adjustment amount is the result of the historical height compensation benchmark value after scaling the deformation influence index and further refined by the smoothness parameter. This adjustment amount directly indicates the displacement amplitude required for synchronous and equal lifting and lowering of the bearing seats on both sides of the roller. Its fundamental purpose is to accurately compensate for the elastic sinking of the conveyor frame caused by the cargo weight, providing a stable and flat support plane for the cargo.

[0051] For example, the first height adjustment amount Satisfy the following formula 1:

[0052] Formula 1

[0053] in, It represents the historical adjusted average of goods of the same category, and is an empirical benchmark for highly compensating for similar goods; The deformation effect index; The normalization function is used to eliminate the interference of different dimensions of parameters. The calculation results are uniformly mapped to a fixed value range [0, 1] by normalizing the maximum and minimum values. The maximum and minimum values ​​are determined based on the system's preset empirical extreme values ​​(such as the maximum design load of the conveyor and the maximum allowable mechanical offset) or the statistical extreme values ​​of the historical database, rather than based solely on the current real-time data.

[0054] The larger (heavier) the cargo, the more pronounced the rack subsidence, requiring a larger amount of compensation. The compensation amount is determined by historical experience values ​​to ensure that it conforms to the adjustment rules of similar goods. The product of these three factors, including the overall balance of weight influence, accuracy requirements, and historical experience, yields a precise compensation amount. The height compensation value of the bearing housing in the Z-axis direction was quantified to counteract the frame sinking caused by heavy load, so that the roller axis returns to the ideal position.

[0055] S202. Based on the cumulative offset, offset trend, smoothness parameters, and historical angle compensation benchmark values, determine the second height adjustment amount.

[0056] The second height adjustment is used to differentially adjust the height of the bearing seats on both sides of the roller; the historical angle compensation reference value is determined based on the historical angle compensation value in the historical adjustment record.

[0057] In one possible implementation, the historical angle compensation benchmark value serves as the initial reference scale for adjustment. Its value is extracted from historical adjustment records of goods of the same category (identified by identity data) and is statistically validated as a typical angle compensation amount. The cumulative offset serves as the core compensation requirement base, directly representing the total amount by which the cargo's center of gravity has deviated from the ideal transport trajectory; the larger the offset, the larger the required axis angle correction benchmark. The offset trend is incorporated into the calculation as a dynamic predictive factor, reflecting the instantaneous direction and rate of change in the cargo's center of gravity offset. A larger trend value (i.e., the offset is rapidly intensifying) necessitates the introduction of a predictive adjustment component aimed at suppressing future offsets, building upon the compensation based on existing offsets. The smoothness parameter again serves as a key sensitivity correction factor, reflecting the frictional characteristics of the cargo surface. For goods with smooth surfaces, their center of gravity is more prone to sliding on inclined roller surfaces, making them extremely sensitive to small angular deviations. Therefore, the adjustment amount needs to be further amplified based on the compensation calculations using offset and trend to achieve more proactive and timely intervention. The final second height adjustment is the result of sensitivity-weighted correction of the historical angle compensation baseline value under the combined effect of cumulative offset and offset trend (providing the static compensation base and dynamic predictive compensation increment, respectively) and smoothness parameters. This adjustment directly indicates the displacement difference that requires differential and unequal lifting of the bearing seats on both sides of the drive roller. Its fundamental purpose is to actively and accurately correct the tilt angle of the roller axis to counteract the lateral slippage trend caused by the offset of the cargo's center of gravity and guide the cargo back to a stable conveying trajectory.

[0058] For example, the second height adjustment amount The following formula 2 is satisfied:

[0059] Formula 2

[0060] in, The normalized centroid shift index is a value determined by comprehensively weighting the shift trend, and is used to characterize the overall strength of the shift trend. This is the normalized centroid offset, representing the actual offset distance; Surface smoothness characterizes the risk of slippage; This is a historical adjustment coefficient for goods of the same category, representing historical adjustment experience. The maximum allowable angle correction amount limits the correction range; This is a normalization function used to eliminate interference from different parameters in terms of dimensions. It normalizes the calculation results to a fixed range [0, 1] by normalizing the maximum and minimum values; where, This is a positive gain coefficient preset based on historical experience (e.g., 0.5). In the figure, 1 guarantees the basic gain. When the cargo is very rough (S approaches 0), the gain factor is 1. The system provides a baseline adjustment based on the offset to ensure that the correction action is not suppressed. When the cargo surface is smooth (S is large), since smooth cargo is more sensitive to tilting and has a higher risk of slippage, more active and larger-amplitude intervention is needed to stabilize it quickly. The gain factor is greater than 1, and the system will calculate a larger angle correction amount.

[0061] To accumulate the offset and offset trend, the larger the product, the higher the offset risk; multiplying by This indicates that the smoother the surface (the larger S), the higher the risk of slippage, and the greater the correction amount required; multiplied by In order to incorporate historical experience and avoid over-correction; multiply by In order to map the dimensionless composite coefficients to actual angle values ​​and ensure that the correction range is reasonable. The angular correction value of the roller axis was quantified to compensate for the axis tilt caused by the shift in the center of gravity.

[0062] The technical solution provided by the above embodiments can bring at least the following beneficial effects: This embodiment refines the generation logic of adjustment parameters. By introducing historical height compensation benchmark values ​​and historical angle compensation benchmark values, the adjustment decision is not only based on real-time sensing data, but also incorporates past successful adjustment experience. This design enables the system to have self-learning and optimization capabilities, and can form increasingly precise adjustment strategies for different types of goods. The distinction between the first height adjustment amount and the second height adjustment amount decouples the two core tasks of compensating for frame deformation and correcting center of gravity offset, realizing the precision of the adjustment target and the differentiation of the strategy, thereby maintaining the accuracy and effectiveness of adjustment even under complex working conditions.

[0063] Please see Figure 3The diagram shows a flowchart of an adjustable bearing housing for a conveyor roller according to an embodiment of the present invention. The method includes the following steps S301-S302, which will be described in detail below.

[0064] S301. Based on the first height adjustment amount, control the adjustment mechanism to drive the bearing seats on both sides of the roller to perform synchronous equal height adjustment.

[0065] In one possible implementation, the adjustment mechanism comprises at least two independent sets of high-precision linear drive units, such as servo electric cylinders, rigidly connected to the bottom of the bearing seats on both sides of the roller. Upon receiving the first height adjustment amount, the control system resolves it into a pair of equal and unidirectional displacement commands. Subsequently, using a synchronous motion control algorithm, these two displacement commands are sent to the servo drives on both sides, driving the servo motors in the corresponding servo electric cylinders to rotate. The rotational motion is then converted into linear lifting motion via a precision ball screw pair. Throughout the execution process, linear encoders embedded in each servo electric cylinder monitor the actual displacement of their output shafts in real time and independently, comparing this displacement feedback signal with the target displacement command to form a closed-loop position control. Through this closed-loop control, the system can dynamically adjust the motor torque and speed to ensure that the bearing seats on both sides overcome load resistance and complete the specified lifting displacement with highly synchronized accuracy and speed, thereby achieving synchronous equal-height adjustment.

[0066] S302. Based on the second height adjustment amount, control the adjustment mechanism to drive the bearing seats on both sides of the roller to perform differential height adjustment.

[0067] In one possible implementation, the adjustment mechanism includes at least two sets of linear drive units with independent closed-loop control capabilities, such as miniature servo electric cylinders or linear motors, respectively connected to bearing seats on both sides of the roller. Upon receiving the second height adjustment amount, the control system interprets this amount as a pair of motion commands with a specific displacement difference (or opposite signs). For example, if the second height adjustment amount is positive, it may indicate that one bearing seat needs to be raised while the other needs to be lowered, or both sides need to be raised but with different displacements; if it is negative, it executes a differential combination in opposite directions. In the specific control process, the central controller sends independent control signals to the servo drivers of the two drive units based on the interpreted displacement commands from both sides. The drivers drive the corresponding servo motors, converting the rotational motion into linear displacement through a precision transmission mechanism (such as a ball screw). To ensure precise coordination of the differential motion and accurate achievement of the final roller axis angle, the system employs dual closed-loop feedback control. The inner loop is a position closed loop for each drive unit, with its actual displacement fed back in real time by a high-resolution linear encoder. The outer loop is an angle closed loop, with the roller axis inclination angle calculated in real time by independently installed angle sensors or by data from high-precision displacement sensors on both sides as the feedback quantity. The controller continuously compares the target angle with the actual angle and dynamically adjusts the displacement commands of the drive units on both sides until the actual angle matches the target angle, thereby accurately completing the differential height adjustment.

[0068] The technical solution provided in the above embodiments can bring at least the following beneficial effects: This embodiment clarifies the specific execution method of the adjustment action, transforming the adjustment parameters into two coordinated control actions. Synchronous height adjustment can quickly and stably compensate for the overall sinking of the frame caused by the weight of the goods, maintaining the flatness of the conveying plane; differential height adjustment can flexibly and accurately adjust the roller axis angle, actively correcting the tendency of the goods to deviate. The combination of these two adjustment actions enables the system to simultaneously cope with two types of problems: static load deformation and dynamic center of gravity shift, realizing composite control of the bearing seat position and attitude, fundamentally enhancing the straightness and stability of the conveying trajectory.

[0069] In one possible implementation, please refer to Figure 4 The diagram shows a flowchart of an adjustable bearing housing for a conveyor roller according to an embodiment of the present invention. The method includes the following steps S401-S403, which will be described in detail below.

[0070] S401. Based on the axial and lateral pressure components in the mechanical data, as well as the displacement data, determine the deformation influence index.

[0071] In one possible implementation, the deformation impact index is used to quantify the degree of deformation impact of the target cargo weight on the conveyor structure. Specifically, for each monitoring roller through which the target cargo passes, the range of displacement change in the height direction under the action of the cargo is first calculated based on the displacement data of the bearing seats on both sides; simultaneously, strain characteristics reflecting the local stress on the frame are extracted based on the mechanical data collected at the corresponding position of the roller. Then, during the complete conveying period of the target cargo, the displacement change characteristics and mechanical strain characteristics obtained from each roller are fused and statistically analyzed across rollers. The deformation impact index is finally calculated as a comprehensive characterization quantity, which is positively correlated with the cargo weight. The larger the value, the more significant the elastic deformation (such as sinking or deflection) of the conveyor frame caused by the current cargo, and therefore the system needs to perform a larger amount of bearing seat height compensation to maintain the flatness of the roller plane.

[0072] For example, the height displacement signal of the bearing seats on both sides of the current roller and the strain value signal of the frame are obtained. The interval corresponding to the previous Radio Frequency Identification (RFID) module and the current RFID identification module is selected within the two signals. The least squares method is used to fit the signal, and the signal is divided into monotonically increasing and monotonically decreasing intervals by the first derivative. The longest monotonically increasing interval is selected, the range of each signal in the interval is calculated, and the mean of the displacement range on both sides of the roller and the mean of the strain value range are calculated. Their product is used as the deformation influence index of the current goods.

[0073] S402. Determine the basic friction parameters based on the axial pressure component and the lateral pressure component.

[0074] In one possible implementation, the axial pressure component reflects the normal pressure exerted by the cargo on the roller surface, while the lateral pressure component reflects the tangential frictional resistance occurring between the cargo and the roller surface. The core of determining the basic friction parameters lies in establishing the correlation between these two force components to characterize the inherent frictional characteristics between the cargo and the roller under specific contact conditions. Specifically, the basic friction parameters can be obtained by calculating the ratio between the lateral pressure component and the axial pressure component, or their corrections (e.g., the effective axial pressure component considering the real-time tilt angle of the roller), or by performing statistical analysis based on this ratio, according to the principles of classical friction mechanics.

[0075] S403. Determine the smoothness parameters based on the basic friction parameters and surface texture parameters.

[0076] In one possible implementation, surface texture parameters are obtained through texture analysis of visual data of the cargo surface. First, a surface image of the target cargo is acquired using a visual acquisition component deployed upstream of the transport path. Then, a pre-defined image processing algorithm (e.g., a semantic segmentation model based on deep learning) is used to extract the cargo region from the image, resulting in a pixel-level mask for the cargo region. To quantify the roughness of the cargo surface texture, features are extracted from two dimensions: gradient distribution features, where the gradient values ​​of each pixel within the cargo region mask are calculated, and the ratio of its mean to variance is calculated, denoted as t; the larger this value, the more significant and uniform the gradient changes of the pixels within the mask region, and the more consistent it is with surface roughness characteristics; and texture statistical features, where gray-level co-occurrence matrices are constructed in the horizontal and vertical directions for the cargo region mask. For each gray-level co-occurrence matrix, its contrast and entropy are calculated, and their product is used as the texture roughness information of that matrix. Contrast reflects the clarity and local variations of the texture, while entropy reflects the complexity and randomness of the texture; their product comprehensively characterizes the roughness of the texture.

[0077] In one possible implementation, the longitudinal displacement of the rollers during the conveying process may interfere with the extraction of texture features in the vertical direction; therefore, a longitudinal displacement suppression factor is introduced. (Value range [0,1]), this factor is calculated based on the degree of change in the displacement signal of the roller where the current cargo is located. The larger the value, the more severe the longitudinal displacement, and the lower the reliability of the vertical texture features, which should be suppressed in the final evaluation.

[0078] For example, surface texture parameters Satisfy the following formula 4:

[0079] Formula 4

[0080] in, This is the ratio of the mean to the variance of the pixel gradient of the cargo area mask (the variance is not zero because the cargo surface is rough and the light effect causes different gray levels). A larger value indicates a more uniform pixel gradient and a rougher surface. The degree of longitudinal displacement of the roller represents the degree of distortion of texture information in the vertical direction. The larger the value, the more severe the distortion. It is a surface roughness feature in the horizontal direction; It is a surface roughness feature in the vertical direction.

[0081] The larger, the horizontal direction The greater the weight (in the vertical direction of suppressing distortion) The smaller the value of h, the greater the weight in the vertical direction (the more reliable the texture information); and Multiplication: through pixel gradient uniformity ( ) Enlarge the roughness features, The larger, The larger.

[0082] For example, smoothness parameter The following formula 5 is satisfied:

[0083] Formula 5

[0084] Among them, μ is the basic friction parameter, which characterizes the frictional properties at the mechanical level; The surface roughness coefficient of the goods represents the texture roughness characteristics at the visual level. is an exponential function with the natural constant e as the base, used to map the calculation results to the interval (0, 1); S is a smoothness parameter.

[0085] Understandably, the basic friction parameter is defined as the ratio of frictional force to normal force; both are units of force, therefore the basic friction parameter is a dimensionless pure number. Calculation... The eigenvalues ​​themselves are dimensionless mathematical statistics, therefore It is also a dimensionless pure number.

[0086] It is a comprehensive roughness index combining mechanical and visual properties. and The larger the value, the higher the overall roughness; after negating the overall roughness index, through the exp function mapping, the larger the overall roughness index, the higher the overall roughness. The closer to 0, the rougher the surface; the smaller the overall roughness index, the closer the result is to 1 (the smoother the surface).

[0087] The technical solution provided in the above embodiments can bring at least the following beneficial effects: This embodiment specifies the construction process of physical attribute vectors, and its beneficial effect lies in realizing deep digitization and integrated perception of cargo characteristics. By combining mechanical data (axial / lateral pressure) and displacement data to evaluate the deformation influence index, the actual impact of cargo weight on the frame can be reflected more accurately, rather than simply relying on nominal weight. By integrating the basic friction parameters derived from mechanics with the surface texture parameters from visual analysis to determine the smoothness parameters, the surface friction characteristics of the cargo are objectively quantified in a creative way. This makes the system's understanding of the physical characteristics of the cargo more comprehensive and accurate, laying a solid data foundation for subsequent differentiated adjustments.

[0088] Please see Figure 5 The diagram shows a flowchart of an adjustment method for an adjustable bearing housing for a conveyor roller according to an embodiment of the present invention. The method includes the following steps S501-S503, which will be described in detail below.

[0089] S501. Based on the height change of the bearing seats on both sides of the roller in the displacement data, determine the displacement difference sequence corresponding to the target cargo transportation process.

[0090] In one possible implementation, for the target cargo, starting from its entry into the conveying monitoring section, the displacement changes in the height direction of the bearing seats on both sides of the roller below it (e.g., marked as side A and side B) are continuously collected. These displacement changes are pre-processed relative displacement data directly related to the load, with reference offset eliminated. The displacement changes on side A and side B acquired at the same sampling time are paired, and the displacement difference at that moment is calculated by subtracting the data from side B. On the complete conveying timeline of the target cargo, the displacement differences calculated at all sampling times are arranged and connected in chronological order to generate a displacement difference sequence. Each data point in this sequence visually represents which side of the roller the cargo's center of gravity projection is biased towards at the corresponding moment and the instantaneous degree of deviation; its continuous change completely depicts the dynamic process of the lateral shift of the center of gravity.

[0091] S502. Integrate the displacement difference sequence to determine the cumulative offset.

[0092] In one possible implementation, the object of the integration operation is a time-varying sequence of displacement differences. Physically, this represents the summation of the lateral offset components that continuously exist within the center of gravity of the cargo during transport. Specifically, the transport time is used as the integration variable, and the function expression of the displacement difference sequence is used as the integrand. A definite integral is calculated over the entire time interval from the target cargo's entry into the monitoring starting point to its current position. The result of the integration is a scalar value with a sign (positive or negative), which is the cumulative offset. Its absolute value directly represents the total lateral offset of the cargo's center of gravity relative to the ideal centerline over history; its sign (positive or negative) indicates which side of the roller the cumulative offset is biased towards. For example, if the difference is set as the displacement on side A minus the displacement on side B, a positive cumulative offset indicates that the historical trajectory of the center of gravity is biased towards side A, while a negative one indicates bias towards side B.

[0093] S503. Perform trend fitting analysis on the displacement difference sequence to determine the offset trend.

[0094] In one possible implementation, data points within a time window adjacent to the current moment are selected from the displacement difference sequence (e.g., data segments corresponding to the most recent seconds or the passage of goods over the most recent rollers). A linear fitting algorithm is used to fit a curve to this data segment, constructing an approximate function that characterizes its recent change pattern. The offset trend is defined as the first derivative or instantaneous slope of this fitted function at the current moment (or at the end of the time window). This trend value is a scalar with a sign (positive or negative) and a magnitude. Its absolute value directly characterizes the instantaneous rate of change of the center of gravity offset in the recent period (i.e., the speed at which the offset is accelerating or decelerating), while its sign clearly indicates the instantaneous direction of the center of gravity offset (i.e., which side of the roller the center of gravity is shifting towards). For example, under the current fitted model, a large positive trend value indicates that the center of gravity is shifting towards side A at a relatively fast speed.

[0095] For example, offset trend Satisfy the following formula 6:

[0096] Formula 6

[0097] in, The current center of gravity offset rate is determined by the mean derivative of the signal fitted to the displacement difference sequence in the most recent time interval. Its physical dimension is length / time (e.g., millimeters per second), which characterizes the speed and direction of the current instantaneous offset of the center of gravity. The angular similarity coefficient represents the similarity between the current roller angle and the historical similarity interval. The historical center of gravity shift rate serves as a reference value for characterizing historical shift trends; This is a normalization function used to map the result to the (0, 1) interval by normalizing it through maximum and minimum values. k is a preset scaling factor, a constant with time / length dimensions (e.g., seconds / millimeter). Its core function is to combine rates with actual physical dimensions. Linear mapping to a A suitable numerical range that matches the function's characteristics ensures that the input value primarily falls within a certain range. Within a linear interval where the function changes significantly, this avoids premature saturation of the function output or sluggish response, ensuring... Sensitivity and discrimination to changes in offset rate.

[0098] It is a core item, directly reflecting the strength of the current center of gravity shifting towards side A; It is a historical correction item. The larger the value (the more similar the current value is to the historical value), the greater the impact of the historical offset rate, and the more significant the correction; sum of inputs The larger the sum, The closer to 1 (the stronger the offset trend); the smaller the sum, The closer to 0, the weaker the offset trend. The continuous trend of the center of gravity of the quantified goods shifting towards side A. The larger the value, the higher the risk of deviation and the more likely it is to continue to worsen.

[0099] The technical solution provided in the above embodiments can bring at least the following beneficial effects: This embodiment specifies the analysis process of the center of gravity offset vector, realizing real-time monitoring and trend prediction of cargo transportation dynamics. The cumulative offset, through the integration of displacement differences, objectively reflects the historical cumulative effect of center of gravity offset, providing the system with information on the total offset that needs to be compensated. The offset trend, through trend analysis of the difference sequence, can keenly capture the direction and acceleration of center of gravity offset, giving the system a certain degree of predictability. Combining the two into a vector supports more timely and forward-looking adjustment intervention, effectively preventing the continuous expansion of offset.

[0100] Please see Figure 6 The diagram shows a flowchart of an adjustment method for an adjustable bearing housing for a conveyor roller according to an embodiment of the present invention. The method includes the following steps S601-S602, which will be described in detail below.

[0101] S601. Based on the identity data, query the preset adjustment database to determine the historical adjustment records.

[0102] The historical adjustment records include: historical adjustment data for goods of the same category as the target goods.

[0103] In one possible implementation, after obtaining the identity data of the target cargo, it is used as a query keyword to access a pre-established and continuously updated adjustment database. This database stores the adjustment parameters and corresponding operational data (such as the physical attribute vector and center of gravity offset vector at the time) successfully applied to different categories of cargo (typically categorized by type, size, material, etc., and associated with the identity data) when they passed through this transport system. A query operation is performed to search the database for a set of historical entries that match the identity data of the current target cargo or belong to the same pre-defined category. This set of retrieved historical adjustment data (including historical height compensation values, historical angle compensation values, and possibly other relevant parameters) is identified as the historical adjustment record. This record provides a historically experienced and practically validated reference benchmark for current adjustment decisions.

[0104] S602. Based on historical adjustment data, determine the historical height compensation benchmark value and the historical angle compensation benchmark value.

[0105] In one possible implementation, historical adjustment data includes valid adjustment parameters recorded and ultimately adopted during the historical transport of multiple goods similar to the current target cargo. The system first preprocesses this dataset, such as removing obviously outlier or invalid records to ensure data quality. Then, it extracts the height compensation value field and angle compensation value field from all records in the preprocessed dataset. For the height compensation value set, the system processes it using a preset statistical algorithm (such as calculating the arithmetic mean, weighted average, or median), and determines the calculated statistical result as the historical height compensation benchmark value. Similarly, for the angle compensation value set, the same or similar statistical algorithm is used to process it, and the statistical result is determined as the historical angle compensation benchmark value. These two benchmark values ​​represent typical empirical values ​​for compensating for frame deformation and correcting axis angles for similar cargo during historical adjustments, respectively.

[0106] The technical solution provided in the above embodiments can bring at least the following beneficial effects: This embodiment establishes an experience learning and recall mechanism based on identity data. By associating identity data with and querying historical adjustment records, the system can identify the cargo category and quickly recall the optimal adjustment experience (i.e., historical compensation benchmark value) for that type of cargo. This allows adjustment decisions to be initialized and optimized based on historical successful experience when facing repetitive or similar transportation tasks, significantly shortening the adjustment convergence time, improving the initial adjustment success rate, and reducing the instability period caused by trial and error, thereby improving the overall intelligence level and operating efficiency of the system.

[0107] Please see Figure 7 The diagram shows a flowchart of an adjustment method for an adjustable bearing housing for a conveyor roller according to an embodiment of the present invention. The method includes the following steps S701-S704, which will be described in detail below.

[0108] S701: Collects mechanical data through force sensing components.

[0109] The mechanical data includes the axial pressure component and the lateral pressure component acting on the roller shaft.

[0110] In one possible implementation, the acquisition of mechanical data is accomplished through a specially deployed force sensing assembly. This assembly includes at least two sets of high dynamic response pressure sensors, such as piezoelectric pressure sensors, integrated and mounted inside the roller's bearing housing at key stress locations or at the support points at both ends of the roller shaft. One set of sensors has its main sensing axis arranged along the roller's axial direction to measure and output the axial pressure component of the cargo acting on the roller in real time; the other set of sensors has its main sensing axis arranged perpendicular to the roller's axial direction to measure and output the lateral pressure component generated by the relative motion between the cargo and the roller surface. When the target cargo passes through the roller, its weight and motion state cause dynamic changes in the force on the roller bearings. These changes are captured by the force sensing assembly and converted into corresponding analog electrical signals. These analog signals are then transmitted to a signal conditioning unit for amplification and filtering, and then converted into digital signals via an analog-to-digital converter, thus forming time-series mechanical data available for subsequent processing. The time series of the axial and lateral pressure components together constitute the mechanical data, which is the core raw input for subsequent calculations of basic friction parameters, deformation influence index, and smoothness parameters.

[0111] S702. Displacement data is acquired through the displacement measurement component.

[0112] In one possible implementation, the displacement data acquisition process is accomplished using a high-precision displacement measurement component. This component typically includes at least two sets of non-contact displacement sensors, such as laser displacement sensors, mounted on a stable reference measurement frame independent of the conveyor frame. These two sensors are symmetrically arranged, with their laser beams vertically illuminating the outer surfaces of the bearing seats on both sides of the roller, or on specially designed reflective targets, from top to bottom. When the target cargo passes by, the bearing seats undergo a slight vertical displacement relative to this independent reference frame due to the load. The displacement sensors measure and output the change in distance between the sensor probe and the measured surface of the bearing seat in real time; this change represents the absolute displacement of the bearing seat. The data acquired synchronously by the sensors on both sides, after signal processing, forms two time-series sequences corresponding to the height displacement changes of the bearing seats on sides A and B of the roller, respectively. These data together constitute the displacement data, which serves as the core raw input for subsequent reference differential analysis to obtain relative displacement, generate displacement difference sequences, and ultimately calculate the cumulative offset and offset trend.

[0113] S703: Collect visual data through the visual acquisition component.

[0114] In one possible implementation, the process of acquiring visual data is accomplished by a visual acquisition component deployed upstream of the transport path. This component typically includes at least one industrial camera and a corresponding lighting unit, positioned to ensure that images of one or more of the target cargo's outer surfaces are captured completely and clearly before it enters the core adjustment section. When the RFID unit reads the cargo's identification data and triggers the acquisition command, the industrial camera takes pictures under suitable lighting conditions, acquiring raw digital images containing the target cargo's surface. These images constitute the raw visual data. This raw data is then transmitted to an image processing unit for preprocessing. Preprocessing operations may include automatic white balance, illumination uniformity correction to eliminate ambient light interference, and background subtraction positioning to crop images containing only the region of interest (ROI) of the cargo itself. The preprocessed image data, as valid visual data, will be used for subsequent surface texture analysis to extract surface texture parameters.

[0115] S704. Read the information identifier of the target goods through the identity recognition component to obtain identity data.

[0116] In one possible implementation, the process of acquiring identity data is accomplished through an identification component deployed upstream of the transport path. This identification component is typically an UHF RFID reader, its installation location ensuring effective reading of passive RFID tags attached to the cargo before the target cargo enters the core sensing and control section. When the cargo moves with the conveyor into the reader's effective identification area, the reader emits a radio frequency signal that activates the electronic tag, causing the tag to backscatter its internally stored unique encoded information back to the reader. The reader receives and decodes this signal, and the acquired encoded information is the identity data. This data is transmitted in real-time to the control system, serving as a core index key for querying historical databases and relating to control experience with similar cargo.

[0117] The technical solution provided in the above embodiments offers at least the following beneficial effects: This embodiment clarifies the specific methods for acquiring multi-source sensing data, and its beneficial effect lies in constructing a reliable and efficient data sensing layer. Through clearly defined force sensing, displacement measurement, visual acquisition, and identity recognition components, the system can synchronously and efficiently acquire the required raw information stream. This modular and collaborative sensing design ensures the accuracy and stability of upstream data sources, providing reliable input for all subsequent advanced analysis and decision-making.

[0118] Please see Figure 8 The diagram shows a flowchart of an adjustable bearing housing for a conveyor roller according to an embodiment of the present invention. The method includes the following steps S801-S802, which will be described in detail below.

[0119] S801. When the conveyor is unloaded, collect the reference displacement data of the roller bearing housing.

[0120] In one possible implementation, when the conveyor is not carrying any goods and is operating under stable no-load conditions, the absolute position of the bearing seats on both sides of the roller in the height direction is measured and recorded using a displacement measurement component. The collected data characterizes the inherent static position reference of the conveyor frame, bearing seat mounting surfaces, and the displacement measurement system itself under no external load. This reference displacement data will be stored in the control system as a reference zero point for displacement measurements during all subsequent load operations.

[0121] S802. Based on the displacement data of the roller bearing housing collected under load, the difference between the displacement data and the reference displacement data is used to determine the displacement data.

[0122] In one possible implementation, raw displacement data, including inherent system biases, acquired in real-time during the target cargo loading process, is compared point-by-point with pre-stored reference displacement data corresponding to the same measurement point, and then subjected to a subtraction operation. This operation essentially subtracts the corresponding reference position value under no-load conditions from the instantaneous absolute position measurement under load. Through this differential processing, common pattern biases introduced by mechanical installation, sensor zero-position, and environmental steady-state factors are systematically eliminated, thereby extracting the dynamic deformation component in the height direction purely caused by the target cargo load. The processed data, i.e., the difference result, is defined as the effective displacement data for all subsequent analysis and adjustment decisions.

[0123] Understandably, this step is the core signal processing step for achieving high precision and robustness in displacement measurement. By differentially analyzing the real-time acquired absolute displacement data with a static reference, this invention achieves software zeroing of the measurement system, effectively eliminating irrelevant system errors.

[0124] The technical solution provided by the above embodiments can bring at least the following beneficial effects: This embodiment introduces a benchmark calibration step, the core beneficial effect of which is to significantly improve the accuracy and reliability of displacement measurement data. By pre-collecting benchmark displacement data under no-load conditions and subtracting it from the load data in subsequent processing, fixed system errors such as conveyor frame installation errors and sensor zero-point drift can be effectively eliminated. This makes the displacement data relied upon by subsequent analysis purely reflect the dynamic deformation caused by the cargo load, greatly improving the accuracy of center of gravity offset analysis and deformation impact assessment, and improving the accuracy of the entire adjustment system from the source.

[0125] Please see Figure 9 The diagram shows a flowchart of an adjustment method for an adjustable bearing housing for a conveyor roller according to an embodiment of the present invention. The method includes the following steps S901, which will be described in detail below.

[0126] S901. Perform data preprocessing on the collected mechanical data, displacement data, and visual data.

[0127] Data preprocessing includes at least one of the following: filtering and denoising, outlier removal, and data alignment.

[0128] In one possible implementation, preprocessing operations vary depending on the data type. For mechanical data, the main focus is on filtering and denoising, such as using low-pass filters or moving averages to suppress high-frequency noise introduced by mechanical vibrations or electrical interference, thereby extracting a stable and smooth force signal sequence. For displacement data, after obtaining effective displacement data through benchmark differencing, outlier removal may be further performed, for example, by identifying and removing obvious outliers caused by instantaneous external impacts or measurement interruptions using statistical methods, ensuring the continuity and rationality of the displacement sequence. For visual data, preprocessing mainly includes data alignment and enhancement, such as timestamp alignment to ensure that the image is synchronized with the mechanical and displacement data in the time domain, and performing image enhancement operations (such as contrast adjustment and deblurring) to optimize image quality. Simultaneously, region of interest extraction may be performed to focus on the surface area of ​​the cargo and exclude irrelevant background.

[0129] The technical solution provided by the above embodiments can bring at least the following beneficial effects: This embodiment adds a data preprocessing stage, which enhances the robustness and anti-interference ability of the system. By performing operations such as filtering and denoising, outlier removal, and data alignment on the original mechanical, displacement, and visual data, noise caused by on-site environmental interference, random sensor errors, and data asynchrony can be effectively suppressed.

[0130] Please see Figure 10 The diagram illustrates a flowchart of an adjustment method for an adjustable bearing housing for a conveyor roller according to an embodiment of the present invention. The method includes the following steps S1001-S1002, which will be described in detail below.

[0131] S1001. The adjustment parameters are associated with and stored in relation to the identity data, physical attribute vector, and center of gravity offset vector.

[0132] In one possible implementation, the adjustment parameters that were finally determined and successfully executed in this adjustment (including the first and second height adjustment amounts), the target cargo identification data that triggered this adjustment process, and the physical attribute vectors and center of gravity offset vectors relied upon in the adjustment decision are packaged as a complete historical record entry. These data are logically closely linked through a pre-defined data structure and are assigned a unified timestamp and batch identifier. Subsequently, this entry is written to a non-volatile storage medium, such as local solid-state storage or a specific data table in a remote database, forming a traceable and queryable historical adjustment data record.

[0133] S1002. Update the data in the associated storage to the adjustment database.

[0134] In one possible implementation, a new data record containing identity data, physical attribute vectors, center of gravity offset vectors, and adjustment parameters, which has already been associated and stored, is added or merged into a pre-defined adjustment database as a new valid historical entry. After the update, the content of the historical adjustment records is expanded and enriched, enabling subsequent queries and calculations of historical height compensation benchmarks and historical angle compensation benchmarks based on this database to be built on a broader and more timely data foundation.

[0135] The technical solution provided by the above embodiments can bring at least the following beneficial effects: This embodiment establishes a closed loop of data feedback and knowledge updating, which enables the system to continuously self-optimize and evolve its performance. After each adjustment is completed, the system associates and stores the adjustment parameters, cargo characteristics, and working condition data and updates them to the database, so that historical adjustment records can be continuously enriched and iterated.

[0136] It should be noted that the order of the above embodiments of the present invention is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0137] The various embodiments in this specification are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.

Claims

1. A method for adjusting an adjustable bearing housing for a conveyor roller, characterized in that, The method includes: Acquire multi-source sensing data of the target cargo during the transportation process; the multi-source sensing data includes: mechanical data, displacement data, visual data, and identity data; the mechanical data is used to reflect the interaction force between the cargo and the conveyor rollers; the displacement data is used to reflect the temporal changes in the position of the conveyor roller bearing seats; the visual data is used to reflect the surface properties of the cargo; and the identity data is used to determine the historical adjustment records of the target cargo. Based on the mechanical data and the visual data, a physical attribute vector of the target cargo is determined; the physical attribute vector includes: a deformation influence index and a smoothness parameter; the deformation influence index is used to characterize the degree of influence of cargo weight on the deformation of the conveyor frame; the smoothness parameter is used to characterize the frictional characteristics of the cargo surface; Based on the displacement data, the center of gravity offset vector of the target cargo during the transportation process is determined; the center of gravity offset vector includes: cumulative offset and offset trend; the cumulative offset is used to characterize the total amount of center of gravity offset; the offset trend is used to characterize the rate and direction of change of the center of gravity; The adjustment parameters are determined based on the physical property vector, the center of gravity offset vector, and the historical adjustment records of the target cargo; the adjustment parameters are used to adjust the bearing seat position and attitude. The bearing housing is adjusted according to the adjustment parameters. The determination of adjustment parameters based on the physical attribute vector, the center of gravity offset vector, and the historical adjustment records of the target cargo includes: Based on the deformation influence index, the smoothness parameter, and the historical height compensation benchmark value, a first height adjustment amount is determined; the first height adjustment amount is used to synchronously adjust the height of the bearing seats on both sides of the roller; the historical height compensation benchmark value is determined based on the historical height compensation value in the historical adjustment record. Based on the cumulative offset, the offset trend, the smoothness parameter, and the historical angle compensation reference value, a second height adjustment amount is determined; the second height adjustment amount is used to differentially adjust the height of the bearing seats on both sides of the roller; the historical angle compensation reference value is determined based on the historical angle compensation value in the historical adjustment record. Determining the physical attribute vector of the target cargo based on the mechanical data and the visual data includes: Based on the axial and lateral pressure components in the mechanical data, and the displacement data, the deformation influence index is determined. The basic friction parameters are determined based on the axial pressure component and the lateral pressure component. The smoothness parameter is determined based on the basic friction parameters and surface texture parameters; the surface texture parameter is determined based on the visual data. Determining the center-of-gravity offset vector of the target cargo during transportation based on the displacement data includes: Based on the height change of the bearing seats on both sides of the roller in the displacement data, a displacement difference sequence corresponding to the target cargo conveying process is determined; The cumulative offset is determined by integrating the displacement difference sequence. The displacement difference sequence is subjected to trend fitting analysis to determine the offset trend.

2. The adjustment method for an adjustable bearing housing for a conveyor roller according to claim 1, characterized in that, The adjustment of the bearing housing according to the adjustment parameters includes: Based on the first height adjustment amount, the control adjustment mechanism drives the bearing seats on both sides of the roller to perform the synchronous equal height adjustment; Based on the second height adjustment amount, the adjustment mechanism is controlled to drive the bearing seats on both sides of the roller to perform the differential height adjustment.

3. The adjustment method for an adjustable bearing housing for a conveyor roller according to claim 1, characterized in that, The method further includes: Based on the identity data, a preset adjustment database is queried to determine historical adjustment records; the historical adjustment records include: historical adjustment data of goods of the same category as the target goods; Based on the historical adjustment data, the historical height compensation benchmark value and the historical angle compensation benchmark value are determined.

4. The adjustment method for an adjustable bearing housing for a conveyor roller according to claim 1, characterized in that, The acquisition of multi-source sensing data of the target cargo during the transportation process includes: The mechanical data is acquired by a force sensing component; the mechanical data includes the axial pressure component and the lateral pressure component acting on the roller shaft. The displacement data is acquired using a displacement measurement component; The visual data is acquired using a visual acquisition component; the visual data includes at least an image of the surface of the target cargo. The identity data is obtained by reading the information identifier of the target goods through the identity recognition component.

5. The adjustment method for an adjustable bearing housing for a conveyor roller according to claim 4, characterized in that, The method further includes: Under no-load conditions, the reference displacement data of the roller bearing housing is collected; The displacement data is determined by the difference between the displacement data of the roller bearing housing collected under load and the reference displacement data.

6. The adjustment method for an adjustable bearing housing for a conveyor roller according to claim 5, characterized in that, The method further includes: The collected mechanical data, displacement data, and visual data are preprocessed; the data preprocessing includes at least one of the following: filtering and noise reduction, outlier removal, and data alignment.

7. The adjustment method for an adjustable bearing housing for a conveyor roller according to claim 3, characterized in that, The method further includes: The adjustment parameters are associated and stored with the identity data, the physical attribute vector, and the center of gravity offset vector; Update the associated stored data to the adjustment database.

Images